Recurrent pneumonia among Japanese adults: disease burden and risk factors

Abstract

Background: In Japan and other societies with rapidly aging populations, recurrent pneumonia (RP) is a major clinical problem yet only limited information exists regarding the burden of this disease.

Methods: A prospective study of adult pneumonia was conducted to investigate the incidence of RP and potential risk factors. From February 1, 2012 to January 31, 2013, patients aged ≥ 15 years who were diagnosed with pneumonia were prospectively enrolled in a representative community hospital located in central Japan. Patients were followed for one-year to evaluate the recurrence of pneumonia and characteristics associated with RP. Cox proportional hazards models were constructed to compute adjusted hazard ratios (aHR) and ascertain risk factors significantly associated with RP.

Results: In total, 841 patients with a median age of 73 years (range 15-101 years) were enrolled totaling 1,048 person-years of observation with a median follow-up time of 475 days. A total of 137 patients had at least one recurrent episode with an incidence rate of 13.1 per 100 person-years (95% confidence interval: 11.1-15.5). In multivariate analysis, a past history of pneumonia (aHR 1.95, 95% CI: 1.35-2.8), chronic pulmonary disease (aHR 1.86, 1.24-2.78) and inhaled corticosteroid usage (aHR 1.78, 1.12-2.84) and hypnotic/sedative medication usage (aHR 2.06, 1.28-3.31) were identified as independent risk factors for recurrent pneumonia, whereas angiotensin converting enzyme-inhibitors usage was associated with a reduction of the risk of RP (aHR 0.22, 0.05-0.91). The detection of P. aeruginosa was significantly associated with RP even after adjusting for chronic pulmonary diseases (aHR = 2.37).

Conclusions: Recurrent pneumonia constitutes a considerable proportion of the pneumonia burden in Japan. A past history of pneumonia, chronic pulmonary disease, inhaled corticosteroid and hypnotic/sedative medication usage and detection of P. aeruginosa were identified as independent risk factors for recurrent pneumonia and special attention regarding the use of medications in this vulnerable population is needed to reduce the impact of this disease in aging populations.

Keywords: Community-acquired pneumonia; Elderly; Epidemiology; Recurrent pneumonia.

Fig. 1

Kaplan-Maier survival curve according to presence of recurrent pneumonia. Patients with recurrent pneumonia were significantly more likely to have fatal outcomes than patients without recurrent pneumonia (Hazard Ratio = 2.81, P < 0.001)