PR interval and survival in diabetic patients with acute myocardial infarction.

In previous reports, we retrospectively studied data concerning a cohort of diabetic patients with myocardial infarction.1, 2, 3 We now studied the relation between ECG data at admission and survival. Particular attention was paid to the PR interval, since conflicting data exist in the literature concerning the prognostic significance of a prolonged PR interval.4, 5

The present study was retrospective and part of the methods have been described in the previous reports.1, 2 The survival of patients was established by the study of electronic health records.

From the ECG at admission in the coronary care unit, PR interval, corrected QT interval, heart rate, QRS axis and QRS duration were the ECG data obtained automatically that were used for further study. The presence of ST segment elevation was evaluated taking the whole set of ECGs from each patient. Pearson correlation was calculated for heart rate and PR interval.

Cox-proportional hazards survival modelling was used. Kaplan-Meyer study was carried out, taking different values of the PR interval as the cut-off value. The comparison between groups was made using the log-rank test. A level of 0.05 was considered statistically significant. The present protocol was approved by the local ethics committee.

The ECG was available for 192 out of 195 initial different patients (123 male/69 female). The median age was 68.5 years (interquartile range 61–76 years). ST segment elevation myocardial infarction was present in 67 patients. Primary coronary angioplasty was carried out in 44 patients. After a period not inferior to 48 months and up to 61 months after each admission (median 49.2 months, interquartile range 33.1–55.4 months), the retrospective analysis of electronic records showed that 57 of the 193 patients had died (29.7%).2 The PR interval was evaluated in 170 patients. The median was 169 ms and the interquartile range was 154–190 ms. A negative correlation was seen between the PR interval and heart rate (Pearson correlation coefficient −0.292; significance level <0.001).

Cox analysis, using age, gender and the six ECG variables, showed that heart rate and the PR interval were predictors of mortality (Table 1). However, a new Cox analysis, adding analytical data (peak plasma cardiac troponin, plasma creatinine, plasma alkaline phosphatase and hemoglobin), as well as beta-blocker therapy at discharge, showed that heart rate and the PR interval were no longer predictors of mortality (age, gender, plasma alkaline phosphatase and hemoglobin were; Table 1).

Table 1

Cox regression analysis of survival of 192 patients with Diabetes mellitus admitted for acute myocardial infarction. Retrospective analysis of survival based on electronic health records after no less than 48 months after admission for each patient, and up to 61 months.

Parameter	Model 1Significance level	Model 2Significance level
Age	0.001	0.011
Gender	0.888	0.030
Heart rate	0.026	0.063
PR interval	0.012	0.409
Corrected QT interval	0.937	0.468
ST segment elevation in electrocardiogram	0.675	0.178
QRS axis	0.299	0.263
QRS duration	0.750	0.934
Peak troponin I plasma level	–	0.945
Plasma creatinine at admission	–	0.335
Plasma alkaline phosphatase	–	0.044
Hemoglobin	–	<0.001
Beat blocker at discharge	–	0.557

Kaplan-Meyer analysis was carried out with different cut-off values for the PR interval: 155 ms, 170 ms, 190 ms, 200 ms, 220 ms. Unlike what happened with the first four cut-off values, Kaplan-Meyer analysis showed that a PR interval ≥220 ms (seen in 14 patients, versus 156 patients with a shorter PR interval) was associated with a decreased survival (p in log rank test of 0.009) (Fig. 1).

Fig. 1

Kaplan-Meier survival curves for 170 patients with Diabetes mellitus and acute myocardial infarction, according to the PR interval in the electrocardiogram at admission. Time measured in months. Lower line—patients with PR interval = />220 ms; Upper line—remaining patients. Significance level in log-rank test 0.009.

We conclude that the PR interval was associated to survival only if analytical variables were not taken into consideration. A prolonged PR interval (=/>220 ms) was associated with markedly decreased survival.

Conflict of interest

None to be declared.