Severe acute respiratory syndrome: radiographic evaluation and clinical outcome measures.

Ooi, Clara G C; Khong, Pek L; Ho, James C M; Lam, Bing; Wong, Wai M; Yiu, Wai C; Wong, Poon C; Wong, Chi F; Lai, Kar N; Tsang, Kenneth W T

PURPOSE To evaluate the relationship among chest radiographs, oxygen supplementation requirement, and treatment response in severe acute respiratory syndrome (SARS). MATERIALS AND METHODS Forty patients (20 women, 20 men; mean age, 42.90 years +/- 14.01 [SD]; median age, 41.5 years; age range, 25-82 years) with SARS were evaluated. Daily chest radiographs were graded according to percentage of lung involvement during 20.15 days +/- 5.56 (median, 20 days; range, 14-38 days). Times between symptoms and treatment and time to reach maximal radiographic score from admission and treatment day were determined. Daily oxygen saturation (Sao2) and oxygen supplementation including mechanically assisted ventilation were recorded. Treatment response was defined as good, fair, and poor. Patterns of radiographic opacity at admission and at maximal radiographic score were noted. Differences in radiographic and clinical parameters with respect to oxygen supplementation and treatment response were respectively evaluated with Mann-Whitney and Kruskal-Wallis tests. RESULTS Larger maximal radiographic scores, lower Sao2 at maximal radiographic change, longer time from treatment to maximal radiographic score (P <.01), and diffuse consolidation at maximal radiographic score were associated with oxygen supplementation. Parameters that influenced treatment response were time from symptom onset to treatment day (P =.003), time from admission to treatment day (P <.001), time to maximal radiographic score from treatment day (P =.001), maximal radiographic score (P =.009), Sao2 at maximal radiographic score (P =.13), and treatment radiographic score (P =.03). Fair responders had shorter time between admission and treatment than did either good (P <.001) or poor responders (P =.002) and shorter time between symptoms and treatment (P <.001) and lower treatment radiographic score (P =.012) than did good responders. Good (82%), poor (36%), and fair (33%) responders developed maximal chest radiographic scores within 4 days of treatment (P =.008). Radiographic patterns at both admission and maximal radiographic score did not influence treatment response. CONCLUSION There are significant relationships among radiographic parameters, oxygen supplementation, and treatment response, and these relationships appear to be clinically useful in the treatment of SARS.