Age at menarche and risk of all-cause and cardiovascular mortality: a systematic review and dose–response meta-analysis. Issue 6 (June 2019)

• Record Type: Journal Article
• Authors: Chen, Xu; Liu, Yu; Sun, Xizhuo; Yin, Zhaoxia; Li, Honghui; Liu, Xuejiao; Zhang, Dongdong; Cheng, Cheng; Liu, Leilei; Liu, Feiyan; Zhou, Qionggui; Wang, Chongjian; Li, Linlin; Wang, Bingyuan; Zhao, Yang; Liu, Dechen; Zhang, Ming; Hu, Dongsheng
• Is Part Of: Menopause. Volume 26:Issue 6(2019)
• Publication Date: 2019-06
• Subjects: Cohort studies -- Menarche -- Meta-analysis -- Mortality; Menopause -- Periodicals; 618.175005
• DOI: 10.1097/GME.0000000000001289
• Languages: English
• ISSNs: 1072-3714

Abstract

Abstract: Objective: The evidence between age at menarche and mortality risk is controversial. We aimed to quantify the dose–response association of age at menarche and risk of all-cause and cardiovascular disease (CVD) mortality based on cohort studies. Methods: PubMed, EMBASE, Web of Science, and Scopus databases were searched up to March 15, 2018 for relevant articles. Random-effects models and restricted cubic splines were used for this meta-analysis. Results: Twelve cohort studies, with 79, 363 deaths and 2, 341, 769 participants, met the inclusion criteria. With each 1-year increase in menarche age, the relative risk (RR) was reduced for all-cause mortality (RR: 0.977, 95% confidence interval [CI]: 0.970-0.984), CVD mortality (RR: 0.993, 95% CI: 0.975-1.011), ischemic heart disease (IHD) mortality (RR: 0.969, 95% CI: 0.947-0.993), and stroke mortality (RR: 0.983, 95% CI: 0.954-1.012). We found a nonlinear dose–response association ( P nonlinearity = 0.001) between age at menarche and all-cause mortality, with the lowest risk observed at menarche age 15 years (RR: 0.849 95% CI: 0.800-0.901), but no evidence of a nonlinear association between menarche age and CVD mortality ( P nonlinearity = 0.543), IHD mortality ( P nonlinearity = 0.310), or stroke mortality ( P nonlinearity = 0.824). Conclusions: Age at menarche is inversely associated with all-cause and IHD mortality. Abstract : Supplemental Digital Content is available in the text