The significance of anemia in patients with acute myocardial infarction (AMI) treated with percutaneous coronary intervention (PCI) remains controversial. The aim of the present study was to evaluate the effect of anemia on the short- and long-term prognosis of patients with AMI treated with PCI, including high-risk subgroups. The study group consisted of 1,497 consecutive patients with AMI treated in the acute phase with PCI. Anemia was defined using World Health Organization criteria (hemoglobin level <13 g/dl for men and <12 g/dl for women). The study population was divided into 2 major groups (patients with [n = 248, 16.6%] and without [n = 1,249, 83.4%] anemia) and 6 subgroups (diabetes mellitus, impaired renal function, age >70 years, left ventricular dysfunction, incomplete revascularization, and multivessel disease). A comparative analysis was performed between both groups within the whole population and within the particular subgroups. Significantly greater 30-day (13.2% vs 7.3%), 1-year (20.5% vs 11.3%), and total (24.1% vs 12.7%; all p <0.05) mortality rates were observed in the anemic group. Multivariate analysis identified anemia as an independent predictor of any-cause death in the whole population during the observation period (covariate-adjusted hazard ratio 1.46, 95% confidence interval 1.31 to 1.61, p <0.05). Anemia was significantly associated with excessive long-term mortality in the multivessel disease group (adjusted hazard ratio 1.54, 95% confidence interval 1.34 to 1.74) and in the incomplete revascularization group (hazard ratio 1.67, both p <0.05). In conclusion, anemia on admission in patients with AMI treated in the acute phase with PCI was independently associated with increasing short- and long-term mortality, especially in the subgroups with incomplete revascularization and multivessel disease.
Anemia is a common co-morbidity in patients with cardiovascular disease. It can be found in about 15% of patients with acute myocardial infarction (AMI) and in ≤43% of elderly patients with AMI. Some data have suggested that anemia might be well tolerated in subjects without coronary artery disease, even at hemoglobin levels as low as 7 g/dl. In the case of AMI, anemia can directly provoke ischemia or arrhythmia and can potentially increase infarct size. Moreover, anemia, by stimulating greater cardiac output, increases the myocardial oxygen demand. Despite many studies that have examined the effect of anemia on the short- and long-term prognosis in patients with heart failure, only a few data exist on the role of anemia in patients with AMI treated with percutaneous coronary intervention (PCI). The aim of our study was to evaluate the effect of anemia on short- and long-term prognosis in patients with AMI treated with PCI, including high-risk subgroups, such as patients with a low ejection fraction, diabetes mellitus, renal insufficiency, advanced age, multivessel disease, or incomplete revascularization.
Methods
The study group consisted of 1,497 consecutive patients, who were admitted to our department from September 2004 to December 2007 with AMI and treated in the acute phase with PCI. The clinical criteria of AMI evaluated on admission were chest pain persisting for >20 minutes, ST-segment elevation of ≥0.1 mV in 2 continuous electrocardiographic leads, or non–ST-segment elevation and enzymatic confirmation of AMI. Patients were mainly admitted from referral hospitals. All patients were included in the analysis. For all patients, coronary angiography and PCI of the infarct-related artery (IRA) were performed as soon as possible after admission using the femoral approach and standard techniques with 6F guiding catheters. PCI of IRA was performed after diagnostic angiography. The goal of PCI was to restore Thrombolysis In Myocardial Infarction (TIMI) grade 3 flow with residual stenosis <30%, which was defined as a successful procedure. The decision to administer glycoprotein IIb/IIIa inhibitors or to perform PCI (including stenting) of the non-IRA and its mode (during the same initial procedure [acute PCI] or in a staged fashion within the subsequent days, but during the index hospitalization [delayed PCI]) was left to the discretion of the operator. After the intervention, all patients received 150 mg/day of aspirin indefinitely and β blockers, angiotensin-converting enzyme inhibitors, and statins, if these agents were not contraindicated. Patients who received stents were given a loading dose of 300 or 600 mg of clopidogrel before PCI, followed by 250 mg of ticlopidine twice daily or 75 mg/day of clopidogrel orally.
The clinical data from all patients were prospectively recorded in a computerized database as a single-center AMI registry. The recorded data included demographics, laboratory values, the presence of concomitant diseases, AMI characteristics, angiographic findings, revascularization procedure, in-hospital complications, and mortality. Data regarding long-term outcomes (mortality and major adverse cardiovascular events [MACE]) were extracted from a main database of the National Fund of Health. The reasons for hospitalization and outcomes were classified on the basis of the codes of the International Classification of Diseases , ninth and tenth revisions. All hospitalizations and outcomes were subsequently reviewed to determine whether the prespecified MACE criteria were fulfilled. Thus, it was possible to collect the data of 99% of all patients. The mean follow-up period was 18.7 months. The study population was divided into 2 major groups: patients with anemia (anemia group, n = 248, 16.6%) and without anemia (nonanemic group, n = 1,249, 83.4%). A comparative analysis was performed between both groups within the whole population and within particular subgroups, including diabetes mellitus (n = 538, 35.9%), impaired renal function (n = 326, 21.8%), age >70 years (n = 422, 28.2%), left ventricular dysfunction (ejection fraction <40%, n = 408, 27.2%), incomplete revascularization (n = 749, 50.3%), and multivessel coronary artery disease (n = 974, 65.1%). Anemia was diagnosed in 117 patients (21.7%) in the diabetic group, 93 (28.5%) in the impaired renal function group, 40 in the left ventricular dysfunction group (19.5%), 103 in the advanced age group (24.5%), 130 with incomplete revascularization (17.3%), and 170 in the multivessel coronary artery disease group (17.4%). The total number of patients within all subgroups exceeded the sum of the whole population because many patients were included in ≥2 high-risk subgroups.
Blood tests were performed just after admission, before the patient’s transfer to the catheter laboratory. The hemoglobin concentration, platelet count, and white blood cell count were measured using automated instruments at a local hematology laboratory. Anemia was defined using the World Health Organization criteria as a hematocrit value <39% (or hemoglobin level <13 g/dl) for men and <36% (hemoglobin level <12 g/dl) for women at the initial presentation. The alternative definition of anemia, with a cutoff for the hemoglobin level of <11 g/dl, was also used for all patients >70 years old. Multivessel coronary artery disease was defined as the presence of >2 major epicardial coronary arteries or their major branches with stenosis of ≥70%, as assessed during the initial coronary angiography. Incomplete revascularization was defined when total occlusion or residual stenosis >70% was present in any of the major coronary arteries or their major branches at discharge. Impaired renal function was defined as an estimate of the glomerular filtration rate of <60 ml/min/1.73 m 2 during hospitalization, as calculated using the simplified Modification of Diet in Renal Disease formula, including age, race, gender, and serum creatinine. Diabetes mellitus was diagnosed using the patients’ history and oral glucose tolerance test results, performed routinely during the index hospitalization for every patient with AMI but without a medical history of diabetes. Contrast-induced nephropathy was defined as an increase in serum creatinine of 44.2 μmol/L (0.5 mg/dl) or a 25% increase from the baseline value within 48 hours after PCI. MACE were defined as a composite of death, myocardial infarction, repeated coronary intervention, coronary artery bypass grafting, and stroke.
All statistic analyses were performed using the Statistica, version 6.1, software program (StatSoft, Tulsa, Oklahoma). p Values <0.05 were considered significant for all tests. Continuous parameters are expressed as the mean ± SD, unless otherwise specified. Categorical variables are presented as the number and percentage. A comparative analysis between groups was performed using Student’s t test for continuous variables and the chi-square test or Fisher’s exact test, as appropriate, for dichotomous parameters. Long-term mortality and MACE were plotted as Kaplan-Meier curves and compared using the log-rank test. Independent predictors of death and MACE were identified on a multivariate Cox regression model and are expressed as the hazard ratio (HR) with 95% confidence interval. Parameters considered as covariates on multivariate Cox analysis included advanced age, diabetes mellitus, incomplete revascularization, impaired renal function, multivessel coronary disease, impaired left ventricle function, and anemia. All these factors were also used to stratify the study population into high-risk subgroups. The same factors were also used as covariates on multivariate Cox analysis within particular subgroups (excluding the presence of diabetes mellitus in the diabetic group, glomerular filtration rate level in the impaired renal function group, ejection fraction in the group with an ejection fraction <40%, age in the group aged >70 years, and so forth). A second regression model was developed that included as covariates the baseline parameters that differed significantly between the anemic and nonanemic groups and important factors potentially influencing the outcome after AMI. A third regression model was used to identify independent predictors of mortality within the anemic group and included as covariates all the factors potentially influencing the outcome after AMI.
Results
The patients with anemia were older, more likely to be women, and to have a lower body mass index and greater baseline serum creatinine. They were more likely to have history of myocardial infarction, PCI, and diabetes and were less likely to have history of smoking. They presented with a higher Killip class and longer lasting symptoms of AMI before admission. The clinical characteristics of the study population are listed in Table 1 .
| Variable | Patients With Anemia (n = 248) | Patients Without Anemia (n = 1,249) | p Value |
|---|---|---|---|
| Age (years) | 65.8 ± 11.5 | 61.4 ± 11.2 | <0.001 |
| Men | 162 (65.3%) | 896 (71.7%) | <0.05 |
| Smokers | 96 (38.7%) | 707 (56.6%) | <0.001 |
| Hypertension | 140 (56.4%) | 659 (52.7%) | NS |
| Body mass index (kg/m 2 ) | 27.0 ± 4.56 | 28.2 ± 4.64 | <0.05 |
| Diabetes mellitus | 117 (47.1%) | 421 (33.7%) | <0.001 |
| Previous myocardial infarction | 62 (25.0%) | 240 (19.2%) | <0.05 |
| Previous coronary artery bypass grafting | 14 (5.6%) | 40 (3.2%) | NS |
| Previous percutaneous coronary intervention | 38 (15.02%) | 111 (8.9%) | <0.05 |
| Killip class on admission | 1.61 ± 0.98 | 1.33 ± 0.72 | <0.001 |
| Shock on admission | 36 (14.2%) | 67 (5.3%) | <0.001 |
| Hyperlipidemia | 85 (34.1%) | 521 (41.7%) | NS |
| Anterior wall myocardial infarction | 77 (31.0%) | 423 (33.9%) | NS |
| Pain duration (hours) | 9.1 ± 11.8 | 7.7 ± 9.3 | <0.05 |
| High-density lipoprotein (mmol/L) | 1.23 ± 0.36 | 1.34 ± 0.40 | <0.001 |
| Low-density lipoprotein (mmol/L) | 3.16 ± 1.27 | 3.65 ± 1.22 | <0.001 |
| Glycoprotein IIb/IIIa inhibitor use | 35 (14.1%) | 168 (13.4%) | NS |
| Glucose on admission (mmol/L) | 9.45 ± 5.01 | 8.96 ± 4.47 | NS |
| Ejection fraction (%) | 42.3 ± 9.1 | 43.1 ± 8.7 | NS |
| Creatinine on admission (μmol\L) | 116.6 ± 103.2 | 88.9 ± 39.6 | <0.001 |
| Number of affected coronary arteries | 1.71 ± 1.5 | 1.11 ± 1.1 | <0.05 |
| Number of chronic total occlusions | 1.31 ± 1.6 | 0.34 ± 0.67 | <0.001 |
| Thrombolysis In Myocardial Infarction before percutaneous coronary intervention | 0.52 ± 0.89 | 0.78 ± 3.33 | NS |
| Thrombolysis In Myocardial Infarction after percutaneous coronary intervention | 2.45 ± 1.08 | 2.61 ± 0.94 | <0.05 |
| Myocardial blush grade after percutaneous coronary intervention | 2.22 ± 1.13 | 2.32 ± 1.03 | NS |
| Contrast-induced nephropathy | 77 (31.0%) | 346 (27.7%) | NS |
| Incomplete revascularization | 137 (55.2%) | 636 (50.9%) | NS |
| Hospitalization (days) | 9.66 ± 10.13 | 8.54 ± 5.4 | <0.05 |
Patients with anemia had more advanced coronary artery disease than the nonanemic group, manifesting as a greater number of affected coronary arteries (p <0.05) and a greater number of chronic total occlusions (p <0.001). However, no intergroup differences with respect to the preprocedural TIMI flow in the IRA were found. In contrast, the TIMI flow in the IRA after PCI was worse in anemic patients (p <0.05). However, no significant differences in successful IRA and non-IRA revascularization rates were found between the 2 groups. Patients with anemia had a longer in-hospital stay (p <0.05) and a greater mortality rate during the index hospitalization than the nonanemic group (p <0.001). The rates of transfusion and serious gastrointestinal bleeding during hospitalization were also greater in the anemic group (11.29% vs 6.56%; p <0.05; and (1.2% vs 0.16%; p <0.05, respectively).
During a mean follow-up period of 18.5 months (range 6 to 48), 222 patients (14.8%) of the whole population died. Significantly greater 30-day, 1-year, and total mortality rates (all p <0.05) were observed in the anemic than the nonanemic group. Similarly, anemic patients were significantly more likely to experience a major adverse cardiac event during the first 30 days and 1 year after AMI and during the whole follow-up period (all p <0.05).
Multivariate analysis identified anemia as an independent predictor of any-cause death in the whole population during the observation period (p <0.05). Anemia was also significantly associated with excessive long-term mortality in the multivessel coronary artery disease group and the incomplete revascularization group (both p <0.05), but not in advanced age, diabetic, impaired renal function, or left ventricular dysfunction groups ( Table 2 ). The analyses performed for the alternative anemia definition (hemoglobin <11 g/dl for those aged >70 years) also revealed greater total mortality rates for the anemic patients within the whole population (adjusted HR 1.77, 95% confidence interval 1.59 to 1.95; p = 0.0013), as well as within the multivessel coronary artery disease and incomplete revascularization groups.
| Variable | Unadjusted HR (95% CI) | p Value | Adjusted ⁎ HR (95% CI) | p Value |
|---|---|---|---|---|
| In-hospital death | ||||
| Whole population | 2.01 (1.80–2.21) | <0.001 | 1.30 (1.09–1.21) | NS |
| Diabetes mellitus | 0.99 (0.77–1.29) | NS | 0.62 (0.22–1.02) | NS |
| Incomplete revascularization | 2.01 (1.86–2.26) | <0.05 | 1.31 (0.95–1.67) | NS |
| Multivessel disease | 2.25 (2.01–2.49) | <0.001 | 1.24 (0.91–1.57) | NS |
| Impaired renal function | 1.15 (0.89–1.41) | NS | 0.82 (0.42–1.22) | NS |
| Advanced age | 1.50 (1.21–1.79) | NS | 1.37 (0.93–1.81) | NS |
| Left ventricular dysfunction | 1.29 (0.96–1.62) | NS | 1.22 (0.87–1.57) | NS |
| Death within 30 days | ||||
| Whole population | 1.87 (1.67–2.07) | <0.05 | 1.19 (0.98–1.40) | NS |
| Diabetic mellitus | 1.05 (0.76–1.34) | NS | 0.79 (0.41–1.20) | NS |
| Incomplete revascularization | 1.69 (1.44–1.94) | <0.05 | 1.11 (0.76–1.46) | NS |
| Multivessel disease | 2.08 (1.85–2.31) | <0.05 | 1.19 (0.87–1.51) | NS |
| Impaired renal function | 1.14 (0.89–1.39) | NS | 0.87 (0.49–1.25) | NS |
| Advanced age | 1.26 (0.98–1.54) | NS | 1.08 (0.66–1.50) | NS |
| Left ventricular dysfunction | 1.33 (1.00–1.66) | NS | 1.15 (0.80–1.50) | NS |
| Death within 1 year | ||||
| Whole population | 1.92 (1.76–2.08) | <0.001 | 1.31 (1.14–1.48) | NS |
| Diabetic mellitus | 1.08 (0.84–1.32) | NS | 0.81 (0.52–1.10) | NS |
| Incomplete revascularization | 1.78 (1.59–1.97) | <0.05 | 1.43 (1.19–1.67) | NS |
| Multivessel disease | 1.90 (1.72–2.08) | <0.001 | 1.35 (1.13–1.57) | NS |
| Impaired renal function | 1.01 (0.79–1.24) | NS | 0.82 (0.52–1.12) | NS |
| Advanced age | 1.21 (0.98–1.44) | NS | 1.10 (0.80–1.40) | NS |
| Left ventricular dysfunction | 1.67 (1.39–1.95) | NS | 1.52 (1.26–1.78) | NS |
| Death, total | ||||
| Whole population | 2.03 (1.88–2.18) | <0.001 | 1.46 (1.31–1.61) | <0.05 |
| Diabetic mellitus | 1.29 (1.08–1.50) | NS | 1.11 (0.86–1.31) | NS |
| Incomplete revascularization | 1.91 (1.63–2.09) | <0.001 | 1.67 (1.45–1.89) | <0.05 |
| Multivessel disease | 2.02 (1.85–2.19) | <0.001 | 1.54 (1.34–1.74) | <0.05 |
| Impaired renal function | 1.11 (0.91–1.31) | NS | 1.06 (0.80–1.32) | NS |
| Advanced age | 1.29 (1.07–1.51) | NS | 1.30 (1.03–1.57) | NS |
| Left ventricular dysfunction | 1.82 (1.56–2.08) | <0.05 | 1.61 (1.36–1.86) | 0.0574 |
⁎ Adjusted for presence of diabetes mellitus, incomplete revascularization, multivessel disease, impaired renal function, advanced age, and left ventricular dysfunction.
The second regression model with baseline intergroup differences (anemia vs no anemia) and other known risk factors included as covariates showed that anemia was still an independent predictor of any-cause mortality for the whole population (p <0.05; Table 3 ). A similar analysis performed for MACE did not reveal that anemia was an independent predictor of adverse cardiac events during 1 year of follow-up or long-term observation for the whole study population or for the subgroups ( Table 4 ). The cumulative survival rates and MACE occurrence in the whole population and the survival rates for the subgroups are plotted in Figures 1 and 2 .
| Variable | Adjusted ⁎ HR (95% CI) | Wald Chi-Square | p Value |
|---|---|---|---|
| Shock on admission | 4.91 (4.74–5.08) | 91.74 | <0.001 |
| Glomerular filtration rate <60 ml/min/1.73 m 2 | 2.64 (2.49–2.79) | 40.22 | <0.001 |
| Incomplete revascularization | 2.37 (2.21–2.43) | 28.91 | <0.001 |
| Maximum creatine kinase-MB (1 U/l increase) | 1.001 (1.008–1.012) | 26.40 | <0.001 |
| Ejection fraction <40% | 1.66 (1.51–1.81) | 11.11 | <0.001 |
| Thrombolysis In Myocardial Infarction after percutaneous coronary intervention <3 | 1.63 (1.46–1.80) | 8.12 | <0.05 |
| Age ≥70 years | 1.51 (1.36–1.66) | 7.61 | <0.05 |
| Anemia | 1.45 (1.29–1.61) | 5.76 | <0.05 |
| Previous myocardial infarction | 1.41 (1.26–1.56) | 5.19 | <0.05 |
| Contrast-induced nephropathy | 1.34 (1.20–1.48) | 4.14 | <0.05 |
| Smoker | 0.83 (0.68–0.98) | 1.36 | NS |
| Diabetes mellitus | 1.14 (0.99–1.29) | 0.71 | NS |
| Male gender | 1.09 (0.94–1.24) | 0.34 | NS |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
