A decrease in mechanical complications after ST-elevation myocardial infarction may have contributed to improved survival rates associated with reperfusion by primary percutaneous coronary intervention (PCI). Mechanical complications occurred in 52 of 5,745 patients (0.91%) in the largest reported randomized trial in which primary PCI was the reperfusion strategy. The frequencies were 0.52% (30) for cardiac free-wall rupture (tamponade), 0.17% (10) for ventricular septal rupture, and 0.26% (15) for papillary muscle rupture (3 patients had 2 complications). Ninety-day survival rates were 37% (11) for cardiac free-wall rupture, 20% (2) for ventricular septal rupture, and 73.3% (11) for papillary muscle rupture. These mechanical complications occurred at a median of 23.5 hours (interquartile range 5.0 to 76.8) after symptom onset and were associated with 44% (23 of 52) survival through 90 days, which accounted for 11% of the 90-day mortality. Factors associated with mechanical complications were older age, female gender, Q waves, presence of radiologic pulmonary edema, and increased prerandomization troponin levels. In conclusion, rates of mechanical complications are lower with primary PCI than those previously reported after fibrinolytic therapy.
The increasing use of reperfusion therapies for acute ST-elevation myocardial infarction (STEMI) appears to have decreased the rates of mechanical complications of cardiac free-wall rupture (FWR), ventricular septal rupture (VSR), and acute mitral regurgitation (MR) compared to historical rates. Also, lower rates of mechanical complications appear to be associated with the increasingly widespread use of these primary percutaneous coronary interventions (PCIs). In this setting we examined the rates and outcomes of individual mechanical complications of STEMI in the Assessment of Pexelizumab in Acute Myocardial Infarction (APEX-AMI) trial. This trial is the largest randomized trial to date in which primary PCI was the reperfusion strategy, and high usage rates of evidence-based therapies including aspirin, clopidogrel, statins, β blockers, and angiotensin-converting enzyme inhibitors have been reported. Also, we sought to confirm whether previously reported factors associated with an increased risk of these complications remained prevalent in the era of primary PCI.
Methods
The APEX-AMI trial recruited patients with STEMI from 17 countries and 296 sites from July 2004 to May 2006, who presented within 6 hours of symptom onset (which persisted for ≥20 minutes) and were anticipated to undergo primary PCI as the reperfusion strategy, and had “high-risk” electrocardiographic features; the APEX-AMI study design and inclusion and exclusion criteria have been reported elsewhere. In brief, patients were randomized to receive, over 10 minutes before PCI, an intravenous bolus of pexelizumab 2 mg/kg or matching placebo, followed by a 0.05-mg/kg per hour infusion of pexelizumab or placebo over the next 24 hours, in a double-blind manner. The primary end point of the trial was mortality at 90 days and secondary end points were the composite incidence and individual components of death, cardiogenic shock, or congestive heart failure, which was adjudicated by a clinical events committee, through 30 and 90 days. Mechanical complications rates were determined for cardiac tamponade or FWR, VSR, and papillary muscle rupture.
Comparisons of characteristics between patients with and without mechanical complications and between patients with each type of mechanical complication and those with none used likelihood ratio chi-square tests for categorical variables, shown as percentages (numbers). Continuous variables were shown as median (25th, 75th percentiles) and were compared between groups using Wilcoxon rank-sum tests. For postrandomization measurements and events, associations with mechanical complications were summarized and tested without regard to the temporal ordering of events; assessment merely examined whether the 2 types of events tended to occur in the same patients. A logistic regression model was used to identify baseline potential predictors of mechanical complications, which were limited to those with the most significant univariable associations with mechanical complications due to the small number of affected patients. These were female gender, age, current smoking, time from symptom onset to hospital arrival, (available) prerandomization troponin levels, presence of Q waves on baseline electrocardiogram, and radiologic pulmonary edema. Also, predictors of each mechanical complication using as candidates any variable with a p value ≤0.10 in univariable tests were evaluated. Due to the small number of events for each complication, it was not possible to examine a reasonable list of candidate factors and still adhere to the common guideline of 1 predictor per 10 events, so analyses are considered exploratory. Kaplan-Meier methods were used to generate time-to-event plots. Patients were included in all analyses to the extent that their data allowed; no missing data were imputed.
Results
Baseline clinical, angiographic, laboratory, and MI characteristics of the 5,745 patients with STEMI undergoing primary PCI, according to the presence or absence of each mechanical complication, are presented in Tables 1 to 3 . Of the 52 patients with mechanical complications, the clinical events committee–adjudicated end point of cardiogenic shock occurred in 19 patients (36%) and 4 (8%) had the congestive heart failure end point (without shock); 29 patients (56%) with mechanical complications did not develop the cardiogenic shock or congestive heart failure end points.
| No Mechanical Complications | Cardiac FWR | Acute MR | VSR | |
|---|---|---|---|---|
| (n = 5,693) | (n = 30) | (n = 15) | (n = 10) | |
| Age (years) | 61 (52, 71) | 74 (61, 78) ⁎ | 68 (62, 74) ⁎ | 72 (56, 78) |
| Male gender | 4,397 (77%) | 12 ⁎ (40%) | 9 (60%) | 3 ⁎ (30%) |
| Transfer admission | 2,051 (36%) | 13 (43%) | 7 (47%) | 6 (60%) |
| History of myocardial infarction | 690 (12%) | 2 (7%) | 2 (13%) | 0 |
| History of hypertension | 2,811 (49%) | 18 (60%) | 6 (40%) | 7 (70%) |
| History of diabetes | 904 (16%) | 6 (20%) | 2 (13%) | 2 (20%) |
| Current smoker | 2,464 (43%) | 6 ⁎ (20%) | 5 (33%) | 4 (40%) |
| History of hypercholesterolemia | 2,163 (50%) | 11 (50%) | 3 (30%) | 4 (50%) |
| History of heart failure | 206 (4%) | 1 (3%) | 0 | 1 (10%) |
| Current renal dialysis | 15 (<1%) | 1 (3%) | 0 | 0 |
| History of peripheral vascular disease | 245 (4%) | 0 | 1 (7%) | 0 |
| Previous angioplasty | 559 (10%) | 3 (10%) | 0 | 0 |
| Previous coronary surgery | 127 (2%) | 0 | 0 | 1 (10%) |
| Baseline systolic blood pressure (mm Hg) | 133 (117, 150) | 116 (100, 145) ⁎ | 116 (100, 131) ⁎ | 121 (111, 135) |
| Baseline heart rate (beats/min) | 75 (65, 86) | 78 (64, 96) | 90 (70, 103) ⁎ | 89 (86, 110) ⁎ |
| Baseline Killip class III or IV | 116 (2%) | 2 (7%) | 1 (7%) | 4 ⁎ (40%) |
| Total ST deviation (mm) | 13.0 (9.0, 18.5) | 17.0 (11.0, 21.5) ⁎ | 15.0 (10.0, 22.0) | 18.0 (17.0, 19.0) |
| Anterior myocardial infarction | 3,371 (59%) | 17 (57%) | 7 (47%) | 4 (40%) |
⁎ Different from group with no mechanical complications at p <0.05.
| No Mechanical Complications | Cardiac FWR | Acute MR | VSR | |
|---|---|---|---|---|
| (n = 5,693) | (n = 30) | (n = 15) | (n = 10) | |
| Initial creatinine (μmol/L) | 89 (80, 106) | 97 (80, 115) | 97 (80, 120) | 92 (88, 115) |
| Initial hemoglobin (g/L) | 148 (136, 157) | 139 (129, 151) ⁎ | 152 (133, 161) | 136 (121, 152) |
| Baseline troponin >×1 upper reference limit | 2,122 (44%) † | 17 ⁎ (68%) | 7 (58%) | 8 ⁎ (100%) |
| Highest creatinine kinase level (IU/L) | 1,797 (862, 3,310) | 3,073 (803, 4,072) | 2,036 (367, 3,072) | 3,112 (1,779, 3,349) |
| Pulmonary edema on x-ray | 131 (2%) | 2 (7%) | 3 ⁎ (20%) | 1 (10%) |
| New Q waves in 2 leads | 3,062 (56%) | 20 (71%) | 9 (69%) | 6 (67%) |
| New left or right bundle branch block | 426 (7%) | 3 (10%) | 1 (7%) | 1 (10%) |
| Anterior myocardial infarction | 3,371 (59%) | 17 (57%) | 7 (47%) | 4 (40%) |
| Myocardial onset to first medical contact (h) | 2.2 (1.3, 3.3) | 3.0 (1.9, 4.3) ⁎ | 2.0 (1.2, 2.9) | 3.8 (1.8, 5.0) ⁎ |
| Myocardial onset to first balloon inflation (h) | 3.3 (2.5, 4.5) | 4.6 (3.0, 5.8) ⁎ | 3.2 (2.5, 5.2) | 4.7 (3.1, 6.0) |
| Cardiogenic shock end point | 177 (3%) | 7 ⁎ (23%) | 8 ⁎ (53%) | 6 ⁎ (60%) |
| Heart failure end point | 270 (5%) | 2 (7%) | 5 ⁎ (33%) | 0 |
⁎ Different from group with no mechanical complications at p <0.05.
† Percentages are in patients with data (e.g., 8 of 10 patients with VSR had troponin data available, so 8 troponin-positive patients are 100% rather than 80%).
| No Mechanical Complications | Cardiac FWR | Acute MR | VSR | |
|---|---|---|---|---|
| (n = 5,693) | (n = 30) | (n = 15) | n = 10 | |
| Angiographic left ventricular ejection fraction (%) | 50 (40, 59) | 36 (29, 48) ⁎ | 35 (28, 47) | 50 (30, 70) |
| Culprit vessel | ||||
| Left main coronary artery | 36 (1%) | 0 | 0 | 0 |
| Left anterior descending coronary artery | 2,902 (51%) | 14 (50%) | 4 (27%) | 4 (40%) |
| Left circumflex coronary artery | 595 (11%) | 5 (18%) | 4 (27%) | 0 |
| Right coronary artery | 1,931 (34%) | 8 (29%) | 5 (33%) | 6 (60%) |
| Saphenous venous graft | 63 (1%) | 0 | 0 | 0 |
| Unknown/none | 133 (2%) | 1 (4%) | 2 (13%) | 0 |
| Thrombolysis In Myocardial Infarction grade 0 or 1 flow in culprit vessel before angioplasty | 4,077 (74%) | 26 ⁎ (93%) | 10 (71%) | 8 ⁎ (100%) |
| Thrombolysis In Myocardial Infarction grade 0 or 1 flow in culprit vessel after angioplasty | 183 (3%) | 4 ⁎ (15%) | 1 (7%) | 2 ⁎ (25%) |
⁎ Different from group with no mechanical complications at p <0.05.
Frequencies of individual mechanical complications were 0.52% (30 patients) for cardiac FWR (tamponade), 0.17% (10 patients) for VSR, and 0.26% (15 patients) for MR secondary to papillary muscle rupture; 2 patients had FWR and VSR and another had FWR and MR. Mechanical complications occurred at a median of 23.5 hours (IQR 5.0 to 76.8) after symptom onset, 20.7 hours (5.0 to 86.0) for FWR, 7.7 hours (5.5 to 23.5) for VSR, and 31.0 hours (4.7 to 72.0) for MR. Patients with mechanical complications had longer times to first balloon inflation and less frequently had Thrombolysis In Myocardial Infarction grade 2 or 3 flow after PCI compared to patients without mechanical complications. Ten patients who had mechanical complications did not undergo primary PCI, and in another 5 patients the mechanical complication occurred before PCI ( Figure 1 ).
