ANESTHESIA CLINICAL PATHWAYS FOR SAME-DAY ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION DECREASE TIMES TO DISCHARGE HOME

Abstract

S22 Understanding which activities add value to patient care will likely decrease costs generated by process variability. A useful indicator of the cost-effectiveness of an anesthetic in the ambulatory setting is the recovery time, or the time from OR exit to discharge from the facility. In studying this quality indicator, we developed anesthesia clinical pathways (CP) in conjunction with surgeons who routinely care for outpatients undergoing anterior cruciate ligament (ACL) reconstruction of the knee. We demonstrate that these CPs decrease the total time for phase I and phase II recovery, when compared with time data for historical controls who underwent non-standardized anesthesia and postoperative care plans. Methods: After IRB approval, we reviewed the medical records of consecutive outpatients undergoing ACL reconstruction from July 1995 to June 1997. Patients in academic year (AY) 1995-96 were not subjected to standardized anesthesia care. AY 1996-97 patients (in whom no contraindications existed) selected epidural anesthesia (EA), general anesthesia (GA), or GA with femoral nerve block (GFNB). EA consisted of lumbar epidural catheters placed preoperatively after sedation with fentanyl and midazolam, epidural dosing with lidocaine and fentanyl, and intraoperative sedation with propofol. Both GA and GFNB included induction with propofol and muscle relaxation with succinylcholine; maintenance with endotracheal desflurane, nitrous oxide, and oxygen; and opioid analgesia with fentanyl. Intraoperative fentanyl doses were decreased in the GFNB patients, who had femoral nerve blocks (with bupivacaine) placed preoperatively. All CP patients were given intraarticular morphine (10 mg) and bupivacaine (0.5%, 30cc) at the end of the surgical procedure. Patients at risk by history for postoperative nausea/vomiting (PONV) were given empiric droperidol and metoclopramide intraoperatively, and patients with refractory PONV were given a sequence of metoclopramide, droperidol, ondansetron, and perphenazine postoperatively before admission was considered. Duration of recovery data from these patients were compared with historical controls, who had a non-standardized intraoperative and postoperative anesthesia care plan. The only similarity between the historical controls and the CP patients was the intraarticular injection at the end of the procedure. Recovery time values were analyzed using one-way analysis of variance (ANOVA), with significant differences in pairwise comparisons determined by independent-sample t-tests. P < 0.05 was considered statistically significant. Results: Time data from 150 ACL anesthesia CP patients (31 CP-GA, 62 CP-EA, 57 CP-GFNB) and 122 historical (H) controls (26 regional, 96 general) were examined. Patients who were known in advance to be admitted postoperatively, who were admitted because of an extension of the surgical procedure, or who were admitted for intractable pain or PONV were excluded from this analysis. CP patients (207 +/- 20 min, mean +/- std. error of mean) were discharged home 35 minutes sooner than were historical controls (242 +/- 14 min, P = 0.008). No significant differences in discharge times were seen within the CPs. Discussion: For ACL reconstruction, the anesthesia process influenced recovery times. Whether anesthesia CP for other surgical procedures would benefit or adversely affect recovery times, and whether decreases in recovery times would create scheduling opportunities with respect to recovery nurse staffing, [1] requires further study.