Over the last decade, a wealth of clinical experience with reverse shoulder arthroplasty has been published. The introduction of the reverse arthroplasty has had a major impact on the management of the rotator-cuff-deficient shoulder. However, reverse shoulder arthroplasty has demonstrated all of the drawbacks inherent in any artificial joint replacement, including durability. Patients with rotator cuff insufficiency have diverse presentations, and such insufficiency presents a unique challenge in young, higher-demand patients who are not ideal candidates for reverse arthroplasty. Since its initial presentation in 1988, latissimus dorsi transfer has gained wide acceptance for the management of such patients with irreparable rotator cuff tears. Longer-term follow-up studies such as this one by Gerber et al. are critical to our understanding of the role of latissimus dorsi transfer in this population, in terms of both clinical longevity and the effect of the transfer on the glenohumeral articulation.
The original series presented in 19881 and the follow-up study in 19922 helped define the clinical indications for this procedure, and those indications are reflected in the present study’s inclusion criteria. The present series demonstrates the long-term clinical effectiveness of latissimus dorsi transfer in a well-defined subset of patients with posterosuperior rotator cuff insufficiency. The clinical results achieved in the present series are comparable with those previously published with respect to improvement of shoulder motion and patient satisfaction. Importantly, these results at intermediate and long-term follow-up in the present study did not show appreciable deterioration over time.
The association of rotator cuff tears with subsequent glenohumeral joint degeneration has been systematically evaluated by many authors. Several classification systems have been proposed and have demonstrated progression of arthropathic changes in the glenohumeral joint in the presence of large defects in the rotator cuff. In the present series, the mean stage of osteoarthritis as classified by Samilson and Prieto3 progressed by 0.6 stages over the twelve-year follow-up period. Using the same classification in an evaluation of nineteen patients with a mean age of sixty-four years who had chronic tears of the rotator cuff, Zingg et al. demonstrated a mean progression of 1.1 stages at four years of follow-up4. This demonstrates that, although latissimus dorsi transfer does not prevent degeneration of the glenohumeral joint in the presence of a massive rotator cuff defect, it does appear to slow the progression of joint degeneration compared with that in patients undergoing no treatment. This has important implications when considering treatment options for the younger, high-demand patients who are the target population for this procedure.
Many patients who are considered candidates for this procedure have had prior failed operative attempts at surgical reconstruction of the rotator cuff defect. In the present series, the proportion of revision procedures in the group with an unsatisfactory result was greater than that in the group with a satisfactory result, but this difference did not reach significance (p = 0.09). The literature remains confusing regarding the results of latissimus dorsi transfer in the setting of revision repair. Buchmann et al. reported no difference in clinical results when comparing latissimus dorsi transfer in primary and revision cases in a retrospective matched-pair study with twenty months of follow-up5. This is in contrast to the experience of Warner and Parsons, who found a mean Constant score of 55% in the revision setting compared with 70% in primary procedures (p < 0.05) at nineteen months of follow-up6. Surgeons contemplating the use of latissimus dorsi transfer in the revision setting must carefully consider the strict inclusion and exclusion criteria presented in each of these studies.
The major limitation of the present series is the loss of 16% of the original patients. The authors noted that the information obtained at the intermediate follow-up examination did not differ significantly between those who were evaluated again at the time of final follow-up and those who were lost to follow-up after the intermediate examination. Although this does reduce the possibility of selection bias, the possibility of a greater number of prior procedures or more advanced degeneration in this group cannot be eliminated.
As surgical options for the management of the rotator-cuff-deficient shoulder expand, the role of latissimus dorsi transfer continues to evolve. We have seen the role of reverse shoulder arthroplasty expand exponentially over the last decade because of its clinical success. The present study provides a unique window into a series of patients with a minimum of ten years of follow-up after latissimus dorsi transfer. It demonstrates an effective and durable biologic solution to a challenging clinical problem. Until our ability to manipulate tendon healing improves, latissimus dorsi transfer will remain an important tool in the management of the young rotator-cuff-deficient patient.