Association between platelet/high-density lipoprotein cholesterol ratio and incidence of coronary heart disease: insights from the UK Biobank

The platelet-to-high-density lipoprotein cholesterol ratio (PHR) integrates platelet count and HDL-C concentration and may reflect thrombo-inflammatory and lipid-related cardiovascular risk. However, its association with incident coronary heart disease (CHD) and its incremental value beyond its individual components remain uncertain. We included 326,803 UK Biobank participants without baseline CHD. PHR was calculated as platelet count (×109 cells/L) divided by HDL-C concentration (mmol/L) and analyzed in quartiles and as a log-transformed continuous variable. Incident CHD was defined as the first occurrence of angina pectoris, acute myocardial infarction (AMI), or chronic ischemic heart disease. Cox proportional hazards models were used to estimate hazard ratios (HRs) after adjustment for demographic, socioeconomic, lifestyle, and clinical covariates. Restricted cubic spline models were applied to examine non-linearity. Discriminatory performance and incremental model information were evaluated using ROC/C-index analyses and likelihood ratio tests. Exploratory mediation-oriented analyses examined hs-CRP, HbA1c, and neutrophil count. During a median follow-up of 13.0 years, 20,340 incident CHD events occurred. Compared with participants in the lowest PHR quartile, those in the highest quartile had a higher hazard of incident CHD in the fully adjusted model (HR 1.36, 95% CI 1.31–1.42; P < 0.01). The association was also observed for CHD subtypes and appeared stronger for AMI (Q4 vs. Q1: HR 1.68, 95% CI 1.56–1.81) than for chronic ischemic heart disease (Q4 vs. Q1: HR 1.38, 95% CI 1.32–1.45). Restricted cubic spline analysis suggested a non-linear association between log-transformed PHR and incident CHD (P for non-linearity < 0.001), with a relatively flat association at lower values and a steeper increase at higher values. PHR showed moderate discrimination for incident CHD (AUC 0.716, 95% CI 0.713–0.720), and adding PHR to the fully adjusted clinical model produced a small increase in C-index (0.714 to 0.720), larger than that observed when platelet count or HDL-C was added separately. Exploratory analyses suggested that hs-CRP, HbA1c, and neutrophil count statistically accounted for part of the observed association, although these findings may reflect shared inflammatory and metabolic background rather than causal mediation. Higher PHR was associated with increased risk of incident CHD in this large prospective cohort, particularly for AMI. PHR may serve as a simple adjunctive marker for CHD risk stratification when interpreted alongside established clinical risk factors.