Microfracture of chondral lesions of the glenohumeral joint

Objective: To determine if microfracture is successful in treating chondral lesions of the shoulder. Design: Case series. Setting: Tertiary referral practice. Patients: From June 2005 to November 2006, eight patients underwent shoulder arthroscopy with arthroscopic microfracture to treat full-thickness chondral lesions of less than 4 cm 2 size. The study group consisted of six men and two women. The mean age at surgery was 37 years (range: 27–55 years). One patient (12.5%) had an isolated chondral defect and seven patients (87.5%) had associated conditions treated simultaneously: two patients had arthroscopic subacromial decompressions, two had capsular plications for multidirectional instability, and three had anterior stabilization done (one with an associated superior labrum anterior to posterior repair and one with repair of a small rotator cuff tear). Five patients had humeral head defects and three had glenoid defects. Intervention : Microfracture. Main outcome measures: Constant score and Oxford score. Results: The mean follow-up period was 15.4 months, with a range of 12–27 months. The mean preoperative Constant score was 43.88 (range: 28–70) and at ﬁ nal follow-up the mean Constant score was 90.25 (range: 85–100); this difference was signiﬁ cant ( P <0.005). The mean preoperative Oxford score was 25.75 (range: 12–37) and the mean postoperative Oxford score at ﬁ nal follow-up was 17 (range: 11–27); the difference was signiﬁ cant ( P <0.005). There were no complications. Two patients underwent reoperation which allowed assessment of the lesion; in both cases the lesions showed good ﬁ lling with ﬁ brocartilage. Conclusion : Microfracture has been shown to be a reliable method of treatment for chondral lesions within the knee. We believe that this technique may also be applied to the shoulder; however, further study is required to assess its efﬁ cacy in this joint.


INTRODUCTION
Chondral lesions of the glenohumeral joint pose a signifi cant clinical problem. The natural history of such lesions is unclear; there is no established treatment mentioned in the literature, and nonoperative treatment may not provide sufficient relief from pain. Often, Outerbridge grade IV osteochondral changes of either the glenoid or humeral articular surface may be impossible to detect with contemporary imaging techniques or clinical examination. [1][2][3] In these cases, the pain originating from the unrecognized osteochondral defect may be attributed to associated pathology, specifically impingement syndrome, [2] labral tears, or other concomitant intra-or extra-articular abnormalities. [4,5] The advent of shoulder arthroscopy has enhanced the surgeon' s ability to diagnose and treat abnormalities of the glenohumeral joint. [1,3,[6][7][8] Few studies have evaluated arthroscopy in the management of grade IV chondral changes of this joint. [4] Currently, the indications, benefi ts, and usefulness of arthroscopy in the management of grade IV osteochondral lesions are relatively unclear.
Treatment of full-thickness chondral lesions with microfracture has become established practice for the knee joint. The intervention leads to a spontaneous repair response, which is based upon therapeutically induced bleeding from the opened subchondral bone spaces and subsequent blood clot formation. [9] This clot contains pluripotent, marrow-derived mesenchymal stem cells, which produce a fi brocartilage repair with varying amounts of type II collagen content. [10] Recent studies on the knee have shown that the functional results at 10 years are equivalent to that of autologous chondrocyte implantation. [11] The aim of our study was to determine if microfracture was effective in the treatment of type IV chondral defects within the shoulder joint.

PATIENTS AND METHODS
From June 2005 to November 2006, eight patients underwent shoulder arthroscopy with arthroscopic microfracture to treat full-thickness chondral lesions of less than 4 cm 2 size. All procedures were carried out by the senior author. All patients included in the study had a postoperative diagnosis of a chondral lesion and had undergone treatment with microfracture. Only grade IV lesions (full-thickness cartilage loss with exposed subchondral bone) of less than 4 cm 2 size were considered appropriate for treatment with microfracture. Patient data was recorded prospectively.
Chondral defects were classifi ed according to their size and site. Microfracture of the cartilage lesions was performed according to the method described by Steadman et al. [12] This included debridement of the cartilage lesion to stable cartilage margins, careful removal of the calcifi ed cartilage layer, and micropenetration of the subchondral bone using commercially available instrumentation (Linvatec, Largo, Florida). The 4-mm-wide subchondral bone bridges were carefully maintained between each microfracture hole so as to ensure the preservation of subchondral bone-plate integrity and function. Release of blood and marrow fat droplets from the microfracture holes was confi rmed by eliminating arthroscopic pump pressure.
Postoperatively, the rehabilitation program was dependent on associated pathology and procedures carried out. In cases where only an isolated chondral defect was present, patients were advised to undertake immediate active movements, with avoidance of movement against resistance. Strengthening exercises began at 12 weeks.
All complications or reoperations were documented. Outcome was measured using the Constant and Oxford shoulder scores by a physiotherapist with an interest in shoulder rehabilitation. Scores were recorded preoperatively and at 3-monthly intervals. Statistical analysis was carried out using the paired Student' s t test.

RESULTS
The study group consisted of six men and two women. The age at surgery ranged from 27 to 55 years, with an average age of 37 years. The dominant extremity was involved in 56% of the cases. Four patients were professional rugby players.
One patient (12.5%) had an isolated chondral defect and seven patients (87.5%) had associated conditions treated simultaneously: two patients had arthroscopic subacromial decompression, two had capsular plications for multidirectional instability, and three had anterior stabilization done (one with an associated superior labrum anterior to posterior (SLAP) repair and one with repair of a small rotator cuff tear). Ellman [14] described grade II/III chondromalacia of the humeral head and glenoid. Arthroscopic debridement led to early pain relief in short-term follow-up. Ogilvie-Harris and Wiley [6] reported on their 10-year experience with shoulder arthroscopy; a subgroup of their patients with degenerative changes of the shoulder joint were treated with arthroscopic debridement with good results. Both Cofi eld [15] and Johnson [16] also reported on small groups of patients treated with arthroscopic lavage and debridement with subsequent relief of symptoms, but their results were not quantifi ed. Weinstein [4] found that there was an improvement in pain following debridement. Postoperative function and pain levels have been shown to improve in 60-80% of patients with mild degenerative changes (Outerbridge grade I-III). [7,9] Favorable outcomes have been reported to be much lower (around 30%) when advanced degenerative changes (Outerbridge grade IV) were treated with arthroscopic debridement alone. Mean pain relief lasted 28 months. The addition of a capsular release may be of benefi t in lesions less than 2 cm 2 in size. [17] In full-thickness cartilage lesions, repair of the lesion cannot be expected because of the limited restorative capacity of cartilage. [18][19][20] It has been shown within the knee that this often leads to worsening symptoms and joint deterioration. [21] Therefore, the focus of recent studies has been to fi nd a treatment method to encourage the repair of cartilage or to produce a substitute tissue that can function as cartilage.
Urist [22] reported in 1958 that intra-articular lesions could heal by the formation of hyaline-like cartilage and it was hypothesized that the process of microfracture allows pluripotent mesenchymal cells and growth factors from the marrow space to gain access to the defect. [23,24] The scapula and humerus have an excellent vascular supply and so there appears to be no reason why this technique should not be applied to the shoulder.
The technique of microfracture is ideally suited to the shoulder. It is minimally invasive because it is arthroscopic through standard portals. In contrast to abrasion chondroplasty, the subchondral bone plate is not completely destructed but is Five patients had humeral head defects [ Figure 1a] and three had glenoid defects [ Table 1 and Figure 1b].
There were no complications. Two patients underwent reoperation which allowed assessment of the lesion; one of these patients had a revision of a superior labrum anterior to posterior (SLAP) repair following another signifi cant tackle injury, and the second patient needed a subsequent subacromial decompression. On assessment of the chondral lesions, both glenoid and humeral lesions had healed well. There was good fi ll and continuity of all lesions [ Figures 2a and 2b].

DISCUSSION
It has been reported that coexisting Outerbridge grade II-IV chondral lesions are detected as incidental fi ndings in 5% of arthroscopic shoulder procedures. [13] Various attempts have been made to treat these lesions to reduce pain and also on the assumption that such treatment will prevent continued deterioration. Preliminary studies have shown that arthroscopy may play a role in the treatment of glenohumeral osteoarthritis. partially preserved between the microfracture holes, improving load-bearing ability following healing. [25] In contrast to Pridie drilling, microfracture avoids any heat necrosis or polishing of the subchondral bone. [26,27] The equipment is standardized and the costs are minimal since expensive cell cultures are not necessary. Unlike osteochondral, perichondral, periosteal, or chondral autograft procedures, the problems of harvest site morbidity and arthrotomy are avoided. [28] The hemarthrosis resulting from the microfracture procedure may have further benefi ts. It is generally accepted that meniscal repair healing is better when carried out in association with anterior cruciate ligament reconstruction. [29] It is thought that the pluripotent cells may improve the repair process. Given that chondral lesions in the shoulder commonly occur in association with other pathology, there may be potential for improved healing of simultaneous labral or rotator cuff repairs.
It is diffi cult to quantify the effect of microfracture on patient outcomes in this study. All but one patient had associated pathology with simultaneous procedures carried out. Thus one would, in any case, expect improvement in pain and function following surgery. The lone patient with an isolated chondral defect improved signifi cantly postoperatively, especially with respect to pain. Reoperation on two patients demonstrated evidence of successful fi brocartilage regeneration. It would appear reasonable to assume that this regrowth contributed, at least in part, to pain reduction and functional improvement in our patients. Regenerating fi brocartilage, it is hoped, will provide more prolonged symptomatic relief as compared to debridement.
Similar results have been shown by Siebold et al. [30]  The mean follow-up period is relatively short at 15.4 months. Thus, while our follow-up is sufficient to show initial improvement, it remains to be seen if this improvement will be maintained.
We did not obtain postoperative follow-up MRI scans in our patients. Preoperatively, only 2 patients had a MRI. Both these patients were professional rugby players and had moved to another location, thus they were not available for follow-up.
A further weakness of our study could be that we did not obtain biopsies. We, however, felt that this would add little to our study as the histology of the regenerate following microfracture is well documented.
We only had eight patients in our case series. These lesions are relatively uncommon and thus it is diffi cult to obtain large numbers without multicenter collaboration.
We believe that microfracture is a useful technique in the shoulder, and may prevent the natural deterioration of grade IV lesions which is seen in other joints. Isolated grade IV chondral lesions are a rare fi nding at arthroscopy and therefore it would be diffi cult to study these lesions exclusively. Multicenter, prospective randomized controlled trials are needed to fully investigate the effect of microfracture within the shoulder. Long-term follow-up is necessary to determine if continued pain relief, beyond that obtained with debridement alone, can be achieved via means of fi brocartilage regeneration.