The present study consists of a prospective analysis of 1,054 patients at the Institute of Cardiology of Rio Grande do Sul/University Foundation of Cardiology (Instituto de Cardiologia do Rio Grande do Sul/Fundação Universitária de Cardiologia - lClFUC). Study sample consisted of patients who underwent valve surgery and/or myocardial revascularization at the ICIFUC, and was collected during two periods, between 2002 and 2005 (n = 448 patients) and between 2010 and 2011 (n = 606 patients).

This was an observational prospective cohort study.

Patients aged 18 years or older who underwent valve surgery (mitral and/or aortic valve replacement) and/or myocardial revascularization surgery were included. Patients who underwent additional procedure associated with any of these surgeries were also included.

Patients who underwent emergency procedures, and patients with AF (detected by standard electrocardiography and 24-h ambulatory electrocardiographic monitoring) were excluded.

The following variables were evaluated in the validation cohort: gender, age, presence of mitral valve disease (severe regurgitation and/or severe stenosis), use of beta-blockers in the preoperative period, discontinuation of beta-blocker therapy in the postoperative period, presence of a positive water balance greater than 1,500 mL within 48 hours after surgery, duration of hospitalization and in-hospital mortality.

The diagnosis of POAF was considered an outcome in the perioperative period. POAF was desfined as an episode of arrhythmia, with electrocardiographic tracing with an irregular baseline secondary to disorganized atrial activity, which is referred to as the so-called “f” waves generating variable RR cycles. Episodes of POAF lasting longer than 15 minutes or requiring medical intervention were considered in the study due to the symptomatology or hemodynamic instability. Patients were monitored continuously for 72 hours, and daily electrocardiograms were obtained during hospitalization. In case of cardiovascular symptoms, additional monitoring was performed.

Anesthesia and cardiopulmonary bypass (CPB) were performed according to local standard protocols. After cardiac surgery (immediate postoperative period), the patient remaining for 48 hours or longer in the intensive care unit.

Continuous variables with normal distribution were described as means and standard deviations. The hypothesis of normality was verified by the Kolmogorov-Smirnov test. Categorical (or categorized continuous) variables were described as counts and percentages and compared using the chi-square test or Fisher’s exact test, when necessary. For construction of the risk score, a derivation cohort was collected between 2002 and 2005, and a validation cohort was collected during 2010-2011. Multivariate analysis with backward selection was applied. Statistical significance was set at p < 0.05.

Variables used for analysis were selected based on biological plausibility (association with POAF) and data from literature on POAF. A total of 67 variables were studied in the derivation cohort14 of 448 patients included between 2002 and 2005.

Variables associated with POAF were selected in a multiple logistic regression model with backward selection, and those with p-values close to 0.05 in the model were maintained. Then, b coefficient of the logistic equation was used for construction of a weighted risk score; when transformed into odds ratios, the values were rounded to compose the score.

The preliminary risk score (obtained in the derivation cohort) was applied in the validation cohort, and the following performance statistics were applied: C statistics (area under the receiver operating characteristic [ROC] curve), the chi-square of goodness-of-fit of Hosmer-Lemeshow (HL) and consequently Pearson›s correlation coefficient between the observed events and those predicted by the model.

Values for the area under the ROC curve greater than 0.70 indicate that the model has good discriminatory power. A chi-square of HL with p > 0.05 indicates good calibration of the model. A Pearson correlation coefficient r ≥ 0.7 indicates a strong association between observed and predicted values.

After obtaining adequate performance in the validation, both databases (derivation and validation) were combined for the formulation of the final score. The variables were the same as those previously studied to achieve more accurate statistics for the coefficients. Performance statistics were obtained as described above.

In addition to the final score, a logistic model (formula below) was generated, which allows direct estimates of the probability of outcome occurrence. The use of a mathematical model is considered by some authors to be more appropriate for obtaining event estimates since a complex formula would limit the use among physicians. In individuals considered to be at high risk in the additive model, the use of the logistic model is the most adequate in determining the individual occurrence of the clinical outcome.15

P event = 1 { 1 + exp ( − ([ 30 + 31 x 1 + ... + kak )

Data were processed and analyzed using the Statistical Package for the Social Sciences (SPSS), version 22.0.

This research project was submitted to the Research Ethics Committee of the ICIFUC (registration number 2345902).

In the total sample (n = 1,054), 272 patients had POAF (25.7%). When considering only myocardial revascularization surgeries (63.8%), the POAF rate was 20.3%. In cases of valve intervention alone (23.9%), the frequency of POAF was 34.3% and, in combined surgeries (7.3%), the highest prevalence was observed - 36.6%.

Mean age of the patients was 60.1 ± 12.1 years old, and 26.6% of the patients were 70 years old or older. Most of the patients (65.2%) were men (Table 1).

POAF was associated with longer hospital stay compoared with patients without POAF (median of 10 days vs. 7 days, respectively, p < 0.05) and increased in-hospital mortality (5.5% vs. 1.0%, respectively; p < 0.001). In addition, with a mean follow-up of five years, we observed higher late mortality rate for patients with POAF compared with those without POAF (6.5% vs 1.4%, respectively, p = 0.002).

The multivariate analysis of the predictors in the derivation cohort (n = 448) is described in Table 2. Based on their statistical significance, the predictors selected for the construction of the score included age (≥ 70 years), mitral valve disease, the non-use or discontinuation of beta-blocker therapy and a positive water balance greater than 1500 mL.

Points were assigned to each variable according to the odds ratio obtained (Table 2).

The area under the ROC curve of the obtained model was 0.77 (95% confidence interval [CI] 0.73 to 0.81).

External validation was performed in 606 patients of the validation cohort. The risk model had an accuracy of 0.78 (95% CI 0.73 to 0.82) measured by the area under the ROC curve, thus exhibiting a good discriminatory ability. Good correlation was noted between expected and observed POAF: r = 0.99, with chi-square = 1.73 (p = 0.94) (Hosmer-Lemeshow test).

The model was then edited using a combination of the developed score and data from the derivation and validation cohorts. Using the variables described, multiple logistic regression was performed, resulting in a recalibrated risk score based on the importance of the coefficient β of the logistic equation (Table 3). Variables related to the development of POAF included age (≥ 70 years), mitral valve disease, the non-use or discontinuation of beta-blockers and a positive water balance greater than 1500 mL.

The area under the ROC curve of the obtained model was 0.76 (95% Cl 0.72- 0.79) (Figure 1).

Figure 1

Table 4 and Figure 2 present the risk of POAF according to the score and the classification of this risk (additive score). There was a progressive increase in the proportion of the event, exhibited by an increase in the score: very low risk (score 0) = 0.0%; low risk (score 1 and 2) = 3.9%; intermediate risk (score 3 to 5) = 10.9%; and high risk (score 6 to 8) = 60.0%; p < 0.0001. The logistic equation should be used for risk assessment in the development of POAF individually (Table 3).

Figure 2

In the total sample, 46.8% of the operated patients had low and medium risk. The score predicted POAF in 7.4% of individuals at low risk and 11.7% of those at medium risk; 17% of the total sample was classified as very high risk. To test the calibration of the final score, the observed POAF was compared with that predicted in each of the four classification intervals of the score, resulting in a predicted/observed correlation coefficient of 0.99 with x² = 0.98 (p = 0.98) (Hosmer-Lemeshow test) (Figure 3).

Figure 3