Two other studies sought to determine an association between CAP and cardiac complications differently; not by following CAP patients prospectively for complications but by retrospectively evaluating patients for a respiratory infection among those who were admitted for a cardiovascular complication (MI or stroke). A study of over 35,000 first-time admissions for either an MI or a stroke were evaluated for a respiratory infection within the previous 90 days [24]. The incidence rates were statistically significant for every time period up to 90 days. The preceding 3 days was the time period with the highest frequency for a respiratory infection preceding an event. When the event was an MI, the incident rate was 4.95 (95% CI, 4.43–5.53). A similar study of over 20,000 first-time admissions for either an MI or stroke were evaluated for a preceding primary care visit for a respiratory infection [25]. An infection preceded 2.9% of patients with an MI and 2.8% of patients with a stroke. Statistical significance was found for the group of patients who had a respiratory infection within 7 days preceding an MI (OR 2.10 [95% CI 1.38–3.21]) or preceding a stroke (OR 1.92 [95% CI 1.24–2.97]). In fact, every time period analyzed for both complications (MI and stroke) was significant up to 1 year. Because the timing of a cardiac complication varies and can occur up to 90 days or even a year after acute infection, physicians should maintain vigilance in suspecting and screening for them.
Predictors of Cardiac Complications During CAP
Recently, Cangemi et al reviewed mortality in 301 patients admitted for CAP 6 to 60 months after they were discharged [26]. Mortality was compared between patients who experienced a cardiac complication—atrial fibrillation or an ST- or non-ST-elevation MI—during their admission and those who did not. A total of 55 (18%) patients had a cardiac complication while hospitalized. During the follow-up, 90 (30%) of the 301 patients died. Death occurred in more patients who had had a cardiac complication while hospitalized than in those who did not (32% vs 13%; P < 0.001). The study also showed that age and the pneumonia severity index (PSI) predicted death in addition to intra-hospital complication. A Cox regression analysis showed that intrahospital cardiac complications (hazard ratio [HR] 1.76 [95% CI 1.10–2.82]; P = 0.019), age (HR 1.05 [95% CI 1.03–1.08]; P < 0.001) and the PSI (HR 1.01 [95% CI 1.00–1.02] P = 0.012) independently predicted death after adjusting for possible confounders [26].
The PSI score was published in 1997, and it instructed that patients with a risk class of I or II (low risk) should be managed as outpatients. Data eventually showed that there is a portion of the population with a risk class of I or II whose hospital admission is justified [4]. Among the reasons found was “comorbidity,” including MI and other cardiac complications. The PSI prediction rule was found to be useful in novel ways, and being associated with a risk of MI in patients with CAP was one of them. The propensity-adjusted association between the PSI score and MI was significant ( P < 0.05) in an observational study of the CAP Organization (CAPO) [20]. Knowing that a PSI of 80 is in the middle of risk class III (71–90), it was noted that below 80 the risk for MI was zero to 2.5%, while above 80 the risk rose from 2.5% to 12.5%. A later study using the same statistical method showed a correlation between the PSI score and cardiac complications (MI, arrhythmias and CHF) with a P value of < 0.01 [21]. Determining the probability for the combination of complications, rather than just an MI, yielded an unsurprisingly higher range of risk for the PSI below 80, which was zero to 17.5%, while risk for a PSI above 80 was 17.5% to 80%.