番茄社区app

For cardiovascular disease (CVD) prediction in women, female-specific factors hardly improved risk stratification over traditional risk factors alone, according to a retrospective study.

No female-specific factor individually added to the accuracy of three established CVD risk calculators to predict 5- or 10-year risk in women in the U.K. Biobank who were 45-69 years old and free of CVD at baseline.

Sex-specific factors included early menarche (<11 years), endometriosis, excessive or irregular menstruation, miscarriage, stillbirth, infertility, preeclampsia or eclampsia, gestational diabetes, premature (<40 years) or early menopause (<45 years), reported researchers led by Jenny Doust, BMBS, PhD, of the University of Queensland in Herston, Australia.

"Our study shows that none of the female-specific risk factors included in this analysis improved the ability to determine who is and is not at risk of CVD when added to the risk factors included in contemporary CVD risk calculators when assessed using either the C-index or the more sensitive [net reclassification index]. Measures of calibration were also not improved," Doust's group wrote in .

Only by pooling all available female-specific factors did the investigators increase the C-index for the CVD risk algorithms tested:

• Pooled Cohort Equation-Atherosclerotic Cardiovascular Disease (PCE-ASCVD) for 10-year risk: improvement in C-index from 0.710 to 0.712
• Qrisk2 from England and Wales, also for 10-year risk: improvement from 0.713 to 0.715
• PREDICT tool from New Zealand, for 5-year risk: improvement from 0.718 to 0.720

The findings were consistent with prior work showing that adding female-specific factors -- and a wider set of -- has generally not helped cardiovascular risk prediction, the researchers noted.

"Although several female-specific risk factors are independent risk factors for CVD in epidemiological cohort studies, most of these studies have not adjusted for the clinical measures that would be used by clinicians when considering blood pressure or lipid-lowering medication," Doust's team explained.

The present report may be timely as American cardiovascular societies have recently discussed possibly transitioning to yet another CVD risk model: the PREVENT score.

, PREVENT reportedly gives more accurate and precise risk estimates, having been derived without race as part of the calculation and with more contemporary derivation populations compared with PCE-ASCVD. The American College of Cardiology and American Heart Association have not endorsed replacing existing models with PREVENT in their guidelines.

The PCE-ASCVD, Qrisk2, and PREDICT scores take into account traditional CVD predictors such as diabetes, hypertension, obesity, and smoking. Though not female-specific, some female-predominant risk factors associated with CVD are also included in the Qrisk2 calculator.

"While these calculators are easy to use, they do not incorporate female-specific risk factors regardless of whether they are considered risk-enhancing. This may lead to an underestimation of CVD risk in women who have female-specific risk factors," commented Setareh Salehi Omran, MD, and Michelle Leppert, MD, MBA, MS, both of the University of Colorado Anschutz Medical Campus, Aurora.

"Although adding female-specific risk factors may not improve CVD risk calculators, these risk factors still play an important role in comprehensive cardiovascular risk stratification in women. Many of these risk factors occur earlier in life and upstream of the causal pathway leading to traditional risk factors and then CVD, which represents a valuable opportunity for earlier intervention in women," stressed Salehi Omran and Leppert in an .

"By focusing on catching and treating traditional risk factors early in their course, we may ultimately lower the high morbidity of CVD in women," the editorialists wrote. "Perhaps, women with female-specific risk factors should be considered for early and more frequent interval screening of traditional risk factors than the average adult. Close collaboration between primary care physicians and obstetricians/gynecologists would ensure that these women are appropriately monitored for the development of traditional risk factors."

U.K. Biobank participants were women who enrolled in the study from 2006-2010. This came out to 135,142 individuals (mean age 57.5 years) whose self-reports largely determined their risk factors.

Median follow-up reached 10.7 years. The investigators stopped follow-ups after 2019 to minimize the impact of the COVID-19 pandemic.

Doust and colleagues noted that the prevalence of several female-specific risk factors was lower than expected based on epidemiological studies -- for example, eclampsia reached <0.1% and gestational diabetes without subsequent type 2 diabetes was 0.3%.

The incidence of CVD was an estimated 5.3 per 1,000 person-years in the PCE-ASCVD and PREDICT models and 5.2 per 1,000 person-years in the Qrisk2 model.

Adding each of the female-specific factors to the three risk scores did nothing to improve their C-index, net reclassification index, integrated discrimination index, slope of the regression line for predicted versus observed events, and Brier score or plots of calibration.

Salehi Omran and Leppert pointed out that the analysis did not include some female-specific risk factors associated with CVD, such as polycystic ovary syndromes, parity, preterm delivery, and small for gestational age babies.

Due to the study's design, investigators were also unable to exclude the possibility that female-specific risk factors may have a causal relationship with CVD that is mediated by traditional risk factors. "However, this does not mean that female-specific risk factors do not ultimately affect CVD occurrence," the editorialists noted.

Comment