The subacromial bursa contains algogenic substances and inflammatory cytokines; therefore, it is believed that bursa removal would decrease symptoms (Fig. 2). Unfortunately, in the subacromial space of shoulder with massive tear, especially if the lesion is not recent, bursa is scarcely represented; therefore, shoulder pain is only in part due to the inflamed bursa. Its removal only marginally affects shoulder pain.

The tear margins are represented by a chronic inflammatory infiltrate, consisting of synovial-like cells, lymphocytes, macrophages, plasmacytes and young fibrocytes. Dystrophic calcifications have occasionally been observed. Adjacent to the inflammatory infiltrate, the tendon appears hypocellular and disorganized, with micro-fragmentation of the normal collagenous architecture. Areas with myxoid or fatty degeneration were often observed [2]. Therefore, it is believed that removal of this tissue may determine pain decrease (Fig. 3a, b).

Acromioplasty is not a standard procedure of the arthroscopic debridement because inevitably it will cause the disruption of the coracoacromial ligament and an incentive for the upper ward migration of the humeral head.

Resection of the distal clavicle (arthroscopic Mumford) (Fig. 4a, b) is not routinely recommended.

Post-operatively, patients’ shoulder is immobilized in a sling for 1 or 2 days; successively, they are allowed immediate active and passive mobilization as tolerated.

Advantages of this technique include: (a) a short operative time; (b) low risk of complications compared with the common cuff repair procedures; (c) uncomplicated rehabilitation protocol.

Patient has to be informed that the main purpose of this procedure is pain relief; shoulder function or strength could only scarcely increase.

In 1991, Burhart [3] attempted arthroscopic cuff debridement and acromioplasty in six patients with massive tear, following strict biomechanical, anatomic and preoperative patient selection guidelines. Five of the six patients had near-normal preoperative active motion, with one having adhesive capsulitis. Pain relief was 90–100 % in all cases. Wiley described superior humeral head migration as a complication of rotator cuff debridement and bursal decompression and concluded that debridement alone may lead to upward migration of the humeral head and an increase in disability [4]. Ogilvie-Harris and Demaziere [5], in 1993, observed that 22 % of patients had moderate loss of function at follow-up after cuff repair but that greater than 64 % had similar loss after arthroscopic debridement. In the same year, Ellman et al. [6] stated that arthroscopic subacromial decompression and debridement of rotator cuff tears has a valuable, but limited, role in selected patients. They noted that patients with irreparable tears did not regain strength or range of motion, but did have significant pain relief. The year after, Zvijac et al. [7] reported on the 3–6-year long-term follow-up of 25 patients treated with arthroscopic debridement associated to acromioplasty alone for full thickness repairs of the rotator cuff. They observed a trend of further deterioration of results with time. There was a significant decrease in rating regarding shoulder pain and function compared with no deterioration with regard to motion and strength. Montgomery et al. [8] submitted 19 patients to arthroscopic debridement and concluded that this procedure, associated with subacromial decompression, was inferior to rotator cuff repair. Furthermore, five of the 19 patients went on to develop cuff tear arthropathy and were treated with hemiarthroplasty.

Rockwood et al. [9], in 1995, proposed the open debridement associated with acromioplasty. After surgery, 50 patients (53 shoulders) were followed (mean follow-up: 6.5 years); out of these, 83 % had satisfactory functional results, with a significant decrease in pain and an average increase of 35° of shoulder forward flexion. Two years later, Gartsman [10] reported similarly encouraging results with open debridement; however, the author a decrease in strength and suggested that this weakness might be due to the incompetent coracoacromial ligament and the loss of superior humeral head containment. The author further demonstrated that irreparable subscapularis or teres minor tendon tear, or both, and superior migration of the humeral head were negative prognostic factors for this procedure. In the same year of Gartsman’s publication, Melillo et al. [11] observed that only 8 % of their patients who were submitted to debridement had had a satisfactory result. Twenty-three of the 25 patients who underwent debridement required additional surgery. Furthermore, in nine of the 25 patients, degenerative changes developed, compared within two of 27 patients submitted to cuff repair. In 1999, Kempf et al. [12] performed a multi-centre study of 210 rotator cuff tears treated by arthroscopic acromioplasty. The preoperative Constant score was 38.2 points (41 % were supraspinatus tears, 40.2 % were supra- and infraspinatus tears, 10.5 % were three-tendon tears and 8.1 % were supraspinatus and subscapularis tears). Acromioplasty was associated to tear debridement in 183 cases (88 %) and to a tenotomy of the long head biceps tendon in 38 cases. Global objective results shown by the corrected Constant score reached 79.7 % and were satisfying in 73 % of cases. Preoperative shoulder stiffness, post-operative painful crises, worker compensation, a preoperative history longer than 4 years, and young age were considered as poor clinical factors; while osteoarthritis, a lesion of the acromioclavicular joint or of the biceps tendon were considered as poor anatomic factors.

Klinger et al. [13], in a nice paper published in 2005, said that arthroscopic debridement of massive and irreparable tears provides reliable expectation for improvement in function, decrease in pain and improvement in shoulder score for most patients. Additional tenotomy of the long head biceps tendon did not significantly influence the post-operative results at the latest follow-up.

In 2008, Liem et al. [14] retrospectively reviewed 31 patients after arthroscopic debridement of an irreparable rotator cuff tear (mean follow-up: 4 years). They observed that the mean ASES score significantly improved from 24 points preoperatively to 70 points at follow-up. Scores for pain were reduced from 8 to 3 points on a visual analog scale ranging from 0 to 10 points.

Barth et al. [15] compared the functional results of arthroscopic debridement with those obtained after partial repair. In their series, consisting of 42 patients, both groups had good or satisfactory results after rotator cuff surgery. Regardless of high rates of structural failures of partial rotator cuff repair, determined using ultrasonographic evaluation, the results of arthroscopic partial rotator cuff repair demonstrated slightly better functional outcome than debridement. A similar study was conducted in 2012 by Franceschi et al. [16]. Authors, after evaluating 68 patients, concluded that both techniques are effective in reducing patients’ symptoms, with higher functional outcomes for partial repair. In particular, those who underwent debridement did not show any increase in strength, maintaining the same preoperative values.

Although a variety of studies have shown that this remains a viable option in the elderly and low-demand patient, it does not slow the progression of osteoarthritis [17]. However, satisfactory results are mostly likely to be achieved in elderly and low-demand patients for whom pain relief is the priority and in patients in whom the integrity of the deltoid is preserved and who have good external rotation strength preoperatively, indicating an intact posterior cuff.

In 2001, Burkhart [18] coined the idiom “margin convergence” to indicate side-to-side closure of massive, U-shaped cuff tears (Fig. 5). The author observed that most apparently irreparable ruptures are not retracted but are L-shaped tears with a vertical split from medial to lateral, which take on a U-shape because of the elasticity of the muscle-tendon unit. Burkhart added that attempts at mobilization of tear margins lead to failure of repair because of tension overload at the apex of the tear, whereas side-to-side repair determines a mechanical advantage because of a biomechanical principle called margin convergence. This technique requires that free margin of the tear converges towards the greater tuberosity as side-to-side repair progresses. As the margin converges, the strain at the free edge of the cuff is reduced significantly, leaving an almost tension-free converged cuff margin overlying the humeral bone bed for repair. Side-to-side repair of two-thirds of a U-shaped tear reduces the strain at the cuff margin to one-sixth of the strain that existed at the pre-converged cuff margin. This surgical approach furnishes a lower probability of failure of fixation to bone, either by anchor or transosseous repair. If a partial defect persists on the superior cuff after margin convergence, it can be repaired to bone using one or two anchors.

Mazzocca and colleagues [19] believe that margin convergence decreases the size of the tear gap and reduces strain with minimal effect on glenohumeral translation and intrinsic tendon strain during knot tying.

In non-mobile tears, an interval slide as described by Tauro [20] may allow an additional 1–2 cm of lateral excursion of a supraspinatus tendon and permit a greater degree of partial repair. The technique requires that the adhesions between the cuff and the superior capsule have to be released with an arthroscopic elevator. Successively, the mobility of the cuff is checked with a grasper inserted through the lateral portal. The interval slide is performed by introducing a basket punch through the lateral portal and through the rotator cuff tear. The capsular attachment to the supraspinatus tendon has to be released. The release is performed from lateral to medial until the articular surface of the glenoid until the tendon is completely detached from the rotator interval capsule (Fig. 6). Tauro suggests performing the release while viewing from the posterior intra-articular portal. In the cases where the biceps tendon is ruptured, the exact position of capsular division must be approximated at the anterior-superior corner of the glenoid. Usually, supraspinatus mobility is greatly improved after the interval slide. A side-to-side repair of the supraspinatus to the infraspinatus is an integral part of the repair. The tendon is then repaired back to the greater tuberosity using the suture anchor technique. If adequate mobilization cannot be achieved, a decision to proceed with debridement or partial repair has to be made.

Concerns regarding this technique include re-vascularization of the superior cuff and de-functioning of an already impaired muscle tendon unit from the interval slide [21].

The advantages of this technique are thought to be a more anatomical and reliable repair. However, studies comparing the results of partial repair with interval slide found no significant difference in outcomes [21, 22].

Burkhart et al. [23], in 1993, first introduced the biomechanical concept of the “suspension bridge” in the rotator cuff. This theory has led to the hypothesis of the functional rotator cuff and furnished a rationale for partial repair of the cuff tear. This procedure involves the restoration of the cables involved in force transmission as well as force couples around the shoulder. The rotator cables have been defined anatomically at the level of the biceps tendon above supraspinatus anteriorly and the lower border of infraspinatus posteriorly. The vast majority of the irreparable tears have an anterior and/or posterior extension that affects the transverse couples. The concept that the balance between the transverse couples (consisting of subscapularis and infraspinatus-teres minor complex) must be restored has been strongly stressed.

After removal of the subacromial bursa, which is often scarcely represented, the mobility of the residual cuff is tested by grasping the edges of the tendons with a clamp and trying to pull it laterally to the footprint. Repair has to include all of subscapularis as well as the inferior half of infraspinatus as a minimum. If this is possible, the transverse force couples and a stable fulcrum for physiologic shoulder kinematics are restored. The supraspinatus footprint may be flattened with a bur. Tenotomy of the biceps tendon and acromioplasty are usually performed before the partial repair.

Using this technique, Burkhart obtained an improvement in forward flexion (from 60° to 150°), strength (from 2.1 to 4.4 in a 0–5 scale). The UCLA score improved from 9.8 to 27.6. All but one of his series of 14 patients obtained satisfactory results. In their series, Duralde and Bair [24], in 2005, had similar results and a significant improvement of the ASES score. Of 24 patients, 11 (46 %) had excellent results; five (21 %) good; seven (29 %) fair and one (4 %) poor. Despite residual defects in the rotator cuff tendon that averaged 1 × 3 cm, patients with a partial repair demonstrated impressive improvements in strength and range of motion at a mean follow-up of 21 months (Burkhart e Duralde). Porcellini et al. [25] observed that the Constant score of their patients improved from 44 to 73 points and the Simple Shoulder Test from 4.6 to 9. The acromion humeral distance increased from 6.1 to 9.1 mm. They stated that “partial” functional repair of the infraspinatus, leaving the greater tuberosity uncovered, gives good results in terms of patient satisfaction and in restoring the acromion humeral distance. Furthermore, they added that complications were rare and in line with the usual sequelae of a rotator cuff repair.