Females Display Lower Risk of Myocardial Infarction From Higher Estimated Cardiorespiratory Fitness Than Males: The Tromsø Study 1994-2014

Abstract

Objective: To examine the dose-response association between estimated cardiorespiratory fitness (eCRF) and risk of myocardial infarction (MI).

Patients and methods: Adults who attended Tromsø Study surveys 4-6 (Janurary 1,1994-December 20, 2008) with no previous cardiovascular disease were followed up through December 31, 2014 for incident MI. Associations were examined using restricted cubic splines Fine and Gray regressions, adjusted for education, smoking, alcohol, diet, sex, adiposity, physical activity, study survey, and age (timescale) in the total cohort and subsamples with hyperlipidemia (n=2956), hypertension (n=8290), obesity (n=5784), metabolic syndrome (n=1410), smokers (n=3823), and poor diet (n=3463) and in those who were physically inactive (n=6255).

Results: Of 14,285 participants (mean age ± SD, 53.7±11.4 years), 979 (6.9%) experienced MI during follow-up (median, 7.2 years; 25th-75th, 5.3-14.6 years). Females with median eCRF (32 mL/kg/min) had 43% lower MI risk (subdistributed hazard ratio [SHR], 0.57; 95% CI, 0.48-0.68) than those at the 10th percentile (25 mL/kg/min) as reference. The lowest MI risk was observed at 47 mL/kg/min (SHR, 0.02; 95% CI, 0.01-0.11). Males had 26% lower MI risk at median eCRF (40 mL/kg/min; SHR, 0.74; 95% CI, 0.63-0.86) than those at the 10th percentile (32 mL/kg/min), and the lowest risk was 69% (SHR, 0.31; 95% CI, 0.14-0.71) at 60 mL/kg/min. The associations were similar in subsamples with cardiovascular disease risk factors.

Conclusion: Higher eCRF associated with lower MI risk in females and males, but associations were more pronounced among females than those in males. This suggest eCRF as a vital estimate to implement in medical care to identify individuals at high risk of future MI, especially for females.

Figure 1

Restricted cubic spline Fine and Grey regressions of eCRF and risk of MI among females and males: the Tromsø Study 1994-2014. Data are shown as subdistributed hazard ratio (line) with 95% CIs (shaded area), adjusted for education, smoking, diet, alcohol, and study survey; age, waist circumference (ie, adiposity), and physical activity are adjusted by inclusion in the eCRF formulas. Reference of the spline is set to the 10th percentile of the distribution of the group (ie, among females and males separately), and values are shown between the first and 99th percentile of the distribution of estimated cardiorespiratory fitness. Frequency refers to the frequency of observed eCRF. eCRF, estimated cardiorespiratory fitness; MI, myocardial infarction.

Figure 2

Restricted cubic spline Fine and Grey regressions of eCRF and risk of MI among females and males with (A) hyperlipidemia (n=2965, MI=400); (B) hypertension (n=8290, MI=834); (C) central obesity (n=5784, MI=379); and (D) metabolic syndrome (n=1410, MI=181): the Tromsø Study 1994-2014. Data are shown as subdistributed hazard ratio (line) with 95% CIs (shaded area), adjusted for education, smoking, diet, alcohol, and study survey; age, waist circumference (ie, adiposity), and physical activity are adjusted by inclusion in the eCRF formulas. Reference of the spline is set to the 10th percentile of the distribution of the group (ie, among females and males separately), and values are shown between the first and 99th percentile of the distribution of estimated cardiorespiratory fitness. Frequency refers to the frequency of observed eCRF. eCRF, estimated cardiorespiratory fitness; MI, myocardial infarction.

Figure 3

Restricted cubic spline Fine and Grey regressions of eCRF and risk of MI among females and males who (A) are smokers (n=3823, MI=372); (B) are physically inactive (n=6255, MI=443); and (C) do not meet any nutritional guideline (n=3463, MI=382): the Tromsø Study 1994-2014. Data are shown as subdistributed hazard ratio (line) with 95% CIs (shaded area), adjusted for education, smoking, diet, alcohol, and study survey; age, waist circumference (ie, adiposity), and physical activity are adjusted by inclusion in the eCRF formulas. Reference of the spline is set to the 10th percentile of the distribution of the group (ie, among females and males separately), and values are shown between the first and 99th percentile of the distribution of estimated cardiorespiratory fitness. Frequency refers to the frequency of observed eCRF. eCRF, estimated cardiorespiratory fitness; MI, myocardial infarction.

Figure 4

Restricted cubic spline Fine and Grey regressions of eCRF and risk of MI among females and males with (A) 2 or more CVD risk factors (n=9720, MI=857); (B) 3 or more CVD risk factors (n=5359, MI=600; and (C) 4 or more CVD risk factors (n=1982, MI=293): the Tromsø Study 1994-2014. The CVD risk factors included are smoking, hypertension, hyperlipidemia, obesity, poor diet, and physical inactivity. Data are shown as subdistributed hazard ratio (line) with 95% CIs (shaded area), adjusted for education, smoking, diet, alcohol, and study survey; age, waist circumference (ie, adiposity), and physical activity are adjusted by inclusion in the eCRF formulas. Reference of the spline is set to the 10th percentile of the distribution of the group (ie, among females and males separately), and values are shown between the first and 99th percentile of the distribution of eCRF. Frequency refers to the frequency of observed eCRF. CVD, cardiovascular disease; eCRF, estimated cardiorespiratory fitness; MI, myocardial infarction.