The Association Between Incentive Spirometry Performance and Pain in Postoperative Thoracic Epidural Analgesia
Introduction: Effective use of postoperative incentive spirometry improves patient outcomes, but is limited by pain following thoracic and upper abdominal surgery. Thoracic epidurals are frequently employed to provide analgesia and attenuate postoperative pulmonary dysfunction. We hypothesized that in patients with thoracic epidural analgesia following thoracic and abdominal surgery, high pain scores would be associated with poorer incentive spirometry performance, even when accounting for other variables.
Methods: Retrospective study of 468 patients who underwent upper abdominal or thoracic surgery utilizing postoperative thoracic epidural analgesia between June 1, 2009 and August 31, 2013 at a single tertiary academic center. Institutional Review Board (University of Michigan, Ann Arbor, MI) approval was obtained. The association between incentive spirometry performance and pain was assessed as the primary outcome. Additional preoperative data analyzed included age, sex, height, weight, planned surgical procedure, smoking history, ASA physical status classification, preoperative numeric rating scale pain score, presence of pain lasting three months or more, and preoperative opioid use. Data were analyzed using R version 3.0.2. Univariate linear regression was used to describe the relationship between incentive spirometry volume outcomes on postoperative days 1 and 2 and the appropriate covariates. Multivariable regression models were built to control for potential confounders (additional preoperative data noted above) and to identify independent associations with incentive spirometry performance. Stepwise variable selection by Akaike information criteria (AIC) provided reduced models from the initial full multivariable models.
Results: Postoperative incentive spirometry (IS) performance was found to be inversely proportional to pain score, which correlated significantly stronger with deep breathing pain compared to pain at rest (-0.33 vs. -0.14 on postoperative day 1, -0.23 vs. -0.12 on postoperative day 2). Pain with deep breathing was independently associated with lower incentive spirometry volumes such that every 3-point increase in pain was associated with approximately 250 mL lower incentive spirometry volume after adjusting for other covariates. Increasing age, female sex, thoracic surgery (vs. abdominal), use of intravenous patient-controlled analgesia, and higher American Society of Anesthesiologists physical status score were also associated with poorer incentive spirometry performance.
Conclusions: To our knowledge, this is the first study to demonstrate an inverse correlation between incentive spirometry performance and pain scores in patients with epidural anesthesia for postoperative pain management. This clinically significant correlation suggests that patients with high pain scores during deep breathing are splinting, which puts them at increased risk for postoperative pulmonary complications. Assuming that one of the primary goals of thoracic epidural analgesia is to attenuate postoperative pulmonary dysfunction and prevent pulmonary complications, pain score with deep breathing is a better indicator of thoracic epidural efficacy than is pain score at rest. In addition to including pain score with deep breathing in the assessment of thoracic epidural efficacy, incentive spirometry performance could be used as a functional measure of thoracic epidural efficacy. The addition of this objective measure of thoracic epidural efficacy may be particularly useful in patients that report high pain scores postoperatively but perform well with incentive spirometry.