Association of LDL Cholesterol and Inflammation With Cardiovascular Events and Mortality in Hemodialysis Patients With Type 2 Diabetes Mellitus
Received 15 September 2008; accepted 15 June 2009. published online 27 September 2009.
Background
In the general population, C-reactive protein (CRP) in addition to low-density lipoprotein (LDL) cholesterol level is useful in predicting cardiovascular events. In hemodialysis patients, the additive value is unknown. The association between LDL cholesterol level and outcome previously was suggested to be inverse and confounded by inflammation.
Study Design
Prospective cohort study.
Setting & Participants
1,255 hemodialysis patients with type 2 diabetes mellitus randomly assigned to atorvastatin versus placebo in the German Diabetes Dialysis Study.
Predictors
Baseline LDL cholesterol level.
Outcomes & Measurements
Combined vascular end point (cardiac death, myocardial infarction, and stroke), mortality, myocardial infarction, sudden death, and stroke.
Results
During 4 years, 465 combined vascular events, 612 deaths, 160 sudden deaths, 200 myocardial infarctions, and 99 strokes occurred. Median LDL cholesterol level was 123 mg/dL. LDL cholesterol level (millimoles per liter and quartiles) was not predictive of outcome. This was analyzed further in patients with and without inflammation. In patients with inflammation (CRP level > 5 mg/L), the adjusted relative risk of combined vascular events was 29% greater compared with those without inflammation and a low LDL cholesterol level (LDL cholesterol ≤ 123 mg/dL). This was irrespective of whether LDL cholesterol level was low or high (hazard ratio [HR] for LDL < 123 mg/dL [HR for LDL≤123mg/dL], 1.29, with 95% confidence interval [CI], 0.98 to 1.70; HRLDL>123mg/dL, 1.29, with 95% CI, 0.99 to 1.69). Similar results were found for all-cause death (HRLDL≤123mg/dL, 1.47 [95% CI, 1.16 to 1.86]; HRLDL>123mg/dL, 1.48 [95% CI, 1.16 to 1.88]), sudden death (HRLDL≤123mg/dL, 1.98 [95% CI, 1.23 to 3.20]; HRLDL>123mg/dL, 1.66 [95% CI, 1.01 to 2.75]), and myocardial infarction (HRLDL≤123mg/dL, 1.74 [95% CI, 1.14 to 2.66]; HRLDL>123mg/dL, 1.54 [95% CI, 0.99 to 2.38]). In patients without inflammation, the respective risks did not differ significantly between patients with varying LDL cholesterol levels. However, there was a trend toward an increased risk of myocardial infarction (HRLDL>123mg/dL, 1.45 [95% CI, 0.95 to 2.21]) in patients with high compared with low LDL cholesterol levels. P values for the interaction between CRP and LDL cholesterol levels were 0.9 (composite vascular end point), 0.5 (mortality), 0.9 (sudden death), 0.09 (stroke), and 0.2 (myocardial infarction).
Limitations
Selected patient cohort, post hoc analysis.
Conclusion
Because CRP level more than LDL cholesterol level determined outcome, the value of regular LDL cholesterol measurements in long-term hemodialysis patients with type 2 diabetes needs reassessment.
4Synlab Center of Laboratory Diagnostics, Heidelberg, Germany
Address correspondence to Vera Krane, MD, University of Würzburg, Department of Medicine, Division of Nephrology, University Hospital, Josef-Schneider-Str 2, D-97080 Würzburg, Germany
ABSTRACT