[A clinical study of rapid-shallow-breathing index in spontaneous breathing trial with pressure support ventilation and T-piece].

OBJECTIVE: To determine the threshold value of rapid-shallow-breathing index (RSBI) and its increased percentage (Delta RSBI) as predictors for successful weaning in spontaneous breathing trial (SBT) with pressure support ventilation (PSV) and T-piece.

METHODS: In 2007, there were 208 patients on ventilators with oral intubation. They were divided into two groups: the PSV group (93 patients) and the T-piece group (115 patients). The duration of SBT was 30 minutes. Weaning indexes [mouth occlusion pressure 0.1 (P0.1), breathing rate (f), tidal volume (V(T))] of SBT 3 minutes and 30 minutes were recorded in each group. The averages of RSBI and Delta RSBI were calculated and compared between two groups.

RESULTS: Among 208 patients, in 168 patients weaning was successful and the ratio was 80.77%. The successful ratios of PSV and T-piece group were 83.87% and 78.26%, respectively (P > 0.05). In PSV and T-piece groups the average of RSBI value (SBT 30 minutes) was (67.18+/-11.55) breaths x min(-1) x L(-1) and (99.11+/-15.53) breaths x min(-1) x L(-1), respectively (P < 0.01); the average of Delta RSBI was (69+/-33)% and (119+/-35)% (P < 0.01). In PSV group, the area under receiver operating characteristic (ROC) curve of RSBI was 0.747+/-0.045 (P = 0.000); when RSBI =75 breaths x min(-1) x L(-1), the diagnostic accuracy was 87%; the area under ROC curve of Delta RSBI was 0.709+/-0.065 (P = 0.001), and when Delta RSBI=90%, the diagnostic accuracy was 82%. In T-piece group, the area under ROC curve of RSBI was 0.821+/-0.049 (P = 0.000); when RSBI =100 breaths x min(-1) x L(-1), the diagnostic accuracy was 82%; the area under ROC curve of Delta RSBI was 0.738+/-0.046 (P = 0.000); when Delta RSBI =130%, the diagnostic accuracy was 77%.

CONCLUSION: The average values of RSBI (SBT 30 minutes) are significantly different between the two groups. In PSV and T-piece groups, using RSBI = 75 breaths x min(-1) x L(-1) and 100 breaths x min(-1) x L(-1) as the threshold value for predicting successful weaning is more valuable than other values. Sequential observation of Delta RSBI is also valuable in predicting a successful extubation.