Abstract
Background
ST-segment resolution has long been used as one of several clinical markers of response to reperfusion therapy. With the use of the DANish trial in Acute Myocardial Infarction-2 (DANAMI-2) database, this case–control study tests the hypothesis that incomplete ST-segment resolution (<70%) will be predictive of the risk of reinfarction in the same myocardial region regardless of the reperfusion therapy employed.
Methods and results
One hundred forty-nine (9.5%) patients with clinical reinfarction were matched to patients with no documented reinfarction (“no reinfarction”). With the use of the initial DANAMI-2 and reinfarction electrocardiograms, 80 patients were found to have reinfarction in the same myocardial region (“reinfarction”). “Reinfarction” and their matched “no-reinfarction” patients were included ( n =160), and prereperfusion and postreperfusion ST segments were measured manually.
Of all “reinfarction” patients, 66% (53 of 80) had incomplete ST-segment resolution [ P =.13; odds ratio (OR)=1.69]. Stratified by the reperfusion strategy employed, this corresponded to 67% (35 of 52) being treated with fibrinolytics ( P =.45; OR=1.33), suggesting that these patients were equally likely to reinfarct in the same myocardial region regardless of the degree of ST-segment resolution. In the primary percutaneous coronary intervention (pPCI) arm, 64% (18 of 28) of patients with reinfarction in the same myocardial region had incomplete ST-segment resolution ( P =.10; OR=6.0). In the pPCI arm, the trend was that patients with incomplete ST-segment resolution were more likely to reinfarct in the same myocardial region than those with complete ST-segment resolution, but these findings did not reach statistical significance. No statistically significant difference was found in the association of reinfarction in the same myocardial region and ST-segment resolution between the two treatment arms ( P =.11).
Conclusion
The results from this study suggest that, although fibrinolytic-treated patients are more likely to reinfarct in the same myocardial region than pPCI-treated patients, there is no clear statistically significant association between ST-segment resolution and reinfarction in the same myocardial region. Trends in the data suggest that fibrinolytic-treated patients are equally likely to reinfarct in the same myocardial region regardless of the degree of ST-segment resolution. However, the trend in the pPCI population is such that incomplete ST-segment resolution is associated with a higher likelihood of reinfarction in the same myocardial region.
1
Introduction
The electrocardiogram (ECG) continues to be a useful tool for the diagnosis of acute myocardial infarction (MI), as well as for the assessment of the success of reperfusion therapy. However, even if reperfusion is successful, outcomes range from reinfarction to stroke and death . These outcomes were studied in the DANish trial in Acute Myocardial Infarction-2 (DANAMI-2), where reduction in reinfarction in the group treated with primary percutaneous coronary intervention (pPCI) helped to establish pPCI as the standard of care in that population. Subsequent DANAMI-2 substudies have shown that both location of the original infarct and ST-segment resolution postreperfusion are major determinants of the incidence of reinfarction . These findings add to those previously documented in the literature — that pPCI-treated patients with poor or partial ST-segment resolution (<70%) are more likely to have higher rates of in-hospital mortality and reinfarction than those with complete ST-segment resolution (≥70%) .
Fibrinolytic-treated patients have been shown to have higher rates of reinfarction compared to patients treated with invasive reperfusion strategies . Using ST-segment resolution as an indicator of the response to reperfusion therapy, this study investigates whether the degree of ST-segment resolution is predictive of the likelihood of reinfarction in the same myocardial region regardless of reperfusion therapy.
This study will focus on those patients with incomplete (<70%) ST-segment resolution. Since fibrinolytic therapy does not alter the underlying pathophysiology at the culprit coronary arterial site, there is the potential for rethrombosis at that same site. Incomplete ST-segment resolution following fibrinolytic therapy might indicate those patients whose nonreperfused myocardium completely infarcted and, therefore, no reinfarction could result from site reocclusion. In contrast, since pPCI therapy isolates the culprit site from the circulation, rethrombosis at that site is unlikely. Therefore, incomplete ST-segment resolution following pPCI therapy might indicate those patients whose myocardium received the least benefit from reperfusion and has a specific quality that makes it vulnerable to reinfarction despite reestablishment of arterial flow.
This study was performed to test the hypothesis that in patients with acute ST elevation myocardial infarction (STEMI), incomplete ST-segment resolution will predict the likelihood of reinfarction in the same myocardial region regardless of the reperfusion therapy employed (fibrinolytics vs. pPCI).
2
Methods
2.1
Patient population
DANAMI-2 patients who had clinical reinfarction in the same myocardial region as the original DANAMI-2 infarct were investigated using a case–control study design. The specific inclusion criteria are detailed in Table 1 .
| Inclusion criteria |
|---|
| 1. Patients with STEMI for whom this was the first MI included in DANAMI-2 |
| 2. ECG data upon admission, posttreatment, upon discharge and at reinfarction |
| 3. With posttreatment ECG data for the entire 3-year follow-up period |
| 4. With ECG-determined reinfarction in the same location as the DANAMI-2 infarct |
A total of 149 (9.5%) DANAMI-1 patients, across both reperfusion arms, had documented clinical reinfarction in the 3-year follow-up period . These 149 patients were matched in a 1:1 ratio by gender, age (within 2 years) and randomized treatment arm to patients who had no documented clinical reinfarction, yielding a total population of 298 patients for ECG review.
2.2
ECG assessment
2.2.1
Initial assessment of reinfarction site
ECG data from all patients with reinfarction ( n =149) and from their matched patients who completed the entire 3-year follow-up and had no documented reinfarction were scanned into portable document format files (PDF) and assessed manually. Of the 149 patients, the subset whose reinfarction occurred in the same myocardial region as the first infarct was determined by K.D. and G.S.W. using the methods described below.
Firstly, in patients for whom the reinfarction ECG was available, reinfarction in the same myocardial region was defined as reelevation of ST segment on reinfarction ECG in the same leads as those in which elevation was seen on the DANAMI-2 admission ECG. ST-segment elevation was defined as in DANAMI-2 .
Secondly, in patients for whom the reinfarction ECG was not available for analysis ( n =41), extensive hospital reports detailing ECG characteristics, patient presentation and hospital course were used to assign reinfarction location. The patients for whom neither reinfarction ECG nor hospital report was available ( n =4) were excluded.
Of the patients with actual ECG data or information on ECG data, 89 patients (61%) had reinfarction in the same myocardial region, 10 patients (7%) had reinfarction location indeterminable by ECG characteristics alone and 46 patients (32%) had reinfarction in a myocardial region different from the region of the DANAMI-2 infarct ( Fig. 1 ). The 89 patients with ECG-determined reinfarction in the same myocardial region (“reinfarction”), along with their corresponding controls (“no reinfarction”) (a total of 178 patients), were included for assessment of ST segment resolution ( Fig. 1 ). Follow-up ECGs of the control patients were also analyzed for evidence of undocumented reinfarction, defined as the development of Q waves of greater than 30 ms duration on ECG corresponding to the same or different myocardial region as the first infarct .
2.2.2
ECG characteristics
For the purposes of this study, ST segment resolution as defined by Schroeder et al. ( ) was modified such that incomplete ST-segment resolution represented a resolution <70% of the original ST elevation above baseline, and such that complete ST-segment resolution represented a resolution ≥70% of the original elevation above baseline. ST-segment analysis of ECG data during hospitalization and across the follow-up period was carried out for all patients; however, ST-segment data on the admission ECG and on the 90-min posttreatment ECG were used to determine ST-segment resolution. Where the 90-min posttreatment ECG was not available, the next earliest ECG (at 4 h posttreatment) was used to determine the percentage of ST-segment resolution from baseline elevation. The lead with the greatest magnitude of ST-segment elevation, and therefore the largest injury vector, was used for analysis in each patient.
“Reinfarction” patients with missing or insufficient DANAMI-2 ECG data were excluded ( n =9). In addition, “no-reinfarction” patients with missing or insufficient data ( n =3) were replaced with equally matched patients whose corresponding “reinfarction” patient had already been excluded from the study. Thus, the total number included in the substudy was 160 patients ( Fig. 1 ).
2.3
Statistical analysis
Conditional logistic regression analysis was performed to account for the case–control study design in which controls were matched one to one according to age, gender and treatment. The major independent variable of interest was ST-segment resolution. Analyses were performed for all patients and then separately for fibrinolytics and pPCI arms. Odds ratios (ORs) with 95% confidence intervals and P values were reported. The Student’s t test was used to compare the regression coefficients for ST-segment resolution in the two treatment groups.
2
Methods
2.1
Patient population
DANAMI-2 patients who had clinical reinfarction in the same myocardial region as the original DANAMI-2 infarct were investigated using a case–control study design. The specific inclusion criteria are detailed in Table 1 .
| Inclusion criteria |
|---|
| 1. Patients with STEMI for whom this was the first MI included in DANAMI-2 |
| 2. ECG data upon admission, posttreatment, upon discharge and at reinfarction |
| 3. With posttreatment ECG data for the entire 3-year follow-up period |
| 4. With ECG-determined reinfarction in the same location as the DANAMI-2 infarct |
A total of 149 (9.5%) DANAMI-1 patients, across both reperfusion arms, had documented clinical reinfarction in the 3-year follow-up period . These 149 patients were matched in a 1:1 ratio by gender, age (within 2 years) and randomized treatment arm to patients who had no documented clinical reinfarction, yielding a total population of 298 patients for ECG review.
2.2
ECG assessment
2.2.1
Initial assessment of reinfarction site
ECG data from all patients with reinfarction ( n =149) and from their matched patients who completed the entire 3-year follow-up and had no documented reinfarction were scanned into portable document format files (PDF) and assessed manually. Of the 149 patients, the subset whose reinfarction occurred in the same myocardial region as the first infarct was determined by K.D. and G.S.W. using the methods described below.
Firstly, in patients for whom the reinfarction ECG was available, reinfarction in the same myocardial region was defined as reelevation of ST segment on reinfarction ECG in the same leads as those in which elevation was seen on the DANAMI-2 admission ECG. ST-segment elevation was defined as in DANAMI-2 .
Secondly, in patients for whom the reinfarction ECG was not available for analysis ( n =41), extensive hospital reports detailing ECG characteristics, patient presentation and hospital course were used to assign reinfarction location. The patients for whom neither reinfarction ECG nor hospital report was available ( n =4) were excluded.
Of the patients with actual ECG data or information on ECG data, 89 patients (61%) had reinfarction in the same myocardial region, 10 patients (7%) had reinfarction location indeterminable by ECG characteristics alone and 46 patients (32%) had reinfarction in a myocardial region different from the region of the DANAMI-2 infarct ( Fig. 1 ). The 89 patients with ECG-determined reinfarction in the same myocardial region (“reinfarction”), along with their corresponding controls (“no reinfarction”) (a total of 178 patients), were included for assessment of ST segment resolution ( Fig. 1 ). Follow-up ECGs of the control patients were also analyzed for evidence of undocumented reinfarction, defined as the development of Q waves of greater than 30 ms duration on ECG corresponding to the same or different myocardial region as the first infarct .
