The pneumonia score: A simple grading scale for prediction of pneumonia after acute stroke Hyung-Min Kwon, MD, a Sang-Wuk Jeong, MD, PhD, b Seung-Hoon Lee, MD, a and Byung-Woo Yoon, MD, PhD a Seoul, Republic of Korea Medical complications are an important clinical problem after acute stroke and present potential bar- riers to optimal recovery. Pneumonia has been esti- mated to occur in one third of all stroke victims and is the most common respiratory complication. 1 Moreover, a recent study showed that pneumonia in- creases the risk of 30-day mortality. 2 In patients with acute stroke, most stroke deaths after the first week and before 30 days were reported to be due to com- plications of immobility (ie, pneumonia or pulmo- nary embolism). 3,4 Because pneumonia is critical in the functional status and prognosis, rapid assessment of a high-risk patient of pneumonia is needed. Some studies found that the neurologic deficit as rated on the National Institutes of Health stroke scale (NIHSS) was correlated with pneumonia 5-7 and observed that dysphagia and mechanical ventilation were also asso- ciated with pneumonia. 2,7 Stroke patients with paral- ysis and those with impaired level of consciousness also had had higher rates of pneumonia. 2,8 Despite the abundant studies of several important risk fac- tors, 1,2,5-9 no grading scale for development of pneu- monia after acute stroke is yet used. The purpose of this study was to define a clinical grading scale for pneumonia, which is predictive of incidence of pneumonia development and can be rapidly and accurately assessed at the early time after stroke onset. METHODS Subjects Stroke patients who were consecutively admitted between February 2001 and October 2002 to Seoul National University Hospital for initial stroke therapy were studied. Patients whose stroke occurred within 4 days of admission were included. This study ex- cluded patients with transient ischemic attacks or sub- arachnoid hemorrhage and those with pneumonia before admission. Pneumonia was diagnosed if a pa- tient had at least 1 of the following: (1) auscultatory res- piratory crackles and fever ($37.7°C in the axillary area), (2) radiographic evidence, or (3) new purulent sputum, as mentioned in the previous report. 9 Patients with a definite pneumonia on admission were ex- cluded to rule out preexisting pneumonia cases. A to- tal of 286 patients–192 (67.1%) male patients and 94 female–were included in this study. Risk factors All patients were examined daily by neurologists from admission to discharge. The following risk factors were subjected to analysis: age, sex, hypertension, dia- betes mellitus, smoking, initial stroke severity, blood pressure, initial body temperature (recorded using an electronic thermometer [TERUMO, Tokyo, Japan] on the axilla), fasting blood glucose level, and high sensi- tivity C-reactive protein (hs-CRP). Age was classified into dichotomized variables ($65 and ,65 years), and body temperature was classified as ,37.0°C and $37.0°C. The following cerebrovascular risk factors were recorded for all patients. Hypertension: Hypertension was considered to be present if a subject had 1 or more of the following conditions: (1) repeated blood pressure readings greater than 140/90 mm Hg at inter- vals of 1 week or more and (2) a history of hypertension and the use of antihypertensives. Diabetes mellitus: a history of diabetes mellitus with or without current treatment or fasting blood glucose levels .6.88 mmol/ dL (125 mg/dL). Smoking: current or exsmoker who had quit smoking less than 5 years before admission. From the Department of Neurology, Seoul National University Hospital, a and Inje University Ilsan Paik Hospital, b Seoul, Republic of Korea. Reprint requests: Byung-Woo Yoon,MD, PhD, Department of Neurol- ogy, Seoul National University Hospital, Yongon-dong 28, Chongno-gu, Seoul, 110-744, Republic of Korea. E-mail: bwyoon@snu.ac.kr . 0196-6553/$32.00 Copyright ª 2006 by the Association for Professionals in Infection Control and Epidemiology, Inc. doi:10.1016/j.ajic.2005.06.011 64