Predictive performances of lipid accumulation product vs. adiposity measures for cardiovascular diseases and all-cause mortality, 8.6-year follow-up: Tehran lipid and glucose study

Abstract

Background: The body mass index (BMI) is the most commonly used marker for evaluating obesity related risks, however, central obesity measures have been proposed to be more informative. Lipid accumulation product (LAP) is an alternative continuous index of lipid accumulation. We sought in this study to assess if LAP can outperform BMI, waist-to-height-ratio (WHtR), or waist-to-hip-ratio (WHpR) in predicting incident cardiovascular disease (CVD) or all-cause mortality.

Results: Among participants of Tehran Lipid and Glucose Study, 6,751 participants (2,964 men), aged ≥ 30 years, were followed for a median of 8.6 years. We observed 274 deaths (men: 168) and 447 CVD events (men: 257). Levels of common CVD risk factors significantly increased across LAP quartiles. Mortality rates did not differ by LAP quartiles. Among participants free of CVD at baseline [6331 (2,741 men)], CVD incident rates per 1000 person increased in a stepwise fashion with increasing LAP quartile values in both men (from 6.9 to 17.0) and women (from 1.3 to 13.0), (Ps < 0.001). Among women, a 1-SD increment in log-LAP conferred a 41% increased risk for CVD (HR 1.41, 95% CIs 1.02-1.96). Among men, however, LAP was not observed to be independently associated with increased risk of CVD; except in a sub-group of men assigned to the lifestyle modification interventions, where, LAP predicted CVD risk. After adjustment with CVD risk factors LAP turned to be inversely associated with risk of all-cause mortality (HR, men 0.74, 95% CIs 0.61-0.90; women, 0.94 95% CIs 0.74-1.20). Among women, magnitude of increased risk of CVD due to LAP was not different from those of anthropometric measures. Among men, however, WHpR was observed to be more strongly associated with increased risk of CVD than was LAP. Among neither men nor women were the predictive performances (discrimination, calibration, goodness-of-fit) of the LAP better than those of different anthropometric measures were.

Conclusions: If LAP is to be used for predicting CVD, it might not be superior to WHtR or WHpR.

Figure 1

Kaplan-Meier curves for incident CVD across LAP quartiles in men with global CVD risk less than 20%.

Figure 2

Kaplan-Meier curves for incident CVD across LAP quartiles in men with global CVD risk 20% or more.

Figure 3

Kaplan-Meier curves for incident CVD across LAP quartiles in women with global CVD risk less than 20%.

Figure 4

Kaplan-Meier curves for incident CVD across LAP quartiles in women with global CVD risk 20% or more.

Figure 5

Kaplan-Meier curves for all-cause mortality across LAP quartiles in men with global CVD risk less than 20%.

Figure 6

Kaplan-Meier curves for all-cause mortality across LAP quartiles in men with global CVD risk 20% or more.

Figure 7

Kaplan-Meier curves for all-cause mortality across LAP quartiles in women with global CVD risk less than 20%.

Figure 8

Kaplan-Meier curves for all-cause mortality across LAP quartiles in women with global CVD risk 20% or more.