Manipulation for Stiffness Following TKR Ann R Coll Surg Engl. 2009 Apr; 91(3): 220–223. LINK Riazuddin Mohammed, Shakir Syed, Royal Orthopaedic Hospital, Birmingham, UK Naveed Ahmed, Alexandra Hospital, Redditch, UK INTRODUCTION Stiffness following knee arthroplasty is a frustrating complication and a significantly disabling problem. We present our experience of knee stiffness requiring manipulation under anaesthesia (MUA) as the first line of treatment following partial or total knee arthroplasty. PATIENTS AND METHODS All stiff knees requiring MUA over a 6-month period from March to August 2007 were retrospectively analysed from theatre records and case notes. Data were collected regarding demographics, range of knee motion before and after manipulation and at subsequent follow-up. Complications of the procedure and outcomes were also analysed. Stiffness following knee arthroplasty is a disabling problem and a frustrating complication both for the surgeon and the patient. The actual definition of what constitutes knee stiffness is variable in the literature. When the problem is identified early, a trial of intensive physiotherapy might help in regaining functional range of movement. When this fails, the various treatment options available are manipulation under anaesthesia (MUA), arthrolysis (open/arthroscopy), exchange poly and revision arthroplasty. We have offered MUA as the first treatment option, after failed physiotherapy, for all stiff knees following arthroplasty at our centre. This study examines the immediate outcome of MUA following knee arthroplasty and its benefit at a short-term follow-up. We retrospectively evaluated all the patients who had MUA after knee arthroplasty between March 2007 and August 2007. For the purpose of study, unicondylar, tricompartmental, patellofemoral and revision arthroplasties were included as knee arthroplasties. The indication for MUA was knee range of movement <70° or flexion contracture >15°. The patients were also reviewed at follow-up to monitor the range of improvement in the short term. The data were gathered from theatre records and case notes. Factors analysed were the relevant predisposing factors, type of implant used, the surgeon grade, the duration between the arthroplasty and the MUA and the type of anaesthesia. The ranges of motion before and after MUA and at follow-up were measured using a pocket goniometer. Statistical analysis was done using the two-sample t-tests for comparison of range of movements before and after manipulation. RESULTS 21 patients (11 female, 10 make) underwent MUA. following knee arthroplasty in the 6 months of the study period. The average age was 62 years (range, 56–80 years) with 11 female and 10 male patients. Primary osteoarthritis was the indication for arthroplasty in all the patients. The relevant past medical history included 7 patients with body mass index (BMI) more than 30 kg/m2, 3 patients with diabetes mellitus and 1 patient on long-term warfarin. 16 primary TKRs (76% of MUA) and 3 revision TKRs (14% of MUA) needed manipulation, and 1 patient each (5% of MUA) for medial unicompartmental knee arthroplasty and patellofemoral joint replacement. All the primary arthroplasties were performed by consultant surgeons except for two procedures done by trainee registrars. Table 1 shows different types of implants used for primary knee arthroplasty in the patients. 13 manipulations (65%) were done under 12 weeks and 8 manipulations after 12 weeks following the primary procedure. The mean duration between arthroplasty and MUA was 13.2 weeks (range, 6–32 weeks). Except for 1 patient who had the manipulation done under spinal anaesthesia (for medical reasons), general anaesthesia with good muscle relaxation was used for all the other patients. There were no complications noted from the procedure. The range of knee movement improved from a mean range of 10.4°–71.2° in the pre-MUA period to 2.1°–94.0° post-MUA and at follow-up was 2.3°–91.9. Manipulation improved the range of movement in all cases, except in 1 patient. The mean improvement in knee movement from the pre-MUA at follow-up was 30.2°. (Fig. 1) TECHNIQUE OF MUA Under adequate muscle relaxation, the manipulation was done with the ipsilateral hip flexed to 90°. With the leg held close to the knee joint (to decrease the lever arm), steady progressive loading was applied till palpable and audible break of adhesions were heard. Beyond this firm point, no further pressure was applied. Postoperatively, active assisted physiotherapy was instituted. The use of a CPR or cryotherapy was advocated on an individual patient basis. The patients were discharged from physiotherapy when satisfactory range of movement was obtained. CONCLUSIONS MUA has a role in the treatment of early stiffness with excellent immediate outcomes. We advocate that MUA should be the first line of management for stiff knee arthroplasties after failed physiotherapy. Stiffness following knee arthroplasty is a disabling problem and a frustrating complication both for the surgeon and the patient. The actual definition of what constitutes knee stiffness is variable in the literature. Kim et al.1 describe stiffness as flexion contracture of >15° and/or <75° of flexion of the knee, whereas Christensen et al.2 have defined it as an arc of knee motion <70°. When the problem is identified early, a trial of intensive physiotherapy might help in regaining functional range of movement. When this fails, the various treatment options available are manipulation under anaesthesia (MUA), arthrolysis (open/arthroscopy), exchange poly and revision arthroplasty. We have offered MUA as the first treatment option, after failed physiotherapy, for all stiffness in knees following arthroplasty at our centre. This study examines the immediate outcome of MUA following knee arthroplasty and its benefit at a short-term follow-up. DISCUSSION Manipulation for stiffness following total knee replacement has a reported low prevalence rate (1.3%).1 Our study included 21 manipulations done in a 6-month period. In the same period, the total number of knee arthroplasty performed at our centre was 519 (429 total knee replacements, 70 unicompartmental knee replacements and 20 patellofemoral joint replacements). This roughly estimates the prevalence of MUA to be 4%. Our study also included MUA performed for stiff revision knee arthroplasty and partial knee arthroplasty. The aetiology for stiffness following knee arthroplasty is poorly understood and is, therefore, described as being multifactorial. Pre-operative, intra-operative and postoperative causes have been cited as reasons for the stiffness.3,4 In our study group, the relevant pre-operative factors were present in a total of 8 patients. These included 3 patients with high BMI and diabetes, 4 patients with high BMI and 1 patient who was on long-term warfarin for atrial fibrillation. The mean duration between knee arthroplasty and the manipulation was 13 weeks in our study. In 13 patients, MUA was done within 8 weeks of the primary procedure. Our aim was to recognise knees with poor range of motion at the first follow-up (6 weeks) and offer manipulation. Though the verdict is not clear, various studies have shown no difference in outcomes from early or late manipulation.5,6 However, an early (at 2 weeks' postoperatively) manipulation study of patients with rheumatoid arthritis showed that MUA failed to increase the ultimate flexion of the knee in 70% of the cases.7 The immediate results from manipulation in our group were satisfactory. Only 1 patient did not benefit from MUA. This lady had a painful revision knee arthroplasty with only 20–65° movement. She did not improve following manipulation either immediately or on follow-up. 7 patients in our study had flexion contractures of more than 15°. Four of these improved to full extension after manipulation which was maintained at follow-up. In 3 patients, only moderate improvement in extension could be retained at the follow-up. Namba et al.6 have shown that flexion contractures are not significantly improved after late MUA (after 90 days). The reported complications from MUA in the literature are fracture, wound dehiscence, patellar ligament avulsions, haemarthrosis, heterotopic bone formation and pulmonary embolism.7–11 We did not encounter any complications during or after the manipulation in any of the patients. It is important to recognise extrinsic factors (e.g. neurologically related muscle rigidity, muscle tightness due to injury or ipsilateral hip osteoarthritis) before excluding the intrinsic causes for stiffness.12 MUA will only be useful for intra-articular adhesions and not for oversized or mal-aligned components and soft tissue imbalances. The aim of MUA is to break the intra-articular adhesions; any further manipulation beyond this can result in untoward complications. Not all stiff knees require or undergo a manipulation procedure. Patients with pain-free joints choosing to accept their restricted joint motion or medical reasons that prevent further procedures under anaesthesia are examples of such cases. It would be of benefit to know if stiff knees, if left to themselves and continued with physiotherapy, would improve in due course. However, a literature search did not reveal any such case-control studies. Therefore, it is our recommendation that a large prospective case-control study be conducted to look into this aspect. Conclusions Manipulation for knee stiffness is notorious for the various complications associated with the procedure. However, with careful attention to technique, it does offer a safe and effective solution to the problem. We recommend MUA as the first line of treatment after failed physiotherapy for stiffness following knee arthroplasty. Josephine edit: Whilst not wishing to demean the quality of this piece of research, it should be noted that a cohort of 21 makes a very small study so should not be taken vertbatim.