Risk factors for pneumonia in patients with acute stroke.

Acute Stroke To the Editor: In their recent study, Hilker et al found a 21% incidence of nosocomial pneumonia in acute stroke patients treated on a neurological intensive care unit.1 Independent risk factors were mechanical ventilation, multiple lesion and vertebrobasilar stroke, dysphagia, and abnormal chest x-ray on admission. Patients with intercurrent pneumonia had a higher mortality and a poorer functional outcome than patients without pneumonia. Although these results are of great significance and will have a substantial impact on the acute care of stroke patients, we nevertheless would like to address 2 important points. First, although the authors applied different scores for evaluating the patients’ clinical status, they did not use these findings when determining independent risk factors for pneumonia. One of the most important clinical features putting patients at an increased risk of pneumonia is a decreased level of consciousness.2–4 Pathophysiologically, this condition leads to (1) an attenuation of protective reflexes, (2) an impaired functioning of the lower esophageal sphincter and a delayed gastric emptying, and (3) a worsening of the coordination of breathing and swallowing, thereby predisposing the individual to aspiration independent of the underlying disease.5 In our opinion it would therefore be interesting to investigate whether a decreased level of consciousness was also independently associated with pneumonia in the present study. Additionally, it would be of clinical relevance to compare the risk attributable to a decreased level of consciousness with the risk attributable to the overall clinical stroke severity, which was measured by the authors with the NIH Stroke Scale. Second, there is a continuing debate about risks and benefits of tube feeding in dysphagic stroke patients. On the one hand, nasogastric tubes are generally recommended as a safe way to supply nutrition in these patients.6–8 Thus, in a study on postacute stroke patients, Nakajoh et al found that the frequency of pneumonia was significantly higher in dysphagic patients with oral feeding than in those with tube feeding.9 On the other hand, there is contrary evidence that nasogastric tubes may predispose patients to aspiration.10 This is believed to be secondary to increased oropharyngeal secretions, impairment of laryngeal elevation, and disruption of the upper and lower esophageal sphincters.11 Contributing to this important controversy, the authors of the present study could investigate whether dysphagic patients with tube feeding had a higher risk to acquire pneumonia than those with oral or parenteral feeding.