Introduction

Total knee arthroplasty (TKA) has become one of the most frequent and successful surgeries in our specialty. However, pain and recovery after surgery still remain a pitfall. There are many publications stating that rapid recovery or fast track pathways may increase the functionality while reducing costs, but to our knowledge, no clinical trials have been published regarding these protocols.

MATERIALS
We performed a randomized clinical trial (NCT03823573) in patients undergoing total knee replacement after a diagnosis of knee osteoarthritis. Intervention group was included in a Rapid Recovery (RR) protocol, with local infiltration of levo-bupivacain in the periarticular tissue and ambulation was performed by a physiotherapist 4-6 hours after surgery (no drains were used). Control group received a femoral nerve block with levo-bupivacain by the anesthesiologist, whilst a drain was used. Ambulation started when the drain was retired, at 24-48 hours. All the patients received 2 g of intraarticular tranexamic acid after closure. Prosthesis model was at the election of the surgeon (Zimmer Persona or Optetrak Logic Fit. Zimmer NexGen was used for patients with history of metal allergy)
Collected data included patients’ demographics, hospital stay, time until ambulation, haemoglobin drop and need of red cell transfusión, flexion-extension at the beginning and on discharge of rehabilitation as well as complications and readmissions. All the patients received an Oxford Knee Score (OKS) prior to surgery and 6 months after discharge and were performed an eco-doppler to assess the presence of deep vein thrombosis). All the patients were followed up a minimum of 6 months.
Inclusion criteria included: patients between 55 and 80 years old undergoing TKA. Patients personal history of deep vein thrombosis or pulmonary embolism and allergy to any of the provided medicines were excluded.

Results

A total of 175 patients were included in the trial (92 patients in the control (C) group, 83 patients in the intervention (RR) group). There were no differences in sex (C: 38% male, 62% female; RR: 39% male, 61% female), side (C: 51% left, 49% right; RR: 49% left, 51% right), age (C: 71.3 years, RR: 70.7 years, p=0.524) or implanted prosthesis (C: 60% Persona, 35% Logic, 5% NexGen of which 78% were posterior-stabilized; RR: 68% Persona, 31% Logic, 1% NexGen, of which 75% were posterior-stabilized).
There were not significant differences on time of ischemia (C: 84.09 minutes; RR: 85.35 minutes; p=0.565), haemoglobin drop (C: 2,58 g/dL; RR: 2,49 g/dL; p=0.534), need for transfusion (C: 3,3% of patients; RR: 0%; p=0.248), need for morphine shots (C: 1.55 mg; RR: 1.81 mg; p=0.543), range of motion (ROM) at the beginning of the rehabilitation: flexion (C: 92.6 degrees, RR: 93.2 degrees; p=0.816) extension (C: -10.1 degrees, RR: -8.3 degrees; p=0.215) and at the end of the follow up: flexion (C: 109.7 degrees, RR: 109.8 degrees, p=0.961), extension (C: -4.7 degrees; RR: -3.6 degrees; p=0.172), OKS improvement (C: 18.14 points; RR: 19.74 points; p=0.429), patient satisfaction (above 10): (C: 8.35; RR: 8.33; p=0.942) or admissions at the emergency department (C: 8.7%; RR: 10.8%; p=0.632).
The only significant variables were hospital stay (C: 3.84 days; RR: 2.54 days, p<0.0001) and time until ambulation (C: 2.46 days; RR: 0.23 days; p<0.0001)

Conclusion

Rapid Recovery protocols can reduce hospital stay and costs without increasing complications or need for readmission. Although non statistically significant, our protocol improved all the clinical variables registered. Further studies with bigger samples could obtain statistically significant results.