Mark N. Rubin, MD

Ji R, Haipeng H, Pan Y, Du W, Wang P, Liu G, Wang Y, et al. Risk Score to Predict Hospital-Acquired Pneumonia After SpontaneousIntracerebral Hemorrhage. Stroke. 2014


Unequivocally, our patients with intracerebral hemorrhage are quite ill. Clinical experience with this highly morbid and potentially fatal disease demonstrates that patients with this condition are at multisystem risk, including worsening hemorrhage, thromboembolism, cardiac complications, and infections including pneumonia. While it comes as no surprise that patients with intracerebral hemorrhage, who are often neurologically disabled if not comatose, experience hospital-acquired pneumonia on an all-too-frequent basis, it seems to happen in spite of our best multidisciplinary efforts to screen for dysphagia and adhere to infection precautions. This is a particularly important issue for hemorrhagic stroke patients because the development of pneumonia is a predictor of worse outcome overall and increased mortality. This is also of major health economic concern, as the development of hospital-acquired pneumonia is a heavily scrutinized aspect of hospital performance in the United States and tied to reimbursement for services. A group of Chinese investigators – who already brought us a pneumonia prediction score for acute ischemic stroke quite recently – have collaborated again with an eye toward pneumonia prediction in patients with spontaneous intracerebral hemorrhage.



The investigators put together a cohort of nearly 5000 patients with spontaneous intracerebral hemorrhage across many institutions in China, collectively referred to as the China National Stroke Registry. This cohort was split for derivation and validation of clinical features that might predict the development of pneumonia based on previously published risk factors. In brief, they derived a 20+ point risk score that reasonably predicted the development of pneumonia after intracerebral hemorrhage, irrespective of hematoma volume, particularly in patients admitted for greater than 48 hours. They found that older age, current smoking, chronic obstructive pulmonary disease, pre-stroke dependence, low admission Glasgow Coma Score, high National Institute of Health Stroke Scale score, and dysphagia were independently and significantly associated with the development of hospital-acquired pneumonia after intracranial hemorrhage.

The natural goal after the development of such scores is to predict so that we may prevent. This score requires  external validation, but the hope is that the score can inform clinical trials aimed at the prevention of pneumonia in these very ill patients, thus averting the addition of the proverbial insult to an already grave injury.