Background: We sought to explore the frequency, risk factors, and clinical consequences of post-stroke infections (PSI).
Methods: We assessed consecutive patients admitted with suspected stroke. On day 1, we recorded the patients' baseline characteristics, stroke subtype and severity, physiological parameters, and laboratory results. On day 5, we recorded the occurrence of complications including infections, and functional outcome.
Results: We recruited 439 patients [398 with stroke and 41 with transient ischaemic attack (TIA)], of whom 73 (17%) had PSI (67 with stroke and 6 with TIA) by day 5. Compared to patients without PSI, those with PSI were significantly more likely to: (i) be older (mean age 78 vs 73 years, P=0.001); (ii) be dependent before admission (47% vs 24%, P<0.001); (iii) have total anterior circulation strokes (40% vs 20%, P<0.001); (iv) be hypoxic on arrival (35% vs 17%, P=0.001); (v) have urinary catheters (43% vs 16%, P<0.001), and (vi) develop pressure sores (7% vs 2%, P=0.022) and seizures (12% vs 2%, P<0.001). After adjusting for case mix, PSI was significantly associated with in-hospital death (OR 2.50, 95% CI 1.27-4.90, P=0.008) and institutionalization on discharge (OR 1.86, 95% CI 1.00-3.43, P=0.049).
Conclusions: Post-stroke infection can affect patients admitted with TIA as well as acute stroke, and it may be associated with poor short-term outcomes.