Johnstone Jennie, Eurich Dean T, Majumdar Sumit R, Jin Yan, Marrie Thomas J
From Department of Medicine (JJ, SRM, TJM), Faculty of Medicine and Dentistry, and Department of Public Health Science (DTE, SRM), School of Public Health, University of Alberta; and Alberta Health and Wellness (YJ), Edmonton, Alberta, Canada.
Medicine (Baltimore). 2008 Nov;87(6):329-334. doi: 10.1097/MD.0b013e318190f444.
Little is known about the long-term sequelae of community-acquired pneumonia (CAP). Therefore, we describe the long-term morbidity and mortality of patients after pneumonia requiring hospitalization. We specifically hypothesized that the Pneumonia Severity Index (PSI), designed to predict 30-day pneumonia-related mortality, would also be associated with longer-term all-cause mortality. Between 2000 and 2002, 3415 adults with CAP admitted to 6 hospitals in Edmonton, Alberta, Canada, were prospectively enrolled in a population-based cohort. At the time of hospital admission, demographic, clinical, and laboratory data were collected and the PSI was calculated for each patient. Postdischarge outcomes through to 2006 were ascertained using multiple linked administrative databases. Outcomes included all-cause mortality, hospital admissions, and re-hospitalization for pneumonia over a maximum of 5.4 years of follow-up. Follow-up data were available for 3284 (96%) patients; 66%were > or =65 years of age, 53% were male, and according to the PSI fully 63% were predicted to have greater than 18% 30-day pneumonia-related mortality (that is, PSI class IV-V). Median follow-up was 3.8 years. The 30-day, 1-year, and end of study mortality rates were 12%, 28%, and 53%, respectively. Overall, 82(19%) patients aged <45 years died compared with 1456 (67%) patients aged > or =65 years (hazard ratio [HR], 5.07; 95% confidence interval [CI], 4.06-6.34). Male patients were more likely to die than female patients during follow-up (971 [56%] vs. 767 [49%], respectively; HR, 1.20; 95% CI, 1.13-1.37). Initial PSI classification predicted not only 30-day mortality, but also long-term postdischarge mortality, with 92 (15%) of PSI class I-II patients dying compared with 616 (82%) PSI class V patients (HR, 11.80; 95% CI, 4.70-14.70). Of 2950 patients who survived the initial CAP hospitalization, 72% were hospitalized again (median, 2 admissions over follow-up) and 16% were re-hospitalized with pneumonia. In conclusion, long-term morbidity and mortality are high following hospitalization for pneumonia and are strongly correlated with initial PSI class. This suggests that patients with pneumonia, especially those with PSI class IV and V at admission, might need better attention paid to preventive strategies and much closer follow-up due to their elevated risk of subsequent adverse events and increased health resource utilization.
关于社区获得性肺炎(CAP)的长期后遗症,人们了解甚少。因此,我们描述了因肺炎住院患者的长期发病率和死亡率。我们特别假设,旨在预测30天肺炎相关死亡率的肺炎严重程度指数(PSI)也将与长期全因死亡率相关。在2000年至2002年期间,加拿大艾伯塔省埃德蒙顿市6家医院收治的3415例成年CAP患者被前瞻性纳入一项基于人群的队列研究。在入院时,收集了人口统计学、临床和实验室数据,并为每位患者计算了PSI。通过多个相互关联的行政数据库确定了截至2006年的出院后结局。结局包括全因死亡率、住院次数以及在最长5.4年的随访期间因肺炎再次住院的情况。3284例(96%)患者有随访数据;66%年龄≥65岁,53%为男性,根据PSI,整整63%的患者预计30天肺炎相关死亡率大于18%(即PSI IV-V级)。中位随访时间为3.8年。30天、1年和研究结束时的死亡率分别为12%、28%和53%。总体而言,年龄<45岁的患者中有82例(19%)死亡,而年龄≥65岁的患者中有1456例(67%)死亡(风险比[HR],5.07;95%置信区间[CI],4.06-6.34)。男性患者在随访期间比女性患者更易死亡(分别为971例[56%]和767例[49%];HR,1.20;95%CI,1.13-1.37)。初始PSI分级不仅可预测30天死亡率,还可预测出院后的长期死亡率,PSI I-II级患者中有92例(15%)死亡,而PSI V级患者中有616例(82%)死亡(HR,11.80;95%CI,4.70-14.70)。在最初因CAP住院存活的2950例患者中,72%再次住院(随访期间中位住院2次),16%因肺炎再次住院。总之,肺炎住院后的长期发病率和死亡率很高,且与初始PSI分级密切相关。这表明肺炎患者,尤其是入院时为PSI IV级和V级的患者,由于其后续不良事件风险升高以及卫生资源利用增加,可能需要更关注预防策略并进行更密切的随访。