Results of Austin Moore replacement.AP Jadhav, SS Kulkarni, SV Vaidya, MM Divekar, SP Suralkar
Department of Orthopaedics, KEM Hospital & Seth GSM College, Parel, Mumbai.
Forty cases of Austin Moore Replacement done for transcervical fractures of the femur in patients were reviewed after a period of 12 to 48 months postoperatively (mean 26 mth). 30 cases (75%) had mild to severe pain of non-infective origin, starting as early as 6 months postoperatively. This was irrespective of the make, size or position (varus/valgus) of the prosthesis. Though the Aufranc and Sweet clinical scoring was satisfactory in 65% cases, radiological evidence of complications like sinking, protrusion, etc. were seen in majority of the cases. Calcar resorption was seen in 34 cases (85%) as early as 4 months postoperatively. Results of THR and bipolar replacement done for transcervical fractures in recent literature show 85% pain-free cases at 5 years. We feel that Austin Moore Replacement should be reserved for patients more than 65 years of age and those who are less active or debilitated because of other factors, because of increased acetabular wear with time in the younger individual. This is corroborated by unsatisfactory results in patients less than 65 years of age (p < 0.05).
Keywords: Activities of Daily Living, Aged, Aged, 80 and over, Arthroplasty, Replacement, Hip, adverse effects,Female, Femoral Neck Fractures, radiography,surgery,Follow-Up Studies, Human, Male, Middle Age, Pain, etiology,Prosthesis Design, Prosthesis Failure, Range of Motion, Articular, Treatment Outcome,
Credentials of hemiarthroplasty as a treatment modality for displaced transcervical fractures in the elderly have been under cloud since the advent of newer types and designs of bipolar and total hip arthroplasties. It has been a common fact that the more "active" patients often complain of pain early in their postoperative period. Hemiarthroplasty has been advocated as the best treatment for transcervical fractures in the elderly, for early rehabilitation, by permitting early mobilisation and preventing the dreaded complications of non-union and avascular necrosis. Currently two types of endoprostheses are in common use - the Thompson type and the Moore type. The Moore type is more popular as it distributes stresses over a wide area in the proximal femur, minimising shear stress. We have studied the results of the Moore type of hemiarthroplasty in fresh and old transcervical fractures of the femur operated at our institute over the past 5 years. We have evaluated our results with respect to pain, range of motion, and status of ambulation, all of which contribute to success or failure of hemiarthroplasty. We have also reviewed recent literature comparing results of cemented and uncemented hemiarthroplasties,,. bipolar hemiarthroplasties,, and total hip arthroplasty. We have also studied the results with respect to age, and correlation of osteolysis and other radiological changes with pain.
Over 150 cases of Austin Moore hemiarthroplasty have been performed at our institute during the past 5 years, for displaced transcervical fractures in patients with a physiological age > 60 years.
Using a Moore's posterior approach in the lateral position, the gluteus maximus is split, and the short external rotators are cut close to their insertion. A posterior capsulotomy exposes the fracture site. The head is extracted using a corkscrew, and the head size is measured using a ring calliper and confirmed by placing it in the acetabulum. The acetabulum is packed during the reaming of the medullary canal. A notch is made in the posterosuperior portion of the neck to help maintain anteversion of 10?- 15?. The prosthesis is inserted into the canal by gentle hammering to seat it well over the calcar. The prosthesis is reduced by gentle traction in the extended position of the knee, without rotational forces. Difficult reductions are achieved using the Murphy's skid. Stability is confirmed by putting the hip through a range of motion and traction. The external rotators are reattached to the greater trochanter and the wound is closed over a negative suction drain.
Postoperatively, the limb is maintained in the abducted position. Partial weight bearing ambulation is started as early as the 2nd postoperative day depending on the general condition of the patient. Patients are advised to use a cane in the opposite hand.
Cases with any form of infection were excluded from this study. 40 cases (40 hips) welt followed up personally by the authors after a minimum period of 12 months to a maximum period of 44 months. Besides routine demographic data, clinical examination was carried out and the range of motion was assessed as compared to the opposite (normal) hip. Any pain or limp was noted and the necessity of using any walking aid was inquired. Any discrepancy in limb length was measured and the overall functional status of the patient was assessed. The criteria for grading the results were according to the grading adopted by the Committee on fracture and traumatology of the American Academy of Orthopaedic Surgeons 10 and those of Autranc & Sweet. [Table - 1]. The patients were divided into three age groups - < 65 years, 65-70 years, and > 70 years for the purpose of analysing the results.
Immediate post-operative X-rays and those taken at the final follow-up were compared to assess the position of the implant (version), osteolysis, subsidence and protrusio acetabulae. Given below, are the details of the study patients.
Total number of cases - 40
Sex distribution - M : F - 10 : 30
Age wise distribution at operation
Minimum age - 50 years
Maximum age - 82 years
Average age - 65.7 years
< 65 years - 15 cases
65-70 years - 13 cases
> 70 years - 12 cases
Average period between injury & surgery - 4.7 days
(Range 1 day to 6 weeks)
Head size (for AMR) - varied from 39 to 49
(most commonly 41 & 43)
Post-operative mobilisation started on -
Earliest - Day 2
Latest - Day 16
Average - 4.7 days
The patients were followed up for an average of 26 months (Range: 18 - 48 months)
Shortening was noted to a maximum of 7 cms. In 10 cases, no shortening was seen 24 patients had 1-2 cms shortening whereas the remaining 6 cases had 3-7 cms shortening
Pain and osteolysis noted in these cases can be seen in [Table - 2].
Pain was noted as early as 6 months postoperatively, even with out any radiological abnormality.
Limp was present in as many as 35 cases. (87.5%)
Calcar resorption (Subsidence) noted in these cases are as shown in [Table - 3].
Calcar resorption was seen as early as four months postoperatively
Protrusio acetabulae was noted in 4 cases (10%)
The position of the prosthesis as doted in the study was valgus - 34 cases; neutral - 3 cases and varus - 3 cases. There was no correlation of pain with position of the prosthesis [Figure - 1]
Ectopic ossification was seen in 10 cases (25%) [Figure - 2]
The final results are as shown in [Table - 4].
Results for age statistically significant by the chi squared test (p < 0.05).
Final results were unsatisfactory (Fair, Poor & Failure classes) in 15 cases (35%). The percentage of unsatisfactory results quoted in literature is variable ranging from 13%, 15%, 29%, 34% to 48% in Western series and 9% to 36% in Indian series. To improve the percentage of satisfactory results the changes proposed by many authors,,, include cementing the stem for secure fixation and reduction of the friction between the prosthetic head and the acetabulum by total hip arthroplasty or bipolar prosthesis. Comparable series of bipolar and unipolar hemiarthroplasty have either favoured the bipolar, or have shown that bipolar prostheses have, no definite disadvantage with probable benefits in younger more active patients. Gebhard et al have compared THR with cemented and uncemented unipolar prostheses and have shown highest functional scores with THR and about 85% patients pain-free at the end of on an average 56 months follow-up. They recommend hemiarthroplasty only for the older patient, who may be occasionally active outside the home. Cement fixation in bipolar hemiarthroplasty and total hip arthroplasty reduces the incidence of prosthetic stem loosening and hence pain. Our results in the younger patients (<65 years) have been comparatively poorer i. e. only 40% satisfactory results, which are statistically significant by the chi squared test (p < 0.05).
The average age of our patients at surgery was 65.7 years. Some authors have advocated hemireplacement in patients over 70 years of age,,,,. This age limit is arbitrary and can be justifiably reduced when the younger patient is in poor health or has a low activity level. The average age of patients in our series is lower as compared to those reported in Western literature viz. - 79 yrs, 75 yrs, 79 yrs, 77 yrs, 81 yrs, but comparable to other Indian series viz. 66 yrs, and 66.3 yrs. This can be explained on the basis of a lower life expectancy amongst the Indian population as compared with the West. The average age in Lestrange's series (bipolar) was 79.67 years and in the Nottage - McMaster series it was 65 years for bipolar, 72 years for Thompson hemiarthroplasty and 73 years for Moore hemiarthroplasty. In Gebhard's series, the average age for the THR group was 75.2 years and that of the hemiarthroplasty group was 76.2 years. Thus in the West, patients of a -comparable age group are relatively more active and hence probably treated with bipolar prosthesis or total hip arthroplasty. Women outnumber men significantly in our series in keeping with the fact that femoral fractures are more common in females due to osteoporosis.
Shortening was seen in 75% cases ranging from 1 to 7 cm. A limp was seen in 35 cases (87.5%) either due to pain, shortening or abductor, muscle weakness. In Andersson's series  only 6% cases walked without a limp.
Post-operatively all except 10 cases had pain varying from mild to extremely severe. The pain was noted as early as 6 months postoperatively, even in the absence of any radiological abnormality. Jensen and Holstein's series had 18% cases without pain and 21% cases with moderate to severe pain. In our series the two patients who had severe pain had osteolysis [Figure - 3] and protrusio acetabulae [Figure - 4]. It has been shown that the two main causes of pain are erosion of the acetabulum and loosening of the prosthesis. Pain significantly affects the end results and both patients with severe pain had a poor result. Osteolysis around the femoral stem was noted in 32% cases by Anderson & Nielsen, and Whittaker et a1, and 22% cases by Jensen and Holstein. The interpretation of osteolysis as a sign of loosening of the prosthesis is controversial. Jensen and Holstein could not correlate it to pain unlike Whittaker et al. We found a significant correlation between osteolysis and pain as has been shown in our results [Table - 2]. Acetabular erosion was seen in 4 cases (10%) which is in agreement with the reports of Meyer (10%), D'Arcy & Devas (11%), but lower than that reported by Whittaker (24%). Sarmiento stated that clinical difficulties were due to acetabular erosion and stem loosening. In his post-mortems of 24 cases there was "noticeable or excessive motion of stem in the canal and failure of cancellous bone to fill the entire fenestrations in the stem". Efthekar states that, "pressure brought by the femoral prosthesis upon the acetabular cartilage makes subsequent migration of the prosthesis inevitable."
Though calcar resorption was seen in 85% of the cases it does not affect the end result [Figure - 5]. There was no clinicoradiological correlation of pain, and most patients had satisfactory function despite severe changes in the X- Rays. [Figure - 6], [Figure - 7] On the other hand, early pain was noticed without radiographic changes in the younger patients who were more active.
It was noted:
1. Higher incidence of unsatisfactory results (60%), are seen in patients less than 65 years of age who are more physically active.
2. Older patients (> 70 years) have a high incidence of satisfactory results (83.3%), and hence the procedure is advocated in such individuals with a low activity level.
3. There is a high incidence of early (as early as 6 months) postoperative pain, which correlates well with osteolysis.
4. Calcar resorption with sinking of the prosthesis was seen in most of the cases (85%), but this does not always affect the clinical (functional) picture
5. Though poorer results are seen in the younger age group (< 65 years) with hemiarthroplasty as compared to bipolar and total hip arthroplasty, the technical ease to perform the procedure by even the averagely trained surgeon should not make it obsolete.
[Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4], [Figure - 5], [Figure - 6], [Figure - 7][Table - 1], [Table - 2], [Table - 3], [Table - 4]