In arthritic knees with severe valgus deformity Total Knee Arthroplasty (TKA) can be performed through medial or lateral parapatellar approaches. Many orthopaedic surgeons are apprehensive of using the lateral parapatellar approach due to lack of familiarity and concerns about complications related to soft tissue coverage and vascularity of the patella and the overlying skin. However surgeons who use this approach report good outcomes and no added complications. The purpose of our study was to compare outcomes following TKA performed through a medial parapatellar approach with those performed through a lateral parapatellar approach in arthritic knees with severe valgus deformity. We conducted a retrospective review of patients from two consultants using computer navigation for all their TKAs. All patients with severe valgus deformities (Ranawat 2 & 3 grades) operated on between January 2005 and December 2011 were included. 66 patients with 67 TKAs fulfilled the inclusion criteria. Patients were group by approach; Medial = 34TKAs (34 patients) or Lateral = 33 TKAs (32 patients). Details were collected from patients' records, AP hip-knee-ankle (HKA) radiographs and computer navigation files. Outcome measures included lateral release rates, post-operative range of knee movements, long leg mechanical alignment measurements, post-operative Oxford scores at six weeks and one year, patient satisfaction and any complications. Comparisons were made between groups using t-tests. The total cohort had a mean age of 69 years [42–82] and mean BMI of 29 [19–46]. The two groups had comparable pre-operative Oxford scores (Medial 41[27–56], Lateral 44 [31–60]) and pre-operative valgus deformity measured on HKA radiographs (Medial 13° [10°–27.6°], Lateral 12° [6°–22°]). Three patients in the Medial group underwent intra-operative lateral patellar release to improve patellar tracking. Seven patients in the Lateral group had a lateral condyle osteotomy for soft tissue balancing (one bilateral). There was no statistically significant difference between groups at one year follow up for maximum flexion (Medial 100° [78°–122°], Lateral 100° [85°–125°], p=0.42), fixed flexion deformity (Medial 1.2° [0°–10°], Lateral 0.9° [0°–10°], p=0.31) or Oxford score (Medial 23 [12–37], Lateral 23 [16–41], p=0.49). Similarly there was no difference in the patient satisfaction rates between the two groups at one year follow up. However there was a statistically significant difference in the mean radiographic post-operative alignment angle measurement (Medial 1.8° valgus [4° varus to 10° valgus], Lateral 0.3° valgus [5° varus to 7° valgus], p=0.02). One patient in the Medial group had a revision to hinged knee prosthesis for post-operative instability. There was no wound breakdown or patellar avascular necrosis noted in either of the groups. The lateral parapatellar approach resulted in slightly better valgus correction on radiographs taken six weeks post-operatively. We found no major complications in the Lateral parapatellar approach group. Specifically we did not encounter any difficulties in closing the deep soft tissue envelope around the knee and there were no cases of patellar avascular necrosis or skin necrosis. Hence we conclude that lateral parapatellar approach is a safe and reliable alternative to the medial parapatellar approach for correction of severe valgus deformity in TKA.
Post operative warfarinisation of elective arthroplasty patients delays their discharge. We retrospectively analysed all patients who required warfarinisation post surgery from April to September 2011. We identified the number of extra days stayed for the sole purpose of warfarinisation (i.e. after discharge by Physiotherapy and Occupational Therapy) and estimated the cost implications of this extended stay. 76 patients were warfarinised post operation, mean age 70.6 years (50–87) with 42 females and 34 males, 37 THR and 33 TKR. The mean extra days stayed was 3.1 (range 0 to 9). Atrial fibrillation and previous venous thromboembolism (DVT/PE) were the most common indication, 78%, followed by a current episode of DVT/PE, 11%. The nature of joint replacement made no difference to the extra days stayed (3.1 for THR and 2.9 for TKR) or the INR (2.27 in both groups) at discharge. Random loading dose instead of the recommended 5 mg of warfarin resulted in prolonged stay, 4.5 days compared to 3 days otherwise. The approximate cost per inpatient day is £500 (£137 nursing, £163 medical and £200 for facilities). From our results this amounts to £1500 per patient and £228,000 a year. In addition, there is a loss of income as the bed occupancy means not being able to undertake another arthroplasty surgery (£3,600 per patient) and possible failure to achieve waiting time targets. We conclude that substantial financial and resource savings can be made if warfarinisation is undertaken at the community level.
