Increased platelet activity is associated with adverse cardiovascular events. The mean platelet volume (MPV) correlates with platelet activity; however, the relation between the MPV and long-term mortality in patients undergoing percutaneous coronary intervention (PCI) is not well established. Furthermore, the role of change in the MPV over time has not been previously evaluated. We evaluated the MPV at baseline, 30 days, 60 days, 90 days, 1 year, 2 years, and 3 years after the procedure in 1,512 patients who underwent PCI. The speed of change in the MPV was estimated using the slope of linear regression. Mortality was determined by query of the Social Security Death Index. During a median of 8.7 years, mortality was 49.3% after PCI. No significant difference was seen in mortality when stratified by MPV quartile (first quartile, 50.1%; second quartile, 47.7%; third quartile, 51.3%; fourth quartile, 48.3%; p = 0.74). For the 839 patients with available data to determine a change in the MPV over time after PCI, mortality was 49.1% and was significantly greater in patients with an increase (52.9%) than in those with a decrease (44.2%) or no change (49.1%) in the MPV over time (p <0.0001). In conclusion, no association was found between the baseline MPV and long-term mortality in patients undergoing PCI. However, increased mortality was found when the MPV increased over time after PCI. Monitoring the MPV after coronary revascularization might play a role in risk stratification.
Platelets play a pivotal role in the development of atherosclerotic lesions, plaque destabilization, and atherothrombosis. Antiplatelet therapy reduces both procedural-related complications and subsequent ischemic cardiovascular events after coronary revascularization. However, despite targeted therapy, ≤1/4 of patients after percutaneous coronary intervention (PCI) demonstrate high platelet activity and a greater rate of major adverse cardiovascular events compared to patients without high platelet activity. The methods of testing platelet activity can be time consuming, expensive, and technically difficult. The mean platelet volume (MPV), a marker of platelet size and activity, is a predictor of cardiovascular risk and quantified on automated hemograms routinely measured before coronary revascularization. In the setting of PCI, a high preprocedural MPV is associated with increased mortality in patients with ST-segment elevation myocardial infarction and an increased rate of restenosis after coronary angioplasty. In the present study, we aimed to determine whether the preprocedural MPV or the change in MPV over time could be a predictor of long-term mortality in unselected patients undergoing PCI.
Methods
The present study was a retrospective analysis of 1,625 patients referred from 8 facilities in Pennsylvania, New Jersey, and New York, who had undergone PCI from January 1993 to October 2005 at the Manhattan Campus of the Veterans Affairs New York Harbor Healthcare System. The baseline MPV was measured within 2 days of the procedure and at 30, 60, and 90 days and 1, 2, and 3 years after PCI, where available. Patients with missing baseline MPV data (n = 86), a baseline platelet count of <100,000 (n = 15) or >500,000 (n = 7), or unknown survival status after PCI (n = 5) were excluded, leaving data from 1,512 patients (93%) for analysis. Of this cohort, 673 (44.5%) had insufficient data to determine whether the MPV had changed over time after PCI. The 839 patients with sufficient data (≥3 post-PCI MPV values) to determine the direction of change in MPV over time after PCI were divided into 3 groups: increasing MPV (n = 157), decreasing MPV (n = 120), and no change in MPV (n = 562) over time.
The baseline characteristics and laboratory data were obtained from the cardiac catheterization laboratory database and from a chart review of the Department of Veterans Affairs computerized patient records. The patients were classified as having diabetes mellitus if they had ever been treated with insulin or oral hypoglycemic agents, had a fasting glucose level >126 mg/dl on 2 separate occasions, or a random glucose value of >200 mg/dl had been documented on ≥1 occasion before the present admission. All other co-morbidities shown as baseline characteristics were ascertained by chart review of physician documentation of the patients’ medical history. The primary end point, defined as all-cause mortality after PCI, was determined by query of both the Social Security Death Benefit Index and the Veterans Affairs computerized patient records.
The patients were grouped according to the baseline MPV quartile (first quartile, MPV <8.2 fL; second quartile, MPV 8.2 to 8.8 fL; third quartile, MPV 8.8-9.4 fL; fourth quartile, MPV >9.4 fL). Differences in the baseline characteristics were evaluated using the chi-square test for categorical variables. Normality was examined using histograms, quantile-quantile plots, and the Shapiro-Wilk test. Normally and non-normally distributed continuous variables were analyzed using 1-way analysis of variance and the Kruskal-Wallis test, respectively. Long-term mortality, stratified by the MPV quartile, was compared using the chi-square test.
To examine the trend in post-PCI MPV over time, the speed of change in the MPV was estimated using the slope of a linear mixed effects regression model. The analysis was performed on a log scale to ensure that the corresponding residuals were gaussian. An increase or decrease in the MPV was inferred from the analysis of the slopes, with p = 0.05 as the cutoff. Because the estimates of mortality corresponding to increasing, decreasing, and no change in MPV did not change when the p value cutoff was relaxed to 0.1, the estimates of slope that passed the 0.1 significance level were considered statistically significant. To examine the sensitivity of the results of this trend analysis, we examined a more stringent definition of the cutoff by requiring ≥4 post-PCI MPV values. Long-term mortality was compared between the increasing MPV group and decreasing or no change MPV groups using a binomial test of proportions. The survivor distribution is presented using Kaplan-Meier curves and compared using a 1-sided log-rank test.
To explore the relation between increasing MPV and nonincreasing MPV (decreasing or no change in MPV) and long-term mortality, logistic regression analysis was used, with mortality as the dichotomous outcome and increasing MPV as the main predictor. We built models hierarchically to include clinical factors that affect mortality and/or MPV. First, we added the demographic variable of age, followed by race (gender was not included because 99% of the study population were men), into the logistic regression model. This was followed by sequentially adding diabetes and peripheral vascular disease to the logistic regression model and, finally, creatinine and tobacco use. Because the baseline MPV was not different among the increasing, decreasing, and no change groups, and baseline MPV was not a predictor of mortality in this cohort, it was not included in the model.
Statistical analysis was conducted using the Statistical Package for Social Sciences, version 19.0 (SPSS, Chicago, Illinois) and R program for Scientific Computing (available at: www.r-project.org ). The Veterans Affairs human studies committee approved the present protocol.
Results
The baseline characteristics by MPV quartile are listed in Table 1 . Mortality during follow-up after PCI was 49.3% (median follow-up 8.7 years). No significant difference was found in long-term mortality when stratified by baseline MPV quartile (first quartile, 50.1%, second quartile, 47.7%, third quartile, 51.3%, and fourth quartile, 48.3%; p = 0.744). Similarly, the 1-year (first quartile, 8.1%, second quartile, 5.7%, third quartile, 7.0%, and fourth quartile, 7.2%; p = 0.625) and 3-year (first quartile, 16.5%, second quartile, 14.2%, third quartile, 16.3%, and fourth quartile, 15.7%; p = 0.821) mortality did not differ across the baseline MPV quartiles.
| Variable | Total Cohort (n = 1,512) | MPV <8.2 fL (n = 369) | MPV 8.2–8.8 fL (n = 386) | MPV 8.8–9.4 fL (n = 355) | MPV >9.4 fL (n = 402) | p Value ∗ |
|---|---|---|---|---|---|---|
| Age (yrs) | 66 (57–74) | 65 (57–74) | 66 (57–75) | 66 (57–75) | 67 (58–74) | 0.76 |
| Men | 1,499 (99%) | 367 (99.5%) | 382 (99%) | 351 (99%) | 399 (99%) | 0.82 |
| Race | ||||||
| White | 979 (65%) | 267 (72%) | 258 (67%) | 217 (61%) | 237 (59%) | 0.002 |
| Black | 304 (20%) | 55 (15%) | 70 (18%) | 74 (21%) | 105 (26%) | |
| Hispanic | 197 (13%) | 37 (10%) | 52 (14%) | 57 (16%) | 51 (13%) | |
| Body mass index (kg/m 2 ) | 28.2 (25.0–31.8) | 27.1 (24.3–31.3) | 28.2 (25.0–31.5) | 28.4 (25.3–31.9) | 28.5 (25.1–32.2) | 0.02 |
| Diabetes mellitus | 603 (40%) | 120 (33%) | 138 (36%) | 150 (42%) | 195 (49%) | <0.001 |
| Hypertension | 1,158 (77%) | 280 (76%) | 287 (74%) | 280 (79%) | 311 (77%) | 0.49 |
| Previous myocardial infarction | 455 (30%) | 119 (32%) | 105 (27%) | 104 (29%) | 127 (32%) | 0.27 |
| Previous coronary artery bypass surgery | 231 (15%) | 56 (15%) | 58 (15%) | 53 (15%) | 64 (16%) | 0.96 |
| Previous percutaneous coronary intervention | 246 (16%) | 67 (18%) | 58 (15%) | 61 (17%) | 60 (15%) | 0.61 |
| Heart failure | 166 (11%) | 31 (8%) | 41 (11%) | 40 (11%) | 54 (13%) | 0.16 |
| Peripheral vascular disease | 226 (15%) | 58 (16%) | 52 (14%) | 54 (15%) | 62 (15%) | 0.82 |
| Hyperlipidemia | 870 (58%) | 207 (56%) | 234 (61%) | 202 (57%) | 227 (57%) | 0.44 |
| Current smoker | 409 (27%) | 95 (26%) | 109 (28%) | 97 (27%) | 108 (27%) | 0.94 |
| Previous aspirin use | 862 (57%) | 203 (55%) | 231 (60%) | 195 (55%) | 233 (58%) | 0.87 |
| Previous statin use | 632 (42%) | 142 (39%) | 167 (43%) | 153 (43%) | 170 (42%) | 0.54 |
| Vessels narrowed (n) | 0.46 | |||||
| 1 | 464 (31%) | 107 (29%) | 118 (31%) | 103 (29%) | 136 (34%) | |
| 2 | 399 (26%) | 107 (29%) | 99 (26%) | 92 (26%) | 101 (25%) | |
| 3 | 369 (24%) | 87 (24%) | 102 (26%) | 93 (26%) | 87 (22%) | |
| Platelet count (× 10 9 /L) | 212 (172–257) | 239 (196–292) | 216 (184–260) | 207 (173–249) | 184 (152–222) | <0.001 |
| Urea (mg/dl) | 17 (13–21) | 16 (13–21) | 17 (13–21) | 17 (13–22) | 17 (13–22) | 0.71 |
| Creatinine (mg/dl) | 1.1 (0.9–1.3) | 1.1 (0.9–1.3) | 1.1 (0.9–1.3) | 1.1 (0.9–1.3) | 1.1 (1.0–1.3) | 0.031 |
| Glucose (mg/dl) | 108 (95–137) | 105 (93–126) | 106 (94–132) | 110 (94–140) | 114 (97–153) | <0.001 |
| Cholesterol (mg/dl) | 179 (156–209) | 183 (157–212) | 181 (158–214) | 176 (156–209) | 176 (153–202) | 0.11 |
| High-density lipoprotein cholesterol (mg/dl) | 37 (32–44) | 36 (32–44) | 39 (32–45) | 37 (32–44) | 37 (31–44) | 0.13 |
| Low-density lipoprotein cholesterol (mg/dl) | 109 (89–132) | 112 (90–135) | 112 (89–137) | 108 (88–126) | 107 (84–129) | 0.08 |
| Triglycerides (mg/dl) | 144 (99–209) | 147 (101–210) | 143 (101–218) | 141 (94–213) | 144 (100–195) | 0.86 |
Data were available to determine the direction of change in MPV over time after PCI in 839 patients (55.5%). When comparing the baseline characteristics between patients with (n = 839) and without (n = 673) data available to determine the changes in MPV over time after PCI, the patients with available data were older (67 vs 65 years, p = 0.003), with a greater prevalence of diabetes mellitus (43% vs 37%, p = 0.02) and hypertension (79% vs 74%, p = 0.017). The patients with available data also had a lower prevalence of tobacco use (26% vs 32%, p = 0.021) and lower platelet levels (205 vs 217 × 10 9 /L, p = 0.014) compared to the patients without available data. Other baseline characteristics and mortality during follow-up after PCI (49.1% vs 49.5%, p = 0.918) were similar for the patients with versus without sufficient data available.
The baseline characteristics of the patients with available data to determine a change in MPV over time after PCI are listed in Table 2 (19% had increasing MPV, 14% had decreasing MPV, and 67% had no change in MPV). Mortality during follow-up was significantly greater in the patients with an increase in the MPV over time after PCI (increase in MPV, 52.9%; decrease in MPV, 44.2%; and no change in MPV, 49.1%; p <0.0001 between patients with an increase vs no increase in MPV). When evaluated over time, mortality remained significantly greater in patients with increasing versus nonincreasing (decreasing or no observed change) in MPV ( Figure 1 ). The average annual increase in MPV was 1.02 ± 0.12 fL.
| Variable | Increasing MPV (n = 157) | Decreasing MPV (n = 120) | No Change in MPV (n = 562) | p Value |
|---|---|---|---|---|
| Age (yrs) | 65 (56–75) | 68 (60–75) | 67 (58–75) | 0.58 |
| Men | 154 (98%) | 120 (100%) | 556 (99%) | 0.31 |
| Race | ||||
| White | 100 (64%) | 69 (58%) | 384 (68%) | 0.24 |
| Black | 33 (21%) | 33 (28%) | 107 (19%) | |
| Hispanic | 22 (14%) | 15 (13%) | 60 (11%) | |
| Body mass index (kg/m 2 ) | 28.2 (25.4–31.9) | 28.2 (25.4–31.9) | 28.1 (24.9–32.2) | 0.90 |
| Diabetes mellitus | 65 (41%) | 48 (40%) | 244 (43%) | 0.75 |
| Hypertension | 119 (76%) | 100 (83%) | 444 (79%) | 0.31 |
| Previous myocardial infarction | 52 (33%) | 29 (24%) | 172 (31%) | 0.38 |
| Previous coronary artery bypass surgery | 27 (17%) | 13 (11%) | 100 (18%) | 0.19 |
| Previous percutaneous coronary intervention | 27 (17%) | 13 (11%) | 95 (17%) | 0.27 |
| Heart failure | 15 (10%) | 14 (12%) | 73 (13%) | 0.46 |
| Peripheral vascular disease | 40 (26%) | 18 (15%) | 77 (14%) | 0.003 |
| Hyperlipidemia | 96 (61%) | 65 (64%) | 332 (59%) | 0.38 |
| Current smoker | 41 (26%) | 33 (28%) | 134 (24%) | 0.59 |
| Previous aspirin use | 82 (52%) | 71 (59%) | 307 (55%) | 0.14 |
| Previous statin use | 68 (43%) | 48 (40%) | 246 (44%) | 0.59 |
| Vessels narrowed (n) | 0.17 | |||
| 1 | 51 (33%) | 29 (24%) | 166 (30%) | |
| 2 | 39 (25%) | 41 (34%) | 145 (26%) | |
| 3 | 36 (23%) | 24 (20%) | 149 (27%) | |
| Baseline mean platelet volume (fL) | 8.6 (8.1–9.4) | 8.9 (8.2–9.3) | 8.8 (8.2–9.5) | 0.22 |
| Platelet count (× 10 9 /L) | 218 (172–279) | 204 (173–243) | 203 (166–253) | 0.08 |
| Urea (mg/dl) | 17 (13–24) | 16 (13–20) | 17 (13–22) | 0.23 |
| Creatinine (mg/dl) | 1.1 (0.9–1.4) | 1.0 (0.9–1.3) | 1.1 (1.0–1.3) | 0.07 |
| Glucose (mg/dl) | 114 (97–137) | 110 (93–138) | 109 (95–141) | 0.62 |
| Cholesterol (mg/dl) | 178 (156–213) | 179 (151–207) | 179 (154–207) | 0.94 |
| High-density lipoprotein cholesterol (mg/dl) | 37 (32–44) | 38 (32–44) | 37 (32–44) | 0.98 |
| Low-density lipoprotein cholesterol (mg/dl) | 108 (83–135) | 110 (80–132) | 109 (87–132) | 0.90 |
| Triglycerides (mg/dl) | 142 (93–211) | 140 (97–197) | 145 (101–213) | 0.78 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
