Arq Bras Cardiol abc Arquivos Brasileiros de Cardiologia Arq. Bras. Cardiol. 0066-782X 1678-4170 Sociedade Brasileira de Cardiologia - SBC Kvph8bzDdpB9Qq3DSf3Dgjt S0066-782X2025001100303 00303 10.36660/abc.20250176 Artigo Original Associação da Osmolalidade Plasmática com o Risco de Doença Cardiovascular: Evidências do Banco de Dados do National Health and Nutrition Examination Survey Gao Feng Obtenção de financiamento e Redação do manuscrito 1 Shi Yicheng Obtenção de dados e Análise estatística 1 Chen Yi Análise e interpretação dos dados 1 0009-0002-8358-3792 Lu Shu Concepção e desenho da pesquisa e Revisão crítica do manuscrito quanto ao conteúdo 1 Nanjing University of Traditional Chinese Medicine Wuxi Hospital Cardiovascular Wuxi China Nanjing University of Traditional Chinese Medicine Affiliated Wuxi Hospital – Cardiovascular, Wuxi – China Shu Lu Nanjing University of Traditional Chinese Medicine Affiliated Wuxi Hospital – Wuxi City, Zhongnan West Road,No. 8 Wuxi, 214000 – China E-mail: wxzy001@njucm.edu.cn

Gláucia Maria Moraes de Oliveira

Não há conflito com o presente artigo

 09 01 2026 11 2025 122 11 e20250176 03 04 2025 01 08 2025 20 08 2025 Este é um artigo de acesso aberto distribuído sob os termos da licença de atribuição pelo Creative Commons Resumo Fundamento

A doença cardiovascular (DCV) continua sendo a principal causa de morte no mundo. Diante de seu elevado impacto em saúde pública, ainda são necessários estudos que definam fatores de risco ao mesmo tempo modificáveis e mensuráveis.

 Objetivo

Investigar a associação entre a osmolalidade plasmática (Osm) e DCV — especificamente insuficiência cardíaca, doença coronariana, angina e infarto do miocárdio — utilizando dados do National Health and Nutrition Examination Survey (NHANES).

 Métodos

Foram analisados nove ciclos do NHANES (1999-2018). As associações entre Osm e DCV foram avaliadas por regressão logística ponderada. Estimaram-se razões de chances (OR) e IC 95%, além de análises por subgrupos. Empregaram-se modelos restricted cubic spline (RCS) para investigar possíveis relações não lineares. Adotou-se significância estatística de p < 0,05.

 Resultados

O aumento de Osm associou-se a maiores chances de insuficiência cardíaca, doença coronariana, angina e infarto do miocárdio nos modelos não ajustados e parcialmente ajustados. Após o ajuste completo, as associações permaneceram significativas para todos os desfechos, exceto angina. As análises com RCS indicaram relação em U significativa para insuficiência cardíaca e angina em todos os modelos, com menor risco em torno de 278 mmol/kg. As análises de subgrupos foram consistentes com os resultados principais.

 Conclusão

O aumento de Osm associa-se de forma significativa ao risco de DCV. A Osm pode ser um biomarcador útil para avaliação e manejo do risco cardiovascular na prática clínica, sujeito à confirmação em estudos de coorte prospectivos.

Associação da Osmolalidade Plasmática com o Risco de Doença Cardiovascular: Evidências do Banco de Dados do National Health and Nutrition Examination Survey NHANES: National Health and Nutrition Examination Survey.

Palavras-chave Doenças Cardiovasculares Insuficiência Cardíaca Doença das Coronárias Infarto do Miocárdio Wuxi Municipal Health and Wellness Commission Major Research Project 2024 Z202306 Fontes de financiamento: O presente estudo foi financiado por Wuxi Municipal Health and Wellness Commission Major Research Project 2024 (No. Z202306). Introdução

A doença cardiovascular (DCV) abrange condições que afetam o coração e os vasos sanguíneos, incluindo doença arterial coronariana (DAC), hipertensão e insuficiência cardíaca (IC).1-5 A Organização Mundial da Saúde reconhece a DCV como a principal causa de morte no mundo, responsável por cerca de 17,9 milhões de óbitos por ano,6 com impactos profundos sobre pacientes e famílias e um substancial ônus econômico para a sociedade. Apesar dos avanços nas terapias farmacológicas, intervencionistas e cirúrgicas, persistem limitações importantes — especialmente na estratificação de risco e no tratamento personalizado.5,7-9 Há, portanto, uma necessidade urgente de pesquisas que aprofundem a compreensão dos fatores de risco para DCV.

A osmolalidade plasmática (Osm) regula o equilíbrio hídrico entre os compartimentos intra- e extracelular e entre os espaços intra- e extravascular, sendo amplamente determinada por Na+, K+ e Cl–. A Osm é calculada como: Osm =2×Na++qlicose/18 + nitrogênio ureico no sangue (BUN)2,8, com faixa normal de 275-295 mOsm/kg.10 Medir e monitorar a Osm é essencial na avaliação e no manejo de diversas condições. Alterações de Osm relacionam-se estreitamente à evolução de distúrbios como hiponatremia, hipernatremia, hiperglicemia, osteoartrite, nefropatia e doença pulmonar obstrutiva crônica.11 Evidências emergentes também sugerem relação entre Osm e desfechos cardiovasculares. Em pessoas com diabetes, tanto valores baixos quanto elevados de Osm associaram-se ao aumento da mortalidade por todas as causas, indicando uma associação em U.12 A razão BUN/albumina (BAR) — sendo o BUN um dos principais componentes da Osm — prediz mortalidade em DAC e insuficiência cardíaca congestiva (ICC) graves.13,14 Além disso, a Osm é um importante indicador prognóstico em ICC, correlacionando-se com a mortalidade por todas as causas em 28 dias.15 No entanto, a maior parte dos estudos prévios concentrou-se no prognóstico de pacientes com DCV estabelecida, e ainda são limitadas as avaliações abrangentes da relação entre Osm e risco de DCV na população geral.

Utilizando dados do National Health and Nutrition Examination Survey (NHANES), examinamos a associação entre Osm e o risco de ICC, DAC, angina e infarto do miocárdio, com o objetivo de informar a avaliação clínica de risco.

 Métodos Participantes do estudo

Utilizamos dados do NHANES (https://www.cdc.gov/nchs/nhanes/index.html), um programa nacionalmente representativo que avalia o estado de saúde e nutricional da população dos EUA por meio de amostragem probabilística contínua em múltiplos estágios.16 As entrevistas domiciliares foram seguidas de exames físicos e laboratoriais realizados em Centros de Exames Móveis (CEMs). Todos os protocolos do NHANES foram aprovados pelo National Center for Health Statistics Ethics Review Board, e todos os participantes assinaram termo de consentimento livre e esclarecido. Todas as análises seguiram as diretrizes e regulamentações do NHANES.

Foram analisados os nove ciclos mais recentes (1999-2018), totalizando 91.351 participantes. Indivíduos sem mensuração de Osm ou sem dados do questionário de DCV foram excluídos.

 Exposição e desfechos

A Osm foi medida nos CEMs usando a variável do NHANES LBXSOSSI, com variação de 201 a 323 mmol/kg. Os participantes foram categorizados em quartis (Q1-Q4), sendo Q1 o menor e Q4 o maior valor de Osm.

Os desfechos primários foram as prevalências de ICC, doença arterial coronariana (DAC), angina e infarto do miocárdio, definidos por autorrelato no questionário: MCQ160b (“Alguma vez lhe disseram que teve ICC?”), MCQ160c (“Alguma vez lhe disseram que teve DAC?”), MCQ160d (“Alguma vez lhe disseram que teve angina, angina pectoris?”) e MCQ160e (“Alguma vez lhe disseram que teve ataque cardíaco?”). As respostas “Sim” classificaram o participante como portador da condição; respostas “Não” indicaram ausência da condição.

 Covariáveis

Seguindo Bays et al.,17 ajustamos por: idade; sexo; raça/etnia (mexicano-americano, outro hispânico, branco não hispânico, negro não hispânico, outros); histórico de hipertensão e diabetes; pressão arterial sistólica (PAS) e diastólica (PAD); tabagismo; índice de massa corporal (IMC); triglicerídeos (TG); colesterol de lipoproteína de baixa densidade (LDL-C); BUN; e glicose sanguínea. Características demográficas e históricos de hipertensão, diabetes e tabagismo foram obtidos por questionário. SBP, DBP e BMI derivaram dos exames realizados nos CEMs; TG, LDL, BUN e glicose vieram dos dados laboratoriais dos CEMs. O status tabágico foi categorizado como “Todos os dias”, “Alguns dias” ou “Não fuma”. TG foi dicotomizado em ≤ 1,70 vs. > 1,70 mmol/L, e LDL em ≤ 3,37 vs. > 3,37 mmol/L. Para reduzir viés por dados ausentes, variáveis com ≤ 20% de missing foram imputadas por múltipla imputação; para variáveis com > 20% de missing, criaram-se categorias incluídas como variáveis dummy nos modelos.

 Análise estatística

A normalidade foi avaliada pelo teste de Shapiro-Wilk. Como todas as variáveis contínuas apresentaram distribuição não normal, foram reportadas como medianas e intervalos interquartis e comparadas pelo teste de Mann-Whitney U. Variáveis categóricas são apresentadas como contagens (n) e porcentagens (%) e comparadas pelo teste do qui-quadrado.

A associação entre exposição e desfechos foi estimada por regressão logística ponderada (RLP), com inclusão sequencial de covariáveis. Modelo 1: sem ajustes; Modelo 2: ajustado por idade, sexo e raça; Modelo 3: adicionalmente ajustado por hipertensão, diabetes, tabagismo, PAS, PAD, IMC, TG, LDL, BUN e glicose sanguínea. Para avaliar o desempenho discriminativo do Modelo 3, geramos curvas receiver operating characteristic (ROC) e calculamos a area under the curve (AUC) e o pseudo-R2 de McFadden. Para mitigar multicolinearidade, computamos os fatores de inflação da variância (FIV) e excluímos variáveis com FIV > 5. Modelos RCS avaliaram possíveis relações não lineares. Em seguida, realizaram-se análises de subgrupos, apresentadas como forest plots. As análises foram conduzidas no R versão 4.4.2, com p bicaudal < 0,05 considerado estatisticamente significativo.

 Resultados

No total, 44.475 participantes atenderam aos critérios de elegibilidade (Figura 1). Como todas as variáveis contínuas apresentaram distribuição não normal, utilizou-se o teste de Mann–Whitney U. As características basais (Tabela 1) indicam predomínio de mulheres, participantes brancos não hispânicos e indivíduos ≤ 60 anos. Em comparação com Q1 (menor Osm), Q4 (maior Osm) incluiu mais homens e adultos mais velhos (> 60 anos) e apresentou maiores prevalências de hipertensão, diabetes, ICC, DAC, angina e infarto do miocárdio, além de valores mais altos de IMC, PAS, BUN e glicose.

– Fluxograma de inclusão do estudo. DCV: doença cardiovascular; DAC: doença arterial coronariana; ICC: insuficiência cardíaca congestiva; NHANES: National Health and Nutrition Examination Survey; Osm: osmolalidade plasmática.

– Características basais dos participantes por quartil de Osm
Característica Total Q1 Q2 Q3 Q4 Valor p
N (ponderado) 206.406.933 58.579.788 53.539.459 58.183.076 36.104.609
IMC 28 [24-32] 27 [24-32] 27 [24-32] 28 [24-32] 29 [25-33] <0,001***
Hipertensão, n (%) 15.455 (31) 3.205 (25) 3.112 (26) 4.552 (32) 4.586 (46) <0,001***
PAS, mmHg 120 [112-132] 118 [110-128] 118 [110-130] 122 [112-132] 126 [116-136] <0,001***
PAD, mmHg 71 [64-78] 70 [64-78] 71 [64-78] 72 [64-78] 71 [64-78] 0,028*
Sexo (%) <0,001***
Homens 21.482 (48) 4.562 (37) 5.391 (49) 6.753 (55) 4.776 (54)
Mulheres 22.993 (52) 7.895 (63) 5.440 (51) 5.595 (45) 4.063 (46)
Faixa etária (%) <0,001***
< 40 anos 15.265 (37) 6.213 (49) 4.379 (42) 3.460 (32) 1.213 (19)
40-60 anos 15.300 (40) 4.067 (38) 4.020 (42) 4.502 (41) 2.711 (37)
> 60 anos 13.910 (23) 2.177 (13) 2.432 (16) 4.386 (27) 4.915 (45)
Raça/Etnia (%) <0,001***
Mexicano-americano 7.499 (8,3) 2.238 (8,9) 1.880 (8,7) 2.030 (7,8) 1.351 (7,7)
Outro hispânico 3.753 (5,4) 981 (5,5) 987 (5,8) 1.045 (5,1) 740 (5,2)
Negro não hispânico 9.069 (11) 2.639 (12) 2.195 (10) 2.402 (9.7) 1.833 (10)
Branco não hispânico 19.835 (68) 5.397 (66) 4.711 (68) 5.678 (70) 4.049 (69)
Outros 4.319 (7,1) 1.202 (7,2) 1.058 (7,1) 1.193 (6,9) 866 (7,2)
Status de diabetes (%) <0,001***
Sim 5.403 (9,0) 697 (4,3) 816 (5,8) 1.532 (9,3) 2.358 (21)
Não 38.140 (89) 11.581 (94) 9.827 (93) 10.532 (89) 6.200 (76)
Limítrofe 932 (1,9) 179 (1,3) 188 (1,5) 284 (2,0) 281 (3.2)
Tabagismo (%) <0,001***
Todos os dias 7.531 (17) 2.505 (22) 1.985 (18) 1.968 (15) 1.073 (12)
Ausente 24.275 (54) 6.917 (53) 6.033 (56) 6.648 (54) 4.677 (54)
Não fuma 10.939 (25) 2.491 (21) 2.321 (22) 3.293 (27) 2.834 (31)
Alguns dias 1.730 (3,7) 544 (4,2) 492 (4,2) 439 (3,4) 255 (2,8)
Triglicerídeos, n (%) 0,017*
≤ 1.70 mmol/l 15.482 (35) 4.199 (34) 3.880 (35) 4.308 (35) 3.095 (36)
> 1.70 mmol/l 5.877 (13) 1.618 (13) 1.366 (12) 1.636 (13) 1.257 (14)
Ausente 23.116 (52) 6.640 (54) 5.585 (52) 6.404 (52) 4.487 (50)
LDL-C (%) <0,001***
≤ 3.37 mmol/l 14.486 (33) 3.990 (32) 3.526 (32) 3.913 (32) 3.057 (35)
> 3.37 mmol/l 6.271 (14) 1.643 (13) 1.598 (14) 1.876 (15) 1.154 (14)
Ausente 23.718 (53) 6.824 (55) 5.707 (54) 6.559 (53) 4.628 (51)
BUN, mg/dl 13,0 [10,0-16,0] 11,0 [8,0-13,0] 12,0 [10,0-15,0] 14,0 [11,0-17,0] 16,0 [13,0-21,0] <0,001***
Glicose plasmática, mmol/l 5,05 [4,72, 5,61] 4,88 [4,55, 5,27] 5,00 [4,66, 5,44] 5,16 [4,77, 5,72] 5,50 [5,00, 6,61] <0,001***
ICC, n (%) 1.397 (2,3) 216 (1,4) 200 (1,3) 363 (2,2) 618 (5,3) <0,001***
DAC, n (%) 1.815 (3,4) 275 (1.8) 294 (2,3) 511 (3,6) 735 (7,5) <0,001***
Angina, n (%) 1.218 (2,3) 215 (1.6) 198 (1,5) 355 (2,3) 450 (4,7) <0,001***
Infarto do miocárdio, n (%) 1.856 (3,2) 294 (2.0) 292 (2,1) 555 (3,5) 715 (6,5) <0,001***

Dados apresentados como mediana [Q1, Q3] ou n (%). Quartis de osmolalidade plasmática (mOsm/kg): Q1 201-275; Q2 275-278; Q3 278-282; Q4 282-323. N (ponderado) indica o tamanho amostral ponderado. Limiar de valores de p: *p < 0,05; **p < 0,01; ***p < 0,001. BUN: nitrogênio ureico; DAC: doença arterial coronariana; ICC: insuficiência cardíaca congestiva; IMC: índice de massa corporal; LDL-C: colesterol de lipoproteína de baixa densidade; PAD: pressão arterial diastólica; PAS: pressão arterial sistólica; Osm: osmolalidade plasmática.

As análises por RLP (Tabela 2) mostraram que Osm mais elevada associou-se ao aumento substancial do risco de DCV nos modelos não ajustado, parcialmente ajustado e totalmente ajustado. Quando modelada como variável contínua, Osm mais alta associou-se significativamente a maiores riscos de ICC, DAC e infarto do miocárdio nos três modelos. Para angina, observou-se o mesmo padrão nos modelos não ajustado e parcialmente ajustado, mas a associação deixou de ser significativa após o ajuste completo (Tabela 2).

– Associação entre Osm e desfechos cardiovasculares
Característica Modelo 1 Modelo 2 Modelo 3
OR IC 95% Valor p OR IC 95% Valor p OR IC 95% Valor p
DAC
Osm (contínua) 1,116 1,097-1,135 <0,001 1,059 1,042-1,076 <0,001 1,033 1,017-1,049 <0,001
Osm (categórica) Valor de p para tendência < 0,001 Valor de p para tendência < 0,001 Valor de p para tendência = 0,004
Q1 — — — — — —
Q2 0,946 0,745-1,201 0,645 0,816 0,642-1,037 0,096 0,825 0,645-1,055 0,124
Q3 1,627 1,304-2,030 <0.001 1,024 0,820-1,278 0,832 0,972 0,784-1,205 0,794
Q4 4,045 3,267-5,008 <0.001 1,784 1,446-2,201 <0,001 1,354 1,086-1,689 0,008
ICC
Osm (contínua) 1,108 1,092-1,123 <0,001 1,044 1,031-1,056 <0,001 1,025 1,013-1,037 <0,001
Osm (categórica) Valor de p para tendência < 0,001 Valor de p para tendência < 0,001 Valor de p para tendência < 0,001
Q1 — — — — — —
Q2 1,245 1,009-1,537 0.041 1.029 0.832, 1.273 0.791 1.095 0.884, 1.356 0.404
Q3 1,998 1,617-2,468 <0,001 1,145 0,914-1,434 0.237 1.169 0.929, 1.471 0.181
Q4 4,345 3,590-5,259 <0,001 1,724 1,429-2,081 <0,001 1,459 1,207-1,764 <0,001
Angina
Osm (contínua) 1,080 1,062-1,098 <0,001 1,028 1,011-1,045 0,001 1,011 0,995, 1,027 0,169
Osm (categórica) Valor de p para tendência < 0,001 Valor de p para tendência = 0,002 Valor de p para tendência = 0,103
Q1 — — — — — —
Q2 0,929 0,724-1,192 0,559 0,8 0,627-1,021 0,073 0,844 0,659-1,080 0,176
Q3 1,460 1,124-1,896 0,005 0,938 0,722-1,219 0,629 0,952 0,727-1,247 0,72
Q4 2,979 2,350-3,775 <0,001 1,396 1,104-1,765 0,006 1,188 0,924-1,527 0,177
Infarto do miocárdio
Osm (contínua) 1,093 1,078-1,109 <0,001 1,035 1,021-1,049 <0,001 1,02 1,007-1,034 0,002
Osm (categórica) Valor de p para tendência < 0,001 Valor de p para tendência < 0,001 Valor de p para tendência = 0,006
Q1 — — — — — —
Q2 1,068 0,892-1,279 0,471 0,892 0,740-1,074 0,224 0,952 0,785-1,155 0,618
Q3 1,808 1,480-2,209 <0,001 1,099 0,892-1,355 0,371 1,145 0,931-1,408 0,197
Q4 3,46 2,849-4,201 <0,001 1,495 1,224-1,826 <0,001 1,315 1,072-1,614 0,009

Modelo 1: não ajustado. Modelo 2: ajustado por idade, sexo e raça/etnia. Modelo 3: ajustado por idade, sexo, raça/etnia, histórico de hipertensão, PAS, PAD, diabetes melito, triglicerídeos, LDL-C, BUN, glicose plasmática e IMC. BUN: nitrogênio ureico; DAC: doença arterial coronariana; ICC: insuficiência cardíaca congestiva; IMC: índice de massa corporal; PAD: pressão arterial diastólica; PAS: pressão arterial sistólica; LDL-C: colesterol de lipoproteína de baixa densidade; OR: odds ratio; Osm: osmolalidade plasmática.

A análise ROC do Modelo 3 está na Figura 2. O melhor desempenho ocorreu para DAC (AUC = 0,863). Os valores de pseudo-R2 de McFadden para o Modelo 3 foram 0,204 (ICC), 0,240 (DAC), 0,172 (angina) e 0,195 (infarto do miocárdio). Quando Osm foi analisada categoricamente, Q4 apresentou riscos maiores de ICC, DAC e infarto do que Q1. O risco de angina também foi maior em Q4 nos Modelos 1 e 2, mas não no modelo totalmente ajustado (Tabela 2).

– Análise da curva receiver operating characteristic do Modelo 3. DAC: doença arterial coronariana; ICC: insuficiência cardíaca congestiva.

As análises com RCS (Figura 3) identificaram relações não lineares significativas entre Osm e ICC, DAC, angina e infarto do miocárdio nos modelos não ajustado e parcialmente ajustado (valor p para não linearidade < 0,0001). No modelo totalmente ajustado, as associações não lineares para DAC e infarto deixaram de ser evidentes, enquanto as relações em U para ICC e angina persistiram.

– Resultados da análise restricted cubic spline. A) Associações entre Osm e ICC. B) Associações entre Osm e angina. C) Associações entre Osm e DAC. D) Associações entre Osm e infarto do miocárdio. DAC: doença arterial coronariana; ICC: insuficiência cardíaca congestiva; OR: odds ratio; Osm: osmolalidade plasmática.

As análises de subgrupos por idade, sexo, status de diabetes e IMC (Figura 4) mostraram associações positivas entre Osm e ICC, DAC e infarto do miocárdio entre participantes > 40 anos, com achados consistentes entre subgrupos de sexo, diabetes e obesidade. Para angina, Osm associou-se positivamente entre indivíduos > 60 anos, com ou sem diabetes e com ou sem obesidade. Observou-se interação entre Osm e diabetes (p = 0,031).

– Resultados da análise por subgrupos. A) Infarto do miocárdio; B) Angina; C) DAC; D) ICC. DAC: doença arterial coronariana; DM: diabetes melito; ICC: insuficiência cardíaca congestiva; OR: odds ratio.

Conforme resumido na Figura Central, a regressão logística evidenciou uma associação dose–resposta significativa entre Osm e o risco de DCV (ICC, DAC, angina e infarto do miocárdio), enquanto as análises por RCS mostraram associações em U especificamente para ICC e angina.

 Discussão

A partir dos dados do NHANES, examinamos a associação entre Osm e DCV. Entre 44.475 participantes, predominaram mulheres, indivíduos brancos não hispânicos e pessoas ≤ 60 anos. Valores mais altos de Osm associaram-se a risco significativamente maior de DCV. Após o ajuste completo, a associação com angina deixou de ser significativa. As análises com RCS identificaram uma associação não linear: padrão em U para ICC e angina em todos os modelos, com menor risco em torno de 278 mmol/kg. As análises de subgrupos sustentaram esses achados de forma consistente.

Estudos prévios vincularam Osm ao prognóstico em diferentes populações.18-25 Na diabetes, tanto Osm baixa quanto elevada associaram-se a maior mortalidade por todas as causas, sugerindo relação em U.12 Kaya et al. relataram faixa normal de Osm de 275-295 mOsm/kg e faixa “ótima” de 292-293 mOsm/kg; valores fora desse intervalo associaram-se a piores desfechos,18 alinhando-se ao padrão em U observado para ICC neste estudo. Osm elevada frequentemente reflete desidratação, hipernatremia ou hiperglicemia; variações ≤ 1% podem aumentar a liberação de arginina vasopressina (AVP).26 Através dos receptores V1A, a AVP promove vasoconstrição periférica e estimulação miocárdica; pelos receptores renais V2, induz retenção hídrica,27 contribuindo para ICC. Osm baixa tem sido associada a desregulação neuro-hormonal, descompensação e fadiga,28,29 favorecendo retenção de fluidos e aumento da pré-carga, o que pode deflagrar ou agravar ICC.

Observamos também associação em U entre Osm e angina. Embora atenuada após o ajuste completo, ao nosso conhecimento essa relação não havia sido descrita. Estresse hipotônico aumenta volume celular e proliferação — sobretudo de células musculares lisas vasculares — processos implicados em DCV.30,31 Também influencia a agregação plaquetária e eleva o volume plaquetário médio, marcador de ativação plaquetária.32,33 Por outro lado, sódio elevado pode estimular a liberação endotelial de fator de von Willebrand e promover hipercoagulabilidade.34 Condições associadas à hipernatremia — como desidratação, idade avançada, diabetes, dietas ricas em sal e terapia hiperosmolar — podem predispor à trombose.35

As análises de subgrupos indicaram associações amplamente consistentes entre Osm e desfechos de DCV em categorias de idade, sexo, status de diabetes e IMC. Para angina, as associações positivas foram evidentes em adultos > 60 anos, com ou sem diabetes, e com ou sem obesidade. Detectou-se interação significativa entre Osm e diabetes. Uma explicação possível é que pessoas com diabetes estabelecido recebam manejo mais intensivo de fatores de risco (p. ex., controle glicêmico, pressórico e lipídico), enquanto indivíduos sem diabetes ou com pré-diabetes podem não receber intervenções comparáveis, o que influenciaria o risco de angina.

Este estudo faz um exame preliminar da associação entre Osm e DCV, oferecendo uma nova lente para investigar risco cardiovascular. Osm elevada pode atuar como sinal de alerta precoce, especialmente em grupos de alto risco. Incorporar o monitoramento de Osm na avaliação e no manejo de DCV pode ajudar a melhorar o prognóstico e reduzir o risco, e Osm pode integrar programas de rastreio precoce e estratificação de risco.

Algumas limitações merecem consideração. Primeiro, o delineamento transversal impede inferência causal; estudos prospectivos de coorte são necessários para confirmação. Segundo, apesar do grande tamanho amostral, a estratificação limitada de subpopulações específicas pode restringir a generalização. Terceiro, o NHANES oferece um conjunto limitado de variáveis, e fatores-chave — como ecocardiografia, angiografia coronariana, função renal (taxa de filtração glomerular estimada), uso de diuréticos e estado de hidratação — não estavam disponíveis. Quarto, por se tratar majoritariamente de americanos brancos não hispânicos, os achados podem não se aplicar a outras regiões, como África e Ásia. Potenciais vieses incluem viés de seleção decorrente de participação domiciliar e voluntária; viés de informação, pois Osm foi calculada indiretamente e não medida por osmometria; viés de autorrelato, já que os desfechos — ICC, DAC, angina e infarto do miocárdio — vieram de questionários e não foram adjudicados clinicamente; e possível viés de publicação devido ao destaque dado a achados positivos, incluindo associações que se atenuaram até a não significância após o ajuste completo.

 Conclusão

Osm elevada associa-se de forma significativa ao risco de DCV, sugerindo que Osm pode ser um biomarcador útil para avaliação de risco e manejo clínico. No entanto, esses achados derivam de dados transversais e requerem validação em estudos de coorte prospectivos.

 Referências 1. Katta N, Loethen T, Lavie CJ, Alpert MA. Obesity and Coronary Heart Disease: Epidemiology, Pathology, and Coronary Artery Imaging. Curr Probl Cardiol. 2021;46(3):100655. doi: 10.1016/j.cpcardiol.2020.100655. Katta N Loethen T Lavie CJ Alpert MA Obesity and Coronary Heart Disease: Epidemiology, Pathology, and Coronary Artery Imaging Curr Probl Cardiol 2021 46 3 100655 10.1016/j.cpcardiol.2020.100655 2. Carey RM, Moran AE, Whelton PK. Treatment of Hypertension: A Review. JAMA. 2022;328(18):1849-61. doi: 10.1001/jama.2022.19590. Carey RM Moran AE Whelton PK Treatment of Hypertension: A Review JAMA 2022 328 18 1849 1861 10.1001/jama.2022.19590 3. Roger VL. Epidemiology of Heart Failure: A Contemporary Perspective. Circ Res. 2021;128(10):1421-34. doi: 10.1161/CIRCRESAHA.121.318172. Roger VL Epidemiology of Heart Failure: A Contemporary Perspective Circ Res 2021 128 10 1421 1434 10.1161/CIRCRESAHA.121.318172 4. Townsend N, Kazakiewicz D, Wright FL, Timmis A, Huculeci R, Torbica A, et al. Epidemiology of Cardiovascular Disease in Europe. Nat Rev Cardiol. 2022;19(2):133-43. doi: 10.1038/s41569-021-00607-3. Townsend N Kazakiewicz D Wright FL Timmis A Huculeci R Torbica A et al Epidemiology of Cardiovascular Disease in Europe Nat Rev Cardiol 2022 19 2 133 143 10.1038/s41569-021-00607-3 5. Powell-Wiley TM, Baumer Y, Baah FO, Baez AS, Farmer N, Mahlobo CT, et al. Social Determinants of Cardiovascular Disease. Circ Res. 2022;130(5):782-99. doi: 10.1161/CIRCRESAHA.121.319811. Powell-Wiley TM Baumer Y Baah FO Baez AS Farmer N Mahlobo CT et al Social Determinants of Cardiovascular Disease Circ Res 2022 130 5 782 799 10.1161/CIRCRESAHA.121.319811 6. Sajjad MW, Muzamil F, Sabir M, Ashfaq UA. Regenerative Medicine and Nanotechnology Approaches against Cardiovascular Diseases: Recent Advances and Future Prospective. Curr Stem Cell Res Ther. 2025;20(1):50-71. doi: 10.2174/011574888X263530230921074827. Sajjad MW Muzamil F Sabir M Ashfaq UA Regenerative Medicine and Nanotechnology Approaches against Cardiovascular Diseases: Recent Advances and Future Prospective Curr Stem Cell Res Ther 2025 20 1 50 71 10.2174/011574888X263530230921074827 7. Magnussen C, Ojeda FM, Leong DP, Alegre-Diaz J, Amouyel P, Aviles-Santa L, et al. Global Effect of Modifiable Risk Factors on Cardiovascular Disease and Mortality. N Engl J Med. 2023;389(14):1273-85. doi: 10.1056/NEJMoa2206916. Magnussen C Ojeda FM Leong DP Alegre-Diaz J Amouyel P Aviles-Santa L et al Global Effect of Modifiable Risk Factors on Cardiovascular Disease and Mortality N Engl J Med 2023 389 14 1273 1285 10.1056/NEJMoa2206916 8. Zhao D. Epidemiological Features of Cardiovascular Disease in Asia. JACC Asia. 2021;1(1):1-13. doi: 10.1016/j.jacasi.2021.04.007. Zhao D Epidemiological Features of Cardiovascular Disease in Asia JACC Asia 2021 1 1 1 13 10.1016/j.jacasi.2021.04.007 9. Lindstrom M, DeCleene N, Dorsey H, Fuster V, Johnson CO, LeGrand KE, et al. Global Burden of Cardiovascular Diseases and Risks Collaboration, 1990-2021. J Am Coll Cardiol. 2022;80(25):2372-425. doi: 10.1016/j.jacc.2022.11.001. Lindstrom M DeCleene N Dorsey H Fuster V Johnson CO LeGrand KE et al Global Burden of Cardiovascular Diseases and Risks Collaboration, 1990-2021 J Am Coll Cardiol 2022 80 25 2372 2425 10.1016/j.jacc.2022.11.001 10. Rasouli M. Basic Concepts and Practical Equations on Osmolality: Biochemical Approach. Clin Biochem. 2016;49(12):936-41. doi: 10.1016/j.clinbiochem.2016.06.001. Rasouli M Basic Concepts and Practical Equations on Osmolality: Biochemical Approach Clin Biochem 2016 49 12 936 941 10.1016/j.clinbiochem.2016.06.001 11. Jiang Y, Abudurexiti S, An MM, Cao D, Wei J, Gong P. Risk Factors Associated with 28-Day All-Cause Mortality in Older Severe COVID-19 Patients in Wuhan, China: A Retrospective Observational Study. Sci Rep. 2020;10(1):22369. doi: 10.1038/s41598-020-79508-3. Jiang Y Abudurexiti S An MM Cao D Wei J Gong P Risk Factors Associated with 28-Day All-Cause Mortality in Older Severe COVID-19 Patients in Wuhan, China: A Retrospective Observational Study Sci Rep 2020 10 1 22369 10.1038/s41598-020-79508-3 12. Hu T, Li C, Wang T, Liu H, Geng J, Gong A. Serum Osmolality Was Non-Linearly Associated with the Risk of All-Cause and Cardiovascular Mortality in Patients with Diabetes. BMJ Open Diabetes Res Care. 2023;11(6):e003826. doi: 10.1136/bmjdrc-2023-003826. Hu T Li C Wang T Liu H Geng J Gong A Serum Osmolality Was Non-Linearly Associated with the Risk of All-Cause and Cardiovascular Mortality in Patients with Diabetes BMJ Open Diabetes Res Care 2023 11 6 e003826 10.1136/bmjdrc-2023-003826 13. Zhang L, Xing M, Yu Q, Li Z, Tong Y, Li W. Blood Urea Nitrogen to Serum Albumin Ratio: A Novel Mortality Indicator in Intensive Care Unit Patients with Coronary Heart Disease. Sci Rep. 2024;14(1):7466. doi: 10.1038/s41598-024-58090-y. Zhang L Xing M Yu Q Li Z Tong Y Li W Blood Urea Nitrogen to Serum Albumin Ratio: A Novel Mortality Indicator in Intensive Care Unit Patients with Coronary Heart Disease Sci Rep 2024 14 1 7466 10.1038/s41598-024-58090-y 14. Zhang YY, Xia G, Yu D, Tu F, Liu J. The Association of Blood Urea Nitrogen to Serum Albumin Ratio with Short-Term Outcomes in Chinese Patients with Congestive Heart Failure: A Retrospective Cohort Study. Nutr Metab Cardiovasc Dis. 2024;34(1):55-63. doi: 10.1016/j.numecd.2023.10.011. Zhang YY Xia G Yu D Tu F Liu J The Association of Blood Urea Nitrogen to Serum Albumin Ratio with Short-Term Outcomes in Chinese Patients with Congestive Heart Failure: A Retrospective Cohort Study Nutr Metab Cardiovasc Dis 2024 34 1 55 63 10.1016/j.numecd.2023.10.011 15. Zou Q, Li J, Lin P, Ma J, Wei Z, Tao T, et al. Association between Serum Osmolality and 28-Day All-Cause Mortality in Patients with Heart Failure and Reduced Ejection Fraction: A Retrospective Cohort Study from the MIMIC-IV Database. Front Endocrinol. 2024;15:1397329. doi: 10.3389/fendo.2024.1397329. Zou Q Li J Lin P Ma J Wei Z Tao T et al Association between Serum Osmolality and 28-Day All-Cause Mortality in Patients with Heart Failure and Reduced Ejection Fraction: A Retrospective Cohort Study from the MIMIC-IV Database Front Endocrinol 2024 15 1397329 10.3389/fendo.2024.1397329 16. An S, Ahn C, Jang J, Lee J, Kang D, Lee JK, et al. Comparison of the Prevalence of Cardiometabolic Disorders and Comorbidities in Korea and the United States: Analysis of the National Health and Nutrition Examination Survey. J Korean Med Sci. 2022;37(18):e149. doi: 10.3346/jkms.2022.37.e149. An S Ahn C Jang J Lee J Kang D Lee JK et al Comparison of the Prevalence of Cardiometabolic Disorders and Comorbidities in Korea and the United States: Analysis of the National Health and Nutrition Examination Survey J Korean Med Sci 2022 37 18 e149 10.3346/jkms.2022.37.e149 17. Bays HE, Kulkarni A, German C, Satish P, Iluyomade A, Dudum R, et al. Ten Things to Know About ten Cardiovascular Disease Risk Factors - 2022. Am J Prev Cardiol. 2022;10:100342. doi: 10.1016/j.ajpc.2022.100342. Bays HE Kulkarni A German C Satish P Iluyomade A Dudum R et al Ten Things to Know About ten Cardiovascular Disease Risk Factors - 2022 Am J Prev Cardiol 2022 10 100342 10.1016/j.ajpc.2022.100342 18. Kaya H, Yücel O, Ege MR, Zorlu A, Yücel H, Günes H, et al. Plasma Osmolality Predicts Mortality in Patients with Heart Failure with Reduced Ejection Fraction. Kardiol Pol. 2017;75(4):316-22. doi: 10.5603/KP.a2016.0168. Kaya H Yücel O Ege MR Zorlu A Yücel H Günes H et al Plasma Osmolality Predicts Mortality in Patients with Heart Failure with Reduced Ejection Fraction Kardiol Pol 2017 75 4 316 322 10.5603/KP.a2016.0168 19. Rohla M, Freynhofer MK, Tentzeris I, Farhan S, Wojta J, Huber K, et al. Plasma Osmolality Predicts Clinical Outcome in Patients with Acute Coronary Syndrome Undergoing Percutaneous Coronary Intervention. Eur Heart J Acute Cardiovasc Care. 2014;3(1):84-92. doi: 10.1177/2048872613516018. Rohla M Freynhofer MK Tentzeris I Farhan S Wojta J Huber K et al Plasma Osmolality Predicts Clinical Outcome in Patients with Acute Coronary Syndrome Undergoing Percutaneous Coronary Intervention Eur Heart J Acute Cardiovasc Care 2014 3 1 84 92 10.1177/2048872613516018 20. Charkoudian N, Eisenach JH, Joyner MJ, Roberts SK, Wick DE. Interactions of Plasma Osmolality with Arterial and Central Venous Pressures in Control of Sympathetic Activity and Heart Rate in Humans. Am J Physiol Heart Circ Physiol. 2005;289(6):H2456-60. doi: 10.1152/ajpheart.00601.2005. Charkoudian N Eisenach JH Joyner MJ Roberts SK Wick DE Interactions of Plasma Osmolality with Arterial and Central Venous Pressures in Control of Sympathetic Activity and Heart Rate in Humans Am J Physiol Heart Circ Physiol 2005 289 6 H2456 H2460 10.1152/ajpheart.00601.2005 21. Bahri Z, Andra CA, Mukhtar Z, Raynaldo AH, Sarastri Y, Hasan H. High Admission Plasma Osmolality Value as in-Hospital Major Adverse Cardiovascular Events Predictor in Patient with Acute Myocardial Infarction. Int J Res Sci Manag. 2020;7(11):79-85. Bahri Z Andra CA Mukhtar Z Raynaldo AH Sarastri Y Hasan H High Admission Plasma Osmolality Value as in-Hospital Major Adverse Cardiovascular Events Predictor in Patient with Acute Myocardial Infarction Int J Res Sci Manag 2020 7 11 79 85 22. Zhai G, Wang J, Liu Y, Zhou Y. The Association between Plasma Osmolarity and in-hospital Mortality in Cardiac Intensive Care Unit Patients. Front Cardiovasc Med. 2021;8:692764. doi: 10.3389/fcvm.2021.692764. Zhai G Wang J Liu Y Zhou Y The Association between Plasma Osmolarity and in-hospital Mortality in Cardiac Intensive Care Unit Patients Front Cardiovasc Med 2021 8 692764 10.3389/fcvm.2021.692764 23. Wang H, He Z, Li J, Lin C, Li H, Jin P, et al. Early Plasma Osmolality Levels and Clinical Outcomes in Children Admitted to the Pediatric Intensive Care Unit: A Single-Center Cohort Study. Front Pediatr. 2021;9:745204. doi: 10.3389/fped.2021.745204. Wang H He Z Li J Lin C Li H Jin P et al Early Plasma Osmolality Levels and Clinical Outcomes in Children Admitted to the Pediatric Intensive Care Unit: A Single-Center Cohort Study Front Pediatr 2021 9 745204 10.3389/fped.2021.745204 24. Bhalla A, Sankaralingam S, Dundas R, Swaminathan R, Wolfe CD, Rudd AG. Influence of Raised Plasma Osmolality on Clinical Outcome after Acute Stroke. Stroke. 2000;31(9):2043-8. doi: 10.1161/01.str.31.9.2043. Bhalla A Sankaralingam S Dundas R Swaminathan R Wolfe CD Rudd AG Influence of Raised Plasma Osmolality on Clinical Outcome after Acute Stroke Stroke 2000 31 9 2043 2048 10.1161/01.str.31.9.2043 25. Nakagawa A, Yasumura Y, Yoshida C, Okumura T, Tateishi J, Yoshida J, et al. Prognostic Relevance of Elevated Plasma Osmolality on Admission in Acute Decompensated Heart Failure with Preserved Ejection Fraction: Insights from PURSUIT-HFpEF Registry. BMC Cardiovasc Disord. 2021;21(1):281. doi: 10.1186/s12872-021-02098-z. Nakagawa A Yasumura Y Yoshida C Okumura T Tateishi J Yoshida J et al Prognostic Relevance of Elevated Plasma Osmolality on Admission in Acute Decompensated Heart Failure with Preserved Ejection Fraction: Insights from PURSUIT-HFpEF Registry BMC Cardiovasc Disord 2021 21 1 281 10.1186/s12872-021-02098-z 26. Verbalis JG. Disorders of Body Water Homeostasis. Best Pract Res Clin Endocrinol Metab. 2003;17(4):471-503. doi: 10.1016/s1521-690x(03)00049-6. Verbalis JG Disorders of Body Water Homeostasis Best Pract Res Clin Endocrinol Metab 2003 17 4 471 503 10.1016/s1521-690x(03)00049-6 27. Jard S. Mechanisms of Action of Vasopressin and Vasopressin Antagonists. Kidney Int Suppl. 1988;26:S38-42. Jard S Mechanisms of Action of Vasopressin and Vasopressin Antagonists Kidney Int Suppl 1988 26 S38 S42 28. Hoch CR, Klinedinst NJ, Larimer K, Gottlieb SS. Heart Failure Related Fatigue: An Exploratory Analysis of Serum Osmolality from the National Health and Nutrition Examination Survey. Heart Lung. 2024;68:284-90. doi: 10.1016/j.hrtlng.2024.08.012. Hoch CR Klinedinst NJ Larimer K Gottlieb SS Heart Failure Related Fatigue: An Exploratory Analysis of Serum Osmolality from the National Health and Nutrition Examination Survey Heart Lung 2024 68 284 290 10.1016/j.hrtlng.2024.08.012 29. Lohmeier TE. Neurohypophysial Hormones. Am J Physiol Regul Integr Comp Physiol. 2003;285(4):R715-7. doi: 10.1152/ajpregu.00390.2003. Lohmeier TE Neurohypophysial Hormones Am J Physiol Regul Integr Comp Physiol 2003 285 4 R715 R717 10.1152/ajpregu.00390.2003 30. Zhou JG, Ren JL, Qiu QY, He H, Guan YY. Regulation of Intracellular Cl- Concentration Through Volume-Regulated ClC-3 Chloride Channels in A10 Vascular Smooth Muscle Cells. J Biol Chem. 2005;280(8):7301-8. doi: 10.1074/jbc.M412813200. Zhou JG Ren JL Qiu QY He H Guan YY Regulation of Intracellular Cl- Concentration Through Volume-Regulated ClC-3 Chloride Channels in A10 Vascular Smooth Muscle Cells J Biol Chem 2005 280 8 7301 7308 10.1074/jbc.M412813200 31. Chen BY, Wang SR, Lu FT, Lv XF, Chen Y, Ma MM, et al. SGK1 Mediates Hypotonic Challenge-Induced Proliferation in Basilar Artery Smooth Muscle Cells Via Promoting CREB Signaling Pathway. Eur J Pharmacol. 2021;898:173997. doi: 10.1016/j.ejphar.2021.173997. Chen BY Wang SR Lu FT Lv XF Chen Y Ma MM et al SGK1 Mediates Hypotonic Challenge-Induced Proliferation in Basilar Artery Smooth Muscle Cells Via Promoting CREB Signaling Pathway Eur J Pharmacol 2021 898 173997 10.1016/j.ejphar.2021.173997 32. Park Y, Schoene N, Harris W. Mean Platelet Volume as an Indicator of Platelet Activation: Methodological Issues. Platelets. 2002;13(5-6):301-6. doi: 10.1080/095371002220148332. Park Y Schoene N Harris W Mean Platelet Volume as an Indicator of Platelet Activation: Methodological Issues Platelets 2002 13 5-6 301 306 10.1080/095371002220148332 33. Pantazi D, Stratou A, Dounousi E, Petrou A, Tselepis AD. In Vitro Investigation of Platelet Dysfunction Induced by Osmotic Pressure Variations. Med Princ Pract. 2023;32(4):242-9. doi: 10.1159/000533852. Pantazi D Stratou A Dounousi E Petrou A Tselepis AD In Vitro Investigation of Platelet Dysfunction Induced by Osmotic Pressure Variations Med Princ Pract 2023 32 4 242 249 10.1159/000533852 34. Bihari S, Prakash S, Peake SL, Bailey M, Pilcher D, Bersten A. ICU Mortality is Increased with High Admission Serum Osmolarity in All Patients Other than Those Admitted with Pulmonary Diseases and Hypoxia. Respirology. 2017;22(6):1165-70. doi: 10.1111/resp.13055. Bihari S Prakash S Peake SL Bailey M Pilcher D Bersten A ICU Mortality is Increased with High Admission Serum Osmolarity in All Patients Other than Those Admitted with Pulmonary Diseases and Hypoxia Respirology 2017 22 6 1165 1170 10.1111/resp.13055 35. Dmitrieva NI, Burg MB. Secretion of von Willebrand Factor by Endothelial Cells Links Sodium to Hypercoagulability and Thrombosis. Proc Natl Acad Sci U S A. 2014;111(17):6485-90. doi: 10.1073/pnas.1404809111. Dmitrieva NI Burg MB Secretion of von Willebrand Factor by Endothelial Cells Links Sodium to Hypercoagulability and Thrombosis Proc Natl Acad Sci U S A 2014 111 17 6485 6490 10.1073/pnas.1404809111

Não há vinculação deste estudo a programas de pós-graduação.

Este artigo não contém estudos com humanos ou animais realizados por nenhum dos autores.

Os autores não utilizaram ferramentas de inteligência artificial no desenvolvimento deste trabalho.

Este estudo utiliza dados de acesso controlado do banco de dados NHANCE. De acordo com as políticas de acesso e os acordos de uso de dados do banco de dados, os dados brutos originais não podem ser redistribuídos publicamente pelos usuários. Pesquisadores interessados em acessar os dados devem enviar uma solicitação diretamente ao comitê diretor do banco de dados NHANCE para obter permissões de acesso. Os dados processados derivados das análises relatadas neste manuscrito estão disponíveis com o autor correspondente mediante solicitação razoável.

O presente estudo foi financiado por Wuxi Municipal Health and Wellness Commission Major Research Project 2024 (No. Z202306).

 10.36660/abc.20250176i Original Article Association of Plasma Osmolality with Cardiovascular Disease Risk: Evidence from the National Health and Nutrition Examination Survey Database Gao Feng Obtaining financing and Writing of the manuscript 1 Shi Yicheng Acquisition of data and Statistical analysis 1 Chen Yi Analysis and interpretation of the data 1 0009-0002-8358-3792 Lu Shu Conception and design of the research and Critical revision of the manuscript for content 1 China Nanjing University of Traditional Chinese Medicine Affiliated Wuxi Hospital – Cardiovascular, Wuxi – China Shu Lu Nanjing University of Traditional Chinese Medicine Affiliated Wuxi Hospital – Wuxi City, Zhongnan West Road,No. 8 Wuxi, 214000 – China E-mail: wxzy001@njucm.edu.cn

No potential conflict of interest relevant to this article was reported.

Gláucia Maria Moraes de Oliveira

 e20250176i Abstract Background

Cardiovascular disease (CVD) remains the world’s leading cause of death. Because of its substantial public health burden, further research is needed to define risk factors that are both modifiable and measurable.

 Objective

We examined the association between plasma osmolality (Osm) and CVDs — namely, heart failure, coronary heart disease, angina, and myocardial infarction — using data from the National Health and Nutrition Examination Survey (NHANES).

 Methods

We analyzed nine NHANES cycles (1999-2018). Associations between Osm and CVD were assessed with weighted logistic regression. Odds ratios (ORs) and 95% CIs were estimated, and subgroup analyses were conducted. Restricted cubic spline (RCS) models were used to evaluate potential nonlinear relationships. Statistical significance was set at p < 0.05.

 Results

Higher Osm was associated with increased odds of heart failure, coronary heart disease, angina, and myocardial infarction in unadjusted and partially adjusted models. After full adjustment, the associations remained significant for all outcomes except angina. RCS analyses indicated a significant U-shaped relationship for heart failure and angina across all models, with the lowest risk around 278 mmol/kg. Subgroup analyses were consistent with the primary results.

 Conclusion

Increased Osm is significantly associated with CVD risk. Osm may serve as a useful biomarker for assessment and management of CVD risk in clinical practice, pending confirmation in prospective cohort studies.

Association of Plasma Osmolality with Cardiovascular Disease Risk: Evidence from the National Health and Nutrition Examination Survey Database NHANES: National Health and Nutrition Examination Survey.

Keywords Cardiovascular Diseases Heart Failure Coronary Disease Myocardial Infarction Introduction

Cardiovascular disease (CVD) encompasses conditions affecting the heart and blood vessels, including coronary heart disease (CHD), hypertension, and heart failure (HF).1-5 The World Health Organization identifies CVD as the leading cause of death globally, accounting for approximately 17.9 million deaths each year,6 with profound impacts on patients and families and a substantial economic burden on society. Despite advances in pharmacological, interventional, and surgical therapies, substantial limitations remain — particularly in risk stratification and personalized treatment.5,7-9 Consequently, there is an urgent need for research to deepen our understanding of CVD risk factors.

Plasma osmolality (Osm) regulates water balance across intra- vs. extracellular and intra- vs. extravascular compartments and is largely determined by Na+, K+, and Cl–. Osm =2×Na++glucose/18 + blood urea nitrogen (BUN)2,8 , with a normal range of 275-295 mOsm/kg.10 Measuring and monitoring Osm is essential in the assessment and management of numerous conditions. Alterations in Osm are closely linked to the course of disorders such as hyponatremia, hypernatremia, hyperglycemia, osteoarthritis, nephropathy, and chronic obstructive pulmonary.11 Emerging evidence also suggests a relationship between Osm and cardiovascular outcomes. In diabetic patients, both low and high Osm were associated with increased all-cause mortality, indicating a U-shaped association.12 The blood urea nitrogen-to-albumin ratio (BAR) — with BUN as a major Osm component — predicts mortality in severe CHD and CHF.13,14 Osm is furthermore a key prognostic indicator in CHF, correlating with 28-day all-cause mortality.15 However, most prior work has emphasized prognosis among patients with established CVD, and comprehensive evaluations of Osm and CVD risk in the overall population remain limited.

By using data from the National Health and Nutrition Examination Survey (NHANES), we examined the association between Osm and the risk of CHF, CHD, angina, and myocardial infarction, with the goal of informing clinical risk assessment.

 Methods Study participants

We used data from the NHANES (https://www.cdc.gov/nchs/nhanes/index.html), a nationally representative program that assesses the health and nutritional status of the U.S. population through continuous, multistage probability sampling.16 Home interviews were followed by physical and laboratory examinations conducted at Mobile Examination Centers (MECs). All NHANES protocols were approved by the National Center for Health Statistics Ethics Review Board, and written informed consent was obtained from all participants. All analyses adhered to NHANES guidelines and regulations.

We analyzed the most recent nine NHANES cycles (1999-2018), comprising 91,351 participants. Individuals with missing Osm measurements or missing CVD questionnaire data were excluded.

 Exposure and outcomes

Osm was measured at MECs using the NHANES variable LBXSOSSI and ranged from 201 to 323 mmol/kg. Participants were categorized into quartiles (Q1-Q4), with Q1 denoting the lowest and Q4 the highest Osm values.

The primary outcomes were the prevalences of CHF, CHD, angina, and myocardial infarction, defined by self-reported questionnaire items: MCQ160b (“Ever told had CHF”), MCQ160c (“Ever told you had CHD”), MCQ160d (“Ever told you had angina/angina pectoris”), and MCQ160e (“Ever told you had heart attack”). Respondents answering “Yes” were classified as having the condition; those answering “No” were classified as not having the condition.

 Covariates

Following Bays et al.,17 we adjusted for age; sex; race/ethnicity (Mexican American, other Hispanic, non-Hispanic White, non-Hispanic Black, other); histories of hypertension and diabetes; systolic (SBP) and diastolic blood pressure (DBP); smoking status; body mass index (BMI); triglycerides (TG); low-density lipoprotein cholesterol (LDL-C); BUN; and blood glucose. Demographic characteristics and histories of hypertension, diabetes, and smoking were obtained by questionnaire. SBP, DBP, and BMI were derived from examination data collected at MECs, and TG, LDL-C, BUN, and glucose were obtained from MEC laboratory data. Smoking status was categorized as “Every day,” “Someday,” or “Not at all.” TG was dichotomized as ≤ 1.70 vs > 1.70 mmol/L, and LDL-C as ≤ 3.37 vs > 3.37 mmol/L. To limit bias from missingness, variables with ≤ 20% missing data underwent multiple imputation; for variables with > 20% missingness, categories were created and included as dummy variables in the models.

 Statistical analysis

Normality was assessed with the Shapiro-Wilk test. Because all continuous variables were nonnormally distributed, they are reported as medians with interquartile ranges and compared using the Mann-Whitney U test. Categorical variables are presented as counts (n) and percentages (%) and were compared using the chi-square test.

Weighted logistic regression (WLR) was used to estimate associations between exposure and outcomes, with covariates added sequentially. Model 1 was unadjusted; Model 2 adjusted for age, sex, and race; Model 3 additionally adjusted for hypertension, diabetes, smoking, SBP, DBP, BMI, TG, LDL-C, BUN, and blood glucose. To evaluate the discriminative performance of Model 3, we generated receiver operating characteristic (ROC) curves and calculated the area under the curve (AUC) and McFadden’s pseudo-R2. To mitigate multicollinearity, we computed variance inflation factors (VIFs) and excluded variables with VIF > 5. Restricted cubic spline (RCS) models assessed potential nonlinear relationships. Subgroup analyses were then performed and displayed as forest plots. Analyses were conducted in R version 4.4.2, with two-sided p < 0.05 considered statistically significant.

 Results

A total of 44,475 participants met the eligibility criteria (Figure 1). Because all continuous variables were nonnormally distributed, the Mann-Whitney U test was used. Baseline characteristics are shown in Table 1 and indicate a predominance of females, non-Hispanic White participants, and individuals ≤ 60 years. Compared with Q1 (lowest Osm), Q4 (highest Osm) included more males and older adults (> 60 years) and showed higher prevalences of hypertension, diabetes, CHF, CHD, angina, and heart attack, as well as higher BMI, SBP, BUN, and glucose levels.

– Study enrollment flow diagram. CVD: cardiovascular disease; CHD: coronary heart disease; CHF: congestive heart failure; NHANES: National Health and Nutrition Examination Survey; Osm: plasma osmolality.

– Baseline characteristics of participants by Osm quartile
Caracteristic Total Q1 Q2 Q3 Q4 p-value
N (weighted) 206,406,933 58,579,788 53,539,459 58,183,076 36,104,609
BMI 28 [24, 32] 27 [24, 32] 27 [24, 32] 28 [24, 32] 29 [25, 33] <0.001***
Hypertension, n (%) 15,455 (31) 3,205 (25) 3,112 (26) 4,552 (32) 4,586 (46) <0.001***
SBP, mmHg 120 [112, 132] 118 [110, 128] 118 [110, 130] 122 [112, 132] 126 [116, 136] <0.001***
DBP, mmHg 71 [64, 78] 70 [64, 78] 71 [64, 78] 72 [64, 78] 71 [64, 78] 0.028*
Sex (%) <0.001***
Male 21,482 (48) 4,562 (37) 5,391 (49) 6,753 (55) 4,776 (54)
Female 22,993 (52) 7,895 (63) 5,440 (51) 5,595 (45) 4,063 (46)
Age group (%) <0.001***
< 40 years 15,265 (37) 6,213 (49) 4,379 (42) 3,460 (32) 1,213 (19)
40-60 years 15,300 (40) 4,067 (38) 4,020 (42) 4,502 (41) 2,711 (37)
> 60 years 13,910 (23) 2,177 (13) 2,432 (16) 4,386 (27) 4,915 (45)
Race/Ethnicity (%) <0.001***
Mexican American 7,499 (8.3) 2,238 (8.9) 1,880 (8.7) 2,030 (7.8) 1,351 (7.7)
Other Hispanic 3,753 (5.4) 981 (5.5) 987 (5.8) 1,045 (5.1) 740 (5.2)
Non-Hispanic Black 9,069 (11) 2,639 (12) 2,195 (10) 2,402 (9.7) 1,833 (10)
Non-Hispanic White 19,835 (68) 5,397 (66) 4,711 (68) 5,678 (70) 4,049 (69)
Other 4,319 (7.1) 1,202 (7.2) 1,058 (7.1) 1,193 (6.9) 866 (7.2)
Diabetes status (%) <0.001***
Yes 5,403 (9.0) 697 (4.3) 816 (5.8) 1,532 (9.3) 2,358 (21)
No 38,140 (89) 11,581 (94) 9,827 (93) 10,532 (89) 6,200 (76)
Borderline 932 (1.9) 179 (1.3) 188 (1.5) 284 (2.0) 281 (3.2)
Smoking status (%) <0.001***
Every day 7,531 (17) 2,505 (22) 1,985 (18) 1,968 (15) 1,073 (12)
Missing 24,275 (54) 6,917 (53) 6,033 (56) 6,648 (54) 4,677 (54)
Not at all 10,939 (25) 2,491 (21) 2,321 (22) 3,293 (27) 2,834 (31)
Some days 1,730 (3.7) 544 (4.2) 492 (4.2) 439 (3.4) 255 (2.8)
Triglycerides, n (%) 0.017*
≤ 1.70 mmol/L 15,482 (35) 4,199 (34) 3,880 (35) 4,308 (35) 3,095 (36)
> 1.70 mmol/L 5,877 (13) 1,618 (13) 1,366 (12) 1,636 (13) 1,257 (14)
Missing 23,116 (52) 6,640 (54) 5,585 (52) 6,404 (52) 4,487 (50)
LDL-C (%) <0.001***
≤ 3.37 mmol/L 14,486 (33) 3,990 (32) 3,526 (32) 3,913 (32) 3,057 (35)
> 3.37 mmol/L 6,271 (14) 1,643 (13) 1,598 (14) 1,876 (15) 1,154 (14)
Missing 23,718 (53) 6,824 (55) 5,707 (54) 6,559 (53) 4,628 (51)
BUN, mg/dL 13.0 [10.0, 16.0] 11.0 [8.0, 13.0] 12.0 [10.0, 15.0] 14.0 [11.0, 17.0] 16.0 [13.0, 21.0] <0.001***
Plasma glucose, mmol/L 5.05 [4.72, 5.61] 4.88 [4.55, 5.27] 5.00 [4.66, 5.44] 5.16 [4.77, 5.72] 5.50 [5.00, 6.61] <0.001***
CHF, n (%) 1,397 (2.3) 216 (1.4) 200 (1.3) 363 (2.2) 618 (5.3) <0.001***
CHD, n (%) 1,815 (3.4) 275 (1.8) 294 (2.3) 511 (3.6) 735 (7.5) <0.001***
Angina, n (%) 1,218 (2.3) 215 (1.6) 198 (1.5) 355 (2.3) 450 (4.7) <0.001***
Myocardial infarction, n (%) 1,856 (3.2) 294 (2.0) 292 (2.1) 555 (3.5) 715 (6.5) <0.001***

Data are presented as median [Q1, Q3] or n (%). Quartiles of plasma osmolality (mOsm/kg): Q1 201-275; Q2 275-278; Q3 278-282; Q4 282-323. N (weighted) indicates weighted sample size. P-value thresholds: *p < 0.05; **p < 0.01; ***p < 0.001. BMI: body mass index; BUN: blood urea nitrogen; CHD: coronary heart disease; CHF: congestive heart failure; DBP: diastolic blood pressure; LDL-C: low-density lipoprotein cholesterol; Osm: plasma osmolality; SBP: systolic blood pressure.

WLR analyses (Table 2) showed that higher Osm was associated with substantially increased CVD risk in unadjusted, partially adjusted, and fully adjusted models. When modeled continuously, higher Osm was significantly associated with greater risks of CHF, CHD, and myocardial infarction across all three models. Angina showed the same pattern in unadjusted and partially adjusted models, but the association was no longer significant after full adjustment (Table 2).

– Association between Osm and cardiovascular outcomes
Caracteristic Model 1 Model 2 Model 3
OR 95% CI p-value OR 95% CI p-value OR 95% CI p-value
CHF
Plasma osmolality (continuous) 1.116 1.097, 1.135 <0.001 1.059 1.042, 1.076 <0.001 1.033 1.017, 1.049 <0.001
Plasma osmolality (categorical) p-value for trend < 0.001 p-value for trend < 0.001 p-value for trend = 0.004
Q1 — — — — — —
Q2 0.946 0.745, 1.201 0.645 0.816 0.642, 1.037 0.096 0.825 0.645, 1.055 0.124
Q3 1.627 1.304, 2.030 <0.001 1.024 0.820, 1.278 0.832 0.972 0.784, 1.205 0.794
Q4 4.045 3.267, 5.008 <0.001 1.784 1.446, 2.201 <0.001 1.354 1.086, 1.689 0.008
CHD
Osm (continuous) 1.108 1.092, 1.123 <0.001 1.044 1.031, 1.056 <0.001 1.025 1.013, 1.037 <0.001
Osm (categorical) p-value for trend < 0.001 p-value for trend < 0.001 p-value for trend < 0.001
Q1 — — — — — —
Q2 1.245 1.009, 1.537 0.041 1.029 0.832, 1.273 0.791 1.095 0.884, 1.356 0.404
Q3 1.998 1.617, 2.468 <0.001 1.145 0.914, 1.434 0.237 1.169 0.929, 1.471 0.181
Q4 4.345 3.590, 5.259 <0.001 1.724 1.429, 2.081 <0.001 1.459 1.207, 1.764 <0.001
Angina
Osm (continuous) 1.08 1.062, 1.098 <0.001 1.028 1.011, 1.045 0.001 1.011 0.995, 1.027 0.169
Osm (categorical) p-value for trend < 0.001 p-value for trend = 0.002 p-value for trend = 0.103
Q1 — — — — — —
Q2 0.929 0.724, 1.192 0.559 0.8 0.627, 1.021 0.073 0.844 0.659, 1.080 0.176
Q3 1.46 1.124, 1.896 0.005 0.938 0.722, 1.219 0.629 0.952 0.727, 1.247 0.72
Q4 2.979 2.350, 3.775 <0.001 1.396 1.104, 1.765 0.006 1.188 0.924, 1.527 0.177
Myocardial infarction
Osm (continuous) 1.093 1.078, 1.109 <0.001 1.035 1.021, 1.049 <0.001 1.02 1.007, 1.034 0.002
Osm (categorical) p-value for trend < 0.001 p-value for trend < 0.001 p-value for trend = 0.006
Q1 — — — — — —
Q2 1.068 0.892, 1.279 0.471 0.892 0.740, 1.074 0.224 0.952 0.785, 1.155 0.618
Q3 1.808 1.480, 2.209 <0.001 1.099 0.892, 1.355 0.371 1.145 0.931, 1.408 0.197
Q4 3.46 2.849, 4.201 <0.001 1.495 1.224, 1.826 <0.001 1.315 1.072, 1.614 0.009

Model 1: unadjusted. Model 2: adjusted for age, sex, and race/ethnicity. Model 3: adjusted for age, sex, race/ethnicity, hypertension history, SBP, DBP, diabetes mellitus, triglycerides, LDL-C, BUN, plasma glucose, and BMI. BMI: body mass index; BUN: blood urea nitrogen; CHD: coronary heart disease; CHF: congestive heart failure; DBP: diastolic blood pressure; LDL-C: low-density lipoprotein cholesterol; OR: odds ratio; Osm: plasma osmolality; SBP: systolic blood pressure.

ROC analysis for Model 3 is shown in Figure 2. Model 3 performed best for CHD (AUC = 0.863). McFadden’s pseudo-R2 values for Model 3 were 0.204 for CHF, 0.240 for CHD, 0.172 for angina, and 0.195 for myocardial infarction. When Osm was analyzed categorically, Q4 had higher risks of CHF, CHD, and myocardial infarction than Q1. The risk of angina was also higher in Q4 for Models 1 and 2, but not in the fully adjusted model (Table 2).

– Receiver operating characteristic curve analysis of Model 3. CHF: congestive heart failure; CHD: coronary heart disease.

RCS analyses (Figure 3) identified significant nonlinear relationships between Osm and CHF, CHD, angina, and myocardial infarction in unadjusted and partially adjusted models (p for nonlinearity < 0.0001). In the fully adjusted model, the nonlinear associations for CHD and myocardial infarction were no longer evident, whereas the nonlinear relationships for CHF and angina persisted.

– Restricted cubic spline analysis results. A) Associations between OSM and CHF. B) Associations between osmolality and angina. C) Associations between osmolality and CHD. D) Associations between osmolality and myocardial infarction. CHF: congestive heart failure, CHD: coronary heart disease; OR: odds ratio; Osm: plasma osmolality.

Subgroup analyses by age, sex, diabetes status, and BMI are presented in Figure 4. Among participants > 40 years, Osm was positively associated with incident CHF, CHD, and myocardial infarction, with consistent findings across sex, diabetes, and obesity subgroups. For angina, Osm was positively associated among individuals > 60 years, those with or without diabetes, and those with or without obesity. An interaction between Osm and diabetes was observed (p = 0.031).

– Results of subgroup analysis. A) MI; B) Angina; C) CHD. D) CHF. CHF: congestive heart failure; CHD: coronary heart disease; DM: diabetes mellitus; OR: odds ratio.

As summarized in Central Illustration, logistic regression demonstrated a significant dose-response association between Osm and CVD risk (CHF, CHD, angina, and myocardial infarction), whereas RCS analyses showed significant U-shaped associations specifically for CHF and angina.

Discussion

By using NHANES data, we examined the association between Osm and CVD. Among 44,475 participants, women, non-Hispanic White individuals, and those ≤ 60 years predominated. Higher Osm was associated with a significantly greater CVD risk. After full adjustment, the association with angina was no longer significant. RCS analyses identified a nonlinear association: a U-shaped pattern for CHF and angina across all models, with the lowest risk around 278 mmol/kg. Subgroup analyses consistently supported these findings.

Prior studies have linked Osm to prognosis across diverse patient groups.18-25 In diabetes, both low and high Osm have been associated with higher all-cause mortality, indicating a U-shaped relationship.12 Kaya et al. reported a normal Osm range of 275-295 mOsm/kg and an optimal range of 292-293 mOsm/kg, with values outside this range associated with worse outcomes,18 which aligns with the U-shaped pattern observed for CHF in our study. Increased Osm often reflects dehydration, hypernatremia, or hyperglycemia; even a ≤ 1% change can increase arginine vasopressin (AVP) release.26 Via V1A receptors, AVP promotes peripheral vasoconstriction and myocardial stimulation, and via renal V2 receptors it induces water retention,27 thereby contributing to CHF. Low Osm has been linked to neurohormonal dysregulation, decompensation, and fatigue,28,29 which may promote fluid retention and increased preload, potentially triggering or worsening CHF.

We also observed a U-shaped association between Osm and angina. Although this association attenuated after full adjustment, to our knowledge it has not been previously described. Hypotonic stress increases cell volume and proliferation — particularly of vascular smooth-muscle cells — processes implicated in CVD.30,31 It also influences platelet aggregation and increases mean platelet volume, a marker of platelet activation.32,33 Conversely, high sodium levels can stimulate endothelial release of von Willebrand factor and promote hypercoagulability.34 Conditions associated with hypernatremia — including dehydration, older age, diabetes, high-salt diets, and hyperosmolar therapy — may predispose to thrombosis.35

Subgroup analyses indicated broadly consistent associations between Osm and CVD outcomes across age, sex, diabetes status, and BMI categories. For angina, positive associations were evident among adults > 60 years and among individuals with or without diabetes or obesity. A significant interaction between Osm and diabetes was detected. One possible explanation is that people with established diabetes may receive more intensive risk-factor management (e.g., glycemic, blood-pressure, and lipid control), whereas those without diabetes or with prediabetes may not receive comparable interventions, which could influence angina risk.

This study offers a preliminary examination of the association between Osm and CVD, providing a new lens for investigating CVD risk. Increased Osm may serve as an early warning signal, particularly in high-risk groups. Incorporating Osm monitoring into CVD assessment and management could help improve prognosis and reduce risk, and Osm could be incorporated into early screening and risk-stratification programs.

Several limitations merit consideration. First, the cross-sectional design precludes causal inference; prospective cohort studies are needed for confirmation. Second, although the sample was large, limited stratification of specific subpopulations may restrict generalizability. Third, NHANES provides a constrained set of variables, and key factors — such as echocardiography, coronary angiography, renal function (estimated glomerular filtration rate), diuretic use, and hydration status — were unavailable. Fourth, because the population was predominantly non-Hispanic White Americans, the findings may not generalize to other regions, such as Africa and Asia. Potential biases include selection bias from household, voluntary participation; information bias, as Osm was calculated indirectly rather than measured by osmometry; self-report bias, because outcomes — namely, CHF, CHD, angina, myocardial infarction — were questionnaire-based rather than clinically adjudicated; and possible publication bias due to emphasis on positive findings, including associations that attenuated to nonsignificance after full adjustment.

 Conclusion

Increased Osm is significantly associated with CVD risk, suggesting that Osm may be a useful biomarker for risk assessment and clinical management. These findings, however, derive from cross-sectional data and require validation in prospective cohort studies.

This study is not associated with any thesis or dissertation work.

This article does not contain any studies with human participants or animals performed by any of the authors.

The authors did not use any artificial intelligence tools in the development of this work.

This study utilizes controlled-access data from the NHANCE database. In accordance with the database’s access policies and data use agreements, the original raw data cannot be publicly redistributed by users. Researchers interested in accessing the data must submit an application directly to the NHANCE database steering committee to obtain access permissions. The processed data derived from the analyses reported in this manuscript are available from the corresponding author upon reasonable request.

This study was partially funded by Wuxi Municipal Health and Wellness Commission Major Research Project 2024 (No. Z202306).