We have conducted a prospective study to assess the mid-term clinical results following arthroscopic repair of the rotator cuff. Patients were evaluated using the Constant score, subjective satisfaction levels and post-operative ultrasound scans.

Of 115 consecutive patients who underwent arthroscopic repair of the rotator cuff at our institution, 102 were available for follow-up. The mean period of follow-up was for 35.8 months (24 to 73). The mean age of the patients was 57.3 years (23 to 78). There were 18 small (≤ 1 cm in diameter), 44 medium (1 cm to 3 cm in diameter), 34 large (3 cm to 5 cm in diameter) and six massive (> 5 cm in diameter) tears. There was a statistically significant increase in the size of the tear with increasing age (p = 0.0048).

The mean pre-operative Constant score was 41.4 points (95% confidence interval, 37.9 to 44.9), which improved to 84.5 (95% confidence interval, 82.2 to 86.9). A significant inverse association (p = 0.0074), was observed between the size of the tear and the post-operative Constant score, with patients having smaller tears attaining higher Constant scores after repair. Post-operatively, 80 patients (78.4%) were able to resume their occupations and 84 (82.4%) returned to their pre-injury leisure activities. Only eight (7.8%) of 102 patients were not satisfied with the outcome.

Recurrent tears were detected by ultrasound in 19 (18.6%) patients, and were generally smaller than the original ones. Patients with recurrent tears experienced a mean improvement of 31.6 points (95% confidence interval, 23.6 to 39.6) in their post-operative Constant scores. Those with intact repairs had significantly improved (p < 0.0001) Constant scores (mean improvement 46.3 points, 95% confidence interval, 41.9 to 50.6). Patient satisfaction was high in 94 cases (92%), irrespective of the outcome of the Constant score. Recurrent tears appear to be linked to age-related degeneration.

Arthroscopic repair of the rotator cuff leads to high rates of satisfaction (92%) and good functional results, albeit with a recurrence rate of 18.6% (19 of 102).

The purpose of this study is to report the 1 to 5 year results of arthroscopic Rotator Cuff repairs.

Between November 2001 to May 2003, 115 consecutive patients were operated (73 males and 42 females) with arthroscopic repair. 13 patients were lost to follow up, leaving 102 patients available for follow up. Patients were evaluated using the Constant score, satisfaction levels and ultrasound scan to evaluate cuff integrity. Failures were defined as dissatisfied patients and those who had had a re-operation. Re-tear rate was recorded.

The mean follow up time was 23.8 months (range 12–61). Mean age was 57.3 years (range 23–78). 47% had a history of trauma. There were 107 patients (95.5%) with full thickness tears and 5 (4.5%) had partial thickness tears. Of the full thickness tears, 8 (7.6%) were massive in size, 36 (34%) large, 44 (41.5%) medium and 18(17%) small. Isolated Supraspinatus (SSP) tear was recorded in 83.5% and subscapularis tear in 7 %. A combination of SSP tear with infraspinatus and teres minor was found in 9.6%.

86% had Acromioplasty (ASD) with or without an AC joint excision arthroplasty. Two patients had Bankart repairs in addition at the time of cuff repair.

The mean pre op Constant score was 40.9 points (95% CI 37.3 to 44.5), which had improved to 84.8 (CI 82.2 to 86.9) at last follow-up. 78% returned to same work and 82% returned to pre injury leisure activity. There were 20 re-tears (19.6%). eight of the 102 patients were not satisfied. Five of these patients had revision operation.

Arthroscopic cuff repair shows high satisfaction rate (92%) and good functional results with 20% re-tear rate, while offering all the advantages of arthroscopic surgery.

Considerable controversy remains in the literature as to whether hemiarthroplasty or total shoulder arthroplasty (TSA) is the better treatment option for patients with shoulder arthritis. Several cohort studies have compared the outcomes of stemmed hemiarthroplasty with those of stemmed TSA and had inconsistent conclusions as to which procedure is best. However, these studies suggest that stemmed TSA provided better functional outcome. 340 CSRA cases were performed between 1987–2003, 218 Hemiarthroplasty – Humeral Surface Arthroplasty (HSA) and 122 TSA. There was very little difference in the functional outcome and pain in patients with and without a glenoid implant early, as well as, later after surgery. Mean post-operative Constant score for TSA was 85.0% (59.8 points) and for HSA patients 86.8% (62.3 points) with no statistically significant differences (t-test, p=0.4821). A highly significant difference between the overall proportions of revised cases was observed, with (21/122) 17.2% and (6/218) 2.8% of TSA and HSA cases revised, respectively (p< 0.0001). Further, HSA prostheses survive significantly longer than TSA prostheses. The difference between the survival curves was highly significant, both in the earlier post-operative period (Wilcoxon’s test, p=0.0053) as well as the later on (Log-rank test, p=0.0028). Long-term survival of total joint replacement is related to polyethylene wear debris, and therefore its use should be avoided if possible. The difference between our series and those with stemmed prostheses may be due to the fact that with surface replacement the normal anatomy for each patient can be mimicked better than with the stemmed prostheses and there is substantially less place for error as in stem positioning, head sizing or wrong version that may lead to glenoid erosion and less favourable result. Our current practice is and we suggest performing Copeland humeral surface replacement without insertion of glenoid prosthesis.

The aim of this study was to define the microcirculation of the normal rotator cuff during arthroscopic surgery and investigate whether it is altered in diseased cuff tissue.

Blood flow was measured intra-operatively by laser Doppler flowmetry. We investigated six different zones of each rotator cuff during the arthroscopic examination of 56 consecutive patients undergoing investigation for impingement, cuff tears or instability; there were 336 measurements overall.

The mean laser Doppler flowmetry flux was significantly higher at the edges of the tear in torn cuffs (43.1, 95% confidence interval (CI) 37.8 to 48.4) compared with normal cuffs (32.8, 95% CI 27.4 to 38.1; p = 0.0089). It was significantly lower across all anatomical locations in cuffs with impingement (25.4, 95% CI 22.4 to 28.5) compared with normal cuffs (p = 0.0196), and significantly lower in cuffs with impingement compared with torn cuffs (p < 0.0001).

Laser Doppler flowmetry analysis of the rotator cuff blood supply indicated a significant difference between the vascularity of the normal and the pathological rotator cuff. We were unable to demonstrate a functional hypoperfusion area or so-called ‘critical zone’ in the normal cuff. The measured flux decreases with advancing impingement, but there is a substantial increase at the edges of rotator cuff tears. This might reflect an attempt at repair.

Introduction Pridie and Steadman independently noticed the development of a smooth layer of fibrocartilage when treating exposed subchondral bone in the knee using their techniques of drilling or microfracture respectively. Since 1997, patients presenting to our unit for a Copeland cementless Surface Replacement Arthroplasty (CSRA) with a congruent glenohumeral joint have routinely undergone biological resurfacing of the glenoid using a technique similar to that described by Pridie and Steadman. We present this technique of glenoid resurfacing, the histological and surgical outcomes in a consecutive group of patients.

Methods/Results Between 1987 and 2002, 218 CSRA were performed without replacing the glenoid. From 1997, 133 CSRA have been performed with multiple drilling of the glenoid face with a guide wire through the subchondral bone in to the underlying soft cancellous bone to stimulate bleeding. This causes formation of a fibrocartilaginous layer – biological resurfacing. 9 (6.8%) of the patients with biological resurfacing have subsequently undergone a shoulder arthroscopy for postoperative impingement pain. This allowed us to evaluate the glenoid surface – macroscopically a layer of cartilage was noted in all patients, intraoperative biopsies have confirmed this layer to be fibrocartilage microscopically. In the biological resurfacing group, the mean postoperative Constant score (CS) is 86.9 (age/sex adjusted), with a mean improvement in CS of 71.0. 3 (2.3%) patients have required revision.

Conclusion Our results confirm that glenoid drilling at the time of CSRA leads to the formation of a fibro-cartilaginous layer over the glenoid, with significant improvements in Constant scores and functional outcomes. These results are comparable to other published results for total shoulder replacement with polyethylene resurfacing of the glenoid and better than patients that have undergone stemmed shoulder hemiarthroplasty.

Arthroscopic rotator cuff repair has evolved significantly in the last decade and has become a standard treatment. Satisfactory results of arthroscopic subacromial decompression (ASD) in the treatment of rotator cuff tears have also been reported (1). The aim of this study was to compare the outcome following arthroscopic repair versus decompression alone in patients with small & medium rotator cuff tears (Classification of Post, Silver & Singh (2). There were 114 patients in the ASD group and 96 in the Arthroscopic rotator cuff repair group (RCR). The groups were statistically comparable in terms of patient age & gender. Clinical follow-up was performed at a minimum of 12 months post -operatively (average 36 months). The average Post-op Constant score was 69.8 for the ASD group and 86.4 for the RCR group. The average post-op pain score (S.D) in the ASD group was 10.9 (± 4.3) and in the RCR group was 13.6 (± 3.1). Post operative strength was 7.6 (± 3.6) in the ASD group and 16.7 (± 5.4) in the RCR group.26 patients (22.8%) in the ASD group and 3 (3.3%) of the RCR group required futher surgery. Patient satisfaction (maximum 10 points) was 7.4% in the ASD group and 8.9 (± 1.4) in the RCR. The results of this study support arthroscopic rotator cuff repair. Shoulder strength is improved and there is significant reduction in the need for revision surgery.

Arthroscopic Rotator cuff repair is gaining popularity in recent years; however, the results of arthroscopic repairs are yet to be reported. Between November 2001 to May 2003, 115 consecutive patients were operated (73 males and 42 females) with arthroscopic repair. 13 patients were lost to follow up, leaving 102 patients available for follow up.

The mean follow up time was 23.8 months (range 12–61). There were 107 patients (95.5%) with full thickness tears and 5 (4.5%) had partial thickness tears. Of the full thickness tears, 8 (7.6%) were massive in size, 36 (34%) large, 44 (41.5%) medium and 18(17%) small.

Mean age was 57.3 years (range 23–78). 47% had a history of trauma.

Mainly Supraspinatus (SSP) tear was recorded in 83.5% and isolated subscapularis tear in 7%. A combination of SSP tear with infraspinatus and teres minor minor (posterior tear) was found in 9.6%.

86% had Acromioplasty (ASD) with or without an AC joint excision arthroplasty. Two patients had Bankart repairs in addition at the time of cuff repair.

Patients were evaluated using the Constant score, satisfaction levels and ultrasound scan to evaluate cuff integrity. Failures were defined as dissatisfied patients and those who had had a re-operation. Re-tear rate was recorded.

The mean pre op Constant score was 40.9 points (95% CI 37.3 to 44.5), which had improved to 84.8 (CI 82.2 to 86.9) at last follow-up. 78% returned to same work and 82% returned to pre injury leisure activity. There were 20 ultrasound demonstrated re-tears (19.6%). However, the majority of patients with radiological re-tears had good function, pain relief and were satisfied. Eight of the 102 patients were not satisfied. Five of these patients had revision operation.

Arthroscopic cuff repair shows high satisfaction rate (92%) and good functional results while offering all the advantages of arthroscopic surgery.

Copeland Surface replacement arthroplasty provides a good outcome in cases of arthritis. However, the limitation of use of this prosthesis is severe bone loss of the humeral head.

We describe the use of Copeland Surface Replacement Shoulder Arthroplasty in cases of severe humeral head destruction extending more than 50% of the humeral head. The bone deficiency was reconstructed with the autograft from the reaming and reshaping of the humeral head and a synthetic bone graft substitute (Tricalcium phosphate granules (Biosorb)) mixed with the patient’s blood.

We report preliminary results in 8 cases. 3 males and 4 females (One patient bilateral). The average age was 48 years ( 22–76 years). 5 cases had Posttraumatic avascular necrosis of humeral head, two patients had locked posterior fracture dislocation with loss of more than 50% of the humeral head and one patient with severe juvenile rheumatoid arthritis.

In this series the remaining bone of humeral head was less than 50% ranging between 30–50% (average 38%).

The patients were followed clinically and radiographically. The average follow–up was 10.6 months (range: 5 to 15 months).

All patients had substantial improvement in there shoulder function and pain. All the patients were very satisfied. Average pre-op constant score was 9.3 points (range: 9–14). At the latest follow-up the average constant score was 59 (range: 36–74). Age and sex adjusted constant score was 68.6 (range: 37– 87).

Radiographically we observed good incorporation of the Bone graft substitute with no signs of loosening or any lucencies under the prosthesis.

These are early and encouraging results in this group of young patients with severe destruction of the humeral head. It seems like it may provide a good bone preserving solution in these cases and extend the frontiers for the surface replacement of the shoulder.

With respect to glenoid neck malunion and rotator cuff injury, conservative treatment is often considered best for extra-articular glenoid neck fractures. More recent studies reveal that outcomes are not uniformly good, but the cause of poor outcomes has not been investigated. We reviewed nine patients who had sustained a glenoid neck fracture of the scapula within the last 10 years and who had been treated conservatively with immobilisation and then early active motion.

Their functional and anatomical outcomes were analysed by clinical examination and validated scoring systems, including the Oxford questionnaire and Constant shoulder score. Plain radiographs and MRI studies were correlated with outcome.

None of the nine patients was free of pain and some had poor Oxford and Constant scores. Pain was associated with glenoid malunion and evidence of subacromial bursitis and/or rotator cuff tendinopathy.

Introduction: Laser Doppler Flowmetry (LDF) is an established technique for the measurement of perfusion in tissues, which has been applied in animal and human studies. Despite many studies the implications of perfusion and blood flow abnormalities in the rotator cuff (RC) in vivo are not clearly understood. This study aims to assess the blood flow in the normal and diseased RC using Laser Doppler Flowmetry (LDF) technique.

Methods: With Ethics approval and informed consent patients undergoing arthroscopy for impingement or cuff tear (diseased cuff) or for shoulder instability (normal cuff) were recruited. Following diagnostic arthroscopy and bursoscopy measurements of the blood flux in the RC using the LDF were made at 6 points. Five of these were in the cuff over a 4 cm² area from the insertion at the greater tuberosity (A,E – @ Tuberosity level, C-Muscelotendinous junction, B,D – between the two) and one (F) in the deep surface. Diseased RC were subgrouped into mild (B1), moderate (B2) and severe (B3- cuff tear) impingement grades (Copeland-Levy Classification). The arthroscopy, grading, and probe placement were made by the two senior authors. LDF flux (unit of measurement of the perfusion – LDFf) was recorded over 30 seconds at each measurement point. The mean of these readings was calculated.

Results: 56 patients were recruited. 35 Males (62.5%) and 21 Females (37.5%).

336 measurements were recorded.

The observed mean LDFf was 32.8 (27.4–38.1; 95% CI) , 25.4 (22.4–28.5) and 43.1 (37.8–48.4; 95% CI) For Normal, Impingement and Tear cases, respectively (p< 0.0001, One-way ANOVA).

The LDFf was lowest in the Impingement without tear grade (B2) with a statistically significant increase at the edges of a cuff tear.

Conclusion: Preliminary recordings of LDF show changes are in keeping with current knowledge of the pathology. LDFf decreases with advancing impingement, being least affected in the musculotendinous junction. There is a substantial increase in LDFf at the edges of RC tears. This might reflect an attempted reparative response.

Introduction: Massive irreparable degenerative rotator cuff tears are amongst the most difficult conditions for treatment in shoulder surgery. These involve usually elderly patients, which present with severely painful and restricted active shoulder movement. These patients have low demand from their shoulders, mainly for pain relief and performing their simple activities of daily living. Major surgery for major tendon transfer will not be advisable in these cases in view of the morbidity involved and the questionable outcome. We suggest a simple non-surgical rehabilitation treatment consisting on anterior deltoid strengthening exercises in the supine position for re-education of the anterior deltoid to compensate for the absent rotator cuff.

Methods: 17 patients with degenerative (non traumatic) Massive irreparable rotator cuff tears were recruited. They were all greater than 70 years of age and of mixed gender. Patients were English speaking, had full mental faculties and gave informed consent. They suffered no other shoulder pathology and were not participants in any other upper limb rehabilitation. All patients complained on severe shoulder pain and severely limited active range of motion with inability to actively elevate the arm to the horizontal. They all had full passive range of motion.

The diagnosis of a Massive irreparable rotator cuff tear was confirmed by diagnostic ultrasound scan. The shoulder function was evaluated using the Constant Score. Patients’ active shoulder ranges of motion were recorded and video-recorded as well. Each participant was taught the initial 6-week of self Deltoid muscle exercise, executed in supine, at least three times a day. They were instructed that when they felt better control on their active shoulder movements to gradually recline up the head of the bed and continue with the same simple exercise. They were reviewed at 6 weeks re-assessed and re-taught the same exercise, with a 2kg weight in their hand. At the 12th week they were reassessed using the constant score, and their active range of motion was video recorded again.

Results: 90% of the participants expressed a significant improvement in their upper limb function already after 6 weeks of treatment. All components of the Constant score (beside the strength) have improved. 90% reported less pain and found general activities of daily living easier to execute and a diminished level of muscle fatigue. 10% of the patients were able to establish a recording of > 1.26kg on the myometer in 90 degrees of abduction. 10% failed to report any benefit.

Discussion and Conclusion: Anterior deltoid strengthening exercises in the supine position for re-education of the anterior deltoid seem to have a significant beneficial effect for restoration of shoulder function and pain relief in the majority of patients with Massive irreparable degenerative rotator cuff tears. Using this simple non-invasive rehabilitation technique helps to re-educate the anterior deltoid to compensate for the absent rotator cuff and restore shoulder function.

The Constant-Murley score has gained wide acceptance for evaluation of shoulder function. The strength component of the Constant score accounts for 25 out of 100 points. It has been criticized for lack of consistency in defined measurement method.

The aim of this study was to evaluate the effect of various variables on the strength component measurement of the Constant score.

Material & Methods We conducted a series of experiments using a digital force gauge (EZ force). We evaluated the effect of strength measurements with 1) patient in sitting & standing positions 2) strength gauge fixed to an immobile platform or hanging free fixed to the floor by the examiners foot 3) patient’s arm in 45 degrees and 90 degrees of abduction 4) plane of elevation in frontal or scapular plane and 5) patient making a fist or keeping the palm open during the test. These experiments were done in groups of 20 patients. We have compared as well this device and the Isobex Myometer.

Results No statistical differences were found between individual measurements with regard to patient’s position (standing-sitting), device setting (Fixed –Hanging), position of the arm in varying degrees of abduction or the plane of elevation. Strength assessment obtained when patient made a fist compared to open palm was found to be higher (p=0.006). The measurements showed good intra-observer reliability.

The readings of the EZ force and the Isobex myometer were comparable.

Conclusions It seems that the shoulder strength measurements as part of the Constant functional score may be performed with the patient sitting or standing, with the arm at varying degrees of abduction and in different planes of elevation without causing any significant deviation in the measurement.

No influence was found as well to the device being either fixed to an immobile platform or fixed to the floor by the examiner’s foot. These make these measurements easy to perform and reproducible using the newly designed digital force gauge (EZ force).

Aim: Despite many studies the implications of perfusion and blood flow abnormalities in the rotator cuff (RC) in vivo are not clearly understood. Laser Doppler Flowmetry (LDF) is an established technique for the measurement of perfusion in tissue, which has been applied in animal and human studies. This study aims to evaluate Laser Doppler Flowmetry (LDF) as a technique for the assessment of blood flow in the normal and diseased RC.

Methods: With Ethics committee approval and informed consent patients undergoing arthroscopy for impingement or cuff tear (diseased cuff) and instability (normal cuff – BO) were recruited. Following diagnostic arthroscopy and bursoscopy measurements of the LDF in the RC tear were made at 6 points. Five of these were in the cuff over a 4 cm. 2 are from the insertion at the greater tuberosity and one in the deep surface.

Diseased RC were sub grouped into mild (B1), moderate (B2) and severe (*B3 – cuff tear) impingement grades (Copeland – Levy Classification). The arthroscopy, grading, and probe placement were made by the senior authors. LDF flux (LDFf) was recorded over 30 seconds at each measurement point. The mean of these readings was then calculated (LDF flux – an arbitrary unit of measurement of the perfusion).

Results: 35 patients were recruited. 210 measurements were recorded, in 6 anatomical locations. The mean LDFf was 34.1 in diseased RC and 55.3 in normal RC (p=0.0002). The mean LDFf was 52.2 in severe (RC tear) impingement and 30.3 in mild and moderate impingement’s (p< 0.0001).

The LDFf was lowest in the moderate grade with a significant increase at the edges of a cuff tear.

Conclusion: Preliminary recording of LDFf show changes are in keeping with current knowledge of the pathology. LDFf decreases with advancing impingement. There is an increase in LDFf at the edges of RC tears. This might reflect a reparative response.

The coraco-acromial ligament (CAL) is partially resected during a subacromial decompression. Clinical studies have reported the regeneration of a structure which appears to be a new CAL. Histological studies of regenerated CAL have demonstrated an abundance of relatively acellular collagen fibrils orientated in the line of a ligament and mechanical testing of the regenerated tissue has properties similar to those of normal CAL. However it is still not clear whether this structure represents scar tissue or truly reformed ligament. Defining the major collagen constituent of this regenerated tissue would allow the distinction between ligament and scar tissue. Therefore the aim of this study was to examine the level of expression of types I and III collagen in regenerated coraco-acromial ligaments (CAL) in humans.

Samples of regenerated CAL were obtained during open surgery for repair of small rotator cuff tears at an average of 24 months (range 14 to 52) after arthroscopic subacromial decompression from 4 men and 3 women with an average age of 58 years (range 44 to 68). A standard protocol radio-active in-situ immunolocalisation technique was used to quantify the ratios of mRNA collagen I to collagen III in the samples.

The results demonstrated that the average ratio of collagen I to collagen III was 6.5. This ratio is similar to the value for normal hip capsule (5–6:1) and human posterior cruciate ligament (8:1).

We conclude that the reformed CALs are ligamentous structures, not scar tissue, and therefore represent truly regenerated ligaments.

Use of shoulder manipulation in the treatment of frozen shoulder (FS) remains controversial. One of the purported risks associated with the procedure is the development of a rotator cuff tear. However the incidence of iatrogenic rotator cuff tears has not been reported. The purpose of the study was to assess the effect of manipulation of the shoulder on the integrity of the rotator cuff.

In a prospective study 20 consecutive patients (21 shoulders) with FS underwent manipulation of the shoulder under anaesthesia (MUA). The average duration of symptoms was 7.3 months (4–18 months). Patients were assessed pre and post manipulation using the Constant score. An ultrasound scan of the rotator cuff was performed before and at 3 weeks after manipulation.

In all patients, pre and post manipulation ultrasound scans showed the rotator cuff to be intact. At 12 weeks after manipulation all patients indicated that they had none or only occasional pain. The mean improvement in motion was 83 degrees (range, 20 – 100°) for flexion, 95 degrees (range, 20 – 120°) for abduction, 58 degrees (range, 0 – 80°) for external rotation and 3 levels of internal rotation (range 3–5 levels). These gains in motion were all significant (p < 0.01). No fractures, dislocations or nerve palsies were observed.

In conclusion manipulation under anaesthesia for treatment of frozen shoulder resulted in significant improvements in shoulder function and pain relief as early as 3 weeks after surgery and was not associated with rotator cuff tears. When performed carefully this procedure is safe and leads to early improvements in pain relief, range of movement and shoulder function.

Avascular necrosis (AVN) of the humeral head is an extremely disabling condition (Gerber et al, JSES 1998. 7(6):586–90). The results of stemmed arthroplasty for this condition are good, with Hattrup and Cofield reporting 79% subjective improvement at nine years (JSES 2000;9:177–82). This study reports the outcomes of surface replacement shoulder arthroplasties for AVN over a 16 year period. Between 1986 and 2001 twenty-seven arthroplasties were performed in patients with advanced avascular necrosis of the humeral head. All patients had secondary degenerative changes. AVN was secondary to corticosteriods and trauma in most cases.

The mean age of the patients was 60 years (range 35 to 86). These included 16 hemiarthroplasties and 7 total shoulder arthroplasties. All prostheses were of the Copeland Surface Replacement Arthroplasty (CSRA) type.

The average follow up period was 6 years (range 1 to 13). The average preoperative Constant score was 17. This improved to 74 at follow-up. Forward flexion improved from 63 degrees preoperatively to 133 degrees at follow-up. Abduction improved from 49 degrees to 118 degrees. External rotation improved from −3 degrees to 61 degrees. Pain scores improved from 0 to 11.7, using a 15 point visual analogue scale. 81% of patients had slight and no difficulty performing their routine activities of daily living. The remaining 19% still had some difficulty with routine activities. Four of the patients performed regular overhead activity and recorded some difficulty in doing this, whereas they had great difficulty pre-operatively. There were no cases of loosening. No difference was seen in any of the results between the hemi-arthroplasty and total shoulder replacement patients.

Surface replacement arthroplasty is a suitable procedure for degenerative disease secondary to AVN of the humeral head, with results similar to stemmed prostheses. It has the advantage of preserving bone stock.

To assess the indication and role of shoulder arthroscopy for the problem shoulder arthroplasty.

Between 1995–2000, 28 patients who had excessive pain or limitation of motion following a shoulder arthroplasty underwent arthroscopy. A pre-operative diagnosis was made in 13 out of the 28 patients.

Of the 13 patients who had a pre-operative diagnosis an impingement syndrome was confirmed and successfully treated by arthroscopic subacromial decompression in 10, a rotator cuff tear was confirmed and debrided in two and in one loose bodies removed. Of the 15 patients who did not have a pre-operative diagnosis a post-arthroplasty capsular fibrosis was found in seven, six undergoing a successful arthroscopic capsular release. Loose or worn components were found in four of the shoulders, a small cuff tear was identified in one, a florid synovitis was present in another, loose cement was found in a further patient and in one no abnormality could be found. During the procedures orientation within the joint was often hindered by the reflection from the prosthesis making it difficult to differentiate between the real and mirror images of both the tissues and arthroscopic instruments. Access was also often compromised.

Arthroscopy following shoulder arthroplasty is useful for the diagnosis and treatment of pain and loss of motion in selected patients, but can be technically demanding. Diagnostic arthroscopy following shoulder arthroplasty should be considered for patients suffering from pain in whom no cause can be found using less invasive investigations.

Purpose: The aims of this study were to describe our technique and outcome of arthroscopic repair of small rotator cuff tears.

Methods: The technique involves two screw-in anchors placed into the footprint area, linked with a double-loop No. 1 Polydioxanone suture. It is quick and simple, providing a similar fixation to other commonly used methods. Between 1999 and 2001, 20 repairs were performed using this technique in patients with rotator cuff tears < 2cm in size. Prospective follow-up was conducted on these patients. Mean age was 56.5yrs (range 37–72yrs) with 10 males and 10 females (8 left and 12 right sided cuff tears).

Results: Mean follow-up period was 13mths (range 10–28mths). Constant scores improved from preoperative mean 34 (range 22–53) to postoperative mean 85 (range 76–96). Pain scores (/15) improved from preoperative mean 3.7 to postoperative mean 13.4. All regained full range of movement. All were satisfied with the operation. There was an average 90% subjective improvement. There were no complications.

Conclusion: Arthroscopic rotator cuff repair for small tears produces satisfactory results using this technique.

Purpose: The Subacromial bursa is the largest bursa in the body. In 1934 Codman described the presence of Subacromial plicae, similar to the suprapatellar plicae found in the knee. This anatomical finding was again described by Strizak et al. in 1982. It is recognised that plicae in the knee can cause anterior knee pain with impingement against the patella in young people. We investigate the possibility that a similar situation exists with plicae of the Subacromial bursa. The aims of this study were to document the incidence of bursal plicae seen at bursoscopy during arthroscopic Subacromial decompressions of the shoulder, and to assess whether there is any pattern to the occurrence of these plicae, and the relation to impingement lesions seen at bursoscopy.

Methods: A review of all patients undergoing Arthroscopic Subacromial Decompression (ASD) of the shoulder between January 1996 and July 2001.

Results: A total of 2043 ASD procedures were performed in the study period. Of these, the number of plicae found was 130, with an incidence of 6.4%. There was a strong age predilection, with a significantly higher incidence in younger age groups. There was no difference between males and females.

Where a plica was present the impingement lesion seen on the cuff side was significantly greater than the lesion seen on the acromial side (p< 0.0001). This suggests that the impingement might be due to the plica itself.

Conclusions: This study is the first to describe the presence of Subacromial plicae in living subjects and correlates with previous anatomical studies. The younger age predominance correlates with the findings of plicae in the knee. Our findings suggest that Subacromial plica may be a cause of impingement in young patients.

Purpose: To evaluate the short to medium term outcome of fifty-four consecutive patients (58 shoulders) treated exclusively with radiofrequency capsular shrinkage for atraumatic instability.

Methods: Twenty-three men and thirty-one women (35 shoulders) with a mean age of twenty-four years (range, 14 to 53), who failed a 12 months course of physiotherapy. Instability was multidirectional in 26 patients (29 shoulders), antero-inferior in 24 patients (25 shoulders) and postero-inferior in 4 patients.

Results: The Rowe score improved from 33.1 to 74.1 points at thirty-two months (range, 19 to 48) (p< 0.001). Instability recurred in 20 of 58 (34.5%) shoulders. Recurrent instability was related to the type of instability (10 of 13 shoulders for voluntary instability, 10 of 33 shoulders for involuntary instability and none of 12 shoulders for instability pain, p< 0.001) and previous instability surgery (7 of 10 shoulders, p< 0.01).

Outcome was not related to the direction of instability, type of radiofrequency probe, age or ligamentous laxity. Two patients had a transient reduction in sensation in the axillary nerve distribution. 22 of 38 (57.9%) patients returned to their pre-instability level of sporting activity.

Conclusion: Radiofrequency capsular shrinkage produces satisfactory results in instability pain and in involuntary instability in patients who had no previous instability surgery.