Arq Bras Cardiol abc Arquivos Brasileiros de Cardiologia Arq. Bras. Cardiol. 0066-782X 1678-4170 Sociedade Brasileira de Cardiologia - SBC n56Yv5jmTrfGcrnrzkCMfTG S0066-782X2024000900303 00303 10.36660/abc.20230813 Artigo Original Duração do Sono e Risco de Aterosclerose: Um Estudo Mendeliano de Randomização Xu Xiaozhuo Concepção e desenho da pesquisa Redação do manuscrito 1 Huang Yilin Obtenção de dados e Análise e interpretação dos dados Revisão crítica do manuscrito quanto ao conteúdo 1 Liu Jing Obtenção de dados e Análise e interpretação dos dados 1 0009-0006-8542-8509 Han Xu Concepção e desenho da pesquisa Revisão crítica do manuscrito quanto ao conteúdo 1 Department of Geriatrics, Affiliated Hospital of Nanjing University of Chinese Medicine Nanjing China Department of Geriatrics, Affiliated Hospital of Nanjing University of Chinese Medicine, Jiangsu Province Hospital of Chinese Medicine, Nanjing – China Xu Han • Department of Geriatrics – Affiliated Hospital of Nanjing University of Chinese Medicine, Jiangsu Province Hospital of Chinese Medicine – No.155 Hanzhong Road, Qinhuai District, Nanjing 210029 – China E-mail: xhan_njgeriatrics@outlook.com

Não há conflito com o presente artigo.

Editor responsável pela revisão: Marcio Bittencourt

 06 09 2024 2024 121 9 e20240813 30 11 2023 07 05 2024 12 06 2024 This is an open-access article distributed under the terms of the Creative Commons Attribution License Resumo Fundamento:

A associação entre a duração do sono e a aterosclerose foi relatada em muitos estudos observacionais. No entanto, pouco se sabe sobre a sua importância como fator de risco para aterosclerose ou como consequência negativa da aterosclerose.

 Objetivo:

Este estudo teve como objetivo avaliar a associação causal entre a duração do sono e o risco de aterosclerose usando estatísticas resumidas de estudos de associação genômica ampla (GWAS) disponíveis publicamente.

 Métodos:

Empregamos um método de randomização mendeliana (RM) de duas amostras com 2 coortes do MRC-IEU (n = 460.099) e do UK Biobank (n = 361.194) para investigar a associação causal entre a duração do sono e o risco de aterosclerose. Três métodos, incluindo a técnica de variância inversa ponderada (IVW), escore de perfil ajustado robusto (RAPS) e abordagem de mediana simples e ponderada, foram usados para obter resultados confiáveis, e uma razão de chances com intervalo de confiança (IC) de 95% foi calculada. P<0,05 foi considerado diferença estatística. Além disso, foram utilizadas análises de regressão: MR-Egger regression, Radial MR, MR-PRESSO e leave-one-out para avaliar os possíveis efeitos de pleiotropia.

 Resultados:

Não foi encontrada associação causal entre duração do sono e aterosclerose [OR (IC95%): 0,90 (0,98-1,00), p = 0,186]. As análises Leave-one-out, MR-Egger, e MR-PRESSO não conseguiram detectar pleiotropia horizontal.

 Conclusões:

Esta análise de RM não indicou nenhuma associação causal entre a duração do sono geneticamente prevista e a aterosclerose nas populações europeias.

 Palavras-chave: Duração do Sono Aterosclerose Análise da Randomização Mendeliana Natural Science Foundation of Jiangsu Province BK20181505 Nanjing University of Chinese Medicine ZYX03KF032 Science and Technology Development of Traditional Chinese Medicine of Jiangsu Province 2020ZX08 Geriatric Clinical Technology Application Research Project of Jiangsu Provincial Health Comission LR202202 Jiangsu Province Posgratuate Practice Innovation Program SJCX22 0766 SJCX21 0670 O presente estudo foi financiado por Natural Science Foundation of Jiangsu Province (No. BK20181505), Phase III Scientific Research Project of Traditional Chinese Medicine Superior Discipline in Nanjing University of Chinese Medicine (No. ZYX03KF032), the Special Plan for the Science and Technology Development of Traditional Chinese Medicine of Jiangsu Province (No. 2020ZX08), the Leader Scientific Research Project of the Geriatric Clinical Technology Application Research Project of Jiangsu Provincial Health Comission (No LR202202), the Jiangsu Province Posgratuate Practice Innovation Program (No SJCX22 0766 and SJCX21 0670). Introdução

A aterosclerose é uma doença multifatorial e a principal causa de eventos cardiovasculares e cerebrovasculares.1 A aterosclerose é uma característica multifatorial complexa com uma etiologia genética enigmática. Sendo uma doença crónica que ameaça gravemente a saúde humana, tem despertado grande atenção, especialmente a aterosclerose coronária. Assistimos a uma "transição epidemiológica".2 O aumento do saneamento e o tratamento de infecções agudas reduziram a prevalência de doenças infecciosas nos países em desenvolvimento, e mais indivíduos sofrem agora de doenças crónicas, como a aterosclerose.3 A aterosclerose pode levar a uma variedade de doenças cardiovasculares (DCV), que têm sido reconhecidas como uma das principais causas de morbidade e mortalidade.4 É urgente interpretar o seu mecanismo, avançar na sua gestão e desenvolver perspectivas para mitigar o seu impacto.

O sono é um processo fisiológico complexo produzido pelo cérebro, que desempenha um papel muito importante na regulação das funções fisiológicas de vários sistemas do corpo. Com a contínua extensão do horário de trabalho na sociedade moderna, a forma como as pessoas trabalham e os hábitos de sono das pessoas também estão em constante mudança, e a redução do tempo de sono está a tornar-se um problema grave.5 Na verdade, descobriu-se que a duração curta e longa do sono está associada ao cálcio da artéria coronária6-8 e à espessura médio-intimal da carótida (CIMT),9,10 que são indicadores de aterosclerose em grandes artérias que alimentam o coração e o cérebro. Além disso, alguns estudos descobriram que um tempo de sono muito longo ou muito curto ainda aumentará a incidência de eventos cardiovasculares após o controle de fatores mistos, como obesidade, hipertensão e diabetes.11 Vários estudos observacionais também mencionaram a relação entre a duração do sono e a aterosclerose subclínica das artérias coronárias ou carótidas.12,13 No entanto, ainda não está claro se dormir pouco ou dormir demais contribui para a ocorrência de aterosclerose.

Até onde sabemos, a associação causal entre a duração do sono e a aterosclerose não foi avaliada. A randomização mendeliana (RM) é um método para verificar a causalidade, evitando confusão residual e superando a causalidade reversa em um cenário retrospectivo, que pode revelar estimativas causais de fatores de risco em doenças complexas usando variantes genéticas como variáveis instrumentais.14,15 Aqui, conduzimos um estudo de RM para avaliar a associação causal entre a duração do sono e o risco de aterosclerose.

 Métodos Design de estudo

Supondo que a estimativa causal dos estudos de RM seja credível. Três suposições cruciais precisam ser atendidas: 1) Deve haver uma forte associação entre as variáveis instrumentais genéticas (IVs) selecionadas e a exposição.16 2) A escolha dos IVs genéticos não influencia o resultado sem considerar a exposição (ou seja, a pleiotropia horizontal é inexistente).17 3) Os IVs genéticos selecionados não estão associados aos possíveis confundidores. A Figura Central fornece uma visão geral. Como a pesquisa foi baseada em conjuntos de dados acessíveis ao público e em estudos publicados anteriormente, a aprovação ética e o consentimento dos participantes não foram necessários para o estudo.

 Fontes de dados

As IVs para a duração do sono foram baseadas numa meta-análise de um estudo de associação genómica ampla (GWAS) de 460.099 pessoas de ascendência europeia. A duração habitual do sono autorreferida foi a principal exposição do presente estudo. Foi obtido a partir de questionários touchscreen na avaliação inicial. A duração do sono foi avaliada de acordo com uma pergunta padronizada: "Quantas horas você dorme a cada 24 horas?". Os participantes que responderam "Não sei" e "Prefiro não responder", e aqueles que forneceram durações de sono implausíveis (< 4 horas ou > 11 horas por dia) foram excluídos para minimizar a duração do sono implausível e potencial confusão por problemas de saúde. Uma descrição completa do desenho do estudo, dos participantes e dos métodos de controle de qualidade (CQ) foi descrita em detalhes anteriormente.18 O UK Biobank recebeu aprovação ética do Comitê de Ética em Pesquisa (a referência REC para o UK Biobank é 11/NW/0382).

A aterosclerose foi identificada com base nas 8ª e 10ª edições da Classificação Internacional de Doenças (CID). Os dados sobre aterosclerose foram coletados de participantes do United Kingdom Biobank (GWAS ID: ukb-d-I9_CORATHER, disponível em https://gwas.mrcieu.ac.uk/datasets/ukb-d-I9_CORATHER/). Este conjunto de dados incluiu 361.194 pessoas de ascendência europeia (um total de 14.334 casos e 346.860 controles) e incluiu 13.586.589 polimorfismos de nucleotídeo único (SNPs). Introduzimos a regressão de pontuação LD ajustada por covariável (cov-LDSC), um método para estimar com precisão a herdabilidade genética (h2g) e seu enriquecimento em populações homogêneas e misturadas com estatísticas resumidas e estimativas de LD na amostra. A divulgação completa dos dados continha a coorte de amostras genotipadas com sucesso (n=488.377). 49.979 indivíduos foram genotipados usando o arranjo UK BiLEVE e 438.398 usando o arranjo axioma UK Biobank. Um total de 9.851.867 SNPs de duração do sono em 460.099 indivíduos foram extraídos do MRC-IEU (GWAS ID: ukb-b-4424, disponível em https://gwas.mrcieu.ac.uk/datasets/ukb-b-4424/). CQ pré-imputação, faseamento e imputação foram conduzidos pelo estudo anterior.19

 A seleção das variáveis instrumentais relevantes

SNPs foram considerados IVs para este estudo.16 Os seguintes critérios foram satisfeitos por cada SNP solicitado: 1. Houve uma correlação substancial com a quantidade de exposição com base na relevância do genoma como um todo; 2. Sem desequilíbrio de ligação (LD) (r2 pareado = 0,001, tamanho da janela = 10.000kb); 3. Não contém estruturas palindrômicas. Um total de 65 SNPs foram encontrados após considerar as três suposições e critérios acima. Não conseguimos encontrar os SNPs apropriados no GWAS de aterosclerose, portanto, para obter estimativas precisas, empregamos SNPs proxy que tinham LD substancial (r2> 0,8) para substituir os SNPs escolhidos, o que nos permitiu obter resultados mais precisos. A regressão de primeiro estágio, ou estatística F, foi utilizada para avaliar a força dos instrumentos e foi calculada pela seguinte equação: F= (R2/k)/ ([1-R2]/[n-k-1]), onde R2 é a proporção da variabilidade da duração do sono contabilizada pelo SNP, k é o número de instrumentos utilizados no modelo e n é o tamanho da amostra.20 Para limitar a influência de um possível viés IV fraco, esperava-se que uma estatística F superior a 10 tivesse força suficiente para o estudo principal.21 O fluxograma para seleção de IVs está representado na Figura 1.

Fluxograma de seleção das variáveis instrumentais. Análise estatística

A abordagem ponderada pela variância inversa (IVW) foi utilizada como principal método para determinar se havia correlação entre a duração do sono e a aterosclerose.22 Se o p do teste Q de Cochran fosse superior a 0,05, optou-se por utilizar um modelo com efeitos fixos; em todos os outros casos, utilizamos um modelo com efeitos aleatórios.23 Se os IVs selecionados não apresentassem pleiotropia direcional (e o p para o intercepto Mr-Egger fosse maior que 0,05), a técnica IVW era considerada a mais confiável.24

Escolhemos a abordagem Mr-Egger para avaliar os possíveis impactos da pleiotropia nas análises de sensibilidade. O termo de interceptação da regressão Mr-Egger, que estimou o efeito causal como a inclinação da regressão ponderada das relações IVs-resultado na relação IVs-exposição, refletiu o efeito pleiotrópico médio.25,26 Para determinar se havia pleiotropia, também utilizamos as técnicas de teste de outlier de mediana básica, mediana ponderada, Radial MR e MR-PRESSO (Mendelian Randomization Pleiotropy Residual Sum and Outlier).26 Se mais de cinquenta por cento dos SNPs estudados fossem IVs eficazes, então a mediana ponderada oferecerá as estimativas mais confiáveis do impacto causal. Além da detecção de pleiotropia, o MR-PRESSO também pode reavaliar estimativas de efeitos e eliminar SNPs discrepantes.26 Para avaliar o impacto dos dados periféricos, foi entretanto realizada uma análise de exclusão. Investigamos ainda a pleiotropia de cada SNP escolhido usando o banco de dados PhenoScanner V2 (//www. pen scanner. medschl. cam. ac. uk/) no nível de significância estatística GWAS (p <5×10-8) para excluir o impacto de outras variáveis.27

Salvo indicação em contrário, todos os testes foram bilaterais e as diferenças foram consideradas estatisticamente significativas (p <0,05). Os pacotes Two Sample MR (V 0.5.6), Radial MR e MR-PRESSO (V 1.0)24 do software R foram usados para todas as análises estatísticas (4.0.5).

 Resultados

Mais informações sobre os SNPs escolhidos são fornecidas na Tabela Suplementar S1-S2. Três SNPs no total (duração do sono: rs1611719, rs17732997 e rs2186122) foram eliminados da pesquisa de RM porque eram palíndromos. No final, 62 SNPs, incluindo 1 SNP proxy, foram escolhidos como IVs (todos p < 5×10-8, r2=0,001).

Estimativas de RM

Os resultados do teste Q de Cochran para o tempo de sono mostraram heterogeneidade mínima (p = 0,108). De acordo com os achados do método IVW, houve pouca evidência de ligação entre a duração do sono e o risco de aterosclerose (OR (IC 95%), 0,992 (0,979-1,004); p = 0,186) (Tabela Suplementar S3).

 Análises de sensibilidade

Os resultados da mediana simples e da mediana ponderada foram comparáveis aos da abordagem IVW. Enquanto isso, a pleiotropia horizontal não foi detectada pela regressão MR-Egger (interceptação p = 0,071 para duração do sono) (Tabela Suplementar S3). Embora o MR Radial sugerisse a existência de outliers (Figura 2), o MR-PRESSO mostrou que os outliers não afetaram os resultados do estudo (Tabela 1). Da mesma forma, a pleiotropia não foi detectada pelo RAPS (Tabela 2) e pelo banco de dados PhenoScanner V2. Quando a pleiotropia horizontal apresentou p > 0,05, o método IVW (efeitos fixos) (Tabela 2) foi utilizado para avaliar os dados. Para a duração do sono, as Figuras Suplementares S1-S4 apresentam gráficos de floresta, gráficos de dispersão, gráficos de funil e gráficos de exclusão de RM.

Uma visão geral do valor discrepante das estimativas radiais de RM. Estimativas MR-PRESSO entre duração do sono e aterosclerose
Característica Estimativa bruta Estimativa corrigida de outlier
N OR IC 95% p N OR IC 95% p
Duração do sono 62 0,991 0,979-1,003 0,163 62 0,991 0,979-1,003 0,163

SNP: polimorfismo de nucleotídeo único; MR-PRESSO: soma residual de pleiotropia de randomização mendeliana e teste de outlier. OR: razão de chances; IC95%: intervalo de confiança de 95%.

 Estimativas de RAPS e IVW (efeitos fixos) entre duração do sono e aterosclerose
Característica RAPS IVW (efeitos fixos)
N OR IC 95% p N OR IC 95% p
Duração do sono 62 0,992 0,980-1,004 0,193 62 0,992 0,980-1,004 0,135

SNP: polimorfismo de nucleotídeo único; RAPS: escore de perfil ajustado robusto; IVW (efeitos fixos), variância inversa ponderada (efeitos fixos). OR: razão de chances; IC: intervalo de confiança.

 Análises de viés e poder

O viés dos instrumentos genéticos foi de 0,000 para a duração do sono. A estatística F dos SNPs selecionados foi de 15.671, o que se esperava ter força suficiente para o estudo principal (Tabela Suplementar S3). A estimativa derivada da técnica de razão foi próxima da razão de chances condicional sob certas condições particulares e se aproximou de uma razão de chances média populacional.28,29 A consistência do estimador sob o valor nulo não foi afetada pela estimativa do odds ratio utilizada. Realizamos os cálculos de potência e o valor do erro Tipo 1 para a duração do sono foi de 0,05. Para o valor do poder estatístico do sono, a duração foi de 95%. De acordo com o tamanho da amostra utilizado na meta-análise de aterosclerose GWAS, houve poder> 80% para identificar a relação entre a quantidade de sono e o risco de aterosclerose para tamanho de efeito (OR) de 0,992 (Tabela Suplementar S4). Em recentes investigações adicionais de RM, todos os IVs para a duração do sono geneticamente prevista foram autorizados e utilizados.30,31 Além disso, nenhum deles teve qualquer influência na hipertensão arterial ou níveis elevados de lipoproteína de baixa densidade (Tabela Suplementar S3).

 Discussão

No presente estudo, tentamos explorar a associação causal entre duração do sono e aterosclerose usando um método de RM. As nossas descobertas não mostraram nenhuma evidência de que a duração do sono geneticamente prevista esteja ligada ao risco de aterosclerose nas populações europeias. Além disso, estudos de sensibilidade mostraram que os resultados eram geralmente confiáveis.

Os escores de cálcio arterial coronariano (CACS), CIMT e velocidade da onda de pulso braquial-tornozelo (baPWV) foram os principais indicadores substitutos de aterosclerose e preditores de eventos cardiovasculares.6,32 Alguns estudos exploraram o efeito da duração do sono na incidência de aterosclerose, analisando a relação entre a duração do sono e CACS, CIMT e baPWV. Um estudo recente com 1.968 homens saudáveis com idades entre 40 e 60 anos indicou que o aumento ou diminuição da duração do sono estava associado a um aumento na incidência de aterosclerose coronariana e avaliou o efeito da duração do sono na incidência de arteriosclerose subclínica medindo a CACS, encontrando pessoas que dormiam por 7 horas teve a menor incidência de aterosclerose coronariana subclínica.33 Para pessoas com fatores de risco para aterosclerose, a duração do sono também foi significativamente correlacionada com a incidência de aterosclerose. Da mesma forma, a CIMT foi mais baixa quando o tempo de sono foi de 7 a 8 horas, e o aumento ou diminuição do tempo de sono levará a um aumento da CIMT.12

Estudos anteriores não mostraram relação entre a duração do sono e marcadores de dano vascular e aterosclerose.9,34-38 Uma pesquisa transversal com 1.093 homens japoneses relatou que a duração do sono autorreferida não estava associada ao aumento da CAC ou da CIMT.33 Souza et al.35 também não encontraram associações independentes da duração objetiva do sono com a CIMT. Nenhuma evidência demonstrou que a associação entre sintomas de insônia e pontuação CAC> 0 diferiu de acordo com o status objetivo de curta duração do sono.36 Além disso, um estudo sobre o Estudo Multiétnico de Aterosclerose (MESA) mostrou que a apneia obstrutiva do sono grave não estava associada a alta carga de CAC ou ITB anormal.37 Os investigadores do MESA não relataram associações entre durações de sono curtas (<6 horas) e longas (>8 horas) e CAC.38 Estes foram consistentes com nossas descobertas.

A ligação clínica e fisiologicamente significativa entre o sono prolongado e o risco de DCV em adultos não é apoiada por dados experimentais suficientes. Nossa hipótese, com base nas informações atuais, é que o mecanismo subjacente era de natureza metabólica e operava por via inflamatória. Especificamente, o sono prolongado pode levar a níveis baixos de HDL,39,40 hiperglicemia, hipertrigliceridemia41 e resistência à insulina,42 os quais podem levar à disfunção endotelial vascular e à inflamação subclínica, promovendo ainda mais a aterosclerose.43,44 Questões sociais, de estilo de vida e comportamentais relacionadas, como o abuso de drogas, a inatividade física ou a falta de acesso a refeições nutritivas, podem exacerbar este ambiente pró-aterogénico.45,46 Independentemente da real relação de causa e efeito, apoiamos o exame da duração do sono nas avaliações clínicas, uma vez que a duração curta ou longa do sono pode indicar risco de doenças crônicas. As DCV e a diabetes são doenças potencialmente fatais que prevalecem na nossa sociedade e podem levar à doença precoce e à morte, pelo que é importante investigar a relação entre o sono e as doenças crónicas ao longo do tempo. Isto inclui encontrar a melhor estratégia de prevenção para alertar contra a aterosclerose, que pode estimular DCV e outras doenças. Este é um passo importante para uma população mais saudável, tanto a nível nacional como global.

De acordo com o nosso conhecimento, este estudo foi a primeira investigação de RM a examinar a associação causal entre a duração do sono e o risco de aterosclerose utilizando conjuntos de dados GWAS disponíveis. Além disso, para esta investigação de RM de duas amostras, escolhemos pessoas da Europa para diminuir o preconceito demográfico. A atual pesquisa de RM também apresentou várias deficiências. Primeiro, como utilizamos dados genéticos acessíveis ao público para a nossa investigação, não fomos capazes de fazer análises estratificadas ou levar em consideração fatores adicionais. Em segundo lugar, os SNPs instrumentais escolhidos como IVs explicaram apenas parcialmente (0,001% -0,01%) a variação na duração do sono. Isto pode resultar de um baixo poder estatístico para identificar relações fracas. Terceiro, numa análise de RM de duas amostras, qualquer viés devido a instrumentos fracos foi na direção do nulo. O viés na direção do nulo foi menos sério do que o viés na direção da associação observacional, pois é conservador e não levará a taxas de erro tipo 1 inflacionadas e a resultados falso-positivos. Havia de fato uma possibilidade de sobreposição entre as duas amostras.28 Em última análise, uma vez que o nosso conjunto de dados era composto por pessoas de herança europeia, as nossas conclusões podem não se aplicar a outros grupos fora da Europa.

 Conclusões

No presente estudo, a duração do sono geneticamente prevista entre as populações europeias não estava causalmente ligada ao risco de aterosclerose. Mais estudos são necessários para investigar a associação causal entre aterosclerose e duração do sono.

O presente estudo foi financiado por Natural Science Foundation of Jiangsu Province (No. BK20181505), Phase III Scientific Research Project of Traditional Chinese Medicine Superior Discipline in Nanjing University of Chinese Medicine (No. ZYX03KF032), the Special Plan for the Science and Technology Development of Traditional Chinese Medicine of Jiangsu Province (No. 2020ZX08), the Leader Scientific Research Project of the Geriatric Clinical Technology Application Research Project of Jiangsu Provincial Health Comission (No LR202202), the Jiangsu Province Posgratuate Practice Innovation Program (No SJCX22 0766 and SJCX21 0670).

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

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

 Agradecimentos

Agradecemos imensamente ao MRC-IEU e ao UK Biobank por fornecerem dados genéticos sobre o estudo de RM.

 *Material suplementar

Para informação adicional da Tabela Suplementar, por favor, clique aqui.

Para informação adicional da Figura Suplementar, por favor, clique aqui.

 Referências Gao W Sun Y Cai M Zhao Y Cao W Liu Z Copper Sulfide Nanoparticles as a Photothermal Switch for TRPV1 Signaling to Attenuate Atherosclerosis Nat Commun 2018 9 1 231 231 10.1038/s41467-017-02657-z 1 Gao W, Sun Y, Cai M, Zhao Y, Cao W, Liu Z, et al. Copper Sulfide Nanoparticles as a Photothermal Switch for TRPV1 Signaling to Attenuate Atherosclerosis. Nat Commun. 2018;9(1):231. doi: 10.1038/s41467-017-02657-z. Dai H Much AA Maor E Asher E Younis A Xu Y Global, Regional, and National Burden of Ischaemic Heart Disease and its Attributable Risk Factors, 1990-2017: Results from the Global Burden of Disease Study 2017 Eur Heart J Qual Care Clin Outcomes 2022 8 1 50 60 10.1093/ehjqcco/qcaa076 2 Dai H, Much AA, Maor E, Asher E, Younis A, Xu Y, et al. Global, Regional, and National Burden of Ischaemic Heart Disease and its Attributable Risk Factors, 1990-2017: Results from the Global Burden of Disease Study 2017. Eur Heart J Qual Care Clin Outcomes. 2022;8(1):50-60. doi: 10.1093/ehjqcco/qcaa076. Libby P Buring JE Badimon L Hansson GK Deanfield J Bittencourt MS Atherosclerosis Nat Rev Dis Primers 2019 5 1 56 56 10.1038/s41572-019-0106-z 3 Libby P, Buring JE, Badimon L, Hansson GK, Deanfield J, Bittencourt MS, et al. Atherosclerosis. Nat Rev Dis Primers. 2019;5(1):56. doi: 10.1038/s41572-019-0106-z. Mensah GA Roth GA Fuster V The Global Burden of Cardiovascular Diseases and Risk Factors: 2020 and Beyond J Am Coll Cardiol 2019 74 20 2529 32 10.1016/j.jacc.2019.10.009 4 Mensah GA, Roth GA, Fuster V. The Global Burden of Cardiovascular Diseases and Risk Factors: 2020 and Beyond. J Am Coll Cardiol. 2019;74(20):2529-32. doi: 10.1016/j.jacc.2019.10.009. Cappuccio FP Cooper D D’Elia L Strazzullo P Miller MA Sleep Duration Predicts Cardiovascular Outcomes: A Systematic Review and Meta-analysis of Prospective Studies Eur Heart J 2011 32 12 1484 92 10.1093/eurheartj/ehr007 5 Cappuccio FP, Cooper D, D’Elia L, Strazzullo P, Miller MA. Sleep Duration Predicts Cardiovascular Outcomes: A Systematic Review and Meta-analysis of Prospective Studies. Eur Heart J. 2011;32(12):1484-92. doi: 10.1093/eurheartj/ehr007. Kim CW Chang Y Zhao D Cainzos-Achirica M Ryu S Jung HS Sleep Duration, Sleep Quality, and Markers of Subclinical Arterial Disease in Healthy Men and Women Arterioscler Thromb Vasc Biol 2015 35 10 2238 45 10.1161/ATVBAHA.115.306110 6 Kim CW, Chang Y, Zhao D, Cainzos-Achirica M, Ryu S, Jung HS, et al. Sleep Duration, Sleep Quality, and Markers of Subclinical Arterial Disease in Healthy Men and Women. Arterioscler Thromb Vasc Biol. 2015;35(10):2238-45. doi: 10.1161/ATVBAHA.115.306110. King CR Knutson KL Rathouz PJ Sidney S Liu K Lauderdale DS Short Sleep Duration and Incident Coronary Artery Calcification JAMA 2008 300 24 2859 66 10.1001/jama.2008.867 7 King CR, Knutson KL, Rathouz PJ, Sidney S, Liu K, Lauderdale DS. Short Sleep Duration and Incident Coronary Artery Calcification. JAMA. 2008;300(24):2859-66. doi: 10.1001/jama.2008.867. Matthews KA Strollo PJ Jr Hall M Mezick EJ Kamarck TW Owens JF Associations of Framingham Risk Score Profile and Coronary Artery Calcification with Sleep Characteristics in Middle-aged Men and Women: Pittsburgh SleepSCORE Study Sleep 2011 34 6 711 6 10.5665/SLEEP.1032 8 Matthews KA, Strollo PJ Jr, Hall M, Mezick EJ, Kamarck TW, Owens JF, et al. Associations of Framingham Risk Score Profile and Coronary Artery Calcification with Sleep Characteristics in Middle-aged Men and Women: Pittsburgh SleepSCORE Study. Sleep. 2011;34(6):711-6. doi: 10.5665/SLEEP.1032. Ma CC Burchfiel CM Charles LE Dorn JM Andrew ME Gu JK Associations of Objectively Measured and Self-reported Sleep Duration with Carotid Artery Intima Media Thickness Among Police Officers Am J Ind Med 2013 56 11 1341 51 10.1002/ajim.22236 9 Ma CC, Burchfiel CM, Charles LE, Dorn JM, Andrew ME, Gu JK, et al. Associations of Objectively Measured and Self-reported Sleep Duration with Carotid Artery Intima Media Thickness Among Police Officers. Am J Ind Med. 2013;56(11):1341-51. doi: 10.1002/ajim.22236. Sands MR Lauderdale DS Liu K Knutson KL Matthews KA Eaton CB Short Sleep Duration is Associated with Carotid Intima-media Thickness Among Men in the Coronary Artery Risk Development in Young Adults (CARDIA) Study Stroke 2012 43 11 2858 64 10.1161/STROKEAHA.112.660332 10 Sands MR, Lauderdale DS, Liu K, Knutson KL, Matthews KA, Eaton CB, et al. Short Sleep Duration is Associated with Carotid Intima-media Thickness Among Men in the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Stroke. 2012;43(11):2858-64. doi: 10.1161/STROKEAHA.112.660332. Kuehn BM Sleep Duration Linked to Cardiovascular Disease Circulation 2019 139 21 2483 4 10.1161/CIRCULATIONAHA.119.041278 11 Kuehn BM. Sleep Duration Linked to Cardiovascular Disease. Circulation. 2019;139(21):2483-4. doi: 10.1161/CIRCULATIONAHA.119.041278. Wolff B Völzke H Schwahn C Robinson D Kessler C John U Relation of Self-reported Sleep Duration with Carotid Intima-media Thickness in a General Population Sample Atherosclerosis 2008 196 2 727 32 10.1016/j.atherosclerosis.2006.12.023 12 Wolff B, Völzke H, Schwahn C, Robinson D, Kessler C, John U. Relation of Self-reported Sleep Duration with Carotid Intima-media Thickness in a General Population Sample. Atherosclerosis. 2008;196(2):727-32. doi: 10.1016/j.atherosclerosis.2006.12.023. Abe T Aoki T Yata S Okada M Sleep Duration is Significantly Associated with Carotid Artery Atherosclerosis Incidence in a Japanese Population Atherosclerosis 2011 217 2 509 13 10.1016/j.atherosclerosis.2011.02.029 13 Abe T, Aoki T, Yata S, Okada M. Sleep Duration is Significantly Associated with Carotid Artery Atherosclerosis Incidence in a Japanese Population. Atherosclerosis. 2011;217(2):509-13. doi: 10.1016/j.atherosclerosis.2011.02.029. Davies NM Holmes MV Smith GD Reading Mendelian Randomisation Studies: a Guide, Glossary, and Checklist for Clinicians BMJ 2018 362 k601 k601 10.1136/bmj.k601 14 Davies NM, Holmes MV, Smith GD. Reading Mendelian Randomisation Studies: a Guide, Glossary, and Checklist for Clinicians. BMJ. 2018;362:k601. doi: 10.1136/bmj.k601. Carreras A Zhang SX Peris E Qiao Z Gileles-Hillel A Li RC Chronic Sleep Fragmentation Induces Endothelial Dysfunction and Structural Vascular Changes in Mice Sleep 2014 37 11 1817 24 10.5665/sleep.4178 15 Carreras A, Zhang SX, Peris E, Qiao Z, Gileles-Hillel A, Li RC, et al. Chronic Sleep Fragmentation Induces Endothelial Dysfunction and Structural Vascular Changes in Mice. Sleep. 2014;37(11):1817-24. doi: 10.5665/sleep.4178. Lawlor DA Harbord RM Sterne JA Timpson N Smith GD Mendelian Randomization: Using Genes as Instruments for Making Causal Inferences in Epidemiology Stat Med 2008 27 8 1133 63 10.1002/sim.3034 16 Lawlor DA, Harbord RM, Sterne JA, Timpson N, Smith GD. Mendelian Randomization: Using Genes as Instruments for Making Causal Inferences in Epidemiology. Stat Med. 2008;27(8):1133-63. doi: 10.1002/sim.3034. Bowden J Smith GD Haycock PC Burgess S Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator Genet Epidemiol 2016 40 4 304 14 10.1002/gepi.21965 17 Bowden J, Smith GD, Haycock PC, Burgess S. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol. 2016;40(4):304-14. doi: 10.1002/gepi.21965. Collins R What Makes UK Biobank Special? Lancet 2012 379 9822 1173 4 10.1016/S0140-6736(12)60404-8 18 Collins R. What Makes UK Biobank Special? Lancet. 2012;379(9822):1173-4. doi: 10.1016/S0140-6736(12)60404-8. Bycroft C Freeman C Petkova D Band G Elliott LT Sharp K The UK Biobank Resource with Deep Phenotyping and Genomic Data Nature 2018 562 7726 203 9 10.1038/s41586-018-0579-z 19 Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank Resource with Deep Phenotyping and Genomic Data. Nature. 2018;562(7726):203-9. doi: 10.1038/s41586-018-0579-z. Palmer TM Lawlor DA Harbord RM Sheehan NA Tobias JH Timpson NJ Using Multiple Genetic Variants as Instrumental Variables for Modifiable Risk Factors Stat Methods Med Res 2012 21 3 223 42 10.1177/0962280210394459 20 Palmer TM, Lawlor DA, Harbord RM, Sheehan NA, Tobias JH, Timpson NJ, et al. Using Multiple Genetic Variants as Instrumental Variables for Modifiable Risk Factors. Stat Methods Med Res. 2012;21(3):223-42. doi: 10.1177/0962280210394459. Pierce BL Ahsan H Vanderweele TJ Power and Instrument Strength Requirements for Mendelian Randomization Studies Using Multiple Genetic Variants Int J Epidemiol 2011 40 3 740 52 10.1093/ije/dyq151 21 Pierce BL, Ahsan H, Vanderweele TJ. Power and Instrument Strength Requirements for Mendelian Randomization Studies Using Multiple Genetic Variants. Int J Epidemiol. 2011;40(3):740-52. doi: 10.1093/ije/dyq151. Burgess S Smith GD Davies NM Dudbridge F Gill D Glymour MM Guidelines for Performing Mendelian Randomization Investigations: Update for Summer 2023 Wellcome Open Res 2023 4 186 186 10.12688/wellcomeopenres.15555.3 22 Burgess S, Smith GD, Davies NM, Dudbridge F, Gill D, Glymour MM, et al. Guidelines for Performing Mendelian Randomization Investigations: Update for Summer 2023. Wellcome Open Res. 2023;4:186. doi: 10.12688/wellcomeopenres.15555.3. Higgins JP Thompson SG Deeks JJ Altman DG Measuring Inconsistency in Meta-analyses BMJ 2003 327 7414 557 60 10.1136/bmj.327.7414.557 23 Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring Inconsistency in Meta-analyses. BMJ. 2003;327(7414):557-60. doi: 10.1136/bmj.327.7414.557. Holmes MV Ala-Korpela M Smith GD Mendelian Randomization in Cardiometabolic Disease: Challenges in Evaluating Causality Nat Rev Cardiol 2017 14 10 577 90 10.1038/nrcardio.2017.78 24 Holmes MV, Ala-Korpela M, Smith GD. Mendelian Randomization in Cardiometabolic Disease: Challenges in Evaluating Causality. Nat Rev Cardiol. 2017;14(10):577-90. doi: 10.1038/nrcardio.2017.78. Bowden J Smith GD Burgess S Mendelian Randomization with Invalid Instruments: Effect Estimation and Bias Detection Through Egger Regression Int J Epidemiol 2015 44 2 512 25 10.1093/ije/dyv080 25 Bowden J, Smith GD, Burgess S. Mendelian Randomization with Invalid Instruments: Effect Estimation and Bias Detection Through Egger Regression. Int J Epidemiol. 2015;44(2):512-25. doi: 10.1093/ije/dyv080. Burgess S Bowden J Fall T Ingelsson E Thompson SG Sensitivity Analyses for Robust Causal Inference from Mendelian Randomization Analyses with Multiple Genetic Variants Epidemiology 2017 28 1 30 42 10.1097/EDE.0000000000000559 26 Burgess S, Bowden J, Fall T, Ingelsson E, Thompson SG. Sensitivity Analyses for Robust Causal Inference from Mendelian Randomization Analyses with Multiple Genetic Variants. Epidemiology. 2017;28(1):30-42. doi: 10.1097/EDE.0000000000000559. Kamat MA Blackshaw JA Young R Surendran P Burgess S Danesh J PhenoScanner V2: An Expanded Tool for Searching Human Genotype-phenotype Associations Bioinformatics 2019 35 22 4851 3 10.1093/bioinformatics/btz469 27 Kamat MA, Blackshaw JA, Young R, Surendran P, Burgess S, Danesh J, et al. PhenoScanner V2: An Expanded Tool for Searching Human Genotype-phenotype Associations. Bioinformatics. 2019;35(22):4851-3. doi: 10.1093/bioinformatics/btz469. Burgess S Sample Size and Power Calculations in Mendelian Randomization with a Single Instrumental Variable and a Binary Outcome Int J Epidemiol 2014 43 3 922 9 10.1093/ije/dyu005 28 Burgess S. Sample Size and Power Calculations in Mendelian Randomization with a Single Instrumental Variable and a Binary Outcome. Int J Epidemiol. 2014;43(3):922-9. doi: 10.1093/ije/dyu005. Harbord RM Didelez V Palmer TM Meng S Sterne JA Sheehan NA Severity of Bias of a Simple Estimator of the Causal Odds Ratio in Mendelian Randomization Studies Stat Med 2013 32 7 1246 58 10.1002/sim.5659 29 Harbord RM, Didelez V, Palmer TM, Meng S, Sterne JA, Sheehan NA. Severity of Bias of a Simple Estimator of the Causal Odds Ratio in Mendelian Randomization Studies. Stat Med. 2013;32(7):1246-58. doi: 10.1002/sim.5659. Cabrera JLR Sotos-Prieto M Ríos AG Moffatt S Christophi CA Pérez-Martínez P Sleep and Association With Cardiovascular Risk Among Midwestern US Firefighters Front Endocrinol 2021 12 772848 772848 10.3389/fendo.2021.772848 30 Cabrera JLR, Sotos-Prieto M, Ríos AG, Moffatt S, Christophi CA, Pérez-Martínez P, et al. Sleep and Association With Cardiovascular Risk Among Midwestern US Firefighters. Front Endocrinol. 2021;12:772848. doi: 10.3389/fendo.2021.772848. Pan XL Nie L Zhao SY Zhang XB Zhang S Su ZF The Association between Insomnia and Atherosclerosis: A Brief Report Nat Sci Sleep 2022 14 443 8 10.2147/NSS.S336318 31 Pan XL, Nie L, Zhao SY, Zhang XB, Zhang S, Su ZF. The Association between Insomnia and Atherosclerosis: A Brief Report. Nat Sci Sleep. 2022;14:443-8. doi: 10.2147/NSS.S336318. Aziz M Ali SS Das S Younus A Malik R Latif MA Association of Subjective and Objective Sleep Duration as well as Sleep Quality with Non-Invasive Markers of Sub-Clinical Cardiovascular Disease (CVD): A Systematic Review J Atheroscler Thromb 2017 24 3 208 26 10.5551/jat.36194 32 Aziz M, Ali SS, Das S, Younus A, Malik R, Latif MA, et al. Association of Subjective and Objective Sleep Duration as well as Sleep Quality with Non-Invasive Markers of Sub-Clinical Cardiovascular Disease (CVD): A Systematic Review. J Atheroscler Thromb. 2017;24(3):208-26. doi: 10.5551/jat.36194. Blasco-Colmenares E Moreno-Franco B Latre ML Mur-Vispe E Pocovi M Jarauta E Sleep Duration and Subclinical Atherosclerosis: The Aragon Workers’ Health Study Atherosclerosis 2018 274 35 40 10.1016/j.atherosclerosis.2018.05.003 33 Blasco-Colmenares E, Moreno-Franco B, Latre ML, Mur-Vispe E, Pocovi M, Jarauta E, et al. Sleep Duration and Subclinical Atherosclerosis: The Aragon Workers’ Health Study. Atherosclerosis. 2018;274:35-40. doi: 10.1016/j.atherosclerosis.2018.05.003. Suzuki S Arima H Miyazaki S Fujiyoshi A Kadota A Takashima N Self-reported Sleep Duration and Subclinical Atherosclerosis in a General Population of Japanese Men J Atheroscler Thromb 2018 25 2 186 98 10.5551/jat.40527 34 Suzuki S, Arima H, Miyazaki S, Fujiyoshi A, Kadota A, Takashima N, et al. Self-reported Sleep Duration and Subclinical Atherosclerosis in a General Population of Japanese Men. J Atheroscler Thromb. 2018;25(2):186-98. doi: 10.5551/jat.40527. Souza SP Santos RB Santos IS Parise BK Giatti S Aielo AN Obstructive Sleep Apnea, Sleep Duration, and Associated Mediators With Carotid Intima-Media Thickness: The ELSA-Brasil Study Arterioscler Thromb Vasc Biol 2021 41 4 1549 57 10.1161/ATVBAHA.120.315644 35 Souza SP, Santos RB, Santos IS, Parise BK, Giatti S, Aielo AN, et al. Obstructive Sleep Apnea, Sleep Duration, and Associated Mediators With Carotid Intima-Media Thickness: The ELSA-Brasil Study. Arterioscler Thromb Vasc Biol. 2021;41(4):1549-57. doi: 10.1161/ATVBAHA.120.315644. Bertisch SM Reid M Lutsey PL Kaufman JD McClelland R Patel SR Gender Differences in the Association of Insomnia Symptoms and Coronary Artery Calcification in the Multi-ethnic Study of Atherosclerosis Sleep 2021 44 10 zsab116 zsab116 10.1093/sleep/zsab116 36 Bertisch SM, Reid M, Lutsey PL, Kaufman JD, McClelland R, Patel SR, et al. Gender Differences in the Association of Insomnia Symptoms and Coronary Artery Calcification in the Multi-ethnic Study of Atherosclerosis. Sleep. 2021;44(10):zsab116. doi: 10.1093/sleep/zsab116. Nagayoshi M Lutsey PL Benkeser D Wassel CL Folsom AR Shahar E Association of Sleep Apnea and Sleep Duration with Peripheral Artery Disease: The Multi-Ethnic Study of Atherosclerosis (MESA) Atherosclerosis 2016 251 467 75 10.1016/j.atherosclerosis.2016.06.040 37 Nagayoshi M, Lutsey PL, Benkeser D, Wassel CL, Folsom AR, Shahar E, et al. Association of Sleep Apnea and Sleep Duration with Peripheral Artery Disease: The Multi-Ethnic Study of Atherosclerosis (MESA). Atherosclerosis. 2016;251:467-75. doi: 10.1016/j.atherosclerosis.2016.06.040. Lutsey PL McClelland RL Duprez D Shea S Shahar E Nagayoshi M Objectively Measured Sleep Characteristics and Prevalence of Coronary Artery Calcification: The Multi-Ethnic Study of Atherosclerosis Sleep Study Thorax 2015 70 9 880 7 10.1136/thoraxjnl-2015-206871 38 Lutsey PL, McClelland RL, Duprez D, Shea S, Shahar E, Nagayoshi M, et al. Objectively Measured Sleep Characteristics and Prevalence of Coronary Artery Calcification: The Multi-Ethnic Study of Atherosclerosis Sleep Study. Thorax. 2015;70(9):880-7. doi: 10.1136/thoraxjnl-2015-206871. Williams CJ Hu FB Patel SR Mantzoros CS Sleep Duration and Snoring in Relation to Biomarkers of Cardiovascular Disease Risk Among Women with Type 2 Diabetes Diabetes Care 2007 30 5 1233 40 10.2337/dc06-2107 39 Williams CJ, Hu FB, Patel SR, Mantzoros CS. Sleep Duration and Snoring in Relation to Biomarkers of Cardiovascular Disease Risk Among Women with Type 2 Diabetes. Diabetes Care. 2007;30(5):1233-40. doi: 10.2337/dc06-2107. Hall MH Muldoon MF Jennings JR Buysse DJ Flory JD Manuck SB Self-reported Sleep Duration is Associated with the Metabolic Syndrome in Midlife Adults Sleep 2008 31 5 635 43 10.1093/sleep/31.5.635 40 Hall MH, Muldoon MF, Jennings JR, Buysse DJ, Flory JD, Manuck SB. Self-reported Sleep Duration is Associated with the Metabolic Syndrome in Midlife Adults. Sleep. 2008;31(5):635-43. doi: 10.1093/sleep/31.5.635. Choi KM Lee JS Park HS Baik SH Choi DS Kim SM Relationship between Sleep Duration and the Metabolic Syndrome: Korean National Health and Nutrition Survey 2001 Int J Obes (Lond) 2008 32 7 1091 7 10.1038/ijo.2008.62 41 Choi KM, Lee JS, Park HS, Baik SH, Choi DS, Kim SM. Relationship between Sleep Duration and the Metabolic Syndrome: Korean National Health and Nutrition Survey 2001. Int J Obes (Lond). 2008;32(7):1091-7. doi: 10.1038/ijo.2008.62. Pyykkönen AJ Isomaa B Pesonen AK Eriksson JG Groop L Tuomi T Sleep Duration and Insulin Resistance in Individuals Without Type 2 Diabetes: The PPP-Botnia Study Ann Med 2014 46 5 324 9 10.3109/07853890.2014.902226 42 Pyykkönen AJ, Isomaa B, Pesonen AK, Eriksson JG, Groop L, Tuomi T, et al. Sleep Duration and Insulin Resistance in Individuals Without Type 2 Diabetes: The PPP-Botnia Study. Ann Med. 2014;46(5):324-9. doi: 10.3109/07853890.2014.902226. Dandona P Aljada A Bandyopadhyay A Inflammation: The Link between Insulin Resistance, Obesity and Diabetes Trends Immunol 2004 25 1 4 7 10.1016/j.it.2003.10.013 43 Dandona P, Aljada A, Bandyopadhyay A. Inflammation: The Link between Insulin Resistance, Obesity and Diabetes. Trends Immunol. 2004;25(1):4-7. doi: 10.1016/j.it.2003.10.013. Tedgui A Mallat Z Hypertension: A Novel Regulator of Adaptive Immunity in Atherosclerosis? Hypertension 2004 44 3 257 8 10.1161/01.HYP.0000140270.26523.9b 44 Tedgui A, Mallat Z. Hypertension: A Novel Regulator of Adaptive Immunity in Atherosclerosis? Hypertension. 2004;44(3):257-8. doi: 10.1161/01.HYP.0000140270.26523.9b. Krueger PM Friedman EM Sleep Duration in the United States: A Cross-sectional Population-based Study Am J Epidemiol 2009 169 9 1052 63 10.1093/aje/kwp023 45 Krueger PM, Friedman EM. Sleep Duration in the United States: A Cross-sectional Population-based Study. Am J Epidemiol. 2009;169(9):1052-63. doi: 10.1093/aje/kwp023. Stranges S Dorn JM Shipley MJ Kandala NB Trevisan M Miller MA Correlates of Short and Long Sleep Duration: A Cross-cultural Comparison between the United Kingdom and the United States: The Whitehall II Study and the Western New York Health Study Am J Epidemiol 2008 168 12 1353 64 10.1093/aje/kwn337 46 Stranges S, Dorn JM, Shipley MJ, Kandala NB, Trevisan M, Miller MA, et al. Correlates of Short and Long Sleep Duration: A Cross-cultural Comparison between the United Kingdom and the United States: The Whitehall II Study and the Western New York Health Study. Am J Epidemiol. 2008;168(12):1353-64. doi: 10.1093/aje/kwn337. 10.36660/abc.20230813i Original Article Sleep Duration and the Risk of Atherosclerosis: A Mendelian Randomization Study Xu Xiaozhuo Conception and design of the research Writing of the manuscript 1 Huang Yilin Acquisition of data and Analysis and interpretation of the data Critical revision of the manuscript for content 1 Liu Jing Acquisition of data and Analysis and interpretation of the data 1 0009-0006-8542-8509 Han Xu Conception and design of the research Critical revision of the manuscript for content 1 Department of Geriatrics, Affiliated Hospital of Nanjing University of Chinese Medicine Nanjing China Department of Geriatrics, Affiliated Hospital of Nanjing University of Chinese Medicine, Jiangsu Province Hospital of Chinese Medicine, Nanjing – China Xu Han • Department of Geriatrics - Affiliated Hospital of Nanjing University of Chinese Medicine, Jiangsu Province Hospital of Chinese Medicine - No.155 Hanzhong Road, Qinhuai District, Nanjing 210029 - China E-mail: xhan_njgeriatrics@outlook.com

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

Editor responsible for the review: Marcio Bittencourt

 Abstract Background:

The association between the length of sleep and atherosclerosis has been reported in many observational studies. However, little is known about its significance as a risk factor for atherosclerosis or as a negative consequence of atherosclerosis.

 Objective:

This study aimed to assess the causal association between sleep duration and the risk of atherosclerosis using publicly available genome-wide association studies (GWAS) summary statistics.

 Methods:

We employed a two-sample Mendelian randomization (MR) method with 2 cohorts from MRC-IEU (n=460,099) and UK Biobank (n=361,194) to investigate the causal association between sleep duration and the risk of atherosclerosis. Three methods including the inverse-variance weighted (IVW) technique, Robust adjusted profile score (RAPS), and simple-and weighted-median approach were used to obtain reliable results, and an odds ratio with a 95% confidence interval (CI) was calculated. P<0.05 was considered as a statistical difference. In addition, MR-Egger regression, Radial MR, MR-PRESSO, and leave-one-out analyses were used to assess the possible pleiotropy effects.

 Results:

No causal association of sleep duration with atherosclerosis was found [OR (95%CI): 0.90 (0.98-1.00), p = 0.186]. Leave-one-out, MR-Egger, and MR-PRESSO analyses failed to detect horizontal pleiotropy.

 Conclusions:

This MR analysis indicated no causal association between genetically predicted sleep duration and atherosclerosis across European populations.

 Keywords: Sleep Duration Atherosclerosis Mendelian Randomization Analysis Natural Science Foundation of Jiangsu Province BK20181505 Nanjing University of Chinese Medicine ZYX03KF032 Science and Technology Development of Traditional Chinese Medicine of Jiangsu Province 2020ZX08 Geriatric Clinical Technology Application Research Project of Jiangsu Provincial Health Comission LR202202 Jiangsu Province Posgratuate Practice Innovation Program SJCX22 0766 SJCX21 0670 Introduction

Atherosclerosis is a multifactorial disease and the leading cause of cardiovascular and cerebrovascular events.1 Atherosclerosis is a complex multifactorial trait with an enigmatic genetic etiology. As a chronic disease severely threatening human health, it has aroused wide attention, especially coronary atherosclerosis. We have witnessed an "epidemiological transition".2 Increased sanitation and the treatment of acute infections have reduced the prevalence of infectious diseases in developing countries, and more individuals are now experiencing chronic diseases such as atherosclerosis.3 Atherosclerosis can lead to a variety of cardiovascular diseases (CVDs), which have been recognized as a major cause of morbidity and mortality.4 It is urgent to interpret its mechanism, advance its management, and develop prospects for mitigating its impact.

Sleep is a complex physiological process produced by the brain, which plays a very important role in regulating the physiological functions of various systems in the body. With the continuous extension of working hours in modern society, the way people work and the sleep habits of people are also constantly changing, and the reduction of sleep time is becoming a severe problem.5 In fact, short and long sleep duration was found to be associated with coronary artery calcium6-8 and carotid intima-media thickness (CIMT),9,10 which are indicators of atherosclerosis in large arteries feeding the heart and brain. In addition, some studies have found that too long or too short sleep time will still increase the incidence of cardiovascular events after controlling for mixed factors such as obesity, hypertension, and diabetes.11 Several observational studies have also mentioned the relationship between sleep duration and subclinical atherosclerosis of coronary or carotid arteries.12,13 However, it remains unclear whether not getting enough or too much sleep contributes to the occurrence of atherosclerosis.

To the best of our knowledge, the causal association between sleep duration and atherosclerosis has not been assessed. Mendelian randomization (MR) is a method for verifying the causality, avoiding residual confounding, and overcoming reverse causality in a retrospective setting, which can reveal causal estimates of risk factors in complex diseases using genetic variants as instrumental variables.14,15 Herein, we conducted an MR study to evaluate the causal association between sleep duration and the risk of atherosclerosis.

 Methods Study design

Assuming the MR studies’ causal estimate is credible. Three crucial assumptions need to be met: 1) There must be a strong association between the selected genetic instrumental variables (IVs) and exposure.16 2) The choice of genetic IVs does not influence the outcome without consideration of exposure (i.e., horizontal pleiotropy is nonexistent).17 3) The selected genetic IVs are not associated with the possible confounders. Central Illustration provides an overview. Since the research was based on publicly accessible datasets and previously published studies, ethical approval and participant consent were not required for the study.

 Data sources

IVs for sleep duration were based on a meta-analysis of a genome-wide association study (GWAS) of 460,099 people of European ancestry. Self-reported habitual sleep duration was the major exposure of the current study. It was obtained from touchscreen questionnaires at baseline assessment. Sleep duration was evaluated according to a standardized question: "How many hours of sleep do you get every 24 hours?". Participants who answered "Do not know" and "Prefer not to answer", and those who provided implausible sleep durations (< 4 h or > 11 h per day) were excluded to minimize implausible sleep duration and potential confounding by poor health. A complete description of the study design, participants, and quality control (QC) methods has been described in detail previously.18 UK Biobank received ethical approval from the Research Ethics Committee (REC reference for UK Biobank is 11/NW/0382).

Atherosclerosis was identified based on the 8th and 10th editions of the International Classification of Diseases (ICD). Data on atherosclerosis were collected from participants in the United Kingdom Biobank (GWAS ID: ukb-d-I9_CORATHER, available at https://gwas.mrcieu.ac.uk/datasets/ukb-d-I9_CORATHER/). This data set included 361,194 people of European ancestry (a total of 14,334 cases and 346,860 controls), and it included 13,586,589 single nucleotide polymorphisms (SNPs). We introduced covariate-adjusted LD score regression (cov-LDSC), a method to accurately estimate genetic heritability (h2g) and its enrichment in both homogenous and admixed populations with summary statistics and in-sample LD estimates. The full data release contained the cohort of successfully genotyped samples (n=488,377). 49,979 individuals were genotyped using the UK BiLEVE array and 438,398 using the UK Biobank axiom array. Totally 9,851,867 SNPs of sleep duration in 460,099 individuals were extracted from the MRC-IEU (GWAS ID: ukb-b-4424, available at https://gwas.mrcieu.ac.uk/datasets/ukb-b-4424/). Pre-imputation QC, phasing, and imputation were conducted by the previous study.19

 The selection of the relevant instrumental variables

SNPs were considered as IVs for this study.16 The following criteria were satisfied by every single SNP that was requested: 1. There was a substantial correlation with the amount of exposure based on the relevance of the genome as a whole; 2. No linkage disequilibrium (LD) (pairwise r2 = 0.001, window size = 10,000kb); 3. Not containing any palindromic structures. A total of 65 SNPs were found after considering the above three assumptions and criteria. We were unable to find the appropriate SNPs in the atherosclerosis GWAS, thus to get accurate estimates, we employed proxy SNPs that had substantial LD (r2>0.8) to stand in for the chosen SNPs, which allowed us to get more accurate results. The first-stage regression, or F statistic, was used to assess the strength of the instruments and was calculated using the following equation: F= (R2/k)/ ([1−R2]/[n−k−1]), where R2 is the proportion of the sleep duration variability accounted for by the SNP, k is the number of instruments used in the model and n is the sample size.20 To limit the influence of possible weak IV bias, an F statistic greater than 10 was expected to be of sufficient strength for the main study.21 The flow chart for the selection of IVs is depicted in Figure 1.

The flow chart of instrumental variables selection. Statistical analysis

The inverse-variance weighted (IVW) approach was used as the main method to determine whether there was a correlation between sleep duration and atherosclerosis.22 If the p from Cochran's Q test was greater than 0.05, we decided to use a model with fixed effects; in all other cases, we used a model with random effects.23 If the selected IVs did not exhibit directional pleiotropy (and the p for the MR-Egger intercept was greater than 0.05), the IVW technique was considered the most reliable.24

We chose the MR-Egger approach to assess the possible pleiotropy impacts in sensitivity analyses. The MR-Egger regression's intercept term, which estimated the causal effect as the slope from the weighted regression of the IVs-outcome relationships on the IVs-exposure relationship, reflected the average pleiotropic effect.25,26 To determine whether there was pleiotropy, we also used the basic median, weighted median, Radial MR, and MR-PRESSO (Mendelian Randomization Pleiotropy Residual Sum and Outlier) outlier test techniques.26 If more than fifty percent of the SNPs being studied were effective IVs, then the weighted median will offer the most reliable estimates of the causative impact. In addition to pleiotropy detection, MR-PRESSO also can reevaluate effect estimations and eliminate outlier SNPs.26 To evaluate the impact of outlying data, a leave-one-out analysis was conducted in the meantime. We further investigated each chosen SNP's pleiotropy using the PhenoScanner V2 database (http://www. pen scanner. medschl. cam. ac. uk/) at the GWAS level of statistical significance (p < 5×10−8) to exclude the impact of other variables.27

Unless otherwise stated, all tests were two-sided, and the differences were regarded as statistically significant (p < 0.05). The R software's Two Sample MR (V 0.5.6), Radial MR, and MR-PRESSO (V 1.0)24 packages were used for all statistical analyses (4.0.5).

 Results

Further information on the chosen SNPs is given in Supplementary Table S1-S2. Three SNPs in total (sleep duration: rs1611719, rs17732997, and rs2186122) were eliminated from the MR research because they were palindromes. In the end, 62 SNPs, including 1 proxy SNP, were chosen as IVs (all p < 5×10−8, r2=0.001).

MR estimates

The results of Cochran's Q test for sleep time showed minimal heterogeneity (p = 0.108). According to the IVW method's findings, there was little evidence of a link between the length of sleep and the risk of atherosclerosis (OR (95% CI), 0.992 (0.979-1.004); p = 0.186) (Supplementary Table S3).

 Sensitivity analyses

The results of the simple median and weighted median were comparable to those of the IVW approach. Meanwhile, horizontal pleiotropy was not detected by the MR-Egger regression (p intercept=0. 071 for sleep duration) (Supplementary Table S3). Although Radial MR suggested the existence of outliers (Figure 2), MR-PRESSO showed that outliers did not affect the study results (Table 1). Similarly, the pleiotropy was not detected by RAPS (Table 2) and PhenoScanner V2 database. When the horizontal pleiotropy showed p>0.05, IVW (fixed effects) method (Table 2) was used to assess the data. For sleep duration, Supplementary Figures S1-S4 present forest plots, scatter plots, funnel plots, and MR leave-one-out plots.

An overview of the radial MR estimates outlier. MR-PRESSO estimates between sleep duration and atherosclerosis
Caracteristic Raw estimate Outlier corrected estimate
N OR 95%CI p N OR 95%CI p
Sleep duration 62 0.991 0.979-1.003 0.163 62 0.991 0.979-1.003 0.163

SNP: single-nucleotide polymorphism; MR-PRESSO: Mendelian randomization pleiotropy residual sum and outlier test. OR: odds ratio; 95%CI: 95% confidence interval.

 RAPS estimates and IVW (fixed effects) between sleep duration and atherosclerosis
Caracteristic RAPS IVW (fixed effects)
N OR 95%CI p N OR 95%CI p
Sleep duration 62 0.992 0.980-1.004 0.193 62 0.992 0.980-1.004 0.135

SNP: single-nucleotide polymorphism; RAPS: Robust adjusted profile score; IVW (fixed effects). Inverse variance weighted (fixed effects). OR: odds ratio; CI: confidence interval.

 Bias and power analyses

The bias of the genetic instruments was 0.000 for sleep duration. The F statistic of the selected SNPs was 15.671, which was expected to have sufficient strength for the main study (Supplementary Table S3). The estimate derived from the ratio technique was close to the conditional Odds ratio under certain particular conditions and approximated a population-averaged odds ratio.28,29 The consistency of the estimator under the null was unaffected by the odds ratio estimate that was used. We performed the power calculations, and the Type 1 error value for sleep duration was 0.05. For the statistical power value for sleep, the duration was 95%. According to the sample size used in the atherosclerosis GWAS meta-analysis, there was >80% power to identify the relationship between the amount of sleep and the risk of atherosclerosis for effect size (OR) of 0.992 (Supplementary Table S4). In recent additional MR investigations, all IVs for genetically predicted sleep duration have been authorized and used.30,31 In addition to this, none of them had any bearing on high blood pressure or elevated levels of low-density lipoprotein (Supplementary Table S3).

 Discussion

In this present study, we attempted to explore the causal association between sleep duration and atherosclerosis using an MR method. Our findings showed no evidence that genetically predicted sleep duration is linked to the risk of atherosclerosis in European populations. Additionally, sensitivity studies showed that the findings were generally reliable.

Coronary artery calcium scores (CACS), CIMT, and Brachial-Ankle Pulse Wave Velocity (baPWV) were major surrogate indicators of atherosclerosis and predictors of cardiovascular events.6,32 Some studies have explored the effect of sleep duration on the incidence of atherosclerosis by analyzing the relationship between sleep duration and CACS, CIMT, and baPWV. A recent study of 1,968 healthy men aged 40 to 60 indicated that increased or decreased sleep duration was associated with an increased incidence of coronary atherosclerosis and assessed the effect of sleep duration on the incidence of subclinical arteriosclerosis by measuring CACS, finding people who slept for 7 hours had the lowest incidence of subclinical coronary atherosclerosis.33 For people with risk factors for atherosclerosis, sleep duration was also significantly correlated with the incidence of atherosclerosis. Similarly, the CIMT was the lowest when the sleep time was 7-8 h, and the increase or decrease in sleep time will lead to an increase of CIMT.12

Previous studies showed no relationship between sleep duration and markers of vascular damage and atherosclerosis.9,34-38 A cross-sectional survey of 1,093 Japanese men reported that self-reported sleep duration was not associated with increased CAC or CIMT.33 Souza et al.35 also found no independent associations of objective sleep duration with CIM. No evidence demonstrated that the association between insomnia symptoms and CAC score >0 differed by objective short sleep duration status.36 In addition, a study on the Multi-Ethnic Study of Atherosclerosis (MESA) showed that severe obstructive sleep apnea was not associated with high CAC burden or abnormal ABI.37 MESA investigators reported no associations between short (<6 hours) and long (>8 hours) sleep durations and CAC.38 These were consistent with our findings.

The clinically and physiologically significant link between extended sleep and CVDs risk in adults is not supported by sufficient experimental data. We hypothesized, based on the current information, that the underlying mechanism was metabolic in nature and operated through an inflammatory route. Specifically, prolonged sleep may lead to low HDL,39,40 hyperglycemia, hypertriglyceridemia,41 and insulin resistance,42 all of which can lead to vascular endothelial dysfunction and subclinical inflammation, thereby further promoting atherosclerosis.43,44 Related social, lifestyle, and behavioral issues, such as drug abuse, physical inactivity, or lack of access to nutritious meals, may exacerbate this pro-atherogenic environment.45,46 Regardless of the actual cause-and-effect relationship, we supported the examination of sleep length in clinical assessments, as short or long sleep duration may indicate the risk of chronic diseases. CVDs and diabetes are life-threatening diseases that are prevalent in our society and can lead to early illness and death, thus it is important to investigate the relationship between sleep and chronic diseases over time. This includes finding the best prevention strategy to warn against atherosclerosis, which can stimulate CVDs and other diseases. This is an important step towards a healthier population both nationally and globally.

According to our knowledge, this study was the first MR investigation to examine the causal association between sleep duration and risk of atherosclerosis using available GWAS datasets. Additionally, for this two-sample MR investigation, we chose people from Europe to lessen demographic bias. The current MR research also had several shortcomings. First, since we used publicly accessible genetic data for our investigation, we were unable to do stratified analysis or take additional factors into account. Second, the chosen instrumental SNPs as IVs only partially (0.001%-0.01%) explained the variation in sleep duration. Low statistical power to identify weak relationships may result from this. Third, in a two-sample MR analysis, any bias due to weak instruments was in the direction of the null. Bias in the direction of the null was less serious than bias in the direction of the observational association, as it is conservative and will not lead to inflated Type 1 error rates and false-positive findings. There was indeed a possibility of overlap between the two samples.28 Ultimately, since our data set was made up of people of European heritage, our conclusions may not apply to other groups outside of Europe.

 Conclusions

In the current study, genetically predicted sleep duration among European populations was not causally linked to the risk of atherosclerosis. Further study is needed to investigate the causal association between atherosclerosis and sleep duration.

This study was funded by Natural Science Foundation of Jiangsu Province (No. BK20181505), Phase III Scientific Research Project of Traditional Chinese Medicine Superior Discipline in Nanjing University of Chinese Medicine (No. ZYX03KF032), the Special Plan for the Science and Technology Development of Traditional Chinese Medicine of Jiangsu Province (No. 2020ZX08), the Leader Scientific Research Project of the Geriatric Clinical Technology Application Research Project of Jiangsu Provincial Health Comission (No LR202202), the Jiangsu Province Posgratuate Practice Innovation Program (No SJCX22 0766 and SJCX21 0670).

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.

 *Supplemental Materials

For additional information Supplemental Table, please click here.

For additional information Supplemental Figure, please click here.

 Acknowledgments

We gratefully thank MRC-IEU and UK Biobank for providing genetic data about the MR study.