[Possible predictors of discontinuation of basal-flow of postoperative patient-controlled analgesia].

Background: The purpose of this study was to identify possible predictors of discontinuation of basal-flow of postoperative patient-controlled analgesia.

Methods: We reviewed postoperative pain assessment records by the postoperative pain service team from April 2010 to July 2011 in which surgical patients were provided with intravenous or epidural patient-controlled analgesia (IV-PCA or Epi-PCA). From these data, we extracted cases with discontinuation of basal-flow of PCA, and candidate variables such as patients' characteristics, preoperative and intraoperative variables were assessed. Predictors with significant univariate association (P < 0.20) with the primary outcome were used to construct multivariable logistic regression models.

Results: We enrolled 685 patients for IV-PCA and 606 for Epi-PCA and obtained discontinuation groups (105 and 73 cases, respectively) with this cohort data. Results of multivariate analysis showed female, non-laparotomy, low body weight, and non-droperidol as independent risk factors for IV-PCA and low body weight, no-co-existing disease, and gastrointestinal surgery for Epi-PCA. There were no significant differences in pain intensity between discontinuation and non-discontinuation cases. The primary cause of discontinuation was PONV for IV-PCA and hypotension for Epi-PCA, respectively.

Conclusions: We should apply IV-PCA for female slender surgical patients undergoing non-laparotomy with great caution and provide prevention for PON. We should pay attention to incidence of postoperative hypotenion when we administer Epi-PCA to slender gastrointestinal surgical patients without co-existing disease.