[The relationship between sputum eosinophils and responses to treatment of inhaled glucocorticoids in patients with persistent asthma].

OBJECTIVE: To evaluate the treatment responses of asthmatics with and without sputum eosinophilia to inhaled glucocorticoids, and therefore to verify whether low sputum eosinophils predict poor response to treatment with inhaled glucocorticoids. METHODS: Forty-two symptomatic asthmatic patients, who had not received glucocorticoid therapy in the 3 months preceding the study, were examined before and 1 month and 3 months after treatment with inhaled glucocorticoids. At each visit, all patients underwent spirometry, symptom scoring and sputum induction. The level of eosinophil cationic protein (ECP) in the sputum supernatants was measured by radioimmunoassay. The patients were divided into 2 groups according to sputum eosinophil (EOS) percentages, an EOS group (EOS > 3%) and a non-EOS group (EOS < 3%). The response to inhaled glucocorticoid therapy (as measured by symptom scores and FEV(1)% pred) and the changes of sputum measurements were compared between the 2 groups. RESULTS: In the EOS group, the baseline EOS [0.080 (0.063 - 0.178)] and ECP level [(324 +/- 149) microg/L] were significantly higher than those of the non-EOS group [0.017 (0.006 - 0.021) and (152 +/- 68) microg/L, respectively, t = 4.40, 3.33, both, all P < 0.01]. Baseline FEV(1), FEV(1)% pred and symptom scores were not different between the 2 groups [EOS group: (1.98 +/- 0.67) L, (65 +/- 20)%, 7.0 (5.0 - 10.0), non-EOS group: (2.07 +/- 1.05) L, (66 +/- 27)%, 5.0 (2.0 - 9.0), t = -0.62, -0.09, 1.32, respectively, all P > 0.05]. After 1 month and 3 months inhaled glucocorticoid therapy, the sputum EOS, ECP, the symptom score, FEV(1) and FEV(1)% pred were [0.019 (0.010 - 0.060), [0.036 (0.006 - 0.070); (173 +/- 153) microg/L, (173 +/- 122) microg/L; 3.0 (1.0 - 6.0), 3.0 (1.0 - 5.0); (2.42 +/- 0.64) L, (2.43 +/- 0.76) L; (77 +/- 13)%, (77 +/- 18)%; respectively in the EOS group, which were significantly different as compared to baseline values (F = 6.73, 6.71, 5.93, 7.38, 5.78, respectively, all P < 0.05). But in the non-EOS group, the sputum EOS, ECP, the symptom score, FEV(1) and FEV(1)% pred were 0.013 (0.000 - 0.025), 0.012 (0.004 - 0.031), (111 +/- 50) microg/L, (117 +/- 50) microg/L; 3.0 (0.0 - 6.0), 3.0 (1.0 - 7.3), (2.22 +/- 0.86) L, (2.21 +/- 0.24) L, (71 +/- 20)%, (65 +/- 21)%; respectively at 1 and 3 months, which showed that the sputum EOS, FEV(1) and FEV(1)% pred did not change (F = 1.98, 0.80, 1.37, respectively, all P > 0.05), but the ECP level and the symptom score improved (F = 3.78, 3.59, respectively, both P < 0.05). Multiple stepwise regression showed that baseline FEV(1), severity degree and sputum EOS correlated significantly with changes in FEV(1) after treatment. Among the baseline indexes examined, sputum EOS had the highest negative predictive value (89.5%) for the response to treatment. CONCLUSIONS: In asthmatics with low sputum EOS, inhaled glucocorticoid therapy for 3 months failed to improve pulmonary function. The result confirmed that low sputum EOS was the best predictor for poor glucocorticoid effect in asthma.