[Management of critical severe acute respiratory syndrome and risk factors for death].

Liu, Xiao-qing; Chen, Si-bei; He, Guo-qing; Li, Yi-min; He, Wei-qun; Chen, Rong-chang; Zhong, Nan-shan

OBJECTIVE To investigate the management of critical severe acute respiratory syndrome (SARS) and possible risk factors for death. METHODS Thirty-three patients with SARS referred to Guangzhou Institute of Respiratory Disease (GIRD) between December 2002 and March 2003 were studied retrospectively. Paired t-test using statistical software SPSS 10.0 was employed to compare the respiratory frequency and pulse O(2) saturation (SpO(2)) before and after noninvasive ventilation with continuous positive airway pressure (CPAP) on the 33 patients. Among them, 18 patients who presented with SpO(2) of 90% - 93% at rest under O(2) inhalation 5 L/min were further recorded for their SpO(2) during slight physical movement and CPAP. The possible death-related risk factors including age, underlying diseases, leucocytosis, thrombocytopenia, and lymphopenia were analyzed by the Mantel-Haenszel chi(2) test. RESULTS The respiratory frequencies were significantly decreased and SpO(2) was improved in the 33 patients after one hour of noninvasive ventilation. The SpO(2) in the latter 18 patients fell during slight physical movement and improved one hour after CPAP (p < 0.01). The RRs of the five death predictors concerning thrombocytopenia, age (> 50 yrs, underlying diseases, leucocytosis, and lymphopenia were 25.83, 8.57, 6.40, 1.64, and 1.17, respectively, with the 95% CI of 3.64 - 183.59, 1.94 - 37.83, 1.75 - 23.33, 0.38 - 7.05 and 0.16 - 8.48, respectively. Management with corticosteroids effectively ameliorated and suppressed the development of pulmonary fibrosis. CONCLUSIONS Noninvasive ventilation relieves dyspnea and SpO(2) in patients with critical SARS, and should also be employed in those with SpO(2) of 90% - 93% at rest under O(2) inhalation 5 L/min. Noninvasive as an add-on management may probably cut down on the dosage and duration of corticosteriod therapy. Among 5 possible risk factors, 3 were recognized as death-related, turning out to be thrombocytopenia, age (> 50 yrs) and underlying diseases.