Frequency, Predictors, Management and Outcome of Post-Stroke Pneumonia

Introduction: Pneumonia is a common complication of acute stroke, occurring in up to one quarter of cases. Those who develop post-stroke pneumonia are at an increased risk of death or poor functional recovery. Despite significant advances in the treatment of stroke in the last few years uncertainty remains regarding the frequency of post-stroke pneumonia. Whether prophylactic antibiotics are effective, and the appropriateness of other management strategies to improve stroke outcomes, remains uncertain. Aims: To systematically review the evidence on post-stroke pneumonia to determine: (i) the frequency, predictors, management and outcome of post-stroke pneumonia, and (ii) whether there are effective and safe interventions to prevent pneumonia and improve the management and outcome of post-stroke pneumonia. Methods: All observational studies published before October 2017 were assessed for studies with adult patients (age ≥18 years) admitted within 30 days of acute ischaemic or haemorrhagic stroke with prospective consecutive recruitment and quantification of the proportion of people who developed pneumonia after acute stroke. Studies were included in the frequency review if they were of adult patients with acute stroke where an assessment of pneumonia was performed within 30 days of ictus (Systematic review registration: PROSPERO CRD20171221). All randomised controlled trials published before December 2016 with adult patients (age ≥18 years) admitted within 30 days of acute ischaemic or haemorrhagic stroke comparing prophylactic antibiotics for the prevention of pneumonia with placebo, no treatment or standard care were included in the management review (Systematic review registration: PROSPERO CRD42016053133). Data were identified from Medline (Medical Literature Analysis and Retrieval System Online) via Ovid, EMBASE (Excerpta Medical Database), CENTRAL (Cochrane Central Register of Controlled Trials), CINAHL (Cumulative index to nursing and allied health literature) and PsycINFO electronic databases. Results: Data were available from 47 observational studies (n=139,432 participants) and 6 trials (n=4,111 participants). The pooled frequency of post-stroke pneumonia was 12.3% (95% confidence interval [CI] 11%-13.6%, I2=98%). In pre-specified subgroup analyses, the pooled frequency of post-stroke pneumonia in 2011-2017 was 13.5% (95% CI 11.8%-15.3%, I2=98%) and this was comparable with earlier periods (P interaction=0.31). The pooled estimate of frequency of post-stroke pneumonia in the studies including only patients treated in stroke units was 7.1% (95% CI 5.8% -8.5%, I2=84%) which was significantly lower than for those treated in other locations (P interaction=0.005). The pooled frequency of post-stroke infection was 21% (95%CI 13%-29.3%; I2=99%) and of post-stroke urinary tract infection was 7.9% (95% CI 6.7%-9.3%; I2=96%). Compared to controls, preventive antibiotics reduced the risk of post-stroke pneumonia (relative risk [RR] 0·75, 95% [CI 0·57-0·99), and all infections (RR 0.58, 95%CI 0.48-0.69). There was no significant difference in the risks of dependency (RR 0.99, 95%CI 0·88-1·11), or mortality (RR 0·96, 95%CI 0·78-1·19) between the preventive antibiotic and control groups. Preventive antibiotics had uncertain effects on the risk of adverse events. Conclusions: Despite advances in the availability of acute stroke interventions, the frequency of post stroke pneumonia remains high at more than 10%. Pooled estimates indicate that the frequency of pneumonia has remained stable over recent decades. The frequency of post-stroke pneumonia is lower in patients who receive stroke unit care as compared to care in other hospital areas. While preventive antibiotics do reduce the risk of post-stroke pneumonia, uncertainty remains over whether this translates into improvements in other clinical outcomes. As such, there is insufficient evidence to recommend routine use of preventive antibiotics in the management of patients with acute stroke