Resistin is an adipokine secreted by macrophages and inflammatory cells linked to insulin resistance and inflammation. Leptin is an adipokine regulator of appetite and obesity. Although circulating levels of both have been associated with atherosclerosis, few data have reported their relation to coronary events in the context of statin therapy. This study measured on-statin levels of both resistin and leptin through enzyme-linked immunosorbent assay in a nested case–control cohort (n = 176 cases with coronary death, myocardial infarction, or unstable angina pectoris observed in follow-up matched 1:1 to 176 controls) derived from the Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis in Myocardial Infarction 22 study, a randomized controlled trial of atorvastatin 80 mg/day versus pravastatin 40 mg/day in patients with a recent acute coronary syndrome. Resistin demonstrated a moderate association with high-sensitivity C-reactive protein (hsCRP; Spearman rho = 0.25, p <0.0001). On-statin resistin levels were linked to recurrent coronary events in conditional logistic regression analysis adjusted for additional risk factors including hsCRP and history of diabetes (tertile 3 vs 1 adjusted odds ratio 2.08; 95% confidence interval [CI] 1.04 to 4.19). An additive risk was noted when patients were stratified by resistin and glycated hemoglobin levels. In contrast, leptin levels were associated with obesity, diabetes, triglycerides, and hsCRP (p <0.001 for each) but demonstrated no association with recurrent coronary events (tertile 3 vs 1 adjusted odds ratio 0.72; 95% CI 0.28 to 1.83). In conclusion, on-statin resistin, but not leptin, is an independent marker of residual risk for recurrent coronary events in patients after hospitalization for an acute coronary syndrome.
Inflammatory pathways and adipocyte biology as they relate to atherosclerosis have generated substantial interest in recent years. Resistin is a secreted protein initially linked to insulin resistance in a rodent model with subsequent analyses suggesting a key additional role in monocyte inflammatory pathways. Leptin is a hormone secreted by adipocytes that is increased by fat storage, thus regulating both appetite and metabolic handling. Both resistin and leptin have been linked to atherosclerotic phenotypes and coronary events. However, little is known about the relation between these markers with recurrent cardiovascular events when assessed in the context of statin therapy. We therefore measured on-statin plasma levels of both resistin and leptin in a nested case–control cohort of patients derived from the Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis in Myocardial Infarction 22 (PROVE IT-TIMI 22) clinical trial to investigate their relation with recurrent coronary events.
Materials and Methods
A nested case–control study was derived from participants in the PROVE IT-TIMI 22 study ( ClinicalTrials.gov : NCT00382460 ), the design and results of which have been published previously. In brief, PROVE IT-TIMI 22 was a double-blinded clinical trial that randomized 4,162 patients hospitalized for an acute coronary syndrome to either intensive (atorvastatin 80 mg/day) or moderate (pravastatin 40 mg/day) statin therapy with mean follow-up of 24 months.
A nested case–control population was created in which participants who subsequently suffered death from coronary heart disease, myocardial infarction, or unstable angina requiring hospitalization after 4-month therapy (n = 176 cases) were matched in a 1:1 ratio to patients who remained free of recurrent coronary events (n = 176 controls). All events were adjudicated by an independent clinical event committee. Cases and controls were matched according to age (within 5 years), gender, smoking status, and treatment group (atorvastatin vs pravastatin).
Venous blood was obtained in ethylenediaminetetraacetic acid–containing tubes at time of randomization and 4 months as part of the study protocol with plasma samples stored at −80°C. Plasma renal function and glycated hemoglobin were measured at baseline; lipid levels and high-sensitivity C-reactive protein (hsCRP) were assessed at the 4-month visit as reported previously.
Plasma levels of resistin and leptin measurements were performed on samples from the 4-month visit using commercially available enzyme-linked immunosorbent assays (Linco Research, St. Charles, Missouri). Review of medication records confirmed that 351 of 352 (99.7%) of the cohort remained on statin therapy at the time of this blood draw. The intra- and inter-assay coefficients of variation (CVs) for pooled human plasma were 3.5% and 4.8% for resistin and 2.5% and 2% for leptin, respectively.
Patient variables were reported as n (%) for categorical variables and median (twenty-fifth to seventy-fifth percentile) for continuous variables. Baseline characteristics were compared using the chi-square test or the Wilcoxon rank sum test for categorical and continuous variables, respectively. Spearman correlation coefficients were used to evaluate the relation between on-statin resistin and leptin with continuous patient variables and biomarkers.
The relation between on-statin resistin and leptin levels and recurrent coronary events was assessed using conditional logistic regression analysis with the cohort divided into tertiles according to each biomarker. Subsequent multivariable adjustment models included (1) index event (STEMI, NSTEMI, or unstable angina), body mass index, history of hypertension, history of diabetes, renal impairment, and on-statin levels of low-density lipoprotein cholesterol (LDL-C) and hsCRP. An additional sensitivity analysis substituted glycated hemoglobin value for history of diabetes. Odds ratios and tests of interaction with additional parameters were additionally calculated with biomarkers dichotomized as the top tertile versus the first and the second tertiles. Given a relation with insulin-resistance phenotypes, the potential additive effects of increased levels of on-statin resistin and baseline glycated hemoglobin were assessed using a 2 × 2 conditional logistic regression analysis after dichotomizing participants as less than or equal to or greater than median value of each biomarker.
All reported p values were 2 tailed with confidence intervals (CIs) computed at the 95% level. Statistical analysis was performed using STATA/SE (StataCorp, College Station, Texas).
Results
The mean age in the substudy population was 61 years, and 77% were men. A total of 36% were current smokers, 56% had a history of hypertension, 39% were obese (body mass index >30 kg/m 2 ), and 24% had a history of diabetes. Compared to matched controls, participants who went on to have a recurrent coronary event had a higher prevalence of baseline diabetes, history of a previous myocardial infarction, and had higher levels of on-statin hsCRP ( Table 1 ).
| Variable | Overall Substudy Population (n = 352) | Case (n=176) | Control (n= 176) | P-Value |
|---|---|---|---|---|
| Age (years) | 61 (53 – 69) | 61 (53 – 69) | 61 (53 – 69) | Matched |
| Men | 272 (77%) | 136 (77%) | 136 (77%) | Matched |
| Atorvastatin Treatment | 146 (42%) | 73 (42%) | 73 (42%) | Matched |
| Smoker | 128 (36%) | 64 (36%) | 64 (36%) | Matched |
| Hypertension | 196 (56%) | 107 (61%) | 89 (51%) | 0.053 |
| BMI > 30 kg/m 2 | 131 (39%) | 63 (38%) | 68 (40%) | 0.67 |
| BMI, kg/m 2 | 29 (26 – 32) | 29 (26 – 32) | 29 (26 – 32) | 0.65 |
| Diabetes Mellitus | 86 (24%) | 54 (31%) | 32 (18%) | 0.006 |
| Glycated Hemoglobin (%) | 5.5 (5.1 – 6.2) | 5.6 ( 5.1 – 6.5) | 5.4 (5.1 – 5.9) | 0.07 |
| Prior Myocardial Infarction | 90 (26%) | 65 (37%) | 25 (14%) | < 0.001 |
| CrCl < 60 | 51 (16%) | 31 (19%) | 20 (12%) | 0.09 |
| Index Event | 0.002 | |||
| STEMI | 115 (33%) | 46 (26%) | 69 (39%) | |
| NSTEMI | 112 (32%) | 52 (30%) | 60 (34%) | |
| Unstable Angina Pectoris | 125 (36%) | 78 (44%) | 47 (27%) | |
| On-statin LDL-C (mg/dl) | 84.0 (63 – 112) | 84 (66 – 108) | 85 (62 – 113) | 0.87 |
| On-statin HDL-C (mg/dl) | 43 (36 – 51) | 41 (35 – 51) | 43 (37 – 51) | 0.16 |
| On-statin TG (mg/dl) | 132 (94 – 204) | 133 (90 – 219) | 131 (99 – 189) | 0.89 |
| On-statin Log hsCRP (mg/L) | 0.2 (0.1 – 0.5) | 0.3 (0.1 – 0.6) | 0.2 (0.1 – 0.5) | 0.01 |
| Recurrent Clinical Event | ||||
| Death from CHD | 41 (23%) | — | ||
| MI | 85 (48%) | — | ||
| Unstable Angina Pectoris | 50 (28%) | — |
Plasma resistin levels, measured after 4 months of statin therapy, were slightly higher in women and those with a history of hypertension but did not differ according to smoking status, treatment arm, or history of diabetes ( Table 2 ). Resistin levels were largely independent of on-statin lipid levels but moderately associated with hsCRP (Spearman rho = 0.25; p <0.0001; Table 3 ). Trends across resistin tertiles of baseline and on-statin biomarkers demonstrated similar trends ( Supplementary Table 1 ).
| Resistin | P-Value | Leptin | P-Value | ||
|---|---|---|---|---|---|
| Sex | Male | 16.7 (13.2 – 22.3) | 0.02 | 8.6 (5.3 – 13.9) | < 0.0001 |
| Female | 18.5 (15.4 – 23.9) | 25 (16.6 – 37.3) | |||
| Treatment Arm | Atorvastatin | 17.3 (13.5 – 22.2) | 0.30 | 10.5 (5.8 – 17.9) | 0.69 |
| Pravastatin | 17.8 (14.2 – 23.0) | 10.2 (6.3 – 19.7) | |||
| Smoker | Yes | 17.7 (12.8 – 22.5) | 0.55 | 9.5 (5.4 – 17.5) | 0.21 |
| No | 17.4 (14.1 – 22.9) | 10.7 (6.4 – 20.0) | |||
| Hypertension | Yes | 18.4 (12.8 – 22.5) | 0.03 | 12.7 (7.4 – 23.4) | 0.0001 |
| No | 16.2 (12.5 – 22.1) | 8.7 (4.7 – 14.8) | |||
| BMI > 30 kg/m 2 | Yes | 18.1 (13.7 – 22.2) | 0.79 | 16.5 (9.7 – 27.4) | < 0.0001 |
| No | 17.2 (14.1 – 23.0) | 12.7 (6.9 – 22.7) | |||
| Diabetes Mellitus | Yes | 17.9 (13.9 – 22.2) | 0.92 | 12.9 (8.1 – 27.6) | 0.0004 |
| No | 17.5 (13.8 – 22.9) | 9.7 (5.5 – 17.7) | |||
| Prior Myocardial Infarction | Yes | 17.8 (14.1 – 23.4) | 0.60 | 11.0 (6.8 – 18.5) | 0.71 |
| No | 17.5 (13.8 – 22.5) | 10.1 (6.1 – 18.4) |
| Resistin | P-Value | Leptin | P-Value | |
|---|---|---|---|---|
| Age | 0.10 | 0.06 | 0.01 | 0.90 |
| BMI (kg/m 2 ) | 0.02 | 0.67 | 0.48 | <0.0001 |
| Glycated Hemoglobin (%) | 0.02 | 0.74 | 0.28 | <0.0001 |
| Creatinine Clearance | −0.14 | 0.02 | 0.09 | 0.11 |
| On-statin LDL-C (mg/dl) | −0.02 | 0.67 | 0.06 | 0.26 |
| On-statin HDL-C (mg/dl) | −0.10 | 0.07 | 0.05 | 0.37 |
| On-statin TG (mg/dl) | 0.01 | 0.83 | 0.23 | <0.0001 |
| On-statin hsCRP (mg/L) | 0.25 | < 0.0001 | 0.27 | < 0.0001 |
| Leptin | 0.16 | 0.003 | — | — |
| Resistin | — | — | 0.16 | 0.003 |
On-statin resistin levels in the third tertile were associated with an increased risk of recurrent coronary events in an unadjusted conditional logistic regression analysis (T2 vs T1 odds ratio 1.19; 95% CI 0.70 to 2.03; T3 vs T1 1.96; 95% CI 1.12 to 3.45, p = 0.02; Table 4 ). Subsequent multivariate analyses showed no significant effect attenuation after additional adjustment for index event, body mass index, history of hypertension, renal impairment, on-statin levels of LDL-C and hsCRP, and either history of diabetes (T3 vs T1 odds ratio 2.08; 95% CI 1.04 to 4.19; p = 0.04) or glycated hemoglobin modeled as a continuous variable (T3 vs T1 odds ratio 2.11; 95% CI 1.05 to 4.26; p = 0.02).
| Range (ng/mL) | Resistin Tertile | ||
|---|---|---|---|
| T1 (n=118) | T2 (n=117) | T3 (n=117) | |
| 2 – < 14.8 | ≥ 14.8 – < 21.1 | ≥ 21.1 – 85 | |
| Model One | Ref | 1.19 | 1.96 |
| 95% CI | (0.70 – 2.03) | (1.12 – 3.45) | |
| P value | P = 0.52 | P = 0.02 | |
| Model Two | Ref | 1.10 | 2.08 |
| 95% CI | (0.55 – 2.20) | (1.04 – 4.19) | |
| P value | P = 0.78 | P = 0.04 | |
| Model Three | Ref | 1.13 | 2.11 |
| 95% CI | (0.56 – 2.26) | (1.05 – 4.26) | |
| P value | P = 0.74 | P = 0.04 | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
