M.S. Partial Resurfacing Device in the Treatment
of Osteonecrosis of the Femoral Head
Original Technique and Report of 35 Cases
M. Siguier*, T. Siguier**, B. Brumpt*, T. Judet**
*Clinique Jouvenet, 6 Square Jouvenet, F-75016 Paris - **Hôpital Tenon, 4 rue de la Chine, F-75020 Paris
The M.S. intraarticular resurfacing device developed by the authors for the treatment of osteonecrosis of the femoral head (ONFH) provides a means of preserving all the healthy articular cartilage in this young patient population. The device replaces only the involved cartilage and the infarcted bone segment, to restore the sphericity of the femoral head (Fig. 1). It is inserted through a small anterior incision that does not involve any muscle division. The procedure is precise but short; the anatomical structures are preserved; and there is no inherent need for blood transfusion.
The postoperative management is straightforward; patients are allowed immediate full weight-bearing.
In the event of failure, revision to THR is as easy as in a hip that has not previously been implanted.
Between April 1991 and February 1996, we treated 35 cases of ONFH in 29 patients; the mean patient age was 43 years. In this series, even very advanced forms of ONFH were included, because of the patients age and the lack of any suitable alternative treatment other than THR in advanced ON. This allowed us to establish the limitations of this technique, and to refine the surgical technique and patient selection criteria.
The M.S. femoral head resurfacing device (Tornier, Saint Ismier, France) is made of a cobalt chromium alloy. It has the shape of a 120° segment of a sphere (Figs. 2 and 3). Fixation in the femoral head is by means of a wide central keel and four peripheral spikes, as well as a small amount of cement. The implants are available in incremental diameters, from 40 mm to 60 mm; this size range permits the restoration of the spherical contour of the femoral head with an appropriately sized device that replaces one third of the femoral head surface in the zone affected by the osteonecrotic collapse.
The procedure is designed to replace the necrotic and detached bone and cartilage segment with a well-centred implant to restore the sphericity of the femoral head. The implant must fit into the normal anatomical contour of the femoral head, without overhanging or standing proud. To this end, the implant comes with precision instruments, which are simple and user-friendly, and which allow the surgeon, throughout the procedure, to appreciate the quality of the fit and, if need be, to take corrective action as he or she sees fit. The instrumentation has been designed to make surgery easier, whilst leaving the surgeon plenty of leeway.
The device is inserted through an anterior incision, which is the preferred approach for a number of reasons:
- It causes minimal trauma, since no muscles need to be divided.
- It gives good exposure of the anterosuperior aspect of the femoral head, where the necrotic segment is usually situated.
- It preserves the medial circumflex vessels.
Details of surgical technique
- The patient is positioned supine on a fracture table (Fig. 4)
- An anterior Hueter incision is made; no muscles are divided (Fig. 5).
- The joint is opened anteriorly, by capsulectomy or an H- or T-shaped capsulotomy well away from the labrum, so as not to damage that structure (Fig. 6).
- The head is dislocated anteriorly to prevent damage to the medial circumflex vessels and so as to obtain optimum exposure of the anterosuperior aspect of the femoral head, where the necrotic segment is usually situated (Figs. 7 and 7a).
- The head diameter is measured, and the instruments and implant chosen to match that diameter or one diameter below; the detached cartilage is removed with a scalpel (Figs. 8 and 8a).
- A pin is inserted exactly in the centre of the head, using a self-centring claw-foot targeting device that grips the healthy cartilage (Figs. 9 and 9a).
- Around the pin, an accurately centred and correctly sized implant bed is created; any mobile necrotic bone is removed (Figs. 10 and 10a).
- The No. 1 Trial (Figs. 11 and 11a) is used to check that the device will not stand proud, which might cause impingement. Ideally, the implant should be recessed some tenths of a millimetre. On no account should the definitive implant, or any cement, protrude above the native femoral cartilage.
In some of our cases, a zone of a few millimetres denuded of cartilage was left at the implant-cartilage border, after the removal of the detached cartilage. In these cases, it was decided to have this small zone anteriorly, in the non-weight-bearing region.
- The No. 1 Trial is used to prepare the holes for the keel and the four spikes; then, the stemmed No. 2 Trial is inserted, for a final check to ensure that, all around its circumference, the implant will be very slightly below the surface of the femoral head cartilage (Figs. 12 and 12a).
- The definitive implant is inserted using an impactor and a small amount of cement (Figs. 13 and 13a).
It should be noted that the implant must be very slightly below the level of the surrounding healthy cartilage. The device is correctly positioned if it is surrounded by a very low wall of healthy cartilage, which must be of even height all around the implant. This way, there will be no risk of implant malposition that could give rise to local pressure peaks between the implant and the acetabular cartilage.
- The cartilage wall is trimmed with a small scalpel, so as to achieve the perfectly flush seating of the device (Fig. 14).
The postoperative management was straightforward:
- Ambulation with full weight-bearing on Day 3;
- Gentle physiotherapy to recover the ROM;
- Mean inpatient stay: 8 days.
Thirty-five osteonecrotic femoral heads were operated on between April 1991 and February 1996.
Three patients were lost to follow-up, which left 32 implants in 26 patients available for review.
In nine of the patients, there was a risk factor for ON: six had previously been treated with steroids; two were chronic alcohol abusers; and one had a history of trauma.
The osteonecrosis was advanced in all cases; 31/32 femoral heads had lost their sphericity. Twelve patients had secondary OA (Arlet & Ficat Stage IV, with slight loss of joint space in eight cases); 19 femoral heads had collapsed without any arthritic changes (Arlet & Ficat Stage III); in only one case was the femoral head not deformed (Arlet & Ficat Stage II).
The depth of the necrosis was assessed in the coronal and the sagittal planes, using as a reference the inferior tip of the necrotic triangle on a grid of four bars of even width placed over the femoral head (Fig. 15). In ten cases, the tip was beyond the third quarter; in 20 cases, it was between the second and the third quarter; while in two cases, it did not reach the second quarter. There was no case of necrosis involving less than a quarter of the femoral head. In angular terms, the mean angle of the osteonecrotic lesion was 129° on the a.p. view, and 118° on the false profile view.
There were no intraoperative complications; blood loss was minimal; and none of the patients required transfusion.
In all cases, the immediate postoperative course was uneventful. There were no surgical or general complications; and the patients were able to ambulate with full weight-bearing on Day 3. One patient with slow ROM recovery (hip flexion 70° at 4.5 months) was successfully managed with mobilization under general anaesthesia.
The longest follow-up to date has been nearly 7 years; mean follow-up has been 3 years and 9 months, for the surviving implants. In nine cases, the partial resurfacing failed, and the patients had to be revised to THR. In seven of these cases, the necrosis had been extensive; in six cases the sphericity of the femoral head that had been restored using the device was secondarily lost, either by subsidence of the implant or as a result of global femoral head collapse. These were cases of deep ON, in which the femoral head went on collapsing after implantation of the partial resurfacing device. In one failure, there was erosion of the acetabular cartilage, without implant subsidence or femoral head collapse. The ninth case was due to technical error: a small amount of the cement used for fixation was protruding at the border between the implant and the healthy femoral head cartilage, and had damaged the acetabular cartilage. In the majority of cases, theses failures were clinically symptomatic at an early stage (from 3 months after surgery); revision was performed between 9 and 40 months after implantation.
There were 20 Good or Excellent results using the Postel-Merle dAubigné scoring system. These good results were obtained early, with patients who had been treated for unilateral ONFH going back to work before the third postoperative month, and with none of these patients having to change jobs. In these 20 hips, there have been no changes in the cement-bone interface. In two cases, the acetabular cartilage was found to be reduced in height; however, clinically these two patients are doing well at 3 and 5 years, respectively, after surgery.
The extent in depth of the necrosis was found to correlate best with the quality of the results: there were five failures in the group of ten hips where the necrosis involved more than three quarters of the depth of the femoral head. Conversely, among the 22 hips where the necrosis did not exceed three quarters of the depth of the femoral head, there were 16 Excellent and Good results.
The angle of the necrotic lesion was sometimes slightly greater than the covering zone of the implant; however, this did not appear to spell failure, providing that the lesion did not extend beyond the three-quarterdepth mark.
In patients who had OA prior to surgery, the implantation of the device did not lead to a flare-up of the condition.
In the group of twelve Ficat Stage IV hips, there were two failures: one was attributable to technical error, as mentioned above; the other occurred in a case of massive ON involving the entire femoral head. Of the ten implants surviving in this group, eight have produced Good or Excellent clinical results. In six cases, no further arthritic changes have been seen on the radiographs; in two cases, there has been slight worsening, without any increase in the loss of joint space; in only one case has there been a further loss of joint space.
The study described in this paper involved a small patient population, and follow-up to date has been short. However, some tentative conclusions may be drawn, and trends identified:
In the majority of the patients in this series, the osteonecrotic lesions were so extensive as to make THR the only alternative to the partial resurfacing performed.
The depth of the osteonecrotic lesion appears to be the most important predictor of outcome: where the depth was greater than three quarters of the femoral head, we had a 50% failure rate; where the depth was less, 75% of the cases of ONFH with collapse of the femoral head were treated with good results.
We have been using this technique chiefly as an alternative to THR in young subjects. We think that it may be offered in less advanced cases, as femoral head preserving treatment, where it would rank with grafts, osteotomies, cement injection, etc. Since at that stage the lesions would be less extensive, there would be a better prospect of a satisfactory outcome. At any rate, the postoperative management should be less cumbersome, with earlier return to work, and easier revision if the partial resurfacing fails.
Figure 16a Minor collapse, but major global ON exceeding three quarters of the femoral head in depth Figure 16b Immediate postoperative view Figure 16c At 7 months: massive collapse of the femoral head, and subsidence of the implant (revised to THR)
Figure 17a Major femoral head collapse, but depth of ON not exceeding three quarters of the femoral head Figure 17b At 6 months: femoral head sphericity restored
Figure 17 c and d A.p. and lateral views at 6 years: excellent clinical and radiological result