One- or two-stage bilateral total hip arthroplasty

Transcription

1 Hip One- or two-stage bilateral total hip arthroplasty A PROSPECTIVE, RANDOMISED, CONTROLLED STUDY IN AN ASIAN POPULATION S. Bhan, A. Pankaj, R. Malhotra From the All India Institute of Medical Sciences, New Delhi, India We compared the safety and outcome of one-stage bilateral total hip arthroplasty with those of a two-stage procedure during different admissions in a prospective, randomised controlled trial in an Asian population. Of 168 patients included in the study, 83 had a single- and 85 a two-stage procedure. Most of the patients (59.9%) suffered from inflammatory arthritis. The intra-operative complications, early systemic complications, the operating time, positioning of the components, the functional score, restoration of limb length and survival rates at 96 months were similar in the two groups. The total estimated blood loss was significantly lower in patients undergoing a one-stage procedure than in patients who had a two-stage procedure, but the transfusion requirements were significantly higher in the former group (p = 0.001). The hospital stay was significantly shorter in the one-stage group, 7.25 days (SD 1.30; 5 to 20) compared with 10 days (SD 1.65; 8 to 24) in the two-stage group (p = 0.023). We believe that a one-stage procedure is safe and appropriate in our population. S. Bhan, MS, FRCS, Professor and Head A. Pankaj, MS, DNB, MRCS(Ed), Senior Resident R. Malhotra, MS, Professor Department of Orthopaedics All India Institute of Medical Sciences, Ansari Nagar, New Delhi , India. Correspondence should be sent to Professor R. Malhotra; rmalhotra62@yahoo.com 2006 British Editorial Society of Bone and Joint Surgery doi: / x.88b $2.00 J Bone Joint Surg [Br] 2006;88-B: Received 4 August 2005; Accepted 26 October 2005 The optimal timing of surgery in patients requiring bilateral hip replacement is controversial. A bilateral procedure is required in approximately 15% to 25% of patients. 1 Bilateral total hip arthroplasty (THA) as a single procedure has been described, 2-5 but concerns have been expressed about the increased incidence of deep-vein thrombosis, 3 heterotopic ossification, 2 a reduced range of movement, 2 a suboptimal gain in walking ability, 6 and the mortality and morbidity in patients with medical problems. 7 The potential advantages of simultaneous bilateral THA include a single anaesthetic, a shorter hospital stay and better functional recovery in patients with severe bilateral deformities. While most published reports have compared the outcome of a single-stage bilateral procedure with that of a unilateral procedure by extrapolation, 2,4-6 studies comparing the outcome of a singlestage bilateral procedure with that of a staged procedure are relatively few and none had prospective randomisation. 3,8 Only by comparing a single-stage with a staged procedure can the safety and superiority of either approach be properly determined. We have compared the peri-operative safety, transfusion requirements, post-operative function, the incidence of complications, the duration of hospital stay and the survival of the implant in a short- to intermediate-term follow-up in patients with simultaneous bilateral THA with those in patients with a staged procedure during two different admissions in a prospective, randomised controlled trial in an Asian population. Patients and Methods Between December 1996 and July 2001 we enrolled 177 patients admitted for bilateral THA in this study. The exclusion criteria included an American Society of Anaesthesiologists (ASA) 9 grade higher than 4, bony ankylosis, bilateral severe deformities, significant involvement of the knee and/or spine, and an inability to obtain informed consent. Patients with bilateral bony ankylosis and bilateral severe deformities were excluded from the study because we considered that a one-stage procedure gave superior functional recovery and randomisation may have deprived this group of this advantage. Patients with significant disability affecting the knee or spine may have encountered difficulties in post-operative functional recovery, making comparison difficult. The regional research ethics committee approved the study protocol. Informed consent was obtained from all patients. They were randomly allocated using sealed envelopes on the day preceding the operation to have surgery in one or two stages. A total of 86 patients 298 THE JOURNAL OF BONE AND JOINT SURGERY

2 ONE- OR TWO-STAGE BILATERAL TOTAL HIP ARTHROPLASTY 299 Table I. Clinical details and pre-operative data on the two groups (n = 85) p value Mean age (SD) in yrs (14.98) (14.35) 0.39 Gender (M:F) 54:29 51: Mean (SD) pre-operative Hb (g/dl) * (0.81) (0.85) 0.88 Mean (SD) pre-operative haematocrit (0.046) (0.058) 0.36 Mean (SD) limb-length discrepancy (cm) 1.24 (1.01) 1.20 (1.1) 0.71 Mean (SD) pre-operative HHS 42 (12.34) 40.5 (13.39) 0.76 * Hb, haemoglobin HHS, Harris hip score Table II. Pre-operative co-morbidities as evaluated by the American Society of Anaesthesiologists (ASA) grading for both groups, by number and percentage (n = 85) Low risk ASA 1 49 (59) 54 (63.5) ASA 2 26 (31.3) 25 (29.4) High risk ASA 3 8 (9.6) 5 (5.9) ASA (1.2) Table III. Aetiology of hip disease in both groups, by number and percentage Diagnosis * (n = 85) Total (n = 168) AS 20 (24.1) 29 (34.1) 49 RA 19 (22.9) 20 (23.5) 39 AVN 16 (19.3) 22 (25.9) 38 OA 12 (14.5) 6 (7.1) 18 DDH 6 (7.2) 6 (7.1) 12 JRA 4 (4.8) 0 (0.0) 4 Others 6 (7.2) 2 (2.4) 8 * AS, ankylosing spondylitis; RA, rheumatoid arthritis; AVN, avascular necrosis; OA, osteoarthritis; DDH, developmental dysplasia of hip; JRA, juvenile rheumatoid arthritis; Others, included psoriatic arthropathy and alkaptonuria had a one-stage bilateral THA and 91 a two-stage procedure with an interval ranging from three to seven months between the two operations. The second operation in the two-stage group was not done earlier than three months after the first in order to allow the physiological parameters to return to normal. Three patients from the one-stage group and six from the two-stage group were not available for regular follow-up and were excluded from the study, leaving 83 patients (166 hips) in the one-stage group and 85 (170 hips) in the two-stage group for further evaluation. The age and gender distribution of patients in the two groups, the level of the haemoglobin and the haematocrit, the limb-length discrepancy and the Harris hip score (HHS) were recorded pre-operatively. No significant differences were found between the two groups (Table I). The levels of haemoglobin and the haematocrit in the two-stage group were calculated as the mean of two values before each operation. The fitness of the patients was categorised according to the ASA system. Both groups were further subdivided according to their pre-operative ASA status. Patients of ASA grade 1 and grade 2 were categorised as low risk and those of ASA grade 3 and grade 4 as high risk (Table II). The cause of the hip disease was noted (Table III). A single surgeon (SB) using a posterior approach performed all the operations eliminating potential bias related to the surgical technique. He had not been involved in randomisation of the patients and was informed of the patients allocation the day before the operation. The most symptomatic hip was always treated first. The procedures were carried out in a laminar airflow operating room under combined spinal and epidural anaesthesia. The decision as to whether cemented or cementless fixation was to be used was made at the time of admission, depending upon the quality of the bone stock and the age of the patient. All patients with good bone stock and less than 65 years of age (106 hips in the one-stage group and 104 hips in the two-stage group) underwent cementless arthroplasty (Harris Galante II and Versys Trilogy; Zimmer, Warsaw, Indiana). A cemented procedure using a CPT femoral stem and ZCA cup (Zimmer) was used in the rest of the patients (60 hips in the one-stage group and 66 in the twostage group). A third-generation cementing technique was used in the cemented hips. The press-fit technique of insertion of the acetabular shell involved under-reaming of the acetabulum by 2 mm before impaction of the implant prior to supplementary fixation by two or three screws in most cases. Before a bilateral procedure, two tables with draping equipment were set up and placed next to the laminar flow unit. One table was left covered during the first operation for later use on the opposite side. As the skin closure with staples was begun on the first hip, the second assistant unscrubbed and then rescrubbed along with a separate nurse for the opposite hip. As soon as the dressing had been applied and the patient repositioned, the second assistant prepared the skin of the second hip. The surgeon and the first assistant rescrubbed and regowned. The time from the last skin staple on the first hip to the skin incision of the second hip was 15 to 20 minutes. One suction drain was placed in each hip for post-operative drainage. The operating time, intra-operative blood loss, and any intra-operative complications were recorded. In the one- VOL. 88-B, No. 3, MARCH 2006

3 300 S. BHAN, A. PANKAJ, R. MALHOTRA stage group the operating time was calculated as the time between the induction of anaesthesia and skin closure on the second hip, and in staged procedures as the sum of the time between the induction of anaesthesia and skin closure in each operation. Intra-operative blood loss was estimated by measuring the volume of blood in the suction bottles, weighing the swabs used and deducting the volume of irrigation fluid. Total blood loss was calculated as the sum of the estimated intra-operative blood loss and the blood collected in the suction drains in the first 24 hours. For the patients in the two-stage group, the total blood loss and the length of hospital stay were calculated as the sum of the values for each operation. The length of hospital stay was calculated from the day of operation until discharge. Each patient received 1 g of Ceftriaxone intravenously before operation followed by 1 g twice daily for three days. Subcutaneous enoxaparin (40 mg once daily) was given to all patients starting on the day of surgery and continuing for seven to ten days. Anti-embolism stockings were also worn. Oral aspirin (150 mg once daily) was given for three to four weeks after the discontinuation of enoxaparin. No routine screening for deep vein thrombosis (DVT) was performed, but Doppler ultrasonography was carried out if it was suspected. Early mobilisation was used both to prevent DVT and to hasten functional recovery, commencing with mobilisation in bed from the second post-operative day in all patients. Full-weight-bearing was allowed from the third day onwards in those with cemented arthroplasty. Patients with cementless components were allowed full weight-bearing after 12 weeks. Those with ankylosing spondylitis received oral indomethacin, 25 mg three times daily for three weeks as prophylaxis against heterotopic ossification. Post-operatively, the haematocrit was measured at eight hours and the haemoglobin levels on the first, second and third post-operative days. Any blood transfusion was also recorded. The transfusion requirement was assessed by the resident on-call guided by intra-operative blood loss, clinical and haematological parameters. No single parameter was used. Complications local to each joint including fracture, dislocation, superficial wound infection, deep wound infection around the prosthesis and the incidence of heterotopic ossification were recorded. Systemic complications including cardiac and gastro-intestinal complications, cerebrovascular accidents, phlebitis/pulmonary embolism, and urinary tract infection were noted and systemic complications which occurred in the same patient during both procedures in the two-stage group were added and the sum used for comparison. Radiological evaluation. An independent investigator, blinded to the study group, evaluated the post-operative radiographs obtained one month after operation and at one month after the second procedure in the two-stage group. The parameters recorded were the angle of abduction of the cup and the alignment of the stem, measured as the angle between the stem and the long axis of the femur on anteroposterior radiographs and classified as varus, neutral or valgus. On the lateral radiograph, the stem alignment was classified as anterior, neutral or posterior. Limb-length discrepancy was determined from measurement of the distance between the upper margin of the lesser trochanter and the interteardrop line on the anteroposterior radiograph of the pelvis. Post-operative follow-up. The patients had clinical and radiological evaluation at one, three, six and 12 months and biennially thereafter following a simultaneous bilateral procedure and the second procedure in the two-stage group. Those patients in the two-stage group were followed at monthly intervals until the second procedure. Since the patients who received a cementless THA were allowed to bear weight only after three months from surgery, the HHS was recorded six months after operation in all patients. They then carried out a timed 10-metre walking test to assess walking speed, examining the overall time to travel ten metres and the time to walk the middle six metres. The use of any walking aid was also noted. The incidence of heterotopic ossification at any point during follow up was recorded using the classification of Brooker et al. 10 Any other complications and the details of any revision procedure were also noted. Statistical analysis. Sample-size estimation showed that 80 patients in each group would be required to show a difference in post-operative haemoglobin level of 0.5 g/dl and a difference in length of stay of half a day, with an alpha level of 0.05 and a beta level of 0.20, that is, a power of 80%. Statistical analysis was undertaken using the SPSS v10.0 program (SPSS Inc., Chicago, Illinois) with the independent sample t-test for continuous variables and the chi-squared test for dichotomous variables. The level of significance applied was 95.0% (p < 0.05). Kaplan-Meier survivorship was calculated using revision of either prosthetic component as the end-point. Results There were no significant differences in age, gender, aetiology, pre-operative haemoglobin, ASA status, pre-operative limb-length discrepancy and the HHS between the two groups (Tables I and II). No significant difference was found between the two groups with respect to the operating time, minutes (SD 37.82) in the one-stage group and minutes (SD 37.39) in the two-stage group (p = 0.16). The mean total blood loss was significantly higher in the two-stage group compared with that in the one-stage group ( ml (SD ) and ml (SD )) respectively (p < 0.001). The mean number of blood units transfused was significantly lower in the two-stage group. The postoperative haematocrit at eight hours was significantly lower in the one-stage group, (SD 0.051) as compared with (SD 0.045) in the two-stage group (p = 0.024; Table IV). THE JOURNAL OF BONE AND JOINT SURGERY

4 ONE- OR TWO-STAGE BILATERAL TOTAL HIP ARTHROPLASTY 301 Table IV. Operative and post-operative haematological data for both groups (n = 85) p value Mean (SD) operative time in mins (37.82) (37.39) 0.16 Mean (SD) total blood loss in ml (517.14) (490.78) < Mean haematocrit (SD) at eight hrs (0.051) (0.045) Mean (SD) total blood transfusion in units 2.37 (0.84) 1.78 (1.08) < Table V. Post-operative complications in both groups, by number and percentage Complication * (n = 85) p value Superficial infection 4 (2.4) 3 (1.76) 0.67 Deep infection 2 (1.2) 1 (0.6) 0.54 DVT 3 (3.6) 2 (2.4) 0.63 PTE 1 (1.2) 0 (0.0) 0.31 GI bleed 1 (1.2) 1 (1.2) 0.98 Dislocation 0 (0.0) 1 (0.6) 0.32 Chest infection 0 (0.0) 1 (1.2) 0.32 MI 1 (1.2) 0 (0.0) 0.31 UTI 5 (6.0) 2 (2.4) 0.23 * DVT, deep-vein thrombosis; PTE, pulmonary thromboembolism; GI, gastro-intestinal; MI, myocardial infarction; UTI, urinary tract infection counted according to the number of joints affected counted according to the number of patients affected Table VI. Alignment of the femoral stem on anteroposterior and lateral radiographs in both groups (n = 162) Alignment on anteroposterior radiographs Varus 10 8 Neutral Valgus 1 0 Alignment on lateral radiographs Neutral Posterior Anterior 4 8 (n = 168) Intra-operative complications. One hip in each group had an intra-operative fracture. A trochanteric fracture occurred in one patient in the one-stage group during preparation of the femoral canal for a cementless stem; it was stabilised by wiring. A patient in the two-stage group had a fracture of the medial acetabular wall during press-fit insertion of the acetabular shell. Neither fracture caused any post-operative problems and the patients were mobilised according to the routine protocol. Early post-operative complications. No significant difference was found in the incidence of complications in the early post-operative period between the two groups (Table V). Superficial infection occurred in four hips in the one-stage group and in three in the two-stage patients. All responded to local wound care and antibiotics. Deep infection was detected in two hips in the one-stage and in one hip in the two-stage group. In all three cases treatment consisted of debridement and irrigation with retention of the prosthesis along with intravenous antibiotic therapy for six weeks. The level of C-reactive protein and the erythrocyte sedimentation rate returned to normal by eight weeks postoperatively. Dislocation occurred in one hip in the twostage group on the third post-operative day after the second-side and was managed by closed reduction under anaesthesia. The patient was mobilised with an abduction brace for three months without recurrence. The overall incidence of systemic complications was similar in the two groups. No deaths occurred. Although in our study group there was no significant difference in the incidence of major complications between high-risk (ASA 3 and 4) and low-risk (ASA 1 and 2) subgroups in both groups, no valid conclusions could be drawn since the number of patients in the high-risk subgroup was small compared with that in the low-risk subgroup. Radiological findings. Complete sets of radiographs were available for 165 patients. Three patients who had only anteroposterior films were excluded from the analysis. The mean angle of abduction of the cup was (SD 2.98 ) in the one-stage group and (SD 2.67 ) in the twostage group (p = 0.35). The alignment of the femoral stem on both anteroposterior and lateral radiographs was similar in both groups (Table VI). The mean stem angle on the anteroposterior radiograph was 0.75 (SD 1.10 ) of varus (2 valgus to 11 varus) for the one-stage group and 0.92 (SD 1.2 ) of varus (neutral to 10 varus) for the two-stage group (p = 0.35). The difference between the two groups with respect to the alignment of the femoral stem on the lateral radiographs was not found to be significant (chisquared test, p = 0.20). The mean pre-operative limb-length discrepancy was 1.24 cm (SD 1.01) (0 to 3.5) in the onestage group and 1.20 cm (SD 1.1) (0 to 4.3) in the two-stage group. This improved to 0.45 cm (SD 0.44) (0 to 1.3) and 0.53 cm (SD 0.46) (0 to 1.5), respectively, after operation (p = 0.239). The incidence of heterotopic ossification was 14.4% in the one-stage and 12.9% in the two-stage group (p = 0.452). None of the patients had heterotopic ossification greater than Brooker grade II. Harris hip score and walking speed. The mean pre-operative HHS was 42 (15 to 60) in the one-stage and 40.5 (20 to 55) in the two-stage group. This improved to 82 (75 to 95) and 83.5 (72 to 98), respectively (p = 0.642). In the timed walking test, the split time for the middle six metres was 35.2 seconds (SD 19.0) in the one-stage and 34.8 seconds (SD 18.3) in the two-stage group (p = 0.78). The ten-metre walking time was 59.3 seconds (SD 30.2) in the one-stage VOL. 88-B, No. 3, MARCH 2006

5 302 S. BHAN, A. PANKAJ, R. MALHOTRA and 58.8 seconds (SD 28.5) in the two-stage group (p = 0.86). Survival of the prosthesis. The mean follow-up was months (SD 15.05; 4 to 8 years) and months (SD 14.54; 4 to 8 years) in the one-stage and two-stage groups, respectively. The Kaplan-Meir survivorship, with revision of either component for any reason as the end-point, was 91.7% (95% confidence interval (CI) 88.1 to 100) in the one-stage group and 93.8% (95% CI 89.2 to 100) at 96 months; the difference was not significant (Wilcoxon test, p = 0.673). Six patients (six hips), three hips in each group, required revision in the short-to intermediate-term. Aseptic loosening of either component necessitated revision in four hips, one was revised because of late infection, and recurrent anterior dislocation in the other was treated by a change of the position of the liner. Hospital stay. The mean duration of hospital stay was significantly shorter in the one-stage group, 7.25 days (SD 1.30; 5 to 20) compared with 10 days (SD 1.65; 8 to 24) in the two-stage patients (p = 0.023). Discussion Our study has shown no detrimental effects when undertaking a single-stage bilateral THA compared with a staged procedure with an interval of at least three months between the sides. Most of our cases were ASA grade 1 or grade 2 and this may have had a favourable effect on the lack of systemic complications. Alfaro-Adrian et al 8 reported a higher incidence of peri- and post-operative complications in patients with significant co-morbidities (ASA grade 3 and grade 4) whether the operation was staged or conducted as a single procedure. Although the total estimated blood loss was significantly lower in patients undergoing a one-stage procedure, the transfusion requirements were significantly higher than in the two-stage group. The need for a transfusion was guided by the post-operative haematocrit which was significantly lower in these patients. Since the total blood loss in the twostage group was the sum of two values during two procedures, carried out at least three months apart, it is easy to understand why the transfusion requirements were lower in these patients. Similar observations were made by Salvati et al 3 and Alfaro-Adrian et al. 8 We did not find any significant difference in the operating time between the two groups. Ritter and Vaughan 11 found a significant increase in the rate of formation of heterotopic bone in patients who underwent a single-stage procedure. They suggested that this may have occurred because there were more patients with gross bilateral osteoarthritis in the one-stage group. Although we had a comparable number of patients with similar aetiologies we did not encounter any difference in the rate of post-operative heterotopic ossification. The incidence of fatal pulmonary embolism has been reported as being between 1.04% and 1.5%, and that of non-fatal pulmonary embolism between 7.89% and 11.5% in patients undergoing elective unilateral THA. 5,12,13 In our study group the incidence of non-fatal thromboembolism was 0.6% with no significant difference between the two groups. There were no fatalities. However, the number of patients was probably too small to detect any difference between groups. The incidence of clinically evident DVT was similar in both. Many earlier studies comparing the incidence of DVT in unilateral THA and single-stage bilateral THA have reported a higher incidence in the latter group. 2,4,12 However, recent studies have compared the incidence of DVT and pulmonary thromboembolism in single-stage and two-stage bilateral THA and have found no difference between the two approaches. 8,14 Ritter and Stringer 4 attributed this to improvement in anticoagulation therapy and to early post-operative mobilisation. Concern has been expressed that there was a suboptimal gain in the range of movement and improvement in gait in patients undergoing bilateral THA. 4,6 Both of these studies, however, had compared the results of bilateral with those of unilateral replacement. The functional outcome in a patient with a unilateral THA who has a normal contralateral hip cannot be compared with a patient with bilateral THA. We did not find any significant difference in the functional outcome of bilateral THA, as evaluated by the HHS and walking capacity at six months, whether done as a single-stage or in two stages. Wykman and Olsson 6 found that in patients with bilateral hip disease, optimal function is not entirely regained until both hips have been replaced. Radiological evaluation at one month after operation showed similar results with regard to the positioning of the acetabular and femoral components and restoration of limb-length in the two groups. These observations suggest that positioning of the component is not compromised during the simultaneous procedure. The survival was also similar in the two groups. The length of hospital stay in the single-stage group was significantly shorter than that in the two-stage patients. Although we did not assess the costs involved, this would certainly have an impact in reducing the economic burden. Similar results have been reported by other authors. 3,7,8 A single-stage bilateral THA is as safe as a two-stage procedure in patients in ASA grade 1 and grade 2 and probably also in grade 3 and 4. Although the need for transfusion is greater in patients undergoing a one-stage procedure there is no increase in the incidence of complications. The length of the hospital stay was reduced in the single-stage procedure and therefore costs are likely to be less. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1. Lazansky MG. A method for grading hips. J Bone Joint Surg [Br] 1967;49-B: Ritter MA, Randolph JC. Bilateral total hip arthroplasty: a simultaneous procedure. Acta Orthop Scand 1976;47: Salvati EA, Hughes P, Lachiewicz P. Bilateral total hip-replacement arthroplasty in one stage. J Bone Joint Surg [Am] 1978;60-A: Ritter MA, Stringer EA. Bilateral total hip arthroplasty: a single procedure. Clin Orthop 1980;149: THE JOURNAL OF BONE AND JOINT SURGERY

6 ONE- OR TWO-STAGE BILATERAL TOTAL HIP ARTHROPLASTY Camissa FP Jr, O Brien SJ, Salvati EA, et al. bilateral total hip arthroplasty: a prospective study of perioperative morbidity. Orthop Clin North Am 1988;19: Wykman A, Olsson E. Walking ability after total hip replacement: a comparison of gait analysis in unilateral and bilateral cases. J Bone Joint Surg [Br] 1992;74-B: Shih CH, Ho WB. versus two-stage bilateral autophor ceramic total hip arthroplasty. Clin Orthop 1985;193: Alfaro-Adrian J, Bayona F, Rech JA, Murray DW. One- or two-stage bilateral total hip replacement. J Arthroplasty 1999;14: American Society of Anaesthesiologists. New classification of physical status. Anaesthesiology 1963;24: Brooker AF, Bowerman JW, Robinson RA, Riley LH Jr. Ectopic ossification following total hip replacement: incidence and a method of classification. J Bone Joint Surg [Am] 1973;55-A: Ritter MA, Vaughan RB. Ectopic ossification after total hip arthroplasty: predisposing factors, frequency, and effect on results. J Bone Joint Surg [Am] 1977;59-A: Bracy D, Wroblewski BM. Bilateral Charnley arthroplasty as a single procedure: a report on 400 cases. J Bone Joint Surg [Br] 1981;63-B: Johnson R, Charnley J. Hydroxychloroquine in prophylaxis of pulmonary embolism following hip arthroplasty. Clin Orthop 1979;144: Eggli S, Huckell CB, Ganz R. Bilateral total hip arthroplasty: one stage versus two stage procedure. Clin Orthop 1996;328: VOL. 88-B, No. 3, MARCH 2006