The ECG analysis may complement cardiologic evaluation and allow the identification of patients at high cardiac risk. The ECG can detect arrhythmias, conduction disorders, previous myocardial ischemia or acute myocardial infarction (MI), ventricular overloads, and changes due to electrolyte disorders or drug effects. In addition, a baseline electrocardiographic tracing is important for perioperative comparative evaluation in patients at high risk for cardiovascular events.

However, routine application of a test with limited specificity may lead to false-positive results in asymptomatic patients, since electrocardiographic abnormalities often concern the surgical and anesthetic staff and may prompt the unnecessary cancelation of the surgery.32 Abnormalities found on the ECG tend to increase with age and the presence of comorbidities, and these electrocardiographic changes usually have a low prognostic value regarding complications.33^,34 In a retrospective study involving more than 23,000 patients, the presence of preoperative electrocardiographic changes was associated with higher incidence of cardiac deaths within 30 days.35 This result was corroborated by two prospective studies that found similar results, where preoperative ECG abnormalities predicted perioperative cardiovascular events.36^,37 However, in the group of patients submitted to low to moderate risk surgery, preoperative ECG presented limited prognostic information.

Therefore, the indication for preoperative ECG depends on clinical history, surgery type, and diseases presented by the patient.

Studies evaluating the routine use of chest radiography (X-ray) in the preoperative evaluation have shown that the result of the test rarely interferes with the management of the anesthetic technique and does not predict perioperative complications. Abnormalities found in the X-ray are usually related to chronic diseases, such as COPD and/or cardiomegaly, and are more frequent in male patients older than 60 years, with a higher cardiac risk and with more comorbidities.40^,41 The indication for preoperative chest X-ray should be based on an initial careful evaluation by using the clinical history and physical exams of the patients. There is no indication for routine chest X-rays in asymptomatic patients as part of the preoperative evaluation.

Several papers in the literature have compared the accuracy of existing indices for different populations of surgical patients.42^-44 Most of these studies show that the various existing indices, although not very accurate, can predict events and should be used in perioperative assessment.

Among the risk indices with cardiovascular outcomes, we highlight the Lee’s Revised Cardiac Risk Index (RCRI),45 the index developed by the American College of Physicians (ACP),46^,47 and the Multicenter Perioperative Evaluation Study (EMAPO-www.consultoriodigital.com.br48 - the Multicenter Perioperative Evaluation Study was developed and validated in the Brazilian population. All indices have advantages and disadvantages that must be considered during their use. When estimating risk, we should consider which outcome we are predicting: the ACP algorithm predicts the occurrence of AMI and cardiovascular death. The RCRI estimates the risk of AMI, acute pulmonary edema, total atrioventricular block, and cardiorespiratory arrest. The RCRI is widely validated in the literature and shows moderate level of accuracy in predicting events in noncardiac surgeries; this index is less accurate in patients undergoing arterial aortic vascular surgeries and peripheral revascularizations.49 Thus, in a specific evaluation guideline on risk assessment in patients undergoing vascular surgeries,50 the VSG-CRI (Vascular Study Group of New England Cardiac Risk Index) is proposed as an alternative to the RCRI; it is adapted from RCRI with additional variables.51

When the aim is to estimate global risk, not only related to cardiovascular morbidity and mortality, the ACS NSQIP Surgical Risk Calculator (www.riskcalculator.facs.org), which has been recently developed by the American College of Surgeons, can be used. This tool was developed using data from more than 1 million surgeries in 393 hospitals in the United States. It had good prediction accuracy in that population. This index includes, in addition to the specific type of surgical procedure, 21 clinical variables, providing a risk assessment of 8 different outcomes.52 On the other hand, this tool presents some limitations related to subjective variables and still needs validation in other populations.

As already discussed, risk indices have advantages and limitations, and none of them is perfect. It should be kept in mind that the risk index selected should be used as a complement, but never a replacement, to the evaluator’s opinion. Data or evidence is not always available in the literature for all situations. Thus, assessment should be customized. In those cases where the evaluating physician considers that the index is underestimating the actual risk, this observation should be mentioned in the evaluation.

In addition to the risk indices already mentioned, other features related to surgical procedure and patient should be considered in the evaluation of the risk of perioperative events. We recommend using a flowchart proposed in this guideline (Figure 1).

Figure 1

In situations when the prognosis of the underlying disease that led to surgical indication demands an emergency intervention, the role of the cardiologist should be restricted to monitoring measures and interventions to reduce the risk in the intra and postoperative periods, with no indication of complementary tests that delay the proposed surgery. For urgent surgeries, there is time to optimize the cardiovascular therapy or to perform complementary tests, such as transthoracic echocardiography, when indicated (item 3.A). On the other hand, the request of functional tests to evaluate myocardial ischemia should not be performed, because the result will not change the plan and the proposed surgery cannot be postponed for 6 weeks (time required for preoperative myocardial revascularization or antiplatelet therapy, if indicated - see items 7.A.V and 7.B).

The first step in elective surgeries is the verification of the patient’s baseline clinical conditions. There are clinical circumstances in which the spontaneous risk of complications is very high, regardless of the surgical procedure. Identification of such conditions is fundamental, because their treatment should take priority over elective surgery, which should be, whenever possible, postponed and reconsidered only after clinical compensation (Chart 2).

The intrinsic risk of the surgical procedure corresponds to the probability of occurrence of perioperative cardiovascular events, independently of the clinical variables of the patients. It is related to the duration of the procedure, hemodynamic stress, and loss of blood and fluids that occurs during the intervention. Patients with stable clinical conditions who do not present high-risk cardiac conditions may be referred for low intrinsic risk procedures without the need for further evaluation. Despite the difficulty in determining a specific risk for surgical procedures, since they occur in different circumstances, a risk classification of cardiovascular events (death or non-fatal AMI) was proposed for noncardiac surgeries (Chart 3).53

Patients with low functional capacity are more prone to perioperative complications.18^,54 Functional capacity can be measured objectively using the exercise stress test (which is not always possible or desirable) or clinical history. Limitations in performing activities of daily living, such as walking quickly, climbing stairs, doing household activities or exercising regularly, are evaluated (Chart 1). In addition to the greater probability of poor perioperative evolution, patients with low functional capacity may have their symptoms underestimated due to their restrictions. Therefore, this can be considered when deciding to request complementary tests, such as, for example, non-invasive testing of ischaemic heart disease.

Based on the above, the III Guideline for Perioperative Cardiovascular Evaluation of the Brazilian Society of Cardiology proposes a flowchart for perioperative evaluation by using the existing risk indices and relevant risk variables for this period (Figure 1). The algorithm contains the conditions that must be analyzed sequentially according to their relevance for the determination of the risk.

Depending on the estimated risk and the nature of the risk, interventions for clinical treatment or additional risk stratification with complementary tests are proposed. This applies to increased risk of ischemic events, decompensation of HF/valvular heart disease, and arrhythmias, considering the current specific guidelines for each case. As an example, if the nature of the risk is ischemic, non-invasive testing of ischaemic heart disease should be considered.

In addition, for patients classified as intermediate or high risk by the algorithms, surveillance for postoperative cardiac events must be performed, including, monitoring in a semi-intensive or intensive care unit (ICU), ECG and troponin being performed once daily up to the third postoperative day.

Perioperative evaluation is a unique opportunity to identify and advice patients about cardiovascular risk factors. During this period, diagnosis of previously unknown diseases, which can be optimized for a better perioperative evolution and, more importantly, for a better long-term prognosis, is often possible.55

The pathophysiology of perioperative MI differs from that of spontaneous MI. Perioperative MI can be caused by plaque rupture in approximately half of the cases or by imbalance between myocardial oxygen supply (anemia, low flow, etc.) and demand (tachycardia and hypertension).68^-70

Exercise ECG testing is a safe, useful, and effective tool to detect myocardial ischemia, which is produced by an imbalance between supply and demand. Therefore, it is reasonable to assume that detection of abnormalities while performing this test may be reproducible during the perioperative period and its varying levels of stress. However, whether this strategy leads to reduction of perioperative cardiovascular events in all cases is unknown. It should also be considered that lower prevalence of coronary disease in a population results in lower positive predictive value of the exercise ECG testing.

Considering that risk stratification aims to reduce perioperative risk, performing the test in a population already stratified as low risk by the recommended algorithms is not logical. Therefore, in cases of low prevalence of coronary artery disease (CAD), the exercise ECG testing would not add value to the perioperative clinical stratification. It could also delay the surgery and require more specific tests to differentiate the true results from the false-positive.71^,72

Even in high-risk individuals, such as those undergoing preoperative vascular surgery, the predictive value, sensitivity, and specificity of the exercise ECG testing (10%, 74%, and 69%, respectively) are low, but with a high negative predictive value (98%).73 On the other hand, in a cohort study, performance of provocative ischemic preoperative tests in high-risk patients, with three or more clinical risk factors, is associated with shorter hospital stays and lower mortality.71 Thus, among asymptomatic individuals with a higher prevalence of the disease, the exercise ECG testing could be requested only if the result would influence the prognosis and, consequently, preoperative decisions, or to provide a more intensive clinical therapy or even a myocardial revascularization procedure.57 In this case, the onset of ischemic response at low load is associated with a higher number of perioperative cardiac events. On the other hand, patients with exercise tolerance up to 4-5 METS have a good perioperative prognosis.65^,74

The exercise ECG testing is safe, useful, and effective to detect myocardial ischemia and has a good cost-risk-benefit ratio,75 but has limitations, such as patients with physical limitations, patients who present abnormal ST-segment changes or left bundle bock in the baseline ECG. The alternative for such patients is an imaging method with pharmacological stress (adenosine, dobutamine, or dipyridamole).

In this context, myocardial perfusion scintigraphy (MPS), when possible associated with exercise and, within the physical limitations, pharmacological stress, has a good accuracy and prognostic value.70 In a meta-analysis of 1,179 patients submitted to vascular surgery, MPS with dipyridamole demonstrated a greater number of perioperative cardiovascular events, proportional to the presence and extent of perfusion defects. Those with reversible ischemia in up to 20% of the left ventricular extension had the same events with those without ischemia. However, when the area affected was 20-29%, 30-49%, and above 50%, the probability of cardiac events was 1.6, 2.9, and 11 times higher, respectively.76

Another meta-analysis with the same method and similar profile of patients showed that patients without perfusion defect, with fixed defect, and with reversible defect presented mortality and nonfatal MI rates of 1%, 7%, and 9%, respectively. Patients with two or more perfusion defects have a high incidence of cardiac events.77

Gated-associated MPS, which allows assessment of both myocardial perfusion and cardiac function, has been recently shown to be a useful tool for risk stratification in vascular surgeries. In one study, abnormal final systolic volume (more than twice the standard deviation) was the only independent variable to predict cardiac events. Patients with normal perfusion but with changes in contractility had significantly more cardiac events than those with normal contractility and perfusion (16% x 2%; p < 0.0001).78

In conclusion, in the perioperative evaluation, the indications for Gated-associated MPS are similar to those of the exercise ECG testing. It is the best option for patients with physical limitations. It is also the best alternative when the ECG is impossible to interpret due to baseline changes of the ST-segment and when the result of the exercise ECG testing is a possible false positive

Stress echocardiography is an accurate and reliable tool to identify patients with CAD and has an important role in the prognosis of cardiac events79^,80

Dobutamine and exercise stress echocardiography have similar diagnostic accuracies, which are higher than that of dipyridamole stress.81 If a dobutamine stress echocardiography does not demonstrate the presence of residual ischemia in a patient with a history of infarction, the prognosis is good and the probability of reinfarction, death, and acute pulmonary edema is low in the perioperative period of a noncardiac surgery.73

The use of dobutamine stress echocardiography in perioperative risk assessment is already well documented in the literature, with a positive predictive value ranging from 25-55% and a negative predictive value of 93-100% for cardiac events in patients undergoing noncardiac surgery,73^,82 The results were generally used to determine preoperative clinical decisions, particularly the decision to perform coronary angiography or myocardial revascularization before or after the elective surgery.

A meta-analysis of 15 studies was performed to compare dipyridamole thallium-201 and dobutamine stress echocardiography in vascular risk stratification before surgery. It demonstrated that the prognostic value of the abnormalities is similar in both imaging methods for perioperative ischemic events.77

Coronary Angio-CT has been increasingly used to evaluate patients with suspected CAD. It presents high sensitivity for the detection of coronary stenosis, including multiarterial disease and lesion in the left coronary trunk.84^-87 However, it has not been extensively investigated in the perioperative period of noncardiac surgeries.

Ahn et al.88 analyzed retrospective data and concluded that angiotomography may be advantageous in reclassifying the risk of patients assessed by Lee’s revised score (RCRI), when submitted to intermediate risk procedures.45^,88 On the other hand, a cohort showed a fivefold higher probability of overestimating the risk of angiotomography in patients who present an event.89

A few studies have shown an association between elevated coronary calcium score and cardiovascular events. Angiotomography could still be applied as an instrument for risk reclassification.88^,90

Nevertheless, the information obtained through such tests was still not correlated with new interventions (revascularization, pharmacoprotection, or monitoring) to reduce perioperative coronary events. Therefore, angiotomography or coronary calcium score is not recommended in the preoperative period.91

The ankle-brachial index (ABI) is one of the preferred methods for the diagnosis of peripheral occlusive arterial disease (PAOD). Values ≤ 0.9 show good accuracy for the diagnosis. ABI is associated with poor cardiovascular prognosis, significantly increasing the risk of amputation, CAD, and cerebrovascular disease.92^,93 It can be used as a risk reclassification tool in conjunction with the Framingham risk score, increasing the mortality risk due to all causes in all risk categories.94 Although ABI is a promising method because of its low cost, rapid standardization, good acceptance by patients, and low intra- and inter-observer variability, it is poorly investigated in the perioperative context.95 Flu et al.96 showed that patients with ABI ≤0.9 submitted to vascular surgery had an odds ratio (OR) of 2.4 for the occurrence of myocardial injury. Two other studies evaluated ABI in patients undergoing noncardiac and nonvascular surgeries and obtained ORs of 10.2 and 7.0 for the occurrence of major cardiovascular events, including isolated increase of troponin.97^,98

Association of this test with perioperative risk scores has not yet been studied, and the risk reclassification capacity is unknown. Therefore, routine use of ABI is not recommended as a risk estimation tool. For patients with previously known vasculopathies, focusing on the pharmacological prevention of cardiovascular events and monitoring, as discussed in other topics of this guideline, is recommended.

Holter is a continuous electrocardiographic monitoring tool that identifies the presence of atrial and ventricular arrhythmias and their complexity. It also identifies dynamic changes of the ST-segment that are indicative of myocardial ischemia. This method is not routinely used in the evaluation of preoperative ischemia, because other diagnostic methods are more sensitive and specific for this purpose. Its possible application during the perioperative period is monitoring ischemic events that occur in the intra- and postoperative periods, which can have a particularly high incidence in some specific groups of patients.

Electrocardiographic monitoring in the postoperative period using Holter was not very sensitive (50%), although very specific (92%), for the diagnosis of reinfarction in patients undergoing noncardiac surgeries and who had a previous history of myocardial infarction.99 Therefore, routine use of this test is not recommended. Requesting a Holter in the preoperative period follows the same indications of other contexts.

The increased sensitivity of the available troponin kits provided a greater accuracy and rapidity in the diagnosis of MI in patients with chest pain in the emergency room.100 On the other hand, in the perioperative period of noncardiac operations, the available scientific evidence does not include all troponins (I and T) or all assays used, which have different detection limits and reference values. For this reason and due to the importance of this diagnostic tool, we decided to include a detailed explanation of the available methods before the recommendations in this guideline.

The detection limit represents the minimum value that is detected by the method. The reference value of normality is determined using the 99th percentile, which is obtained by performing the test in a normal population, and indicates that 99% of normal individuals have values below this cut-off.

Troponin assays can be classified as low sensitivity (conventional), medium sensitivity (contemporary or sensitive), or high sensitivity. This classification is based on the percentage of healthy individuals in whom troponin can be detected. Contemporary troponin assays (medium sensitivity) can detect values above the 99% percentile (altered values); however, troponin is only detected in a few normal individuals. High sensitivity assays can determine values of this marker (limit of detection) in 50-95% of normal individuals.101

Table 1 presents some troponin assays and their classification and respective reference values of normality.101^,102 To properly interpret and request troponin as a biomarker in the perioperative period, the physician must be familiar with the troponin assay used in their hospital. Notably, in the preoperative period, only the Roche high-sensitivity troponin T (hs-TnT) assay has been tested in the available studies and may have clinical applicability. In the postoperative period, most studied troponins are conventional and some are sensitive (as specified in item 7.F).

Some published papers demonstrate the efficacy of high sensitivity preoperative troponin as a marker of perioperative cardiovascular complications and general mortality in noncardiac surgeries. Nagele et al.103 studied 608 patients who underwent noncardiac surgery. They showed that 41% had increased values of hs-TnT above the 99th percentile (> 14 ng/L) in the preoperative period. The increase in preoperative hs-TnT was associated with a higher total mortality during a 3-year follow-up.103 These findings were confirmed in a study involving 455 patients undergoing vascular surgeries. In this study, patients with increased hs-TnT in the preoperative period presented a greater number of cardiovascular events in the postoperative period.104 In a comparison to Lee’s RCRI algorithm in 979 patients aged more than 55 years with at least one cardiovascular risk factor submitted to noncardiac surgeries, preoperative hs-TnT presented an area under the ROC curve (0.78) similar to the RCRI (0.68; p = 0.07) in predicting combined cardiovascular events (mortality, MI, recovered cardiac arrest, and acute HF). In addition, in a multivariate analysis, increased hs-TnT in the preoperative period was an independent predictor of these combined events (HR 2.6; p = 0.008). As for general mortality, hs-TnT was superior to RCRI (area under the curve 0.81 × 0.66, p = 0.006).105 The prevalence of preoperative hs-TnT increases varies between 21% and 41%, depending on the age and risk factors, such as diabetes, CAD, systemic arterial hypertension, and renal failure.103^-107

No study has evaluated the role of high-sensitivity troponin I (hs-TnI) in predicting cardiovascular events in the preoperative period. Contemporary/sensitive TnI was evaluated in 560 patients undergoing noncardiac surgeries, with only 5% presenting preoperative values above the 99th percentile. Its use did not improve the prediction of risk of perioperative cardiovascular events.108

In conclusion, measurement of troponins with conventional or contemporary/sensitive assays is not useful in the preoperative period and should not be performed. On the other hand, the hs-TnT measurement in the preoperative period can be used as a tool for risk stratification associated with the use of the algorithms. In addition, these data help to establish a baseline value in patients with indication for postoperative monitoring, facilitating the interpretation of postoperative values of hs-TnT and the diagnosis of postoperative MI (items 7.F and 8.A).

OThe risk scores usually used in the perioperative evaluation enable to estimate the risk of cardiovascular events in the perioperative period with moderate accuracy. Tests to evaluate ischemia, as well as biomarkers (troponins and natriuretic peptides), allow to refine the risk assessment before surgery.109

Natriuretic peptides are released into the bloodstream by the heart in response to multiple physiological stimuli, such as myocardial stress and ischemia. Several studies demonstrated that high preoperative levels of BNP are potent predictors of perioperative cardiovascular complications.110

The two studies by Biccard et al.111 conducted investigations on patients submitted to arterial vascular surgery. In 2011, they reported that preoperative BNP was an independent predictor of increased postoperative troponin levels in a cohort of 267 patients undergoing vascular surgery. They also reported that the use of this biomarker improved the risk prediction of Lee’s RCRI in 38-70% in patients classified as intermediate risk.111 In 2012, the authors evaluated 788 patients undergoing vascular surgery and showed that increased preoperative BNP levels was an independent predictor of cardiac events in a period of 30 days (OR = 5.0; p < 0.001).108

A meta-analysis involving individual data from six different studies evaluated natriuretic peptides as predictors of events in patients undergoing vascular surgeries. The study confirmed that increased preoperative natriuretic peptide level is an independent predictor of events (cardiac death or nonfatal MI) in up to 30 days after surgery. It was also observed that it improves the predictive value of the RCRI.112

With regard to patients undergoing nonvascular surgeries, there are no studies that exclusively evaluate this population. The vast majority of the studies involve both vascular and nonvascular surgeries.

In a meta-analysis published in 2009, including 15 prospective observational studies and 4,856 patients submitted to vascular or nonvascular surgeries, the authors found that increased preoperative BNP or NT-proBNP levels was associated with a high (nearly 20-fold higher) risk of major cardiovascular events, cardiac mortality, and mortality due to all causes (almost 10-fold) in the perioperative period (< 43 days after surgery).113 However, whether prognostic information was improved in these studies, considering the existing risk indices, were not determined.114

More recently, a prospective multicenter observational study analyzed 979 patients aged more than 55 years with at least one cardiovascular risk factor (hypertension, diabetes, dyslipidemia, smoking, family history) in the preoperative period of noncardiac surgery. The study aimed to assess increases in hs-Tn as a risk predictor compared to the Lee’s RCRI, with 93% of the patients with RCRI < 2. In this study, the authors also evaluated the role of natriuretic peptides as risk predictors in noncardiac surgery. They observed that both hs-Tn levels and NTproBNP levels were higher in individuals with cardiovascular events and seemed to provide additional information to the RCRI.105

Finally, a meta-analysis conducted in 2014 included 18 prospective observational studies. The authors assessed individual data from 2,179 patients undergoing noncardiac (vascular and nonvascular) surgery. They demonstrated that postoperative levels of BNP and NTproBNP provide additional risk prediction of death or nonfatal infarction in a period of 30 and 180 days to a risk prediction model that includes preoperative BNP. They also showed that high values of BNP/NTproBNP both in the postoperative (OR = 3.7; p < 0.001) and preoperative (OR = 1.9; p < 0.001) periods were independent predictors of death or nonfatal MI in a period of 30 days.115

Therefore, we consider that BNP/NTproBNP measurement can provide additional prognostic information on the risk stratification of patients undergoing vascular and nonvascular arterial surgeries.

AoS is the most common valvular heart disease in elderly patients, affecting 2-4% of adults aged more than 65 years. The prevalence of AoS is expected to double in the next 20 years, with the progressive aging of the population.134^,135

Several studies have shown that patients with moderate to severe AoS may have a high risk of cardiac complications during noncardiac surgery.19^,136^-139 However, in many of these studies, the definition and degree of AoS were ambiguous or based on few details. Since most studies included symptomatic or ventricular dysfunction patients, it is unclear whether aortic valve replacement surgery should precede the noncardiac procedure. The study by Calleja et al., which included 30 patients with anatomically relevant AoS and 60 controls (mild to moderate AoS), was the only one to exclude symptomatic patients. They showed similar MI or death rates during noncardiac surgery, mostly with low or intermediate risk.64

For these reasons, Samarendra et al. recommended a new flowchart to assess the perioperative risk of noncardiac surgeries in patients with AoS. Patients with a high risk of cardiac complications are those with an average gradient > 45-50 mmHg and/or valvular area < 0.8 cm2; left ventricular systolic dysfunction; symptomatic AoS; significant mitral regurgitation or other associated valvular diseases; increase in ≥ 18 mmHg in the average gradient during exercise; and associated CAD.67

Furthermore, in a recent study with 218 patients, Mizuno et al.140 demonstrated that patients with AoS who underwent a major noncardiac surgery present a faster progression of aortic valve disease compared to controls with AoS who did not undergo surgical intervention.

For the above reasons, it is recommended to first correct the anatomically relevant AoS, even if asymptomatic, in patients who will undergo intermediate- or high-risk noncardiac surgeries.

On the other hand, some patients with relevant AoS scheduled for noncardiac surgery have clinical indication for correction of valvular heart disease , but present a high risk for cardiac surgery or are ineligible for the conventional cardiac procedure. For these cases, a transcatheter aortic valve implantation (TAVI) is an option preceding noncardiac surgery.141^-143

When patients with relevant AoS are submitted to urgent/emergency noncardiac surgery, preoperative clinical compensation is recommended with the use of diuretics, as well as postoperative period in an ICU, with hemodynamic and electrocardiographic monitoring and serial measurement of myocardial necrosis markers.

Patients with MS and formal indication for surgical or percutaneous correction of valvular heart disease should be submitted to the valve procedure before elective noncardiac surgery.144 If noncardiac surgery is an emergency, it can be performed with invasive hemodynamics monitoring, optimization of blood volume, and prevention of tachycardia and hypotension. Increased heart rate, particularly if there is development of atrial fibrillation (AF), can lead to congestion and pulmonary edema. Therefore, β-blockers and/or diuretics may be used during the perioperative period.

Regurgitant valvular heart disease are associated with increased cardiac risk during noncardiac surgery, but are better tolerated than stenotic valve lesions.145^,146 However, aortic insufficiency (AoI) and mild or moderate mitral insufficiency (MI) do not increase the risk of cardiovascular complications during noncardiac surgery.

On the other hand, patients with symptomatic AoI or MI associated with ventricular dysfunction have a high risk of cardiovascular complications, and valvular heart disease should be performed before elective noncardiac surgery. Urgent or emergency noncardiac procedure should be performed after optimization of pharmacological treatment and optimal hemodynamic stability through the preferential use of vasodilators and diuretics, in addition to the postoperative period in an ICU.

Patients with valve prostheses with normal function, without left ventricular dysfunction, may undergo noncardiac surgery without additional risk. For mechanical prosthesis, oral anticoagulation and bridge with heparin should be performed as described in section 7.D.132

Supraventricular paroxysmal tachycardias are more prevalent in structurally normal youngsters and rarely present hemodynamic intolerance. They can occur by atrial tachycardia, ortho-ventricular atrioventricular tachycardia (in patients with accessory pathway), and nodal reentrant tachycardia. Asymptomatic patients who present ECGs with ventricular pre-excitation also have a low risk of perioperative complication. Attention should be only given to the occurrence of supraventricular tachycardia and pre-excited AF, and treatment follows the ACLS standard care.148

In this period, occurrence of pain, nausea, gagging, hypothermia, sympathetic blockade in anesthesia, laparoscopic insufflation, laryngoscopy, hyperventilation, anesthetics, and cholinergic drugs may precipitate arrhythmia due to autonomic imbalance. Such stimuli may trigger supraventricular tachycardias but do not increase surgical morbidity and mortality.

Controlling triggering factors mentioned above can minimize the occurrence of arrhythmias. In patients taking antiarrhythmic drugs, these drugs should be continued because preoperative interruption may promote arrhythmia.

Detection of extrasystolic arrhythmias in the preoperative period is common in high-risk patients. On the other hand, about 20% of the population can present these arrhythmias in the 24-hour Holter exam performed routinely.

Evaluation of extrasystolic arrhythmias includes collecting personal and family history. The occurrence of symptoms of hemodynamic intolerance (syncope or pre-syncope, precordial pain) may indicate complex or sustained arrhythmias. The presence of family history of sudden cardiac death may indicate the need for specific evaluation. In asymptomatic young patients with no personal or family history of heart disease, isolated monomorphic ventricular extrasystoles may generally be benign and with no implications in the perioperative period. In low-risk surgeries, evaluation of the ECG and cardiac area by chest X-ray may be sufficient and suspension of noncardiac surgery is unnecessary.149

Suspicion of adjacent structural disease can be detected in ECG. ECG can identify ventricular overload, inactive area, conduction system diseases, and other rarer arrhythmic syndromes. Thoracic radiography is an important screening procedure for pulmonary diseases and detection of increased cardiothoracic index. Echocardiogram is a sensitive method that should be used for additional morphological analysis. An exercise ECG testing may be useful to demonstrate the presence of ischemia as a triggering factor for ventricular arrhythmias or to demonstrate benignity when dealing with idiopathic arrhythmias, which are suppressed by sinus tachycardia at the peak of the exercise.149

Monomorphic ventricular tachycardia (VT) is often due to myocardial scarring. Polymorphic tachycardia may indicate ischemia, which reinforces the need for further investigation according to the specific guidelines.149

The first preoperative treatment of ventricular arrhythmias is correction of reversible causes. There is no evidence that unsusceptible extrasystoles or VTs worsen the perioperative prognosis, nor is there is a proven benefit of suppression with antiarrhythmics.150

AF is the most common sustained tachyarrhythmia and its prevalence increases with age. Patients who present a previous diagnosis of AF with adequate clinical control, considering symptoms and basal heart rate, do not need special considerations, except recommendations on anticoagulation (see item 7.D).151

In the preoperative period of elective surgeries, β-blockers or non-dihydropyridine calcium channel blockers (verapamil and diltiazem) are essential to control heart rate when patients have persistent AF/Flutter with high ventricular response. Administration of these drugs should preferably be performed slowly to avoid hypotension, which is known to be deleterious postoperatively. Calcium channel blockers may cause depression of myocardial function, particularly in patients with structural cardiopathy.152 Digoxin may be attenuated by surgical hyperadrenergic condition.65 There are specific situations, such as pre-excited AF, in which ablation can be considered before surgery.153 It should be noted that patients with AF with ventricular response above 120 bpm have a severe cardiovascular condition. In these cases, surgery should be postponed until the heart rate is controlled.

Control of rhythm, i.e., reversal of AF, could be an option before the procedure, depending on the symptoms and specific evaluation by cardiologists based on the current guidelines. However, interruption of anticoagulation for the procedure could only be performed after 4 weeks, and perioperative adrenergic stress facilitates recurrence.152^,154

Regardless of the control strategy selected (rhythm or heart rate), patients should be assessed for risk of cardioembolism and risk of bleeding using the CHA2DS2Vasc and HAS-BLED scores, respectively.152

The occurrence of postoperative AF (POAF) is associated with increased time in ICU, increased morbidity (including stroke, with incidence of 1.3-1.7%) and mortality, and consequently increased hospital costs. The AF preventive measures are adequate hydroelectrolytic control in the pre- and postoperative periods (normovolemia, magnesium and potassium monitoring, and replacement), in addition to maintenance of drugs previously used if hemodynamically tolerated.155

Some drugs have been investigated to reduce the incidence of POAF and its deleterious consequences. Preventive antiarrhythmic therapy with amiodarone or venous magnesium should be discussed individually. Retrospective studies showed that the use of amiodarone for anesthetic induction in patients submitted to esophagectomy may reduce the rate of POAF, but did not reduce the mortality and hospitalization rates. Moreover, its use in patients submitted to pulmonary resection is associated with reduced POAF and decreased time in ICU.156^-158 Riber et al.159 conducted a randomized, double-blind, placebo-controlled study. They demonstrated that administration of 300 mg of intravenous amiodarone in the early postoperative period, followed by 1200 mg of oral drug per day for 5 days, in hemodynamically stable patients reduced the POAF rate (9% versus 32% in the control group). On the other hand, Khalil et al.160 compared the use of amiodarone in the immediate postoperative period for 48 hours (attack of 5 mg/kg, followed by 15 mg/kg) with venous magnesium sulfate (attack of 80 mg/kg, followed by 8 mg/kg/h) and with a control group from a retrospective analysis in patients submitted to pulmonary resection. Their results showed POAF rates of 10, 12.5, and 20.5%.

Other medications were studied to reduce the incidence of POAF. A meta-analysis with statins showed a potential role in the prevention of AF, although, of the 16 trials included, only 4 were conducted in noncardiac surgeries.161 Thus, its benefit to this population remains inconclusive.

Colchicine, an anti-inflammatory drug, is being studied for prevention in high-risk patients submitted to thoracic surgery.155

Genetically determined arrhythmias are a heterogeneous group of diseases and may occur due to defects in ion channels (channelopathies). Protein defects lead to atrial or ventricular arrhythmias, which may be manifested by tachycardic palpitations, arrhythmic characteristic syncope, or sudden death. The most prevalent channelopathies are Brugada Syndrome (1:5,000) and Long QT Syndrome (1:5,000). Other rarer diseases are catecholaminergic polymorphic ventricular tachycardia and short QT syndrome.

The estimated cardiovascular risk in this population is very variable and there is no complementary test that adequately stratifies it. The surgical and perioperative risk of these patients is poorly known, but some recommendations are well established by specialists.162 In a retrospective analysis of 1,700 cases of early sudden death, 50 occurred postoperatively in young patients with no history of cardiopathy and a part may represent primary arrhythmias.163

Symptomatic patients (syncope or palpitation) have a higher risk and sometimes need drugs (quinidine, β-blockers) or pacemaker defibrillators.164 Except for urgent surgeries, these patients should be evaluated by cardiologists before surgical release.

Channelopathies present specific anesthetic indications and immediate therapeutic interventions, which make perioperative management of these patients a challenge for the anesthesiologist.162 In addition to cardiac monitoring and adequate electrolyte balance, all the drugs used in the perioperative period should be checked on specific sites, such as www.brugadadrugs.org, for Brugada Syndrome, and www.crediblemed.org, mainly for Long QT Syndrome, and also for all other channelopathies.164

Most current pacemakers have active fixation electrodes, which enable them to be actively fixed to the endocardium and cardiac veins. Displacement of these electrodes is rare, but it is a possible complication at this stage. The area where the generator is implanted is in the process of surgical recovery. Inflammatory responses, bruises, edema, rejections, and even infections, which could still be subclinical at that stage, can occur with various intensities. The CIED and electrodes are susceptible to infections by organisms originating from other areas and even from surgical manipulations. If possible, it is recommended to wait until the end of the second post-implant month to perform elective noncardiac surgery to minimize the risk of complications.

End-of-life CIEDs due to advanced battery wear should be replaced with newer and more modern units before elective noncardiac surgeries. These devices may present adverse behavior when subjected to extreme operating conditions (repeated interrogations and schedules), which may occur during surgery. In addition, these CIEDs can enter the end-of-life mode, changing the behavior and even disabling several important functions to save battery, which could be important in the perioperative period.

For elective noncardiac surgeries, patients should be evaluated by their CIED physicians, which will perform a complete verification of the stimulation system. The physician will determine the need for special programming, report the care that should be taken by the surgeon and anesthesiologist, and describe the possible behaviors of the CIED during surgery or even indicate the need for a stimulus during the procedure to make necessary perioperative schedules, which is usually required in patients at higher risk and with more complex CIED, such as defibrillators.170

The biggest concern usually involves those patients scheduled for major surgeries with the use of electric scalpels. In such cases, a safety evaluation must be performed always in a pacemaker evaluation unit and by a qualified physician. The physician should program the pacemaker in asynchronous mode only in cases where the patient depends on pacing and has no arrhythmia history (avoiding competition between pacemaker pace and self-pacing). The physician should also advise the surgical team to use bipolar or ultrasonic scalpel when possible, as these types of devices interfere less with the CIED.

The report should contain at least the recommendations described below for surgeries without the presence of the physician who programmed the pacemaker:

Continuous cardiac monitoring with an ECG and a pulse oximeter (heart rhythm monitoring is possible even when using an electric scalpel).

In patients with resynchronizers, the presence of a greater number of electrodes in the heart undeniably increases the possibility of complications due to external interferences on the stimulation system. Most stimulation electrodes used in the venous system of the left ventricle are used in unipolar mode. Therefore, they are more susceptible to external interferences, particularly those produced by the electric scalpel. However, there is a current trend of using bipolar and even multi-polar electrodes. Notably, many unipolar implants that have already been implanted will remain active for many years. The presence of more electrodes and unipolar electrodes forces physicians to take rigorous measures and pay more attention to signs of interference in the multisite stimulation system.

The complexity and diversity of the behavior of ICD, the risk of serious arrhythmias during surgery, and the potential electromagnetic interferences from an electric scalpel with the release of inappropriate shocks lead us to recommend, whenever possible, the presence of the specialist in the operating room with the ICD equipment. The equipment enables to program the ICD during the procedure in accordance to the clinical and metabolic needs of the patient. The antitachycardia function should be turned off, and the patient must be properly monitored. Turning off the function makes the patient unprotected. The physician should be prepared to treat a high-risk arrhythmia by using an external defibrillator and antiarrhythmic drugs. Following the experts’ advice, this type of patient must stay in the ICU during the postoperative period to allow monitoring during the critical stage. At the end of the surgery, the ICD parameters should be re-established and even adjusted to the patient’s clinical condition. The antitachycardia function of the ICD must be switched back on.

During the perioperative period, patients with CIED may present complications that require application of electrical cardioversion or defibrillation. Although generators can theoretically withstand shocks, it is advisable to avoid them whenever possible.172 When necessary, some care should be taken to preserve the pacemaker or defibrillator, the electrodes, and the heart-electrode interface, as described below:

If the patient has an implantable defibrillator, internal cardioversion is recommended because it uses a small amount of energy, biphasic pulse, and internal safety features of the device itself.

Shocks generated by lithotripsy have been related to transitory events of loss of sensitivity and command of the CIED, as well as reversion to the safe mode; however, these situations are extremely rare.173 When lithotripsy is required in patients with pacemaker and/or defibrillator, direct the focus away from the area of the apparatus and electrodes.

Turn off atrial stimulation when ECG-triggered lithotripsy is used to avoid the device from synchronization with the atrium. Programming the atrial channel with less energy and in the bipolar mode may be another option, keeping the bicameral stimulation more physiological. A test can be performed before the application to observe the behavior and the interaction of the devices. Do not immerse the part of the body that contains the pacemaker or ICD when performing immersion lithotripsy.

It is also recommended to monitor the patient throughout the procedure and, whenever possible, the physician should stay in the room with the CIED programmer to adjust the program of the device as needed. Since lithotripsy is usually outpatient, the patient may perform a regular assessment of the CIED and the specialist can be requested to provide specific guidelines for the procedure. Placing a magnet over the CIED during lithotripsy should not be a rule because, as previously discussed, the device may have different behaviors and may deactivate therapies.

Magnetic resonance imaging should not be performed on patients with older CIEDs. There is a risk of dysfunction of the prosthesis, electrodes, and even displacement due to the magnetic field generated.174

The recent advancement of the CIED industry has led to the production of devices that support the field of resonance, including tests of the chest area with a field strength of up to 1.5 Tesla. In this case, both the CIED and electrodes need to be compatible with this technology. Some patients with new devices but with old electrodes (connected or abandoned) could not undergo magnetic resonance imaging.175

Despite the development in this field, even patients with CIED and compatible electrodes require the presence of a physician and a programmer during the test because a specific program is required, which should be deactivated at the end of the procedure. The most appropriate recommendations in these cases are previous evaluation of the patient in a pacemaker evaluation unit and guidance from the specialist on whether the patient should undergo testing. It is important to note that not only the images present artifacts due to the presence of the prosthesis but the patient may also experience local discomfort. This local discomfort is frequently described as burning and palpitations or dizziness related to inhibitions/deflagrations of the CIED.

Radiotherapy can be used if the radiation focus is not directed to the CIED. If the device is close to the radiation focus, the area should be covered with a lead screen. If the irradiated site is located exactly in the region of the implant or very close to it and many radiotherapy sessions are necessary, reimplantation of the CIED away from the irradiation site should be evaluated.

Direct radiotherapy on the CIED may cause transient or definitive dysfunction of the device and premature battery wear.176^,177 Each radiotherapy application may put the CIED in safety mode, even if local protection measures are taken; thus, evaluation of the CIED is required after each radiotherapy session.178

Electrodes may also be affected with radiotherapy, especially at the site of contact with the endocardium, which may suffer fibrosis and loss of control. This phenomenon can occur from days to months after radiotherapy. Therefore, particular attention should be given to these patients, especially those that depend on stimulation. Under these conditions, a greater frequency of electronic evaluations (weekly/monthly) should be stipulated after the procedure. CIEDs that enable remote evaluation can facilitate this monitoring.179

Dental procedures in patients with CIEDs are increasingly common in dental practice. In addition to the risk of infection in prostheses, interaction between the equipment used in the dental treatment and the CIED, particularly the electric scalpel, is also possible.180 In this situation, care taken should be the same as covered in item 4.G.III for general surgeries.

In patients with ICD, the device may interpret the thermocautery interference as an arrhythmia and release low- or high-energy therapy, placing both the patient and the dental surgeon at risk of receiving an inappropriate shock. Placing a magnet on the generator does not provide adequate protection and should be avoided.

In dental procedures, where the use of thermocautery is mandatory and the patient has ICD, it is essential to disable the antitachycardia therapy, keeping only the pacemaker function.

To prevent untreated risk of arrhythmias in this condition, the patient should be monitored and an external defibrillator should be available on site. The decision to perform the procedure in a hospital environment should always be considered due to the presence of necessary equipment for electrical emergencies, besides the support of the arrhythmologist. General, simple, and routine dentistry procedures carried out in dental practices do not require additional precaution, in addition to the usual therapeutic recommendations. Analgesia should be effective, and the use of anesthetics with vasoconstrictors at recommended doses (vasoconstrictor concentration and amount) should not be avoided. According to case studies and systematic reviews, these anesthetics do not interfere with cardiovascular parameters and do not predispose to coronary events, and when they induce arrhythmias, the risk is low.181

Minor surgeries may be performed with the usual care for patients with CIED provided that the thermocautery is not used. Analgesia should be efficient in dental treatments, and local anesthetics with vasoconstrictors at the usual doses can be administered to patients with cardiopathies because of the low risk of complications.

The cardiologist should, whenever possible, provide guidance in advance to the person responsible for the surgical procedure. If this is not possible, the use of the thermocautery should be avoided. Placing a magnet over the CIED does not guarantee protection in all cases and is not recommended in all situations.

The operative period was divided into preoperative evaluation; preoperative preparation; and intraoperative and postoperative care. The recommendations were grouped to facilitate follow-up of patients with CIED. The suggested sequence should be followed for each patient.

Liver transplantation remains the procedure of choice for the treatment of terminal liver disease. However, changes occurring both intraoperatively and post-transplantation have increased the existing cardiovascular morbidity in these patients. This may be due to several risk factors that affect the population of the same age group (such as age, diabetes, male gender, smoking, previous history of CAD) or it may be related to liver disease and its etiology, such as alcohol-related cardiomyopathy, deposition of amyloid substances, and alterations due to cirrhosis-associated cardiomyopathy.184

Studies have shown that cardiac events can occur in up to 70% of post-transplant patients, depending on the criteria.185 Among the most common are arrhythmias, pulmonary edema, and systolic ventricular dysfunction, but sudden death and myocardial infarction may also occur.186

Therefore, careful investigation of heart disease in candidates for liver transplantation is required.187 However, due to liver disease, cardiological investigation in these patients is difficult because the hemodynamic changes and limitations resulting from the disease do not allow the same sensitivity and specificity in cardiology tests, when compared to other populations.188

Approximately 50% of patients with cirrhosis show an increase in the QT interval.189 β-adrenergic receptors respond only slightly to sympathetic stimuli, leading to dubious responses in studies of dobutamine echocardiography.190 The hyperdynamic state and its consequent chronic vasodilatation impair the vasodilator-induced response, such as dipyridamole in myocardial scintigraphy. Terminal liver disease is often accompanied by renal dysfunction, which impairs the use of contrast agents, such as in coronary angiography or coronary Angio-CT.

However, there are tests and approaches that have become routinely used in the preoperative evaluation of liver transplant candidates due to their cost-effectiveness. Moreover, some cardiovascular comorbidity characteristics of cirrhotic patients should be ruled out because they lead to high morbidity and mortality in the perioperative period. The most important characteristics to note are as follows:

It is characterized by the triad: systolic dysfunction, mainly due to a deficit in the contractile response induced by stress, and ejection fraction at rest below 55%; diastolic dysfunction, typically with E/A < 1 and prolonged isovolumetric relaxation time; and electrophysiological changes, particularly increased QT interval, chronotropic deficit and bradycardia, changes in ventricular repolarization, increased left atrium and myocardial mass, and increased levels of BNP, NTpro-BNP, and TnI.187^,188^,191

Although these findings increase morbidity and mortality in liver transplant candidates, no benefit has been demonstrated in the specific treatment of these alterations.

Cardiomyopathy associated with alcohol accounts 21-32% of patients with dilated cardiomyopathy in some medical centers. Excessive ingestion of alcohol leads to myocyte apoptosis, reduced calcium sensitivity, depressed myocyte contractile function, and myocardial fibrosis.192 Considering that alcoholic cirrhosis is among the major causes of liver disease, concomitant occurrence of cirrhosis and dilated cardiomyopathy is relatively common.

Interruption of alcohol intake in the early phase of this cardiomyopathy may lead to partial or total recovery of ventricular function, which reduces cardiovascular morbidity in these patients.193

The hyperdynamic state of the patient with portal hypertension may cause vasoconstriction and remodeling of the pulmonary vessels, leading to pulmonary hypertension (PH). These changes affect 5-10% of the transplant candidates and, depending on the pulmonary artery average pressure, may be mild (> 25 and < 35 mmHg), moderate (> 35 and < 45 mmHg), and severe (> 45 mmHg).194

Although there are studies on endothelin receptor antagonists and phosphodiesterase inhibitors, there are no definitive guidelines for moderate to severe PH. Perioperative mortality in the latter is close to 100%, therefore representing a contraindication for isolated liver transplantation, and combined lung-liver transplantation may be indicated in specific medical centers.195

Although sometimes confused with PH, hepatopulmonary syndrome presents several different characteristics. It is defined as hypoxia in the presence of hepatic disease that worsens with upright posture, with evidence of intrapulmonary vasodilation. Hypoxia is due to the accumulation of pulmonary vasodilators, particularly nitric oxide, leading to intrapulmonary arteriovenous shunt.196

Unlike for PH, the selected treatment for hepatopulmonary syndrome is liver transplantation, although non-definitive data correlates the degree of hypoxia with perioperative mortality.197

As previously mentioned, the risk factors for CAD are either similar or more in cirrhotic patients than in the general population, especially diabetes. Although CAD increases the morbidity and mortality of these patients, the degree of impairment of stenosis does not seem to correlate with worse prognosis.

Computerized coronary tomography with quantification of the calcium score has been recently shown to be useful, and a calcium score above 400 has a high predictive value for early cardiovascular events in these patients.198 However, the use of this test cannot be routinely indicated for this population.

The choice of treatment for these patients should consider that lack of intervention may lead to an excessive risk during and after surgery. However, the best treatment is not established and should be customized for each patient.

An important controversy concerns the use of drug eluting stents that requires antiplatelet agents for a longer time in patients with thrombocytopenia, and the risk of bleeding is always present.199

Coronary artery bypass grafting (CABG) should be, whenever possible, postponed until after transplantation because of the high probability of hemorrhagic events or worsening of the hepatic condition with the surgery.200 Myocardial revascularization before the transplant should be only performed in patients for whom the risk of death due to CAD exceeds the risk of death due to liver disease.

Finally, it is worth mentioning that the cardiologist is part of the multidisciplinary team that accompanies these patients. For this reason, a possible contraindication of the procedure should always be discussed and customized with the multiprofessional team and the patient.

Patients with terminal renal disease are one of the highest cardiovascular risk groups, with mortality rates from cardiovascular disease 5 to 100 times higher than those found in the general population.201 Cardiovascular disease is the leading cause of death after renal transplantation, particularly due to CAD.202 In the first 30 days after successful kidney transplantation, approximately half of the deaths are due to MI.203 Similarly, in the late follow-up, chronic ischemic heart disease accounts for more than a third of deaths in patients with a functioning graft.204

Therefore, the preoperative evaluation of renal transplant candidates aims not only to reduce cardiovascular risk in the short term, related to the surgical procedure, but also to reduce cardiovascular events in late follow-up.205 During the evaluation of renal transplant candidates, the identification and presence of CAD are of fundamental importance because they allow the medical team to establish more precisely the risk-benefit of transplantation, the possible need for coronary intervention in the preoperative period, the use of cardioprotective measures in the perioperative period, and the control of risk factors in the postoperative period.

This section aims to provide the cardiologist with the most appropriate methods to determine cardiovascular risk in a very special population of patients, almost always excluded from the risk stratification studies. The main aim is to identify those most likely to be diagnosed with CAD among kidney transplant candidates. Thus, the recommendations included in this section should be applied only to asymptomatic patients or patients with atypical symptoms. For those individuals with clinical evidence and/or diagnostic investigations suggestive of coronary disease, further investigation and treatment should follow the proposed rules for the general population.

Identification of relevant CAD is challenging in renal transplant candidates. Patients with terminal renal disease often have atypical symptoms, or are even asymptomatic, in the presence of advanced CAD.206 The use of non-invasive methods, such as exercise ECG testing, MPS, and stress echocardiography, all routinely performed in the general population has lower sensitivity and specificity than in individuals with normal renal function, providing numerous false negative results.207^-209 On the other hand, coronary angiography should not be performed indiscriminately because it is an invasive method, with risk of complications and high cost. In addition, the prevalence of significant CAD in patients routinely assessed invasively is less than 50%.207^,209^,210

Observational studies show that patients with terminal chronic kidney disease and CAD undergoing PCI or CABG have a risk of cardiovascular events similar to those with terminal chronic kidney disease without significant CAD. Those with obstructive coronary disease who did not undergo myocardial revascularization had significantly higher rates of cardiovascular events.207^,208

Therefore, it is necessary to define a strategy that allows identification of patients more likely to have relevant CAD and who should therefore be referred for angiographic study. With this, we would reduce the number of patients inadequately classified as being of low cardiovascular risk due to failure in the preoperative risk stratification and consequently their exposure to a higher risk of cardiovascular events.

The clinical parameters most strongly associated with ischemic heart disease after kidney transplantation are age > 50 years, DM, and previous evidence of cardiovascular disease (clinical history and/or tests).211 The prevalence of relevant CAD (stenosis ≥ 70%) increases with the number of risk factors. These three risk factors have been the basis for the formulation of algorithms to investigate coronary disease in patients with chronic kidney disease. Other factors considered as predictors of cardiovascular events in this population are systemic arterial hypertension, left ventricular hypertrophy, smoking, dyslipidemia, and dialysis for more than one year.212

Based on the results of existing studies, we proposed a risk stratification model for asymptomatic chronic renal patients from a cardiovascular perspective, evaluated for renal transplantation, according to the presence or absence of the three previously mentioned risk factors.213^-216 If there is any latency between initial stratification and transplantation, we suggest a period of three years for a new stratification if the patient is stable and without new cardiovascular events or symptoms.

The preparation of dental procedures in patients with heart disease is not based solely on the use of antibiotic prophylaxis, vasoconstrictors, and/or control of postoperative bleeding. The presence of infectious foci in the oral cavity may represent a factor of postoperative complication. The incidence of odontogenic bacteremia increases significantly in the presence of infectious foci, such as in periodontal disease and endodontic lesions.

Although the occurrence of bacteremias is commonly reported during dental procedures, they occur with similar frequency during oral hygiene and chewing.228 For this reason, assessment of oral health with elimination of infectious foci and intensive oral hygiene control of hospitalized patients is advisable whenever possible prior to surgical procedures in patients with or without heart disease to reduce perioperative complications.

In general, patients with controlled heart disease under optimized medication can undergo a dental procedure safely with the usual routine care.

Individuals with pacemakers and implantable automatic defibrillators do not present changes with high or low rotation motors, amalgamators, electric pulp tests, electric toothbrushes, endodontic ultrasound, periodontal ultrasound, and radiography. The use of an electric scalpel has specific guidelines discussed in this guideline (item 4.G.III). Further studies are needed to determine the possible effect of the laser on pacemakers.

The use of local anesthetics with vasoconstrictors in patients with heart disease has generated controversy. The administration of vasoconstrictors in combination with local anesthetics increases the quality and duration of pain control and promotes reduction of bleeding.229 Local anesthetic without vasoconstrictor has a short duration, rapid absorption (high toxic potential), and inadequate pain control. It can also generate hemodynamic changes and even cardiac arrhythmias, besides promoting mild vasodilation, increasing bleeding.

Lidocaine with epinephrine has been the most widely used local anesthetic worldwide. Although the interaction of epinephrine with β-blockers, tricyclic antidepressants, diuretics, and cocaine has been reported in the literature, the use of two to three 2.0% lidocaine tubes with 1:100,000 epinephrine (36-54 µg epinephrine) is well tolerated in most patients. This also applies to individuals with hypertension or other cardiovascular disease, situations where the use of this vasoconstrictor has more benefits than risks.229

Most dental procedures have low risk of bleeding. Therefore, warfarin should not be discontinued in most patients submitted to dental procedures, including dental extractions.230 A meta-analysis231 and smaller studies232^-234 have demonstrated safety in performing dental procedures in anticoagulated patients with international normalized ratio (INR) < 4.0 by using local measures to reduce bleeding. Major procedures, and consequently with greater risk of bleeding, such as extraction > 3 teeth, should have customized discussion based on the thrombotic risk of each patient to determine interruption of therapy and possible bridge therapy, as discussed in specific section in this guideline.

To date, evidence of bleeding risk in patients using new anticoagulants (NOACs) in dental procedures is limited.235 There are also no recommendations available for perioperative measures.

Regarding antiplatelet agents, several studies in the literature show the safety of performing dental procedures usually using aspirin or clopidogrel monotherapy.236^-239 Although bleeding increases, this is easily controlled with local hemostatic measures.240^-242 Therefore, patients in secondary prevention of cardiovascular events using aspirin or clopidogrel monotherapy should keep using the drugs in the perioperative period of the procedures.

Patients on DAPT with a recent PCI (6 weeks after bare metal stent (BMS) and 6 months after drug eluting stent - DES) or acute coronary syndrome for less than one year should maintain the use of the drugs, if dental procedures are required in the period of highest risk of intra-stent thrombosis. There is already evidence in the literature on the safety of this strategy when local hemostatic measures are increased. Studies with ticagrelor or prasugrel remain scarce,243 but the recommendation is to maintain them in these conditions of DAPT in periods of increased risk of intra-stent thrombosis.

When using antithrombotic therapy, dental procedures may be performed following some precautions.

Some precautions and measures may be considered to reduce bleeding in patients taking antithrombotic drugs.

The use of antibiotics for endocarditis prophylaxis is indicated in patients with previous history of endocarditis or valve disease who will undergo procedures involving manipulation of gingival tissue, periodontal region, or perforation of the oral mucosa, as discussed in a specific section in this guideline (item 7.E.I). Antibiotics often used for this purpose may interfere with the metabolism of oral anticoagulants, particularly warfarin. Patients taking anticoagulants should be alerted to a possible increase in bleeding and control the INR, if necessary. There is no need to change the anticoagulant regimen when a single dose of prophylactic antibiotic is used.

Dermatological surgical procedures are low-risk procedures for both cardiovascular and bleeding events. Data from the literature suggest that approximately 50% of patients scheduled for dermatological procedures are using antiplatelet or anticoagulant therapy.244^,245 In these cases, the surgical team and the anesthesiologist should be informed about the drugs used and the necessary care, including a more time-consuming and cautious hemostasis, because in most cases the risk associated with discontinuation of antithrombotic therapy outweighs the risk of bleeding inherent in the procedure.

Suspension of the use of aspirin for secondary prevention of cardiovascular events is not necessary before performing any dermatological surgical intervention.246^,247 For patients using double antiplatelet therapy for stent implantation who are not in the period of greatest thrombotic risk, the recommendation is to suspend the second antiplatelet drug,248^,249 considering the intervals already described in this guideline (see section on antiplatelet agents - item 7.A.V).

For individuals taking warfarin, the recommendation is to continue its use and adjust the INR to ≤ 3.5 to minimize the risk of bleeding.248 However, some studies have not demonstrated the correlation between INR levels and the risk of increased bleeding in patients taking warfarin.250 Although evidence is scarce, it is recommended that patients taking one of the new oral anticoagulants (NOACs) can perform most of the dermatological procedures during the medications.248 This is to ensure that the surgical intervention is scheduled, whenever possible, a few hours before the next dose to avoid the peak serum level of the drug.

Considering the risk analysis of cardiovascular events, endoscopic procedures are low risk.251 Thus, suspension of the procedure is not required for cardiovascular intervention, except in severe cardiovascular conditions already mentioned in the section of perioperative evaluation algorithms of this guideline. In addition, most drugs that are included in the cardiovascular therapy do not need to be discontinued and can be ingested with minimal water. In fact, the most important issue is whether the patient makes use of antithrombotic drugs due to the potential risk of endoscopic bleeding and thromboembolic events caused by discontinuation of these drugs.

Endoscopic procedures have different bleeding potentials, which is very important to determine the strategy to be used. The risk varies with the type of procedure and is mainly related to the existence of therapeutic interventions. Chart 4 presents the risks of bleeding attributed to common endoscopic procedures in clinical practice.252 The risk of thromboembolic events with discontinuation of antithrombotic therapy varies with the therapy proposed and individual patient conditions.

For endoscopic procedures classified as low risk of bleeding, antiplatelet therapy may be maintained, either in the form of monotherapy (independent of the agent) or DAPT.252^-255 For procedures considered as high risk of bleeding, some points should be considered.

Patients taking DAPT for a recent PCI (6 weeks after BMS stent and 6 months after DES) or acute coronary syndrome in the past year present the highest risk if antiplatelet therapy is discontinued. Therefore, elective high-risk bleeding endoscopic procedures should be postponed, whenever possible, until the end of this period of increased risk. However, for procedures that require to be performed during this period, the most accepted strategy is to maintain aspirin and withdraw the second antiplatelet drug,255^,256 although evidence for this strategy is limited.

Patients taking aspirin monotherapy for secondary prevention of cardiovascular events may maintain the treatment in the perioperative period of endoscopic procedures, even in those considered to be high risk for bleeding, because most evidence in the literature shows a low risk of significant bleeding in these situations.257^-266 Some studies have demonstrated increased bleeding in procedures, such as submucosal dissection in patients with gastric neoplasia267 and mucosectomy in colonic tumors larger than 20 mm,268 these procedures should be analyzed individually according to the risk of thrombotic events with suspension of aspirin.255 There is some evidence showing the safety of clopidogrel monotherapy during percutaneous endoscopic gastrostomy, and its maintenance may be considered in this situation.258 Evidence for the use of prasugrel and ticagrelor in high-risk bleeding endoscopic procedures is scarce.

If antiplatelet therapy is discontinued, the intervals between the suspension and the procedure should follow the recommendations of this guideline in the antiplatelet management section (item 7.A.V.C). Antiplatelet therapy may be resumed after the procedure as soon as hemostasis is achieved. An attack dose may be considered in patients who are at high risk for cardiovascular events.256

For endoscopic procedures classified as low risk of bleeding, anticoagulant therapy with warfarin may be maintained252^,253^,255^,269 and should be discontinued in those considered to be at high risk of bleeding.260^,266 To date, no evidence has been reported regarding the use of new anticoagulants (NOACs) in these situations, suggesting that they should be maintained in low-risk bleeding procedures and suspended in those with high risk for bleeding.252

Intervals for the suspension and resumption of NOACs and warfarin (including consideration of bridge therapy in those patients considered to be at a greatest risk for thromboembolic events) should follow the guidelines in the perioperative management section of this guideline (section 7.D).

Ophthalmologic surgical interventions are relatively frequent procedures in the elderly population. The presence of cardiovascular comorbidities that require the use of antithrombotic drugs and their associated treatment during the perioperative period is a subject of intense debate between ophthalmological surgeons and cardiologists. In Brazil, the fear of hemorrhagic complications, including bruising in the periorbital region, is responsible for the indiscriminate interruption of aspirin and warfarin in 82.7% of patients who undergo glaucoma surgeries.270

The limited evidence regarding the occurrence of complications shows that this fear is not reasonable. The rate of hemorrhagic complications described in observational studies is low and without major consequences, particularly in cataract surgeries using conventional anesthetic techniques.271^-275

Some ophthalmologic surgical interventions, such as trabeculectomy276^,277 and vitrectomy,278^,279 which are used to treat glaucoma and retinal diseases, respectively, present a greater hemorrhagic risk. Nevertheless, the evidence does not demonstrate an increased risk of significant hemorrhagic complications in these surgeries with the use of aspirin.277^,280^,281 In such cases, the decision should be customized, but maintaining this agent is generally recommended in the perioperative period.282

Patients receiving a DAPT for a recent PCI (6 weeks after BMS stent and 6 months after DES) or acute coronary syndrome in the past year are those at highest risk of events due to discontinuation of antiplatelet therapy. Therefore, ophthalmologic procedures, whenever possible, should be postponed until this period of greatest risk ends.

For procedures that required to be performed in this period, the strategy depends on the hemorrhagic risk of the intervention. For interventions with low hemorrhagic risk (intravitreal injections, cataract, and peribulbar anesthesia), aspirin and P2Y12 receptor inhibitors should be maintained. For interventions with higher hemorrhagic risk, such as vitrectomy and trabulectomy, the most accepted recommendation is the maintenance of aspirin and suspension of the second antiplatelet, considering the intervals already described in a specific section of this guideline (item 7.A.V.C). However, the evidence for this strategy is limited.

Similar to patients taking ASA monotherapy, evidence in the literature favors the maintenance of clopidogrel monotherapy in the perioperative period of cataract surgeries.274^,275 Evidence is more limited in glaucoma and retinal surgeries. The recommendation is to suspend clopidogrel in the perioperative period of these interventions, considering the 5-day period between the suspension and the procedure.

With regard to patients taking warfarin, the evidence in the literature favors their maintenance in surgeries with lower hemorrhagic risk, such as cataract surgeries, ensuring that the INR is in the therapeutic range.271^,272 A meta-analysis of observational studies including patients submitted to cataract surgery and using warfarin found a bleeding incidence of around 10%. This incidence was mostly self-limiting, subconjunctival, and with no visual loss.273 On the contrary, in glaucoma and retinal disease surgeries, warfarin should be discontinued. Perioperative management should follow the strategy described in this guideline in the perioperative anticoagulation management section (item 7.D), considering the individual risk of thrombotic events of the patients.

To date, evidence of risk of bleeding during ophthalmologic surgeries in patients using new anticoagulants (NOACs) is limited. No recommendations have been established for their perioperative management.

Recommendations, particularly for patients with coronary stents and mechanical valvular prostheses, should be customized, considering the relationship between thrombotic and hemorrhagic risks. For patients recommended to maintain anticoagulant and/or antiplatelet agents, the surgeon should be informed of the need to ensure adequate hemostasis. A suggestion that can be considered and discussed with the anesthesiologist, who makes the final decision, is the use of a specific type of anesthesia that is less associated with hemorrhagic complications.275 For antiplatelets, if the decision is to interrupt the drug, it should be restarted postoperatively as soon as possible. In addition, the procedure must be performed in a hospital with competence for urgent hemodynamic intervention (PCI), if necessary.

The recommendations regarding the use of β-blockers in the perioperative period of noncardiac surgical interventions have been under intense debate in recent years because of the results of large clinical studies. These studies presented limitations that generate discussions in the academic and care communities.

Pioneering studies in the 1990s suggested that perioperative use of β-blockers could reduce cardiovascular mortality and morbidity in a broad spectrum of patients. Three randomized trials conducted between 2005 and 2006 did not confirm the protective effect of β-blockers in the vascular perioperative period of low- or intermediate-risk patients, highlighting the potential harm, given the association with a higher incidence of bradycardia and hypotension.289^-291

The benefit of β-blockers was later questioned in meta-analyses.292^,293 On the other hand, the largest retrospective study on the use of β-blockers in the perioperative period, which analyzed more than 780,000 patients submitted to noncardiac surgery, showed that the impact of β-blockers depends on the estimation of cardiac risk. In high-risk patients, β-blockers are associated with lower mortality, whereas in low-risk patients, no benefit was found and the β-blockers could be harmful.294 In 2008, the POISE study was conducted. In this study, 8,351 patients who were mostly having an intermediate risk of complications were randomized to receive metoprolol succinate or placebo starting 2-4 hours prior to noncardiac surgery, with doses up to 400 mg in the first 24 hours. The results showed a lower incidence of MI, reversed cardiac arrest, and cardiac mortality in the group with β-blockers. However, the authors observed two times higher incidence of stroke and greater overall mortality in this group. The high incidence of hypotension and bradycardia was strongly associated with higher mortality and stroke.295

Careful analysis of these data shows a great heterogeneity among studies, mainly regarding the dosage of β-blockers used and time of onset. There are studies that initiated β-blockers a few hours before the surgery, with no time to determine the doses conferring adequate heart rate control.289^-291 In other studies, some patients continued to receive β-blockers despite the occurrence of bradycardia and/or hypotension and most importantly without time for hemodynamic adaptation.295

On the other hand, there are studies that started β-blockers earlier, at least one week before the surgery, to determine the adequate dose.296 These were the studies that showed benefit. In 2008, even before the publication of the POISE study, an interesting publication reviewed data from the two main meta-analyses previously cited292^,293 based on the heart rate control obtained for each study. The authors divided the data into two groups according to degree of heart rate control, and observed that the trials in which patients achieve the most effective control of heart rate were associated with a reduced incidence of postoperative MI, suggesting that effective control of heart rate is important for achieving cardioprotection.297

Thus, once the specific indications have been evaluated, the use of β-blockers in the perioperative period must always comply with safety principles. The time of onset should be as early as possible (at least one week before the surgery) to ensure adequate time to evaluate the hemodynamic response of each patient, avoiding bradycardia and hypotension. Low doses should be prescribed, with progressive titration to a HR of 55 to 65 bpm, without hypotension (SBP > 100 mmHg). During the perioperative period, frequent monitoring of HR and BP must be done. If HR < 50 bpm or SBP < 100 mmHg is detected, the β-blockers should be suspended until hemodynamic and chronotropic balance is restored. On the other hand, from the point of view of effectiveness, the benefit of the β-blockers is associated with adequate heart rate control. Therefore, we should target for a HR of 55 to 65 bpm in the pre- and postoperative periods.

Finally, β-blockers should not be withdraw in the perioperative period of patients who receive them chronically for various indications. Acute β-blocker suspension is associated with a significant increase in postoperative mortality.298

In addition to reducing cholesterol levels, statins have a pleiotropic effect of reducing inflammation and stabilizing plaques of atherosclerosis. The use of statins to prevent cardiovascular events after vascular surgeries is well established and is based on prospective, randomized, placebo-controlled studies.

In 2004, in the first randomized study published, the authors demonstrated that the use of 20 mg of atorvastatin is associated with a large decrease in major cardiovascular events (death, MI, stroke, unstable angina) in the perioperative period and after 6 months of follow-up. This effect occurred regardless of baseline cholesterol levels.299

In 2009, the use of 80 mg of slow-release fluvastatin in 250 patients submitted to vascular surgeries was shown to reduce the occurrence of postoperative myocardial ischemia and the combined outcome of MI and cardiac death in a period of 30 days compared to the placebo group (247 patients).300

This result was confirmed in a recent meta-analysis involving 23,536 patients, in which the use of statins in the perioperative vascular period reduces overall mortality and MI and stroke rates.301 The specific benefits of statins for each type of vascular procedure can be found in a specific guideline for vascular surgeries.50 The administration of 20 mg of atorvastatin (or 40 mg of simvastatin) in patients submitted to vascular surgery should be preferably performed two weeks before the procedure and maintained for 30 days. Subsequently, the dose should be adjusted to the individual LDL goal of each patient.

On the other hand, evidence on the use of statins for the prevention of cardiovascular complications in nonvascular surgeries is obtained from retrospective studies. Lindenauer et al.302 evaluated 780,591 patients submitted to noncardiac surgeries (92% nonvascular surgeries) in a retrospective cohort study with 77,082 patients (9.9%) receiving statins. In this study, the patients who received statins had lower mortality during hospital stays. Another retrospective case-control study with only nonvascular surgeries, including 989 patients who died postoperatively within 30 days and 1879 controls, showed that the use of statins is also associated with a reduction in mortality (OR = 0.4; CI 0.24-0.68).303 In a retrospective cohort that included 752 patients submitted to nonvascular surgeries, the authors demonstrated a reduction in the combined outcome of nonfatal MI, AF, and mortality in a period of 30 days in patients using statins.304 Recently, in an analysis of the patients included in the VISION study,305 Berwanger et al.306 evaluated 2,842 patients receiving statins and 4,492 patients without statins. They compared the occurrence of mortality, isolated elevation of troponin levels (defined as troponin increased levels, without MI or other cause), and stroke in a period of 30 days by using propensity score matching. About 10% of the patients were submitted to vascular surgeries and the rest to nonvascular procedures. Patients receiving statins showed reduction in the risk of the combined outcome (RR 0.83; 95% CI 0.73-0.95; p = 0.007). The use of statins reduced overall mortality (RR 0.58; 95% CI 0.40-0.83; p = 0.003), cardiovascular mortality (RR 0.42, 95% CI 0.23-0.76; p = 0.004), and the occurrence of isolated increase of troponin levels (RR 0.86, 95% CI 0.73-0.98; p = 0.02). There was no reduction in noncardiovascular mortality and in the rate of MI or stroke. It should be noted that, despite propensity score matching, the patients in the statin group had CAD, diabetes, peripheral vascular disease, aspirin use, and ACE inhibitors/angiotensin receptor blockers more frequently than the patients in the group without statins. Although presenting more risk factors, the patients in the statin group had fewer cardiovascular events.306 Basing on these studies, we can conclude that the patients with higher cardiovascular risk and those using statins due to comorbidities (CAD, diabetes, peripheral vascular disease) may benefit with the administration of statins in perioperative nonvascular surgeries.

Statins are often withdraw postoperatively. The main reasons for statin withdrawal are as follows: postoperative ileus and inability to administer oral medications, hemodynamic instability, concern with the occurrence of side effects, and lack of awareness of the importance of maintaining statins.307

Perioperative statin suspension in patients who use this medication on a chronic basis is an independent predictor of cardiovascular events following vascular surgeries.308^,309 The use of statins in the perioperative period is safe. Although patients using statins have a higher baseline CPK level, the occurrence of increases above 5 times their reference value or rhabdomyolysis is rare.310 Therefore, in patients who already use statins, it should be maintained in the perioperative period.

Alpha2-agonists modulate the response of catecholamines to surgery and anesthesia, decreasing the release of noradrenaline and reducing blood pressure and heart rate. The first randomized studies that used clonidine to prevent cardiovascular complications following noncardiac surgeries demonstrated a reduction in myocardial ischemia, but without a reduction in clinical events or mortality.311^,312 On the other hand, a meta-analysis demonstrated that α2-agonists reduce mortality and MI in patients submitted to vascular surgeries, but not in those submitted to nonvascular surgeries.313

The European Mivazerol Trial (EMIT) evaluated the use of mivazerol in 1,897 patients with CAD submitted to noncardiac surgeries. The authors found a decrease in general mortality and MI or cardiac death only in the subgroup of patients submitted to vascular surgeries.314

A randomized study with 190 patients demonstrated reductions in myocardial ischemia and mortality with the use of perioperative prophylactic clonidine in patients with CAD or risk factors for CAD,315 but these results were not confirmed.

Recently, the POISE-2 study included 10,010 patients submitted to noncardiac surgery in 23 countries. The patients were randomized to receive clonidine or placebo in the perioperative period of noncardiac surgeries. The use of clonidine did not reduce the incidence of death or MI in a period of 30 days (HR 1.08, 95% CI 0.93-1.26; p = 0.29). Furthermore, patients on clonidine more frequently presented clinically significant hypotension (HR 1.3, 95% CI 1.24-1.4, p < 0.001) and reversed cardiac arrest (HR 3.2, 95% CI 1.17-8.76, p = 0.02).316

Therefore, the introduction of clonidine in the preoperative period is not recommended to reduce the risk of cardiovascular events.

Evidence for the use of calcium channel blockers with the aim to reduce cardiovascular risk in the perioperative period of noncardiac surgeries is scarce. In a meta-analysis of 11 studies involving 1,007 patients, there was no reduction in mortality or MI with verapamil, diltiazem, or dihydropyridine.317 Another study evaluated 1,000 patients submitted to aortic aneurysm surgeries, and the results demonstrated an increase in perioperative mortality with the use of calcium channel blockers.318

Therefore, the use of calcium channel blockers to prevent cardiovascular events in the perioperative period of noncardiac surgeries is not recommended.

Operating patients who use antiplatelet therapy implies an increased risk of bleeding;247 however, the suspension is known to be associated with rebound effect319 and clinical atherothrombotic events.247^,320 In general, the decision should be based on the discussion between the surgical, clinical, and anesthetic teams. The team should consider the risk of exacerbation of bleeding inherent to the surgical procedure and, on the other hand, the thrombotic burden that led to the prescription of the antiplatelet agent.

POISE-2 study, which was published in 2014, is the largest randomized, placebo-controlled study evaluating the impact of ASA in the perioperative period.321 In this study, 10,010 patients with risk factors for perioperative complications receiving ASA or placebo were evaluated. Patients who never took ASA were included, as well as patients who were already on chronic use, randomized to placebo, or continued ASA at the study doses of 200 mg immediately prior to surgery and 100 mg daily for 30 days. The authors did not show a significant difference in the primary outcome (death or MI) or in the secondary outcome of the study (death, MI or stroke). No difference was observed based on the history of use of pre-randomization ASA. On the other hand, they observed a higher incidence of bleeding in the ASA group: 4.6% × 3.8%, p = 0.04, especially at the surgical site.

Some considerations should be made regarding the clinical profile of patients who mostly (almost 70% of the population) had no history of cardiovascular disease and used ASA for primary prevention. Another extremely relevant fact is the non-inclusion of patients with PCI with DES in the last year or BMS in the last 6 weeks. Most surgeries in the study were orthopedic, general, or gynecological, with 605 vascular procedures, for which the main results of the study are maintained.321 Thus, the most practical applicability of POISE-2 results is to recommend the non-use of ASA in the perioperative period of individuals in primary prevention. For patients who are already using ASA for primary prevention and are scheduled for noncardiac surgery, suspension of antiplatelet treatment 7 days before is recommended.

Oscarsson et al.322 conducted a much smaller study than POISE-2. It included 210 patients. The study design was interesting because it did not investigate the initiation of ASA, but the suspension or maintenance in the noncardiac perioperative period of patients chronically using ASA. Patients scheduled for vascular surgeries were not included (the authors thought that it was unethical to withdraw the antiplatelet treatment in the vascular perioperative period). They observed a lower incidence of cardiovascular events in the group that maintained ASA, without a higher incidence of bleeding. Anecdotally, the subjective notion of the surgeon on the bleeding tendency due to impaired hemostasis during the surgery did not allow discriminating patients receiving placebo or antiplatelets.322

In the STRATAGEM study, the patients using ASA only for secondary cardiovascular prevention were randomized to receive 75 mg of ASA or placebo in the perioperative period. The results showed no increased incidence of bleeding or significant difference in thrombotic complications.323 However, this study included only 20% of the planned patients, which hinders definitive conclusions for patients at higher risk.

In the vascular perioperative period, evidence suggests the beneficial use of ASA for protection of infrainguinal grafts, but without conclusion about systemic outcomes. On the other hand, Calderaro et al.324 analyzed patients in the elective vascular perioperative who were already on chronic use of ASA and observed that those individuals with lower platelet responsivity up to 100 mg daily (according to the aggregability test after stimulation with arachidonic acid) presented more than twice the systemic atherothrombotic events, when compared to the more responsive individuals, without an increase in bleeding rate.324

For patients receiving ASA for secondary prevention, it is recommended to maintain it at a maximum dose of 100 mg daily. Meta-analysis data suggest that this ratio is favorable for most perioperative patients.247 Neurosurgeries due to high morbidity and mortality associated with bleeding, even small ones, represent an absolute indication for ASA suspension 7 days before.247

Patients scheduled for transurethral resection of the prostate using the conventional technique should also suspend ASA owing to the high risk of bleeding.247 Urologists recently acknowledged that ASA can be maintained using the hemostatic technique called laser green-light in patients scheduled for transurethral resection.325^,326 This example demonstrates the benefit of new techniques for more complex patients and the constant need for a multidisciplinary approach to the perioperative decision process. At present, there is no recommendation for routine ASA withdrawal for transrectal prostate biopsy, an extremely common urological procedure.325

There is no recommendation to start ASA before noncardiac surgeries. If we evaluate patients with established vascular disease but who erroneously omit using antiplatelets, it is the opinion of this guideline by consensus of the specialists that this therapy should be implemented at the time of hospital discharge. However, no study has supported the administration of the drug before surgery.

Approximately 20% of patients submitted to PCI will require noncardiac surgery in the subsequent 2 years.327^,328 This implies perioperative management not only of ASA but also of the second antiplatelet agent (clopidogrel, prasugrel or ticagrelor), particularly in cases with less than one year between interventions. This is a common and quite complex situation because evidence on the safety of maintaining DAPT in the noncardiac perioperative period is scarce and indirect. The evidence is mainly extrapolated from the cardiac surgery data, which reveals a great increase in the rate of bleeding.329^,330 On the other hand, the potential for treatment suspension is also quite high, especially after PCI, with DAPT suspension being one of the main predictors of stent thrombosis.331 An interesting study was conducted in France evaluating 1,134 patients with PCI who required subsequent noncardiac surgery. The study identified DAPT suspension for more than 5 days before the surgery as one of the independent predictors of perioperative cardiovascular complications.332

The best way to deal with DAPT in the perioperative period is to maintain the optimal duration of this therapy and not to perform elective surgeries during this period (see the topic of prophylactic myocardial revascularization in this guideline - item 7.B): 6 weeks after PCI with BMS; 6 months after DES or one year after PCI in the context of acute coronary syndromes. Some elective procedures, such as oncological treatment, cannot be postponed without consequences. In this situation, it is recommended to maintain ASA alone and suspend clopidogrel. Clopidogrel should be suspended 5 days before the surgery and restarted as soon as possible, ideally before completing 10 days without DAPT. The postoperative deadline for restarting the drug depends on adequate hemostasis control and should be individually established between the surgical and clinical teams. Basing on the evidence of relative safety of withdrawal of the second antiplatelet in up to 10 days, we recommend not to exceed a total of 10 days without DAPT outside the perioperative context.333

Evidence is even scarcer for the newer antiplatelet drugs. The TRITON-TIMI 38 study included patients receiving prasugrel or clopidogrel associated with ASA and requiring cardiac surgery. Prasugrel showed higher rates of bleeding than clopidogrel, even with suspension of clopidogrel or prasugrel for up to 7 days.334 This observation supports the recommendation to discontinue prasugrel 7 days before noncardiac surgeries.

Although pharmacokinetic data support the suspension of ticagrelor for a shorter period^,335 the current recommendation is still 5 days. Sub-analysis of patients who required myocardial revascularization in the PLATO study (patients randomized to ASA + ticagrelor vs. ASA + clopidogrel) demonstrated less bleeding with ticagrelor than with clopidogrel. This finding is in agreement with the idea of a faster platelet activity recovery after suspension of ticagrelor in comparison with clopidogrel.336^,337

The new drug eluting stents are less thrombogenic; thus, the ideal interval for noncardiac surgery is shortened to 6 months in cases of elective PCI (see specific session of prophylactic myocardial revascularization in this guideline - item 7.B). Notably, maintenance of DAPT can be considered for some procedures performed in compressible sites or by endovascular technique depending on multidisciplinary consensus.

Patients at very high risk of stent thrombosis, such as diabetic or with PCI involving grafts, or in the context of acute coronary syndromes or complicated PCI, may be considered for “bridge” therapy with parenteral antiplatelet consisting of glycoprotein IIb/IIIa inhibitors.338 There is no recommendation for “bridge” therapy with LMWH because recent clinical evidence has demonstrated the harm of such measure, in addition to the pharmacological need of inhibition of platelet activity rather than coagulation.339

We now use more objective scores to assess the risk of thromboembolism associated with each type of surgery to better guide prophylaxis. One of these scores that can stratify the risk for venous thromboembolism with greater accuracy is the Caprini risk assessment model.371^,372 In this model, a score is assigned to each clinical or laboratory variable (Chart 6). Based on the number of these variables and the score obtained, the categories of risk are defined (very low, low, moderate, and high) according to the risk of VTE (Chart 7).367

In addition to the risk of venous thromboembolism, according to risk factors attributed to the condition of the patient or to the surgical procedure, it is important to analyze risk factors for bleeding that may modify the choice of the best thromboprophylaxis. Risk factors for severe bleeding complications are described in chart 8.367

Next, we present the recommendations for specific non-orthopedic surgeries. The recommendations are no longer guided by the Caprini risk score, but according to the risk characteristics of each surgery.

The risk of VTE associated with major orthopedic surgeries (hip and knee prosthesis surgeries and hip fracture surgery) is one of the highest of all surgical specialties. The combined risk of VTE in a postoperative period of 35 days in untreated patients is currently estimated at 4.3%.373 Table 3 describes the components of this risk.373 Next, we present the recommendations for prophylaxis for VTE in major orthopedic surgeries.

For patients submitted to major orthopedic surgeries, regardless of the possibility of using IPC or duration of treatment, LMWH is preferred compared with other antiplatelet agents suggested as alternatives. When using LMWH, it is suggested to start administration at least 12 hours before surgery or at least 12 hours after the surgical procedure.373

Whenever possible, mechanical prophylaxis should be associated with IPC during hospital stay (Class of Recommendation IIa, Level of evidence C). In addition, use of apixaban or dabigatran is preferred for patients rejecting IPC or multiple subcutaneous injections.373

The doses of the new anticoagulants used in the studies for prophylaxis of VTE in major orthopedic surgeries are outlined in chart 9.

Warfarin is a vitamin K antagonist; its anticoagulant effect takes days to disappear (half-life from 36 to 42 hours) and may require similar time to reach adequate levels after surgery. Thus, patients at high risk for thromboembolism may require “bridge therapy” with parenteral antiplatelet agents, such as UFH and subcutaneous LMWHs. These two agents present faster onset and a shorter half-life, which would allow the possibility of suspending warfarin as close to the surgical procedure as possible, minimizing thromboembolic risk as much as possible.

Since the metabolism of warfarin may be influenced by several factors, such as patient age, renal function, and drug interactions, the INR should be measured on the day before surgery to ensure that it is <1.5. If the INR is >1.5, reverse it with oral vitamin K administration (1 to 2 mg) and re-evaluate it the following day.

The decision to suspend or not to suspend warfarin before the surgical procedure will depend on the combined analysis of thromboembolic risk (Table 4), risk of bleeding (Chart 11), and the patient’s own risk.

There is little evidence on the best course of action in patients with moderate risk of thromboembolism regarding whether to use or not to use bridge therapy. Thus, the choice should be based on the individual characteristics of each patient and the proposed surgery. Whether the patient requires bridge therapy is decided by the attending physician.

The therapeutic measures used for the reversal of oral anticoagulation with warfarin will depend on how quickly normalization of prothrombin time, measured by the INR, is reached. For surgeries that can wait 18-24 hours, suspension of warfarin associated with intravenous vitamin K1 at a dose of 2.5-5 mg usually normalizes the INR when it is within the therapeutic range.380

If rapid normalization of INR is needed, it is necessary to replace the deficient coagulation factors with fresh frozen plasma (FFP) and prothrombin complex concentrate. The Resolution - RDC No. 10 of January 23, 2004 from the Brazillian Health Regulatory Agency (ANVISA) states that “for the correction of hemorrhage due to coumarin antiplatelet agents or rapid reversal of the effects of coumarins”, the product of choice is the prothrombin complex. As this type of concentrate is not yet broadly available in Brazilian hospitals, the use of FFP is an acceptable alternative.386

For the FFP, the recommended dose is 15 mL/kg of body weight and volume overload should be avoided.387 No standard procedure has been established for the prothrombin concentrate. Table 5 shows the doses used in health services in the United Kingdom. However, regardless of what is used to replace vitamin K-dependent factors, using vitamin K1 (2.5-5.0 mg, oral or slow venous administration) is necessary to maintain normal prothrombin values during the preoperative period.380

Dabigatran is an anticoagulant drug that acts as a direct inhibitor of thrombin, reversibly blocking the conversion of fibrinogen to fibrin (factor IIa). It is a drug that acts rapidly. Its concentration peaks after 30-120 minutes. Dabigatran has a half-life of 12-17 hours and is predominantly renally excreted (80%).

This drug is approved for preventing stroke in patients with non-valvular AF, in the treatment of VTE (DVT/PE), and for the prevention of recurrent VTE and VTE in major orthopedic surgeries. However, its use is not authorized for the prevention of arterial thromboembolism in patients with mechanical valve prostheses. Because of its rapid action and shorter half-life, there is no need for bridge therapy associated with this drug.

One of the concerns associated with the use of dabigatran is the lack of specific antidotes until recently. The available possibilities were limited to the use of the prothrombin complex and hemodialysis, which had limited success. The first antidote for thrombin inhibitors (dabigatran), the idarucizumab, was FDA approved in the USA in October 2015. Idarucizumab completely reversed the anticoagulant effect of dabigatran in phase I and phase III studies. Another promising agent under study is Arapazine (PER-977), which has been shown to reverse the anticoagulant action of dabigatran, as well as rivaroxaban, apixaban, and LMWH.392

Rivaroxaban is a drug that acts as a factor Xa inhibitor, blocking its enzymatic function of converting prothrombin to thrombin. It is also a fast-acting substance. Its concentration peaks after 2-4 hours and has a short half-life (5-9 hours in young people and 11-13 hours in the elderly). This drug undergoes liver metabolism and renal excretion (66%).

Rivaroxaban is approved for the prevention of stroke in patients with non-valvular AF, in the treatment of VTE (DVT/PE), in the prevention of recurrent VTE, and in the prevention of VTE in major orthopedic surgeries. However, its use is not authorized for the prevention of arterial thromboembolism in patients with mechanical valve prostheses. Since it is fast-acting and has a shorter half-life, there is no need for bridge therapy associated with this drug.

In the past, only the prothrombin complex was available to reverse the effect of rivaroxaban since there were no specific antidotes available. Currently, Andexanet alfa (PRT064445) is a specific antidote against factor Xa inhibitors. It shows a rapid reversal of the anticoagulant effect of apixaban and rivaroxaban in minutes, as observed in two recent parallel phase III studies, ANEXA-A and ANEEXA-R, respectively. Currently, ANNEXA-4, a phase IV study, is being performed. Another promising agent under study is Arapazine (PER-977), which shows an effect in reversing the anticoagulant action of dabigatran, as well as rivaroxaban, apixaban, and LMWH.392

Apixaban is also a factor Xa inhibitor that blocks the conversion of prothrombin to thrombin. It has a rapid onset of action. Its concentration peaks after 3 hours and has a short half-life (8-15 hours). This drug undergoes liver metabolism and renal (27%) and fecal excretion. Apixaban is approved for the prevention of stroke in patients with non-valvular AF, prevention of VTE in major orthopedic surgeries, and treatment of DVT and PE. Its use is not authorized for the prevention of arterial thromboembolism in patients with mechanical valve prostheses. Due to its rapid onset of action and shorter half-life, there is no need for bridge therapy associated with this drug.

Currently, Andexanet alfa (PRT064445) is the specific antidote against factor Xa inhibitors. It shows a rapid reversal of the anticoagulant effect of apixaban and rivaroxaban in minutes, as observed in two recent parallel phase III studies, ANNEXA-A and ANEEXA-R, respectively. A phase IV study, ANNEXA-4, is underway. Another promising agent under study is Arapazine (PER-977), which shows an effect in reversing the anticoagulant action of dabigatran, as well as rivaroxaban, apixaban, and LMWH.392

Edoxaban is also a factor Xa inhibitor. It has a rapid onset of action. Its concentration peaks in 1-2 hours. This drug has a short half-life (10-14 hours) and undergoes renal (50%) and biliary and intestinal (50%) excretion. Edoxaban is indicated for the prevention of arterial thromboembolic phenomena in patients with non-valvular AF and in the treatment of DVT or PE, but it has not yet been released in Brazil.

At present, there are no studies investigating specific antidotes for edoxaban. An option would be to use the prothrombin complex for the occurrence of bleeding that necessitates the reversal of its effect. Because it is the newest oral anticoagulant, studies evaluating its use in the perioperative period are very limited.

In principle, the most accepted approach is interruption of edoxaban 24 hours before surgeries with low risk of bleeding and interruption 48-72 hours before surgeries associated with high risk of bleeding (Chart 10).

Early studies correlated dental extraction with the presence of transient bacteremia.396^,397 Others indicated that endodontic and periodontal manipulation may lead to similar levels of bacteremia.398^-400 Based on this, experimental animal models confirmed the reduction of bacteremia after dental manipulation with the use of prophylactic antibiotic therapy.401^,402 Since then, this recommendation has been established for individuals with an anatomical predisposition to endocarditis.

More recently, the impact of prophylaxis on the prevention of endocarditis has been questioned. Clinical trials showed a low prevalence of infectious endocarditis (IE) presumably related to dental treatments, ranging from 2.7 to 13%.403^-405 Moreover, it has been shown that daily activities, such as mastication, tooth brushing, and flossing are related to transient bacteremia.400^,406^-409 Other arguments against the use of prophylaxis are risk of anaphylaxis associated with the use of penicillin, efficacy proven only in experimental studies, and possibility of induction of bacterial resistance.408^,410

Based on these arguments, the recommendation not to use prophylaxis for endocarditis has been instituted in the United Kingdom by the National Institute for Health and Clinical Excellence (NICE) since 2008.411 In France, prophylaxis has been recommended for high-risk individuals only since 2002.412 The same recommendation has been made by the American Heart Association (AHA) since 2007413 and the European Society of Cardiology (ESC) since 2009.414 The population considered to be at high risk is composed of individuals with a greater chance of developing complications and die due to illness (severe IE risk conditions). The individuals described in chart 12 are considered at risk of IE.413^,414

Recent observational studies have shown no increase in the number of endocarditis following recommendations for prophylaxis in high-risk individuals in France and the United States.415^-417 However, an observational study in the United Kingdom showed an increase in the incidence of infective endocarditis since the NICE recommendations in 2008.418 In this country, a study performed in 2012 revealed that most cardiologists and cardiac surgeons believed that prophylaxis should be performed in cases of valve prosthesis and previous endocarditis.419 In the USA, following the new AHA recommendations in 2007, one study showed an increase in the incidence of hospitalizations for streptococcal endocarditis.420 On the other hand, the limitations imposed on observational cohorts should be considered.

Considering that most patients with valvular heart disease in Brazil present characteristics different from those currently observed in the USA and European countries (young people with rheumatic sequelae and higher lethality due to endocarditis) and the lack of prophylaxis studies in Brazil, prophylaxis is recommended for patients with native valve injury, although they do not have a valve prosthesis. Another difference of our population compared to the American and European populations is the higher prevalence of individuals with low access to health care and therefore with lower dental hygiene and higher risk of bacteremia after invasive dental procedures.400^,421

Although cited in the international literature, a significant adverse effect of antimicrobial therapy is a rare event. Therefore, use of prophylaxis for endocarditis is recommended prior to dental procedures involving the manipulation of gingival tissue, periodontal region, or oral mucosa perforation (Chart 13) for all individuals with anatomically relevant valve disease (Chart 12). The antibiotic should be given as a single dose 30-60 minutes before the procedure (Table 6).

It should be noted that infective endocarditis is a more frequent result of bacteremia from daily activities than after dental procedures. There is no doubt that maintaining good oral health is the best strategy to prevent endocarditis. In individuals with periodontal and endodontic diseases, the incidence and magnitude of bacteremia in daily care and during procedures are higher compared to individuals with healthy teeth.421 Thus, we recommend emphasizing daily dental care and biannual dental evaluation.

Patients submitted to an incision or biopsy of the respiratory tract mucosa, such as otorhinolaryngological surgeries, should receive an antibiotic treatment scheme similar to the one used before dental treatment with high risk of bacteremia. There is no recommendation for prophylaxis for bronchoscopy, laryngoscopy, and orotracheal intubation. For infection treatments, such as abscess drainage, antibiotic prophylaxis with antistreptococcal action should also be administered.414

Despite limited evidence, it is believed that patients at high risk for infective endocarditis (Chart 12) would probably benefit from prophylaxis before genitourinary or gastrointestinal procedures. Patients with non-high-risk valvular heart disease may also benefit from prophylaxis before these procedures (Chart 12). The recommended antibiotic treatment scheme for this group is in table 7.

For treatment of infections, such as abscess drainage, antibiotic prophylaxis should be administered with antistaphylococcal and antistreptococcal action.414

The number of reports of infective endocarditis related to piercing and tattooing has increased, mainly associated with tongue piercing, but the risk was not estimated.422 Therefore, patients should be warned about this risk.

DVT and PTE are two manifestations of the same disease, the venous thromboembolism (VTE). There are clinical probability scores that can be used for the diagnosis of VTE. One of the most used is the Wells score (Table 9 for DVT and table 10 for PTE).460^,461

DVT of the lower limbs is subdivided into two categories, namely, distal venous thrombosis (calf veins) and proximal vein (popliteal, femoral, or iliac veins). Proximal venous thrombosis is the most commonly associated with PTE. The diagnosis is performed with history and clinical examination (edema, pain, and erythema involving the site) and confirmed by imaging tests. The proposed flowchart for the diagnosis of DVT is shown in figure 4.

Figure 4

DVT: deep venous thrombosis.

Acute PTE is a common and often fatal disease. Clinical evaluation and diagnostic tests are required before the start of anticoagulation (Figure 5).

Figure 5

PTE: pulmonary thromboembolism; Chest AngioCT: computed tomography angiography of the chest. * Ventilation/perfusion scintigraphy; contrasted pulmonary angiotomography; serial Doppler of lower limbs; pulmonary angiography by nuclear magnetic resonance.

Arterial blood gas: this test is usually altered. However, it is neither sensitive nor specific for the diagnosis of PTE. Hypoxemia is present in 74% of cases.

BNP and troponin: they may be increased but are not sensitive or specific tests for the diagnosis of PTE. They have prognostic implications, being indicative of the severity of PTE.

D-dimer: D-dimer, as well as for DVT, is a sensitive but not very specific test.

ECG: ECG alterations in patients with PTE are common, although not specific. Tachycardia and ST-segment and T-wave alterations are the most frequent findings (70% of cases of PTE). Classical alterations considered as suggestive of PTE (S1 Q3 T3, right ventricular overload, and incomplete right branch block) are not frequent (< 10%). Electrocardiographic changes that are associated with a worse prognosis are atrial arrhythmias (e.g., AF), bradycardia (< 50 bpm) or tachycardia (> 100 bpm), new right branch block, Q waves in lower leads (DII, DIII, and aVF), ST-segment changes in anterior wall, and T-wave and standard S1 Q3 T3 inversion.

Chest X-ray: Chest X-ray is a common test, but it has low sensitivity and specificity. However, it can detect atelectasis or parenchymal abnormalities (18-69%), pleural effusion (47%), and cardiomegaly (> 50%). Peripheral wedge-shaped opacity in peripheral lung regions and abrupt cut-off of pulmonary arterioles with distal hypoperfusion are rare but are highly suspected of PTE. Chest X-ray may be normal in 12-22% of patients with PTE.

Angio-CT with protocol for PTE: for most patients with PTE, it is the selected diagnostic test due to its high sensitivity (> 90%) and specificity (> 95%) for PTE, especially when associated with D-dimer dosage,471 in patients with moderate to high probability. Demonstration of filling failure in any branch of the pulmonary artery using contrast is a diagnosis of PTE.

Ventilation/perfusion (V/Q) scintigraphy: this test is reserved for patients with suspected PTE when Angio-CT is contraindicated (renal insufficiency, creatinine clearance < 60 mL/min/m2, contrast allergy, or morbid obesity) or when Angio-CT is inconclusive or negative, but in disagreement with the high clinical suspicion. V/Q scintigraphy is a sensitive test for the diagnosis of PTE, but it is not specific due to its high incidence of false-positive results. Accuracy is higher when chest X-ray is normal. V/Q scintigraphy is selected for the diagnosis of PTE during pregnancy. A high probability V/Q scintigraphy is sufficient for the diagnosis of PTE, whereas a normal scintigraphy is sufficient to exclude PTE. Low or intermediate probabilities are not sufficient for diagnosis.

Pulmonary digital angiography with iodinated contrast: this test was the historical gold standard for the diagnosis of PTE. With the development of Angio-CT, it is reserved for patients with suspected PTE when Angio-CT or V/Q scintigraphy is not conclusive. In a retrospective analysis of 20 cases of the PIOPED II study,472 digital angiography was shown to be less sensitive than Angio-CT for the diagnosis of small emboli. It has morbidity of 5% and mortality of < 2%. Exposure to radiation is greater than in Angio-CT. Demonstration of filling failure and abrupt cut-off of pulmonary arterial vessel are diagnoses of embolization.

Pulmonary angiography by nuclear magnetic resonance: this test is not very sensitive (77-84%). It is reserved for cases in which other methods cannot be performed.

Echocardiogram: this test does not diagnose PTE. However, presumptive diagnosis can be performed using the echocardiogram in patients with high clinical suspicion and hemodynamic instability. Approximately 30-40% of patients with PTE have echocardiographic changes, indicative of right ventricular overload, especially in those with massive PTE, such as right ventricular dilatation, right ventricular dysfunction, and tricuspid insufficiency.

There is limited evidence for the best treatment for venous thromboembolism in the perioperative period. This is because surgeries may have different bleeding risks, clinical situations may be extremely heterogeneous, and conventional therapy may not be the most appropriate after a specific surgical procedure. Thus, we will present the usual recommended treatment, regardless of the perioperative context, but it is important to customize the decisions in conjunction with the surgeon.

The main pillar of venous thromboembolism treatment (deep vein thrombosis and PTE) is anticoagulant therapy, which should be a long-term therapy, with a duration of at least 3 months.380 However, there are clinical situations that require the use of anticoagulant therapy for longer periods, which has become known as extended anticoagulant therapy and implies its use for an indefinite period.380

Several recent studies have examined the efficacy of new anticoagulants in acute and long-term treatment of VTE and compared to warfarin. These studies showed that the reduction in the risk of recurrence of VTE is similar with both therapies, including cancer patients.473^-477 The reduction in the risk of recurrence of VTE with the different new anticoagulants (dabigatran, rivaroxaban, apixaban, and edoxaban) is not directly compared among them, but it appeared to be equally effective based on indirect comparisons.477 In fact, in the recommendations, the order of the new anticoagulants cited in the text refers to the chronology of the publication of the phase III studies on VTE and is not the order of preference.

The new anticoagulants have lower bleeding rates and provide greater convenience for patients and health professionals in relation to fixed dose. They have less drug and food interactions and do not require serial blood tests to ensure a specific therapeutic range. Given these advantages, the new anticoagulants are now preferred compared to warfarin for initial and long-term treatments of VTE in patients without cancer.

In cancer patients, a recent randomized study compared the use of LMWH (tinzaparin) and warfarin for the treatment of 900 cancer patients with DVT during the first 6 months. The study demonstrated that LMWH is more effective than warfarin, without alterations in death rates and major bleeding.478 Other studies have also shown that the reduction of the risk of recurrence of VTE in cancer patients is higher with the use of LMWH compared to warfarin.479^,480

Therefore, warfarin is preferentially used in patients with VTE without cancer and LMWH in patients with VTE and cancer. No study has directly compared the new anticoagulants and LMWH in cancer patients. However, based on indirect comparisons, LMWH appears to be more effective than the new anticoagulants in patients with VTE and cancer.479

It is important to emphasize that parenteral anticoagulation was given before the use of dabigatran and edoxaban in previous studies. It was not used prior to rivaroxaban and apixaban. It was used before and for a period with warfarin until the desired INR was reached.

The studies that determined the appropriate duration of treatment for VTE basically compared four treatment duration options: 4 or 6 weeks; 3 months; more than 3 months, although limited to 6 to 12 months; and extended or indefinite duration therapy. These four treatment duration protocols were tested in different available studies in four profiles of VTE patients with different estimates of recurrence risk after suspending anticoagulant therapy:

(1) VTE caused by surgery (major transient risk factor with 3% recurrence in 5 years);481 (2) VTE caused by a transient non-surgical risk factor (estrogen therapy, pregnancy, lower limb lesions, flights > 8 hours, with a risk of recurrence of 15% in 5 years); (3) idiopathic VTE with no transient risk factors or cancer (30% recurrence in 5 years);482^,483 (4) VTE associated with cancer (15% annual recurrence).484^,485

Another important factor that guides the duration of anticoagulant therapy in VTE is the risk of bleeding that can be categorized as low (absence of risk factors for bleeding, with a 0.8% annual risk for major bleeding), moderate (one risk factor for bleeding, with a 1.6% annual risk for major bleeding), or high (two or more risk factors for bleeding, with a ≥6.5% annual risk for major bleeding). The risk factors for bleeding during anticoagulant therapy are described in chart 14.486

It is important to note that for all patients using extended or indefinite anticoagulant therapy, treatment should be reassessed at least annually.

It is still unclear whether the benefits of anticoagulation outweigh the risks of anticoagulant treatment for distal isolated DVT because of the low risk of progression and recurrence of VTE.380 About 15% of distal isolated DVT will develop with thrombus progression to the popliteal vein and risk of PTE.487

The following risk factors favor thrombus extension in distal isolated DVT and support the use of anticoagulant therapy over imaging follow-up: positive D-dimer, extensive thrombosis involving multiple veins, proximal vein thrombosis, absence of reversible trigger factor for DVT, active cancer, previous history of VTE, and hospitalized patient.488^-492

Evidence is scarce with regard to the use of catheter-directed thrombolysis for the treatment of proximal DVT of the lower limb, causing substantial uncertainty that the benefits outweigh the risks associated with the procedure.380

Evidence shows the inconsistent benefit of using the inferior vena cava filter to prevent recurrence of VTE in anticoagulated patients. The most recent randomized study, PREPIC,493 demonstrated that inferior vena cava filter implantation during 3 months does not reduce the recurrence of PTE, including fatal PTE, in anticoagulated patients with PTE and DVT, with additional risk factors for recurrence of VTE

A recent multicenter, placebo-controlled, study has shown that in contrast to two previous smaller studies, the routine use of compression stockings does not reduce the risk of post-thrombotic syndrome nor does it add any other important benefit.494

With technological development of pulmonary angiotomographies, diagnostic identification of subsegmental PTEs increases and the best therapeutic management in these cases is uncertain. Changes are often small and may correspond to false positives, and true subsegmental PTE is generally associated with small DVTs. The risk of progression or recurrence of VTE in the absence of anticoagulation is small in relation to larger PTEs.486

Imaging of DVT should be performed on the lower limbs, as well as on the upper limbs and central venous catheters. If DVT is detected, anticoagulant therapy should be introduced, but if no DVT is detected, the need for anticoagulation in these patients is uncertain.486

In these cases, it is important to evaluate the risk factors for recurrence or progression of VTE, which include the following: patients hospitalized or with reduced mobility for other reasons; patients with active cancer, especially those with metastatic disease or those treated with chemotherapy; or patients with non-reversible risk factors, such as recent surgery. Similarly, important clinical symptoms that cannot be attributed to another cause or a low functional reserve favor the use of anticoagulant therapy, whereas the presence of a high risk of bleeding favors the preference for clinical follow-up.

Recent meta-analyses evaluated the possibility and safety of home treatment for pulmonary embolism.495^-497 It is recommended that patients candidate for home treatment meet all the following criteria:380 clinically stable with good cardiopulmonary reserve; no contraindications, such as recent bleeding, severe renal or hepatic failure, or severe thrombocytopenia (< 70,000/mm3); willingness to follow the treatment; and feel safe to be treated at home.

Systemic thrombolytic therapy is associated with a faster decrease in pulmonary artery pressure, increased arterial oxygenation, and resolution of filling faults on CT, accelerating the resolution of PTE. However, such therapy is associated with increased bleeding risks. Patients who will benefit the most are those who have the highest risk of death associated with PTE and the lowest risk of bleeding.380

Recently, three randomized trials evaluated the use of systemic thrombolytic therapy in 1,200 patients with acute PTE and improved the evidence regarding this topic.498^-500

It is important to note that patients recently submitted to a surgery will always have an at least moderate risk of bleeding (Chart 14) and the possibility of using systemic thrombolytic therapy for the treatment of acute PTE needs to carefully assess the risks and benefits and should be discussed with the surgeon.

DM affects 6.2% of the Brazilian adult population,501 with a progressive increase in prevalence according to age, affecting more than 19% of individuals more than 65 years. These patients have a high incidence of CDs. Glycemic control is one of the most important aspects to be considered in the perioperative evaluation of patients with DM. There is substantial observational evidence that links hyperglycemia to unfavorable surgical outcomes, such as infection, longer hospital stay, disability after discharge, and mortality.

In Brazil, approximately 90% and 73% of type 1 and type 2 DM patients, respectively, are outside the recommended targets for glycemic control (glycated hemoglobin lower than 7.0%). Therefore, it is expected that most individuals in preoperative evaluation need specific guidelines regarding glycemic control.

Preoperative evaluation becomes an additional opportunity to adjust medication doses, educate an individual, and improve metabolic control. A staggered scheme (insulin to correct capillary glycemia) should be avoided as an exclusive therapy for prolonged periods because it is ineffective for most patients. In addition, this scheme favors glycemic variability, attempting to correct the “problem” (hyperglycemia) after it has already occurred and may even be deleterious, predisposing to diabetic ketoacidosis in patients with type 1 DM.

Prandial insulin: dose of fast (regular) or ultrafast (lispro, aspart, glulisine) insulin used to control postprandial blood glucose, used before meals.

A free Brazilian application was developed at the Hospital das Clínicas (HCFMUSP) to assist physicians and nurses in performing intensive glycemic control in hospitalized patients. InsulinApp is a tool developed for smartphones and tablets that calculates hospital doses of insulin necessary for a patient in a few minutes.508 It is available free of charge from both Google Play (Android) and Apple (iOS) under the name InsulinAPP.

Pre-assessment and in-hospital monitoring with specialist is recommended, if available.

Monitor capillary glycemia: pre-meal and at 10 pm while maintaining usual diet; every 4 hours during the fast; and every hour or two hours if using continuous intravenous insulinization.

Never substitute basal-bolus insulin in the preoperative period by staggered scheme alone - risk of diabetic ketoacidosis.

In medium to major surgeries or with a surgical time of more than 1 hour, ideally use continuous intravenous insulin pump as soon as fasting starts or on the morning of surgery, maintaining the therapy during the intraoperative and postoperative periods.

If venous insulinization is not possible to perform, the following can be used:

Evaluate blood glucose before surgery.

Correct hypoglycemia and maintain glucose supply at 5 to 10 g/h of glucose. Preferably, control hyperglycemia with intravenous insulin and maintain blood glucose levels between 100 and 180 mg/dL.

Attention to potassium correction.

In 2001, an important study demonstrated a clinical benefit of strict glycemic control in the postoperative period for the first time in surgical patients: lower rates of in-hospital mortality, polyneuropathy, infections, and acute renal failure, and shorter time of mechanical ventilation and of stay in ICUs.509 Regarding patients with diabetes, the clinical benefit associated with strict glycemic control was also observed, but there was no impact on the reduction of mortality. Based on this study, the recommendation was strict glycemic control in the postoperative period for patients undergoing noncardiac surgery.

Another large randomized multicenter trial (NICE-SUGAR)510 involving more than 6,000 patients, with approximately one-third of surgical patients and two-thirds of clinical patients, compared strict glycemic control (81-108 mg/dL) with conventional glycemic control (144-180 mg/dL). Surprisingly, the group of patients randomized to strict control presented higher mortality rates in a period of 90 days (27.5%) compared to the conventional group (24.9%). No differences were found in other minor outcomes between the groups. The group with the strict glycemic control presented higher hypoglycemia levels (<40 mg/dL) compared to the control group.

In epidemiological studies, the overall incidence of hypothyroidism varies from 0.1 to 2%. The prevalence of subclinical hypothyroidism is higher, ranging from 4 to 10% in the adult population and tending to be higher in women more than 65 years. Most of the population, even asymptomatic, has thyroid alterations. Some clinical conditions present a potential risk for the development of perioperative complications and a rapid decline in thyroid function, such as age > 65 years; hypothalamic or pituitary disease; coexisting autoimmune disease; irradiation of the neck, thyroid surgery, or radioiodine therapy; significant hyperlipidemia; hyponatremia; high levels of muscle enzyme; macrocytic anemia; and pleural or pericardial effusion.

If these risk conditions are present, screening for thyroid disease may be useful in the preoperative period. The recommended test is TSH because 95% of the causes of hypothyroidism are of primary thyroid etiology.

No randomized study has demonstrated the benefit of patients with hypothyroidism being euthyroid in the preoperative period compared to hypothyroid and postoperative morbidity and mortality. Current evidence shows that if there is a prior diagnosis and time, the patient should be euthyroid in the preoperative period. However, if the patient has subclinical or mild hypothyroidism and the operation is urgent, the surgical procedure should not be postponed. In elective surgeries, treatment can begin, but we should not wait until TSH normalized.

Patients with clinical or moderate hypothyroidism and scheduled for urgent surgery should undergo the surgical procedure and initiate treatment in the immediate postoperative period. Patients with moderate hypothyroidism and elective surgeries should wait for euthyroidism to undergo the surgery. These patients do not necessarily need to achieve normalized TSH levels because 10 to 20% have a slow TSH decay. The most important criterion is that progressive increase and normalization of free T4 levels occur, which should exist within seven days, or treatment should continue to further increase the levels.

Patients with severe hypothyroidism or myxedema coma should only be operated if the surgery is an emergency. If the surgery is elective, previous treatment of hypothyroidism and acquisition of normal thyroid function should be considered. Treatment should be given in the form of T4 and T3. The doses used are as follows: T4 with an attack dose of 200-300 mcg intravenously, followed by 50 mcg per day; T3 dose of 5-20 mcg intravenously, followed by 2.5-10 mcg every 8 hours, depending on age and cardiovascular comorbidities.

In the postoperative period of any patient with hypothyroidism, if the patient does not resume eating in 5-7 days, 80% of the total dose of T4 should be given intravenously or intramuscularly once a day. The dose is 20% lower due to bioavailability.

Thyrotoxicosis affects 2% of women and 0.2% of men. The prevalence of clinical and subclinical hyperthyroidism in the USA is 0.2 and 1%, respectively. The most common causes are Graves-Basedow’s disease, toxic nodular goiter, thyroiditis, and iatrogenic. Adrenergic effects pose a high risk for perioperative complications, such as cardiac arrhythmias (8-15% AF). These are related to the increase in the number and/or sensitivity of β-adrenergic receptors. In addition, studies showed that more pronounced hyperthyroidism indicates greater chance of AF. The mortality in hyperthyroidism is related to the occurrence of cardiovascular events^.513^,517^-521

For the diagnosis, there should be laboratory confirmation of clinical suspicion. TSH level should be low, and free T4 level should be normal (subclinical hyperthyroidism) or high. Several situations may increase the total T4 levels by increasing the T4-binding protein. However, they do not affect free T4, which has biological activity: pregnancy, cirrhosis, acromegaly, Cushing’s syndrome, use of lithium, contraceptives, propranolol, amiodarone, and iodinated contrast agents. In these situations, there is no real hyperthyroidism, only a compensatory increase in free T4, a consequence of the increase in TBG, a T4-binding protein.

Cardiovascular: increased cardiac inotropism and chronotropism with decreased systemic vascular resistance (SVR), left ventricular hypertrophy, increased incidence of angina, HF, arrhythmias, and embolic events.

Treatment of thyrotoxic storm includes hydration, cooling, inotropes (if necessary), administration of PTU attack dose (1,000 mg oral) and maintenance (200 mg every 6 hours), ventilatory support, oral metabolic control, hydrocortisone attack dose of 300 mg intravenously and maintenance of 100 mg every 8 hours, iodine as oral Lugol or intravenous iodine at a dose of 1 g every 8 hours, and, if necessary, plasmapheresis, dialysis, or cholestyramine to remove hormones from the circulation.

Hypotension and hemodynamic shock (which may be resistant to vasopressors), with multiple organ dysfunction; hypoglycemia; tachycardia; hydroelectrolytic disorders: hyponatremia, hyperkalemia (in primary adrenal insufficiency - AI), hypercalcemia, acidosis; cardiac hypocontractility; anemia, eosinophilia, and neutropenia; nausea, vomiting, weakness, orthostatic hypotension, dehydration, abdominal or flank pain (acute adrenal hemorrhage), fatigue, weight loss; vitiligo, alteration of skin pigmentation, hypogonadism, hypothyroidism.

Diagnosis of AF should not be trusted if there is unexplained hypotension or refractory shock to volume and drugs in the intraoperative or postoperative period, discrepancy between disease severity and patient condition, high fever without apparent cause (negative cultures) or unresponsive to antibiotic therapy, unexplained mental changes, apathy, or depression without specific psychiatric disorder. In such cases, AAI should be initiated, and subsequent diagnostic confirmation should be obtained.

Patients with diagnosis of AI,523 patients at risk for AI524 and patients with relative hypoadrenalism (limited adrenocortical reserve): pituitary tumors (macroadenomas); radiation therapy of the pituitary region; previous pituitary surgical intervention; postoperative period of Cushing’s disease surgery, bilateral adrenalectomy or unilateral adrenalectomy in case of another adrenal attack; chronic corticosteroid users (> 5 mg of prednisone or equivalent for more than 21 days or dose > 7.5 mg for more than 14 days); patients with type 1 DM or autoimmune diseases (Hashimoto’s thyroiditis, ovarian or primary testicular failure, hypoparathyroidism, vitiligo, autoimmune polyglandular syndrome); individuals with suggestive clinical conditions (darkening of the skin, weakness, fatigue, nausea, vomiting, depression, hypotension, electrolytic imbalances, hypoglycemia, fever).

Evaluation of the hypothalamic-pituitary-adrenal axis for confirmation of AI should be made by measuring serum cortisol level at 8 AM. Perioperative corticosteroid replacement may be indicated depending on the result (Chart 17). If < 5 mcg/dL, replacement should be performed; 5-10 mcg/dL, perform simple cortrosyn test and measure serum ACTH levels to complement evaluation because it can be a false positive, that is, have an acute response and have no reserve. In this case, empirical therapy with steroids should be started. In cortisol level > 10 mcg /dL, replacement is not necessary.

Small (local anesthesia or hernia): maintain usual dose of corticoid in the morning without a new attack dose; maintain regular dose for 24 hours perioperatively.

Clinical history limited by the difficulty to differentiate between dyspnea and cardiogenic and pulmonary origins of the low functional capacity of the obese. Physical examination and detailed analysis of the cardiopulmonary system is limited due to obesity. Few risk scores used in the perioperative evaluation include obesity and quantify the risk associated with this variable.

Higher prevalence of comorbidities that are risk factors for atherosclerosis and myocardial ischemia (hypertension, DM, and dyslipidemia); increased risk of thromboembolic events and infection of the surgical wound; greater difficulty in measuring blood pressure and acquiring venous access; longer mechanical ventilation time and longer hospitalization time; increased risk of renal failure; greater sensitivity to opioids and sedatives; increased risk of aspiration of gastric contents; higher probability of presenting hypoxemia due to hypoventilation, pulmonary restriction, postoperative atelectasis, increased occurrence of central and obstructive sleep apnea and hypercapnia; higher mortality in intensive care in severely obese patients. Specific regimens for venous thromboprophylaxis in obese patients are shown in table 12.

As doenças hematológicas podem aumentar a morbidade e aHematologic diseases may increase the morbidity and mortality of individuals submitted to surgical procedures. Anemia is the most common hematological problem found in the preoperative period. It is generally defined according to WHO criteria:540^,541 hemoglobin concentration < 13 g/dL for men and < 12 g/dL for women. It is often a sign of underlying disease that can affect the surgical outcome. Studies involving many patients indicated that preoperative anemia is an independent risk factor for morbidity, mortality, and transfusion requirement, with an association between extent of anemia and outcome.542^-546 Anemia leads to overload of the cardiovascular system, increasing cardiac output. Individuals with CD have less tolerance to anemia, and its presence can intensify the conditions of myocardial ischemia and underlying HF. Therefore, identification of anemia in the preoperative period assists in the identification of patients at risk of adverse outcome in the postoperative period. Whenever possible, anemia should be identified, investigated, and corrected before surgery, although there is no randomized evidence that its correction alters the perioperative risk. On the other hand, there is randomized evidence that the correction of anemia in the preoperative period decreases the need for blood transfusion and consequently the postoperative transfusion risk.546

The available guidelines for perioperative blood transfusion are limited, but the risks and benefits of this measure should always be questioned.547 Traditional practices, such as correction of preoperative anemia for normal or near-normal values of hemoglobin concentration (Hb ≥ 12 g/dL) to prepare patients for surgery, are not supported in the literature and are not recommended in clinical practice.546

Numerous studies and reviews have attempted to establish transfusion triggers for patients with anemia by evaluating two strategies: “restrictive” (usually Hb < 7.0 g/dL) and “liberal” (usually Hb ≥ 7.0 g/dL).548^-551 Most meta-analyses included perioperative patients of various natures: critical and clinical patients, as well as adults and children. The meta-analysis of Carson et al.548 included 6,264 patients. These surgical and clinical patients involved adults and children. The authors concluded that the existing evidence supported the use of restrictive transfusion therapy in most patients, but that the effects of the restrictive strategy in high-risk groups, such as acute coronary syndrome, needed to be tested in future large studies. In the meta-analysis of Holst et al.,549 9,813 patients from 31 randomized controlled trials were included. These trials consisted of twenty perioperative and acute blood loss studies, eight critical patient studies, two trauma studies, and one study with patients with leukemia undergoing bone marrow transplantation. According to the authors, the results were not affected by the inclusion of studies with high risk or unclear risk. The restrictive strategy was associated with the reduction in the number of transfused red cell concentrate units and the number of transfused patients; however, mortality, overall morbidity, and myocardial infarction remained unchanged. The restrictive strategy is safe in most clinical settings, and liberal transfusion does not show any benefit to the patients analyzed in this review.549

In the meta-analysis of Docherty et al.,551 they specifically investigated the effect of restrictive versus liberal strategy in patients with CD undergoing noncardiac surgery. A total of 3,033 patients were included: 1,514 with restrictive transfusion and 1,519 with liberal transfusion. The risk of acute coronary syndrome was higher in patients with a restrictive strategy compared to patients with a liberal strategy, but the effects on mortality and other outcomes were uncertain in a period of 30 days. The authors concluded that it may not be safe to use a transfusion trigger below 8 g/dL in these patients.551

Perioperative mortality in a period of 90 days was investigated in a recent review with meta-analysis, which included only perioperative adult patients and critically ill patients. Twenty-seven studies with 11,021 patients were included in the review: 17 in the perioperative period (9 in orthopedic surgery, 5 in cardiac, 1 in vascular, 1 in oncologic, and 1 in obstetric) and 10 in critically ill patients. Overall, there was no difference in mortality between the liberal and restrictive strategies. However, in the perioperative period, mortality was reduced in adult patients randomized to receive the liberal strategy compared to those who received the restrictive strategy with 7,552 patients. In critical patients, there was no difference between the groups. The heterogeneity between the studies was low. It was concluded that blood transfusion had a different statistically significant effect on the survival of patients in different clinical contexts.550

In an isolated study, Carson et al.552 included 2,016 patients aged ≥ 50 years. The patients had a history or risk factors for CD, with a hemoglobin concentration below 10 g/dL after hip surgery. They were randomized to liberal (transfusion trigger above 10 g/dL) or restrictive (symptoms of anemia or Hb < 8 g/dL) strategy to determine whether a higher transfusion trigger would improve the recovery of patients submitted to orthopedic hip fracture surgery. The authors concluded that the liberal and restrictive strategies did not reduce death rates, did not improve recovery in a period of 60 days of follow-up, and did not reduce hospital morbidity in older patients at high cardiovascular risk.552

Another isolated study was conducted at John Hopkins Hospital including 10,163 patients submitted to vascular or gastrointestinal cardiothoracic surgery. The authors aimed to determine the transfusion practices and the effect of the use of transfusion on the perioperative outcome. They concluded that the use of the liberal transfusion trigger (Hb ≥ 7.0 g/dL) after major surgeries was more common than the restrictive practice (trigger < 7.0 g/dL) and that patients with restrictive transfusion had no increased risk of complications compared to patients with liberal transfusion.553

Therefore, the optimal transfusion trigger remains undetermined. Hemoglobin level is probably not the answer because some patients require higher values and others tolerate values lower than 7 g/dL.554 The transfusion triggers used in isolated studies and in studies that were part of large reviews with meta-analysis are not homogeneous.

Therefore, the decision on blood transfusion should be based not only on hemoglobin levels, but also on the suspicion of organic ischemia, the risk or presence of bleeding, the status of the intravascular volume, and the susceptibility to complications due to inadequate oxygenation.555 Common sense, careful observation of the patient, and clinical context should guide the decision of the best strategy for each case. It should be noted that blood transfusion is not a risk-free procedure and that a dose-dependent relationship exists between transfusions and complications.545 Thus, even if selecting the liberal strategy, hemoglobin correction towards normal values is not necessary. One unit of erythrocyte concentrate increases the hemoglobin rate by approximately 1.0 g/dL and the hematocrit by approximately 3.0%. The optimal rate of administration of red blood cell concentrate should consider the clinical situation. Most patients can receive a packed red blood cell unit every one to two hours. Patients at risk of volume overload should receive 1.0 mL/kg/h. After transfusion of each unit, the patient should be reevaluated and the hemoglobin level must be determined.547

Several studies have demonstrated a strong correlation between thrombocytopenia and hemorrhagic risk, as well as the effectiveness of platelet transfusion in reducing this risk. However, controversy still exists about the appropriate value to indicate for transfusion of platelet concentrates.562 Patients scheduled for invasive surgical procedures may present benefits with a higher platelet count with 50,000 platelets/mm3.562^-564 In neurosurgeries, there is no evidence-based data to determine a safe minimum platelet count, with several consensuses indicating 100,000 platelets/mm3.565^-567

Antiphospholipid antibodies are a family of autoantibodies directed against plasma phospholipid-binding proteins.569 Antiphospholipid syndrome is characterized by thrombosis (arterial and/or venous) and/or gestational morbidity in patients with persistent antiphospholipid antibodies.570

However, there are patients with a persistent presence of antiphospholipid antibodies without vaso-occlusive manifestations, only gestational morbidity and manifestations are not considered as criteria for the antiphospholipid syndrome (thrombocytopenia, livedo reticularis, cardiac valve disease).569 In addition, not every positive test for antiphospholipid antibodies is clinically significant and not every patient with positive antiphospholipid antibodies has the same thrombotic risk.569 To better estimate the thrombotic risk in patients with positive antiphospholipid antibody tests, some variables should be considered including persistently positive laboratory tests and presence of additional thrombotic risk factors.569

Although several studies have evaluated thrombotic risk in asymptomatic patients with persistent antiphospholipid antibodies, most of them included patients with systemic lupus erythematosus. The annual risk of the first thrombotic event in these laboratory-positive individuals, but without other associated autoimmune diseases and other thrombotic risk factors, is low (less than 1% per year). In the presence of another autoimmune disease, the risk increases to less than 4% per year. Based on the hypothesis of the association of “two injuries”, antiphospholipid antibodies induce a prothrombotic and proinflammatory phenotype in endothelial cells that is not capable of causing thrombosis alone. However, the presence of a triggering event or “second injury”, such as infection, surgeries, estrogen use, and prolonged immobilization, may trigger a vaso-occlusive event.569

Therefore, pharmacological antithrombotic prophylaxis is associated with mechanical measures in patients with positive antiphospholipid antibodies and in a period of greater vaso-occlusive risk (surgeries, immobilization, hospitalization).569 Patients with antiphospholipid syndrome under anticoagulant treatment have a higher thrombotic risk when submitted to surgical procedures.367

The term thrombophilia describes the tendency to develop venous thromboembolism due to a state of hypercoagulability caused by the presence of inherited or acquired abnormalities of coagulation or fibrinolysis.571 Hereditary thrombophilias do not present the same thrombotic risk. Severe thrombophilias are those resulting from deficiencies of natural anticoagulants (antithrombin, protein C, and protein S), abnormalities in homozygous, and presence of multiple defects. Presence of factor V Leiden in heterozygosity and the mutation G20210A in heterozygosis are considered mild thrombophilias. On the other hand, the presence of family history of venous thromboembolism events is a strong risk factor for venous thromboembolism, regardless of the presence of genetic alterations.571 Documentation of the presence of an inherited thrombophilic alteration implies the need for primary anticoagulant prophylaxis in situations where an increased risk of venous thromboembolism exists, such as surgical procedures.571

Surgical procedures should be performed in conjunction with a team experienced in the treatment of hemophilia. Before performing the surgical procedure, ensure that sufficient concentrate of the deficient factor is available.

The procedures must be performed in a medical center with adequate laboratory support, with capacity to monitor the deficient factor. Preoperative laboratory evaluation should always include the search for inhibitors for the deficient factor;

The surgical procedure should be performed at the beginning of the week and at the beginning of the day to provide optimal laboratory and blood bank support. For the intraoperative period, the plasma level of the deficient factor should be corrected for hemostatically safe values through the use of specific factor concentrate.

In the postoperative period, maintain the plasma concentration of the deficient factor for adequate time and concentration according to the type and size of the surgery.

Efficacy of hemostasis should be assessed using the criteria defined by the International Society of Thrombosis and Hemostasis (ISTH).

In the postoperative period, the minimum plasma levels of FVIII:C and von Willebrand factor:ristocetin cofactor (vWF:RCo) will vary according to the type and surgical extent.

Cessation of smoking during hospitalization offers an opportunity to access withdrawal symptoms more readily, titer medication doses more safely, and monitor the effectiveness of the therapeutic program more reliably.

The reasons that drive patients to stop smoking during hospitalization, which are part of health treatment or merely resulting from the condition of staying in a tobacco-free environment, should be seen as an important step. Measures of support and follow-up indispensable for the patient to remain abstinent should be implemented.

If such efforts are not organized in a structured program that involves the identification of smokers at the time of hospitalization, the medical center responsible for therapeutic interventions (informational, cognitive-behavioral, and medication), follow-up during hospitalization, and post-discharge follow-up, such efforts become ineffective in the medium and long term.

The negative effects of smoking on surgical outcomes are multifactorial. However, they are mainly due to the direct effects of nicotine carbon monoxide (CO) and increased oxidative and inflammatory stress. CO and nicotine increase heart rate, blood pressure, and tissue oxygen demand, as well as decrease oxygen transport capacity. Because of the vasoconstricting effect of nicotine, it increases the risk of tissue ischemia in surgery and in other areas, such as the coronary artery.627

The irritant and proinflammatory effect of numerous components of cigarette smoke on the airways also increase the susceptibility of smoking patients to respiratory infections, local healing complications in lung surgeries, and longer periods under mechanical ventilation.628

Smoking is also associated with the need for higher doses of anesthetics and neuromuscular blockers,629 increased incidence of thromboembolic events, and slower repair processes in orthopedic surgeries.630

Patients scheduled for surgeries are usually more motivated to quit smoking and are thus susceptible to a therapeutic approach to quit smoking. With the regulation of hospitals (and other enclosed spaces for public and private use) as tobacco-free environments, and with the increasing availability of effective therapeutic resources to help the patient to stop smoking, the preoperative period becomes a perfect time to stop smoking before elective surgical hospitalization.

The ideal time to stop smoking before surgery had been controversial. This was partly due to the great methodological heterogeneity of studies that evaluated the different periods to stop smoking, the difficulty of controlling confounding variables in the samples of patients, the great variation in the postoperative follow-up time, and the multiplicity of outcomes studied.

A review of prospective studies on the impact of smoking cessation in the preoperative period on the occurrence of postoperative complications (respiratory, infectious, general mortality and length of hospital stay) was conducted by Cropley and Theadom.631 They concluded that although there is great methodological limitation of the studies evaluated, there are several benefits of smoking cessation before surgical hospitalizations. They reported that longer abstinence period results in greater benefit. It should also be mentioned that there is no ideal period to recommend preoperative smoking abstinence, in terms of reduction of surgical complications and risk in the medium and long term. Cessation of smoking should not be postponed due to the unsustainable assumption that risk increases when cessation occurs less than two months before surgery.

In 2009, a retrospective cohort study evaluated data from 7,990 lung resection surgeries due to neoplasia. The study concluded that the risks of in-hospital mortality and respiratory complications after lung resection were higher in smokers and clearly reduced by smoking cessation in the preoperative period. The ideal interval between cessation of smoking and surgery could not be identified, which reinforced the recommendation for counseling (and treatment) for smoking cessation, regardless of the proximity of the surgery. This corroborates the results presented in the study published in 2001 by Nakagawa et al.632 They found a clear and increasing reduction in the risk of postoperative complications after four weeks of preoperative smoking cessation.

As in general situations, the treatment of nicotine dependence in patients scheduled for surgery and in hospitalized patients is based on cognitive-behavioral interventions (brief approach, individual counseling, provision of informational materials, and group therapy), systematized or not, and in the pharmacological support.

Regarding the “intensity” of the non-pharmacological approach, a systematic review published in 2012 evaluated several studies in hospitalized patients.633 The study showed a dose-response relationship between the intervention and the cessation rate. Moreover, structured counseling programs initiated at the hospital and extended for at least one month after discharge are more effective compared to single-point approaches during hospitalization (RR 1.37; 95% CI 1.27-1.48; 25 studies).

Prospective studies evaluating the effectiveness of implementing a structured counseling, cognitive-behavioral approach, pharmacologic support, and post-discharge follow-up of hospitalized smokers showed success rates of about 35-44% in six months634^,635 and approximately 33% after 12 months, with studies showing success rates above 50% after one year in hospitalized coronary patients.636

Given the need to specify, in cases of surgical and hospitalized patients, cessation of smoking and the control of nicotine withdrawal symptoms in a shorter period, nicotine replacement therapy (NRT), isolated or combined, is the most frequent approach selected. The usual transdermal nicotine prescription schemes (6 to 8 weeks of 21 mg/24h or 15 mg/16h, 2 weeks of 14 mg/24h or 10 mg/16h and 2 weeks of 7 mg/24h or 5 mg/16h, depending on the presentation selected) are recommended, in association with rapid forms of ad libitum replacement (in Brazil, chewing gum and tablets are available, both in presentations of 2 and 4 mg per unit) for craving episodes. The addition of NRT to an intensive counseling intervention (more than one session, post-discharge) increased smoking cessation rates compared to intensive counseling alone (RR 1.54, 95% CI 1.34-1, 79, six studies).637

The same systematic review did not show evidence of benefit of adding bupropion (RR 1.04, 95% CI 0.75-1.45, three studies) or varenicline (RR 1.29, 95% CI 0.95-1.76, two studies) to intensive counseling. Regarding bupropion, new studies, such as the randomized placebo-controlled trial conducted by Eisenberg et al.,638 failed to demonstrate the superiority of bupropion over placebo in hospitalized patients. In the same year, a study published by Smith et al.639 evaluated the addition of varenicline to counseling through a randomized, placebo-controlled, protocol. The result showed a significant superiority of pharmacological intervention compared to the control group (RR 1.45; 95% CI 1.03-2.03; p = 0.03).

Although there is potential benefit of other pharmacological approaches to smoking cessation in hospitalized patients, NRT remains the standard of care at the same doses and schemes normally recommended for other clinical situations. The use of customized doses (above 21 mg/day) of nicotine replacement to reach plasma levels of nicotine closer to the arterial concentrations of an active smoker and aimed at better control of withdrawal symptoms in heavy smokers has been tested and is safe up to doses higher than 42 mg per day,640^-643 even in individuals who persisted smoking. However, the heterogeneity of the studies and the small number of volunteers included do not provide sufficient evidence of an increase in long-term smoking abstinence rates.

The possible association of NRT with non-nicotinic drug (such as bupropion) or the choice of varenicline monotherapy is theoretically an acceptable option, but they do not find great support in specific studies in these special situations.640^-643

NRT is not superior over bupropion in patients with a history of recent (below six weeks) high-risk acute coronary syndrome and patients with complex ventricular arrhythmias. Published studies are controversial in identifying additional benefits (in addition to the control of withdrawal symptoms) from the drug treatment compared to the counseling program and behavioral approach alone.

For hospitalized patients, we propose a treatment based on the flowchart in figure 6.

Figure 6

* < 20 CIGARETTES/DAY: 14 mg patches; 20-30 cigarettes/day: 21 mg patches; 31-40 cigarettes/day: 21 mg + 7 mg patches; > 40 cigarettes/day: 21 mg + 14 mg patches. For all cases, consider the association with nicotine gum or tablet of 4 mg ad libitum. † consider titration of transdermal nicotine dose (avoid doses above 42 mg/day) or substitute for varenicline. ‡ outpatient return in a maximum of one month, with follow-up for a time not lesser than one month. Reduce dose of NRT according to the guidelines applicable to general situations.