Date Published: December 21, 2016
Publisher: Public Library of Science
Author(s): Shannon Anjelica Allport, Ngum Kikah, Nessim Abu Saif, Fonkem Ekokobe, Folefac D. Atem, Katriina Aalto-Setala.
The risk for cardiovascular disease (CVD) is higher for individuals with a first-degree relative who developed premature CVD (with a threshold at age 55 years for a male or 65 years for a female). However, little is known about the effect that each unit increase or decrease of maternal or paternal age of onset of CVD has on offspring age of onset of CVD. We hypothesized that there is an association between maternal and paternal age of onset of CVD and offspring age of onset of CVD.
We used the Framingham Heart Study database and performed conditional imputation for CVD-censored parental age (i.e. parents that didn’t experience onset of CVD) and Cox proportional regression analysis, with offspring’s age of onset of CVD as the dependent variable and parental age of onset of CVD as the primary predictor. Modifiable risk factors in offspring, such as cigarette smoking, body mass index (BMI), diabetes mellitus, systolic blood pressure (SBP), high-density lipoprotein (HDL) level, and low-density lipoprotein (LDL) level, were controlled for. Separate analyses were performed for the association between maternal age of onset of CVD and offspring age of onset of CVD and the association between paternal age of onset of CVD and offspring age of onset of CVD.
Parental age of onset of CVD was predictive of offspring age of onset of CVD for maternal age of onset of CVD (P < .0001; N = 1401) and for paternal age of onset of CVD (P = 0.0134; N = 1221). A negative estimate of the coefficient of interest signifies that late onset of cardiovascular events in parents is protective of onset of CVD in offspring. Cigarette smoking and HDL level were important associated confounders. Offspring age of onset of cardiovascular disease is significantly associated with both maternal and paternal age of onset CVD. The incorporation of the parameters, maternal or paternal age of onset of CVD, into risk estimate calculators may improve accuracy of identification of high-risk patients in clinical settings.
Cardiovascular disease (CVD), defined as coronary death, myocardial infarction, coronary insufficiency, angina, ischemic stroke, hemorrhagic stroke, transient ischemic attack, peripheral artery disease, and heart failure, is the leading global cause of death, accounting for over 30 percent of all deaths worldwide– 17.3 million deaths per year. In the United States, someone dies from cardiovascular diseases every 39 seconds . Though the death rate due to CVD has decreased slowly over the last 5 decades, heart disease remains the leading cause of death in the United States, and caring for patients with poor cardiovascular health continues to be one of the largest burdens on the health care system today. From 1990 to 2009, CVD ranked first in the number of days for which patients received hospital care , yet 72% of Americans do not consider themselves at risk for heart disease .
The study was approved by University of Texas Health Science Center in Houston Institutional Review Board and Biologic Specimen and Data Repository Information (BioLINCC). For this study we used The Framingham Heart Study (FHS) database. The data comes from a well-known, longitudinal prospective cohort study that is a premier database for study of cardiovascular diseases and other terminal disease. The FHS Original Cohort was launched at Exam 1 in 1948 and has continued with biennial examinations to the present. The FHS Original Cohort consists of 5,209 respondents (55% females) aged 28–62 years residing in Framingham, Massachusetts, between 1948 and 1951. Nearly all subjects were Caucasians. The Offspring Cohort (FHSO) was launched at Exam 1 in 1971 and has on average been examined every 3 to 4 years since enrollment. The FHSO dataset consists of a sample of 3514 biological descendants of the Original Cohort, 1576 of their spouses and 34 adopted offspring for a total sample of 5124 subjects (52% females). This analysis used values collected during the FHS Original Cohort Exam 12 (1971–1974, n = 3261) and the FHSO Exam 1 (1971–1975, n = 5124). There was a total of 2622 linked parent-offspring pairs. The age of onset of CVD is the age at which either parent or offspring was diagnosed with CVD, including coronary death, myocardial infarction, coronary insufficiency, angina, ischemic stroke, hemorrhagic stroke, transient ischemic attack, peripheral artery disease, and heart failure, by a medically trained professional. This ‘time-to-event’ data was collected based on follow up data of both the original and offspring cohorts. Baseline values of the covariates of interest, collected at first clinical visit, were used in data analysis. Participants were included if they had not yet developed overt symptoms of cardiovascular disease or suffered a heart attack or stroke, and subjects were excluded if they had missing data.
We used Cox’s proportional hazards model  to examine the relationship between offspring age of onset of CVD and parent age of onset of CVD. As an illustration for our paper, we are interested in estimating the parameters of the Cox proportional hazards model of the relationship between parent age of onset of cardiovascular disease as a primary predictor while controlling for offspring baseline cigarette smoking status, diabetes mellitus, BMI, SBP, HDL, and LDL. We used the time of onset of cardiovascular disease in offspring participants as the outcome of interest. The primary variables of interest, paternal and maternal age of onset of CVD, were imputed using the single imputation for a randomly censored covariate. Potential confounding variables (BMI, SBP, HDL, and LDL) were entered as continuous variables while diabetes status and smoking status were entered as categorical variables. Non-diabetics and non-smokers, respectively, were the designated reference groups.
Our primary variable of interest is parental age of onset of CVD, which is potentially right-censored. We controlled for the following confounders: offspring smoking status, offspring diabetes status, offspring BMI, offspring SBP, offspring HDL, and offspring LDL (Tables 3 and 4). The outcome, offspring age of onset of CVD, is survival data, data where the outcome variable is the time until the occurrence of an event of interest. The offspring age of onset is survival data because the variable is recorded as the time from birth to the development of CVD. In this cohort, not all of the offspring developed CVD. The age at which they were diagnosed with CVD by a physician was recorded as ‘age of onset’ while those who did not develop CVD were censored. The group that did not develop CVS is considered censored because we assume that they have a non-zero chance of developing CVD later in life but are CVD free at the present time.
Maternal age of onset of CVD is statistically significantly related to age of onset of CVD in offspring with a hazard ratio of 0.959, while paternal age of onset has a hazard ratio of 0.973. Our analysis describes hazard ratios based on unit increase or decrease of maternal and paternal age of onset of cardiovascular events. Although several papers have evaluated the link between CVD in parents and CVD risk in offspring, they have categorized parental history of CVD as a dichotomous variable using a thresholding approach [26,39] or in 10-year intervals , which decreases the power of the analysis. None have shown the effect of specific age of onset of CVD in parents on the risk of CVD in offspring. Our results are in line with previous knowledge that late onset CVD decreases hazard ratio and is protective of CVD development in offspring [26,27], but also increases power by providing a greater level of detail of this relationship.