Rotator cuff pathology is commonly encountered in an orthopedic practice and can result in significant pain and dysfunction. The prevalence of rotator cuff tears in the general population is reported to be between 9.7% and 62%, and increases with age. Varkey et al. reported 397,116 Medicare patients underwent rotator cuff repair between 2005 and 2012. REFERENCE: Varkey et al JSES 2016 dx.doi.org.easyaccess1.lib.cuhk.edu.hk/10.1016/j.jse.2016.05.001 . The majority of these cases are now performed arthroscopically, given advancements in instrumentation, improved technique, and the advantages over open techniques, which include less soft tissue disruption and an improved ability to treat concomitant pathology. Although arthroscopic repair has been shown to yield excellent outcomes in the hands of an experienced surgeon, there is a learning curve, as well as potential preoperative, intraoperative, and postoperative pitfalls.
The overall morbidity of arthroscopic RCR is low compared with other orthopedic procedures; however, there are a number of complications that can occur and have lasting consequences for patients. These complications can be categorized as occurring intraoperatively or postoperatively and may be secondary to patient-related factors or to surgical technique. This chapter aims to discuss possible complications in arthroscopic RCR. An awareness of potential issues allows surgeons to develop improved practices to avoid these pitfalls, recognize problems when they arise, and be prepared to treat issues in an evidenced based manner. Moreover, in an era of shared decision making, real data about the risks of complications after arthroscopic RCR allow the patients to be better informed about the risks of their procedure.
Patient Selection and Patient-Related Factors
One of the most important factors in avoiding complications is careful and thoughtful patient selection. This includes a robust understanding of the indications for arthroscopic RCR, an assessment of the patient’s risk for undergoing a surgical procedure, and consideration as to whether the pathology can be addressed with the arthroscopic skill that a particular surgeon may possess. Attempting to arthroscopically fix a rotator cuff tear in a patient with advanced rotator cuff tear arthropathy, insufficient quality tissue for repair, advanced fatty infiltration/atrophy, or operating on a medically unstable patient may lead to failure. Although beyond the scope of this chapter, the proper indications should be carefully studied, and surgery should be undertaken when there is a high likelihood of success. Choosing the appropriate procedure in the appropriate patient is of the utmost importance. A preoperative discussion involving appropriate expectations and the risks and benefits of both operative and nonoperative management is critical to allow patients to participate in the decision-making process.
Several patient-related factors have been reported to affect rotator cuff tendon healing, and should be considered. A recent study showed that smoking adversely affected healing failure after arthroscopic RCR. Hyperlipidemia has been identified as a risk factor for retear after arthroscopic RCR. In addition, sustained hyperglycemia was shown to impair tendon–bone healing after RCR in a rodent model, and a higher rate of complications has been noted in the diabetic population. Modification of serum cholesterol levels and blood glucose levels may play a role in improving healing after repairs. Multiple studies have demonstrated inferior clinical outcomes in patients with worker’s compensation claims. , This specific patient population has overall lower clinical outcome scores and a higher rate of noncompliance, which should be factored into decision making.
Most intraoperative problems can be minimized through careful preoperative planning and meticulous operative technique. Some of the common intraoperative issues a surgeon can encounter during RCR are discussed in this section.
Positioning and Anesthesia
Beach chair and lateral decubitus are the most commonly used positions for arthroscopic RCR. Although the preponderance of evidence has not shown any clear superiority in terms of set-up, visualization, or portal placement, there are certain advantages and disadvantages to each. The beach chair position, particularly when a limb positioner is used, allows the surgeon to rotate and position the arm without an assistant to facilitate access for anchor placement and visualization. There are data demonstrating a higher incidence of cerebral desaturation–related effects when in the beach chair position; however, recent studies have shown that this is safe position for shoulder surgery. , The risk of cerebral desaturation may be related to the position of the head and intraoperative management of blood pressure. , The patient’s blood pressure should be closely monitored on the contralateral arm at the level of the heart to avoid errors in pressure measurement, which can potentially produce possible serious complications related to perfusion, including stroke or death. In some centers, a cerebral oximeter has also been used for real-time monitoring of cerebral perfusion, and may be a useful adjunct. Traction injuries specific to the lateral decubitus position may occur from distraction of the extremity. Pressure-related injuries may also be seen in the lateral position, and can be avoided by use of an axillary role and careful padding of the down extremities. Complications innate to general anesthesia may be minimized by careful patient medical optimization preoperatively. Patients with medical comorbidities or prior issues with anesthesia may benefit from a preoperative appointment with anesthesia and/or their primary care provider for medication optimization and anesthetic planning. Regional anesthesia has been shown to be a highly effective adjunct in shoulder surgery, but also carries a low risk of neurovascular injury. The use of multimodal anesthesia, including an interscalene block, may be useful to decrease opioid consumption in shoulder surgery.
Technical Complications: Inability to Anatomically Reduce the Cuff
Especially in chronic retracted rotator cuff tears, mobilization of the rotator cuff tissue to appropriately cover the footprint may be challenging. In our experience, a significant number of massive and retracted rotator cuff tears seen on magnetic resonance imaging may be amenable to primary repair after adequate mobilization. However, in cases with a high degree of retraction (Patte 2), fatty infiltration (Goutallier 4), macerated tissue, and a large or massive tear pattern according to the Cofield classification, the patient should be consented for alternative procedures such as patch augmentation, superior capsule reconstruction, tendon transfer, or reverse total shoulder arthroplasty ( Table 30.1 ). Critical assessment of the preoperative imaging will allow the surgeon to have the proper instrumentation and implants readily available to quickly transition to another procedure if needed.
|Stage 1||Proximal stump close to bony insertion|
|Stage 2||Proximal stump at level of humeral head|
|Stage 3||Proximal stump at glenoid level|
|Stage 0||Normal muscle|
|Stage 1||Some fatty streaks|
|Stage 2||<50% fatty muscle atrophy|
|Stage 3||50% fatty muscle atrophy|
|Stage 4||>50% fatty muscle atrophy|
In cases of massive retracted tears, visualization is paramount to allow for proper mobilization. We typically use four portals (posterior, anterior, anterolateral, posterolateral). After the subacromial bursectomy is performed and the tear is visualized, traction can be applied by pulling the arm downward in a slightly forward flexed and abducted position to increase the subacromial space, yielding an improved view. An arthroscopic tissue grasper or a traction suture can be used to test the tissue excursion and maintain traction while releasing scar tissue. Adhesions must be carefully released in both the subacromial space and intraarticular space on the underside of the cuff to allow full mobilization. The coracohumeral ligament can be released to improve excursion of the supraspinatus ( Fig. 30.1 ), and in certain cases a posterior interval slide ( Fig. 30.2 ) can be of utility. Awareness of the anatomy of the suprascapular nerve (SSN) is essential to avoid damage while performing the releases. The surgeon should be prepared to perform extensive releases and arthroscopic mobilization; however, if the tissue will not fully reduce to the footprint, being prepared with appropriate grafts for a patch augmentation or superior capsular reconstruction is essential. Advanced preparation will allow for the completion of a procedure in one setting rather than requiring a return to the operating room and separate anesthetic. In cases where the surgeon does not have experience with a grafting procedure, the patient can be treated with a debridement and partial repair.
Technical Complications: Anchor Pull-Out or Fracture of the Footprint
Anchor pull-out is rare with modern suture anchor designs; however, when the bone is soft, inadequate purchase can be a problem. We typically use 4.75-mm suture anchors and a punch to assess the bone quality. In cases where the bone is soft, we either will change to a smaller punch (most cases) or place a larger (5.5-mm) suture anchor. Suture anchors should be placed in the proximal and middle parts of the greater tuberosity (rather than distal) to improve load to failure. Double-row fixation using linked constructs also dissipates the force and decreases the risk of anchor failure. In cases where the bone quality is so poor that anchor fixation is impossible to achieve, transosseus repair techniques using cortical augmentation may be considered as well.
The so called “deadman-angle” of suture anchor insertion has been reported to improve pull-out strength and minimize total tension across the sutures. , Other studies have recommended a 90-degree angle of insertion for improved fixation. It is clear that the loading direction and placement of the suture anchor play a role in the performance of the suture anchor suture complex. Moreover, the depth of the anchor is important because Bynum et al. found that the varying the depth of suture anchor insertion changes the mechanical properties and mode of failure of suture anchor constructs. These results were consistent with other data demonstrating that deeper placement of suture anchors for increased purchase caused greater displacement and was not recommended. The best location to place an anchor at is the articular margin or the lateral cortex of the tuberosity. A linked double-row construct can optimize placement of anchors in these positions with tension shared across each of the anchors. We prefer to use a double-row linked construct, to insert the anchors perpendicular to the bone, and to insert the anchor to a depth flush with the bone ( Fig. 30.3 ).
Technical Complications: Issues with the Sutures and Tensioning
Although the first-generation anchors were loaded with sutures of insufficient strength, breakage in modern high-strength sutures or tapes is uncommon. A biomechanical analysis revealed that newer suture products show significant improvements in load to failure values when compared with older braided polyester sutures. Great care should be taken during the process of arthroscopic RCR to visualize the anchor while grasping and shuttling sutures. This will prevent inadvertent unloading. Most anchors come preloaded with two or three sutures, and certain anchors may allow the surgeon to reload a deployed unloaded suture anchor.
Appropriate knot tying and knot security (when required) are essential for appropriate tissue reduction and fixation. Lo et al. demonstrated that making a sliding knot alone without reversing half-hitches on alternating posts leads to both poor loop security and knot security and should not be done. The addition of three reversing half-hitches on alternating posts improved the knot security of all sliding knots to adequately resist predicted in vivo loads. When tying knots we prefer a Weston (sliding locking knot) backed up by a minimum of three half-hitches. The use of suture tape may improve contact area and prevent cutout of the tissue. In most cases, we now use a knotless construct with suture tape because it offers a more consistent and biomechanically stronger fixation.
As per the basic principles of soft tissue repair, insufficient tensioning can cause inadequate reduction of the soft tissue to the footprint, thus impeding proper healing. Although not unique to RCR, overtensioning may present a unique set of complications in this procedure, including medial failure of the repair near the musculotendinous junction. Based on a study by Neyton et al., careful placement of the sutures lateral to the musculotendinous junction is less likely to lead to failure of the repair at the medial row. A musculotendious tear after RCR is a difficult complication, which may not be amenable to a revision repair.
Technical Complications: Dog Ears
Dog-ear deformities, or redundancies of the cuff tissue, occur because of inadequate compression of a section of the rotator cuff onto the footprint. This results in a puckering of the tissue, typically on the anterior or posterior edges of the limbs of the suture bridge. If left untreated, this can cause insufficient footprint restoration and impede healing at the greater tuberosity, and could possibly lead to a retear at the site of the gap formation. All efforts should be made to avoid this deformity through proper spacing of the sutures and anatomic repair of the tendon. A provisional reduction with a grasper will help assess if a dog ear is likely. In this case, an extra suture should be placed into the redundant tissue and incorporated into the lateral row. Double-row techniques such as the transosseous-equivalent RCR or modified suture-bridge technique allow for reduction of redundant lateral tissue.
Heyer et al. reported a 30-day postoperative complication rate of 0.7% among 21,143 patients following arthroscopic RCR. They were able to identify male sex, American Society of Anesthesiologists class over 2, and history of chronic obstructive pulmonary disease and dyspnea as independent risk factors for 30-day complications following arthroscopic RCR.
Arthroscopic RCR offers a minimally invasive approach; however, postoperative pain can still become a difficult issue. Identification of factors that predispose patients to pain can help physicians with appropriate pain management postoperatively. Factors such as subjective pain tolerance, use of preoperative narcotics, smoking, and younger age have been noted to correlate with higher postoperative pain scores. Setting appropriate expectations is extremely important in these cases, and a pain management plan should be discussed with the patient before surgery to avoid complications from excessive use of narcotic medications postoperatively. Kim et al. showed that high initial Visual Analog Scale scores and the acute onset of pain affected the postoperative pain pattern. Additionally, stiffness of internal rotation 3 months postoperatively affected the higher than average intensity pain pattern for each period after arthroscopic RCR.
Several options are available to patients following surgery to manage postoperative pain. Pain modalities include opioid and nonopioid medications, cryotherapy, intralesional analgesia, SSN blocks, interscalene brachial plexus blocks, and indwelling interscalene catheters. The use of multimodal analgesia has reduced the need for opioids postoperatively in shoulder surgery, and is highly encouraged. In the era of opioid crisis in the United States, note that approximately 43% of patients undergoing RCR received opioid medications before RCR. Patients who were prescribed narcotics before RCR were at increased risk of postoperative opioid demand. Moreover, patients with psychiatric diagnoses, myalgia, and lower back pain may be at increased risk of prolonged opioid use after surgery. All methods have pros and cons, and surgeons should critically analyze patients to identify which method is the best for each patient.
Persistent postoperative pain in medium- to long-term follow-up may be attributed to missed pathology at the initial time of surgery, a retear, or an infection (see later). For example, long head of the biceps tendinopathy has been shown to be commonly associated with rotator cuff tearing and, if left untreated, may cause persistent anterior shoulder pain after RCR. Great care should be taken at the time of arthroscopy to diagnose and treat concomitant pathology.
One of the most common complications after RCR is a structural failure of the repair, resulting from either a retorn tendon or insufficient healing. The rate of structural failure has been shown to be approximately 20%, depending on the size of the tear and the quality of the tissue. We believe that, with careful patient selection, the rate of structural failure can be much lower. There are a number of factors that increase the risk of reinjury in patients, including those related to the patient and those related to surgical technique. Studies have shown that preoperative tear size, higher stages of fatty infiltration, older age at surgery, and longer operative times were independent predictors of retear. Different techniques of repair have also been shown to reduce retear rates, with double-row and suture-bridge methods having lower rates of retear compared with single-row. Adequate compression of the tendon to the footprint will increase healing and thus promote greater rotator cuff integrity. Surgeons should be aware of the benefits of different fixation techniques and use one in which they are able to obtain the most anatomic repair with appropriate strength for healing.
Although some patients may experience a structural failure of the repair, note the difference between structural failure and clinical failure. Many patients who sustain a retear of the repaired tissue may not have a clinical failure of the procedure. Patients with intact RCRs have lower VAS scores; however, studies have found that the difference between patients with intact rotator cuffs and patients with retears did not reach clinical significance. ,
Patients who sustain a retear have the option to undergo a revision procedure or to treat the tear nonoperatively. A persistent defect has been shown in some cases to be a well-tolerated condition that only occasionally requires subsequent surgery. This decision is dependent upon the patients’ function, pain, and desire to undergo another procedure. In certain cases of massive rotators cuff tears that tear after repair, consideration of a superior capsular reconstruction or reverse shoulder arthroplasty may be prudent if it is felt the tissue would not be amenable to further repair.
Immobilization of the shoulder after RCR may be beneficial for the healing of the tendon to bone; however, in many cases a longer period of immobilization leads to increased stiffness. McNamara et al. showed that in patients who developed stiffness after surgery, an RCR was more likely to heal. The incidence of stiffness following RCR is 1.5% to 11.1%. Some studies have shown that older age is a risk factor for stiffness, and that preoperative limited range of motion influences the rate of postoperative stiffness. Many patients with mild to moderate postoperative stiffness can be treated with proper physical therapy. In those patients with severe limitation in motion and/or failed attempts at directed therapy, an arthroscopic capsular release and manipulation should be considered. In a series of 489 arthroscopic RCRs, Huberty et al. reported an overall postoperative stiffness rate of 4.9% requiring arthroscopic capsular release. Risk factors for stiffness in this group included a history of adhesive capsulitis, single-tendon cuff repair, partial articular supraspinatus tendon avulsion repair, and having worker’s compensation insurance. Late postoperative stiffness, especially acute onset, may reflect a retear of the repair and should be carefully evaluated.
Conflicting data exists in the scientific literature regarding the appropriate rehabilitation of patients after an arthroscopic RCR. There is a tradeoff between aggressive early motion, which can theoretically cause loosening of the repair, and delayed therapy, which predisposes to stiffness. We prefer initiating early passive range of motion in most cases. In certain circumstances, including massive tears and poor tissue quality, we will delay the initiation of and/or restriction of passive range of motion for 2 to 4 weeks with the aim of minimizing stress placed on the repaired tissues to facilitate early tissue healing ( Table 30.2 ).