Author + information
Hypertensive patients are at increased risk of stroke. Although low baseline HDL levels have been associated with a higher risk of stroke, whether changing levels of HDL over time are more strongly related to the risk of stroke has not been examined.
Incident stroke was examined in relation to baseline and in-treatment HDL levels prior to development of stroke in 8606 hypertensive patients with baseline HDL levels randomly assigned to losartan- or atenolol-based treatment. HDL levels at baseline and each year of testing were categorized into quartiles according to baseline HDL levels.
During 4.7±1.1 years follow-up, a new stroke occurred in 502 patients (5.8%). In univariate analyses, compared with HDL >1.78 mMol/L, patients with in-treatment HDL <1.22 entered as a time-varying covariate had an 80% greater risk of new stroke; patients with in-treatment HDL in the 2nd or 3rd quartiles had intermediate but nonsignificant increased risks of having a stroke. Baseline HDL was only a weak predictor of new stroke. In multivariate Cox analyses adjusting for multiple stroke risk factors, there was no attenuation of the risk of new stroke associated with low in-treatment HDL; baseline HDL was not a significant predictor of stroke in multivariate analysis.
Lower in-treatment HDL is more strongly associated with risk of new stroke than low HDL levels at baseline. These findings suggest that serial assessment of HDL can better estimate stroke risk in hypertensive patients.
|Hazard Ratios (95% Confidence Interval)|
|Analysis||Quartile 1||Quartile 2||Quartile 3||Quartile 4|
|HDL <1.22||HDL 1.22–1.47||HDL 1.48–1.78||HDL >1.78|
|Univariate Cox Model|
|Baseline HDL||1.34 (1.04–1.71)||1.19 (0.91–1.54)||1.11 (0.85–1.45)||1|
|In-treatment HDL||1.80 (1.36–2.38)||1.26 (0.94–1.69)||1.00 (0.73–1.37)||1|
|Multivariate Cox Model*|
|Baseline HDL||1.21 (0.91–1.60)||1.20 (0.91–1.59)||1.13 (0.85–1.50)||1|
|In-treatment HDL||1.78 (1.30–2.43)||1.31 (0.95–1.79)||1.04 (0.75–1.46)||1|
↵* Adjusted for randomized treatment, baseline age, sex, race, diabetes, history of stroke, MI, ischemic heart disease, heart failure or atrial fibrillation, prevalent atrial fibrillation, serum glucose and creatinine, and urine albumin/creatinine ratio treated as standard covariates, and incident MI, in-treatment non-HDL cholesterol, heart rate, Cornell product LVH, diastolic and systolic pressure, and statin use treated as time- varying covariates
Poster Sessions, Expo North
Saturday, March 09, 2013, 10:00 a.m.-10:45 a.m.
Session Title: Coronary Risk Factors and Management
Abstract Category: 28. Quality of Care and Outcomes Assessment
Presentation Number: 1115-105
- 2013 American College of Cardiology Foundation