Total Knee Arthroplasty Performed With Long-Acting Liposomal Bupivacaine Versus Femoral Nerve Catheter
Inadequate pain management after total knee arthroplasty (TKA) can interfere with participation in and progression of physical rehabilitation, and thereby prolong hospital stay, and increase costs and overall dissatisfaction with the procedure. At our institution, TKA traditionally has been performed with femoral nerve catheters (FNCs) for postoperative pain control.
We conducted a retrospective, longitudinal, repeated -measures study to compare FNC and long-acting liposomal bupivacaine (LALB) with respect to pain control, range of motion, ability to ambulate, and hospital length of stay. Twenty-three patients underwent separately staged bilateral TKAs, the first with FNC and the second with periarticular injection of LALB.
Statistically significant differences favoring LALB over FNC were found for hospital length of stay (P < .01), per-attempt walking distance during hospitalization (P < .01), total range of motion (extension plus flexion) at 3-week follow-up (P = .02), and total morphine-equivalent dose during hospitalization (P = .02).
Our results showed that, compared with patients who received FNC, patients who received LALB had comparable pain control, improved knee range of motion, and shorter hospital stays. Additional clinical studies are needed to better determine the efficacy and cost-effectiveness of LALB and other long-acting local anesthetic formulations.
In the first TKAs, FNC-administered ropivacaine 0.2% (2 mg/mL) was maintained at a standard basal rate of 8 mL/h for 48 hours. In the second TKAs, LALB was administered along with bupivacaine/epinephrine. Twenty milliliters of LALB from a single-use vial was diluted in 40 mL of normal (0.9%) saline to obtain a 60-mL solution, and a 25-gauge needle was used to inject this solution into the periarticular soft tissues; another needle was used for PAI of 30 mL of bupivacaine 0.25% with epinephrine.
Continuous passive motion devices were not used. Most patients began therapy on day of surgery. Knee immobilizers were not used in the FNC group.
The same standardized multimodal pain management protocol was used for all TKAs. Non- narcotic medications, including acetaminophen, ketorolac, and celecoxib, were given on a scheduled basis. Tramadol and opioid medications were administered as needed for pain. The attending physician based patient discharge timing on pain control, ability to safely ambulate, and absence of complications.
Outcome Measures
Outcome measures were LOS; extension and flexion at discharge and 3-week follow-up; total ROM (extension plus flexion) at discharge and 3-week follow-up; per-day and total hospital stay morphine -equivalent doses (MEDs); and per-attempt walking distance during gait training.
ROM was measured with a standard goniometer. Flexion was tested with the patient supine and the hip and knee in neutral rotation. The goniometer axis was along the lateral epicondyle of the femur with the proximal arm of the goniometer parallel to the long axis of the femur and pointing at the greater trochanter and with the distal arm parallel to the long axis of the fibula and pointing at the lateral malleolus. The patient was instructed to flex the hip and knee by moving the heel toward the buttock. Expected normal ROM is 135°. The same landmarks were used for extension. The patient was instructed to push the back of the knee toward the plinth/bed, for maximal active extension. The same ROM assessment strategy was used during the hospitalization and at the 3-week follow-up.
Several opioid medications (eg, hydrocodone, oxycodone, tramadol, hydromorphone, morphine) with different dosages were used during hospitalization. Opioid doses were converted to MEDs to permit FNC–LALB comparisons. For each patient, total MEDs were divided by LOS to determine MEDs per day.
Mean per-attempt walking distance was calculated by dividing the total distance walked during hospitalization—the sum of the number of feet walked during each and every attempt, as measured by the treating physical therapist—by the total number of walking attempts.
Data Analysis
A paired-samples t test was used to calculate differences between all outcome measures: LOS; extension and flexion at discharge and 3-month follow-up; per-day and total MEDs; and mean per-attempt walking distance. P < .05 was considered significant. We elected not to adjust our α for a potential familywise error.
Results
Of the 23 patients, 14 were female and 9 were male, and 19 were white and 4 were black. Mean (SD) age was 64.4 (6.4) years for the FNC group and 66.0 (6.0) years for the LALB group. The age difference was not statistically significant.
