- Email: [email protected]
All arthroscopic rotator cuff repairs
ALL ARTHROSCOPIC ROTATOR CUFF REPAIRS GARY M. GARTSMAN, MD
The arthroscopic operation for repair of full-thickness rotator cuff tears is successful and has the advantages of glenohumeral joint inspection, treatment of intra-articular lesions, smaller incisions, no deltoid detachment, less soft tissue dissection, less pain, and more rapid rehabilitation. However, these advantages must be balanced against the technical difficulty of the method, which limits its application to surgeons skilled in both open and arthroscopic shoulder operations. This article contains many technical pearls to, as much as is possible, simplify and improve all arthroscopic cuff repair. KEY WORDS: Arthroscopic rotator cuff repair
The role of the arthroscope in the management of rotator cuff lesions is evolving. In the last decade, shoulder arthroscopy had advanced from its use as a diagnostic tool to an effective treatment option for stage 2 impingement and acromioclavicular joint arthritis. >3Arthroscopic decompression has also been described in the management of more severe rotator cuff lesions such as partial-thickness tears. 4-6 Recently, repair of full-thickness rotator cuff tears by use of arthroscopic technique has been described. 7q3 The advantages of arthroscopic repair are that it allows smaller skin incisions, glenohumeral joint inspection, treatment of intra-articular lesions, no deltoid detachment, less soft tissue dissection, less pain, and more rapid rehabilitation. Arthroscopic repair of full-thickness rotator cuff tears has certain specific advantages for patients who present with athletic injur~Les. Commonly these individuals are younger, more athletic, present earlier, and have smaller tendon tears. These qualities make this patient population ideally suited for arthroscopic rotator cuff repair. Perhaps most importantly, arthroscopy allows the orthopedist to inspect and repair concomitant intra-articular lesions that occur frequently. Sometimes lesions in inaccessible locations, or in younger patients, for example, a patient with posterior cuff tear and an anterior Bankart lesion, would be difficult to manage through a single open incision.
INDICATIONS The indications for arthroscopic rotator cuff repair are identical to those for an open repair as discussed elsewhere in this section and should not in any way be altered or "broadened" in the mistaken view that arthroscopic repair is a lesser procedure. Although the skin incisions may be smaller and the deltoid is left attached, all elements of an open repair are performed arthroscopically, and patients From the Department of Orthopaedic Surgery, University of Texas
Houston Health Science Center, and the Fondren Orthopedic Group, Texas Orthopedic Hospital, Houston, TX. Address reprint requests to Gary M. Gartsman, MD, Fondren Orthopedic Group, Texas Orthopedic Hospital, Houston, TX. Copyright© 2000 by W.B. Saunders Company 1060~1872/00/0804-0006510.00/0 doi:10.1053/otsm.2000.17779
who are unable to tolerate either the surgery or the postoperative rehabilitation of an open procedure are not candidates for an arthroscopic rotator cuff repair. Surgery is performed for persistent pain that interferes with activities of daily living, work, or sports, and is unresponsive to a 6- to 12-month period of nonoperative care.
OPERATIVE
TECHNIQUE
Positioning I prefer the patient sitting, because easy access is afforded to the anterior, lateral, and posterior aspects of the shoulder. It is easier for most surgeons to convert to an open repair when the patient is in this position. Arm positioning during repair is greatly facilitated by the use of a McConnell arm holder (McConnell Co, Greenville, TX). Portals
Posterior. As the complexity of shoulder arthroscopy increases, precise portal placement becomes more critical. I prefer to mark the outlines of the acromion and lateral clavicle. Three portals are used (Fig 1). The posterior portal is 1.5 cm medial and 1.5 cm inferior to the posterolateral acromial border. This portal is made superior to the traditional point of entry and has an important advantage in that the arthroscope enters the subacromial space parallel to the acromial undersurface, allowing the surgeon to "look down" at the rotator cuff tear from a distance, improving perspective. Lateral. The lateral portal is made 10 to 15 m m posterior to the anterior acromial border and approximately 2 to 4 cm lateral to the acromion border. The axillary nerve can be in jeopardy if the lateral portal is placed more than 5 cm lateral to the acromial border (Fig 1). The lateral portal should allow the cannula to enter midway between the humeral head and the acromion. I identify the precise location of the lateral portal with a spinal needle bisecting the distance between the anterior and posterior edges of the tear.
Operative Techniques in Sports Medicine, Vol 8, No 4 (October), 2000: pp 293-300
293
\
\ \
/ t
Fig 1, (A) Portal sites. (B) Cannulas in position.
Anterior. The anterior portal is located approximately 2 to 3 cm anterior to the acromial border midway between the anterolateral border and the AC joint.
Inspection of Glenohumeral Joint The glenohumeral joint is entered and inspected. Because the most common location for partial-thickness tears is in the anterior portion of the supraspinatus, the lesion is well visualized. Shoulder rotation and abduction are then altered so that other areas of cuff tearing are clearly visualized. Frequently the tear will be overlooked unless the arm is placed in 30 ° to 40 ° abduction and 20 ° external rotation. Particular attention should be directed to the attachment and integrity of the glenohumeral ligaments and the biceps-labrum complex. The labrum should be carefully examined for signs of glenohumeral instability. If the posterior aspect of the shoulder cannot be seen clearly, then viewing should be performed from the anterior portal.
PARTIAL-THICKNESS ROTATOR CUFF TEARS With the advent and increased use of shoulder arthroscopy, partial articular surface tears have been frequently described. 5,6,13 The articular surface of the rotator cuff is extremely well visualized arthroscopically, and tears can be precisely measured. I use Ellman and Gartsman's
294
classification system. 14 They classified partial-thickness tears according to the depth of the lesion and the anatomic site. Tears are noted to be present on the articular surface or the bursal surface. Grade 1 lesions show definite disruption of the tendon fibers but are less than one fourth of the tendon (<3 mm) thickness. Grade 2 lesions are less than half of the tendon thickness and are 3 to 6 m m in depth. Grade 3 lesions are more than half the tendon thickness and are more than 6 m m in depth. If the margins and depth of the tear are not well visualized, then a small synovial resector is introduced through the anterior portal and the tear debrided. The decision process in the management of these lesions is complex, and the following are general guidelines. Lower-grade lesions and smaller rotator cuff tears may be treated with decompression, whereas more advanced lesions require repair. Certain bone shapes or changes are consistent with extrinsic tendon compression. Type 3 or hooked acromion, anterior acromial spurs, and inferior acromioclavicular joint osteophytes are radiographic findings that suggest that the patient will respond to operative decompression, whereas patients with benign bone anatomy more likely have intrinsic tendon lesions and require tendon repair. If tendon repair is needed, the area of partial tear is marked. With the tear in direct view, a spinal needle is introduced percutaneously into the center of the tear. A #1 PDS suture (Ethicon, Boston, MA) is passed down the GARY M. GARTSMAN
spinal needle into the joint. The needle is withdrawn; the arthroscope is removed from the joint and inserted into the subacromial space. The blue suture is identified and removed. A motorized shaver is used to convert the tear into a full-thickness lesion that is then repaired arthroscopically, as follows.
very thick and require resection with the electrocautery. This is particularly true in the area of the coracohumeral ligament. I have not found intra-articular division of the joint capsule to be helpful. Remember that tendon reparability is based not only on tendon mobility but also the integrity of the tissue to hold sutures and thus be used in the repair. 15
FULL-THICKNESS ROTATOR CUFF TEARS
Acromioplasty
Visualization
If the tear is reparable, I then evaluate the acromion. The purpose of acromioplasty is to create adequate space for the rotator cuff tendons. If the acromion compromises the space for the cuff or if there is clear evidence of impingement (coracoacromial ligament or acromial undersurface fraying), I perform an acromioplasty. Because the thickness and shape of the acromion varj6 the amount of bone removed during acromioplasty also varies. My goal is to achieve a flat acromial undersurface. If the acromion is flat and there is adequate space for the cuff, then acromioplasty is not performed.
After inspection of the glenohumeral joint, the arthroscope is removed and redirected into the subacromial space. ! establish a lateral portal and use a motorized shaver to remove any bursal tissue that impedes a clear view of the tendon tear. Surprisingly, the subacromial space is easily visualized in full-thickness tears. The thick subacromial bursitis that exists in stage 2 impingement is usually absent.
Tear Classification The arthroscope is then rotated so that it is pointed directly down at the rotator cuff tear. With small to medium tears, the size and geomefry of the tear is easily appreciated. The length of the tear from anterior to posterior as well as the amount of medial retraction is noted. Straight medial retraction or retraction in an elliptical shape are the most common findings. As tear size increases, the surgeon's ability to appreciate tear geometry becomes more difficult. Reverse "U' tears with a longitudinal component along the rotator interval will allow the tear to rotate posteriorly. "U'-shaped tears have a longitudinal limb posteriorly, often at the junction of the supraspinatus and infraspinatus in addition to the lateral detachment at the greater tuberosity. Longitudinal tears may be present in the area of the rotator interval and occasionally within a substance of the supraspinatus. "V"-shaped tears have the longitudinal component in addition to a lateral detachment. Only when tear geometry is appreciated can an effective repair be performed. I use a grasper and pull on the tear edge, attempting to determine its anatomic location. I will pull on the tendon edge while varying elevation and rotation until a "best fit" is obtained. The McConnell is then secured to maintain arm position.
Cuff Mobilization Adhesions may have formed within the subacromial space between cuff and acromion or cuff and deltoid, interfering with tendon mobilization. Adhesions to the coracoid or coracohumeral ligament contracture may falsely give the impression of irreparability. The adhesions must be released to enable mobilization. Posterior adhesions usually are not dense, and often I can release them by inserting a metal trocar and cannula through the lateral portal, placing it superior to the anterior tear edge and sweeping posteriorly directly beneath the arthroscope. Occasionally electrocautery is used to divide adhesions if they are particularly thick. I have found it unwise to attempt to remove the adhesions with a power shaver, because bleeding often will result. Anterior adhesions to the coracoid are usually ALL ARTHROSCOPIC ROTATOR CUFF REPAIRS
Acromioclavicular Joint Inferior acromioclavicular joint osteophytes are removed if necessary, as determined by preoperative radiograph or inspection at the time of surgery. The acromioclavicular joint is removed only if preoperative symptoms and signs are positive.
Repair Site Preparation The next step is preparation of the bone surface at the repair site (Fig 2). A 4-mm round bur is used to prepare a cancellous bed for the tendon. One to two millimeters of bone is removed until the cancellous bone is visible. I do not place the tendon in a trough. The site of bone preparation is based on tendon mobility. If an anatomic repair is possible, then the bone is prepared from the articular surface margin to the greater tuberosity. The length of the bone preparation site is determined by tendon tear length. The width is 1 to 2 cm. If anatomic repair is not possible without excessive tendon tension, then the tendon is repaired medially, and I adjust the bone decortication site accordingly. Anchor
Selection
No currently available suture anchor is ideal. I believe the best choices as of this writing are the ROC 5-mm anchor (Orthopedic Biosystems Ltd, Scottsdale, AZ) and the 5-mm Corkscrew anchor (Arthrex, Naples, FL). Each anchor offers relative advantages and disadvantages when compared with the others, and the surgeon should select the anchor based on personal preference. I use the ROC 5 anchor for rotator cuff repair. Suture
Selection
I use #2 braided, nonabsorbable suture of different colors for rotator cuff repair. The ROC 5 anchor has 2 sutures on each anchor, 1 dark green and the other white. This allows the surgeon to identify at a glance where each suture belongs. 295
/!
Fig 2. (A) Repair site abrasion, superior view. (B) Repair site abrasion, coronal view.
Anchor Placement The number of anchors is dependent on the length of the rotator cuff tear. For all but the smallest tears, I advise using 2 anchors. Tears involving 2 or more tendons generally require 3 to 4 anchors. 1 place the anchors lateral to the greater tuberosity, for the following reasons: (1) The anchor is placed in bone with an intact cortical surface as compared with the prepared bed of the repair site. (2) Lateral placement allows the anchors to be positioned so that the line of tendon pull is 90 ° to the direction of the anchor, which minimizes anchor pullout. (3) The tendon is repaired anatomically. As the anchors are moved medially onto the superior surface of the tuberosity, the tendon healing site also is moved medially. The anchors are placed proportionally along the length of the greater tuberosity corresponding to the length of the torn tendon, beginning anteriorly approximately 1 cm posterior to the bicipital groove and proceeding posteriorly (Fig 3). The anchors are placed through the lateral cannula and their sutures withdrawn out the anterior cannula with a crochet hook. The arm is gradually internally rotated to allow anchor insertion posteriorly. All anchors are placed in bone before any tendon suturing starts. Suture Placement and Passing Once suture anchor placement has been completed, the braided sutures must be passed through the torn tendon (Fig 4). The soft tissue grasper is passed through the lateral cannula, and the precise location for the tendon repair as well as the location and spacing of each suture is estimated. The sutures should be evenly spaced from the anterior tendon edge to its posterior margin. The sutures are placed 296
approximately 5 mm from the tendon edge. I begin suturing anteriorly and work posteriorly. The Caspari suture punch will accept only monofilament suture. I currently employ a 2-0 nylon suture, which I fold in half. The 2 free ends are passed into the suture punch, and the loop end exits from the handle. The tendon is grasped and punctured and the 2 ends advanced into the subacromial space (Fig 4). A grasper hook is used to retrieve the free ends out the anterior cannula and a hemostat applied to the suture ends. The suture punch jaws are opened and the suture punch is withdrawn out the lateral cannula. The crochet hook then is used to retrieve 1 of the sutures from the anterior cannula to thelateral cannula (Fig 5). The suture retriever is passed through the lateral cannula into the subacromial space and the nylon sutures grasped. The retriever is removed from the cannula to assure that the nylon loop and the braided suture have not become entangled. This is a critical step and should be repeated with each suture. The end of the lateral braided suture is placed within the loop of the 2-0 nylon. Traction is placed on the 2 ends of the nylon suture anteriorly, and the braided suture is pulled from the lateral cannula, into the subacromial space, through the tendon and out the anterior cannula (Fig 6A). At this point, a simple suture has been placed through the rotator cuff. This process is repeated until all of the sutures have been placed (Fig 6B). Many surgeons find suture passing with a nylon loop cumbersome and prefer to use instruments that allow the surgeon to pass the braided sutures directly through the cuff tendon. The Cuff Stitch (Orthopedic Biosystems Ltd., Scottsdale, AZ) is such a device. After the suture anchors are placed, the sutures are drawn out the lateral cannula. GARY M. GARTSMAN
Fig 3. (A) Anterior sutures withdrawn through anterior cannula. (B) Anchor angle of insertion.
//
Fig 4. Suture punch piercing anterior tendon. ALL ARTHROSCOPIC ROTATOR CUFF REPAIRS
297
Fig 5. Crochet hook pulling nylon sutures through cannula.
Fig 6. (A) Nylon loop pulling anterior suture through tendon. (B) Multiple sutures through tendon.
298
GARY M. GARTSMAN
One suture limb from the most anterior anchor is pulled back out through the anterior cannula and loaded on the Cuff Stitch. The instrument is introduced into the subacromial space from the anterior cannula and used to pierce the tendon. The suture is withdrawn out the lateral carmula with a loop forceps or crochet hook. This process is repeated until all of the sutures are passed through the tendon edge. Knot Tying
Arthroscopic knot tying can be a frustrating experience that is only overcome with practice. However, a few fundamentals may ease the process. Always make certain that no tangles exist. The suture retrieval forceps should be passed through the knot-tying cannula and used to grasp 1 suture. The suture retrieval forceps is then withdrawn out the cannula. This will separate any suture wrapping. I grasp the suture that exits from the bursal cuff surface. This bursal suture is then passed underneath the other suture limb and a surgeon's knot created with 2 half-hitches in the same direction. The bursal suture is then placed in the knot pusher, which is passed into the subacromial space. If the knot pusher end is placed 1 cm away from the knot, a 180 ° pull can be obtained and a true surgeon's knot tied. Another half-hitch is placed in the same direction, and then a half-hitch is placed in the opposite direction. The suture limb that did not pass through the tendon is then placed in the knot pusher, reversing the suture post. A half-hitch is then placed, followed by a half-hitch in the opposite direction. The knot is then cut. I prefer to use simple sutures for the rotator cuff repair. The knots are tied on the bursal surface (Fig 7), and I have not been aware of any
problems with this approach. Other surgeons may be more comfortable with mattress sutures or buried knots. My operative technique is described in more detail. 7
instruments A number of specialized arthroscopic instruments are useful in this procedure. My preferences include the Modified Caspari suture punch (Linvatec, Naples, FL), crochet hook, suture retrieval forceps, suture grasper, knot pusher, and Cuff Stitch (Orthopedic Biosystems Ltd, Scottsdale, AZ).
POSTOPERATIVE MANAGEMENT A sling maintains the arm in 15 ° of abduction, and an ice pack wrap decreases shoulder swelling and pain. Passive motions in elevation and external rotation are started the afternoon of operation and continued at home for a period of 6 weeks. Patients go home the morning after the operation. At 6 weeks, they begin active range of motion, which is continued until week 12, at which point strengthening is begun. Muscles that are strengthened with rubber tubing include the deltoid, infraspinatus, supraspinatus, scapular rotators, and the biceps. Patients continue range of motion and strengthening exercises for i year. Individuals who have a SLAP repair in addition to a rotator cuff repair follow a modified program. Because external rotation beyond 40 ° causes the SLAP repair to separate, passive external rotation beyond 40 ° is not allowed for 6 weeks. Patients with a rotator cuff repair and a Bankart repair must have their postoperative program individualized. The "safe" zone of movement that does not disrupt the
// \
Fig 7. (A) Posterior knot tied. (B) Completed repair, superior view. (C) Completed repair, coronal view. ALL ARTHROSCOPIC ROTATOR CUFF REPAIRS
299
glenohumeral reconstruction should be d o c u m e n t e d at operation. C o m m o n l y m y postoperative p r o g r a m allows passive m o t i o n 90 ° of elevation and 20 ° of external rotation immediately after operation. After week 2, this is a d v a n c e d to 120 ° of elevation and 40 ° of external rotation. At the start of w e e k 6, unlimited active elevation and external rotation are allowed.
RESULTS Our results 8,9 from arthroscopic repair of full-thickness rotator cuff tears is satisfactory and certainly equals the results reported from open repair. To s u m m a r i z e our experience: 1. At final evaluation, 90% of patients rated satisfaction with the shoulder as good to excellent. 2. Postoperative pain relief was self-assessed b y patients as good to excellent in 78%. 3. Shoulder function was i m p r o v e d in all 10 of the American Shoulder and Elbow Surgeons' activities of daily living functional evaluation, and 90% assessed shoulder function as good to excellent. Patients were particularly pleased with the i m p r o v e d ability to sleep comfortably. 4. Active range of m o t i o n was i m p r o v e d in all m e a s u r e d planes with a preoperative m e a n Constant m o t i o n score of 27.2 (out of a possible 40) and a postoperative score of 37.9. 5. Strength i m p r o v e d from 7.5 p o u n d s preoperatively to 14.0 p o u n d s postoperatively. 6. Patient self-assessed levels of general health were imp r o v e d in all scales and s u m m a r y measures of the SF-36 H e a l t h Survey, with the exception of the General Health and the Mental C o m p o n e n t Summary.
DISCUSSION Miller and Savoie 16 d o c u m e n t e d a 76% incidence of intraarticular p a t h o l o g y in patients u n d e r g o i n g mini-open repair and concluded that g l e n o h u m e r a l arthroscopic examination is useful. However, the significance of the lesions described in that report is unclear. M y experience has been that although g l e n o h u m e r a l joint abnormalities are common, occuring in 68% (137 of 200 patients) of patients with complete rotator cuff tears, most of the abnormalities are minor, and in only 12.5% (25 of 200 patients) were the findings of significance. 17 We used 3 criteria to define a major intra-articular abnormality. The lesion f o u n d (1) required surgical treatment, (2) changed the postoperative rehabilitation, or (3) altered the expected goals of the procedure. Does treatment of the intra-articular lesion make a difference? Yes! W h e n patients with full-thickness rotator cuff tears u n d e r g o i n g arthroscopic repair were analyzed, m e a n preoperative UCLA scores were 23.7 for the normal group (without intra-articular lesions) and 10.9 for the group with a major glenohumeral lesion. Postoperative UCLA scores were 31.2 for the n o r m a l group and 29.9 for the group with a major glenohumeral lesion, differences that are not statistically significant (P = .23). The compari-
300
son indicates that identification and treatment of intraarticular lesions resulted in these patients achieving results comparable to those of patients w i t h o u t intra-articular lesions. 9 Our operative treatment of full-thickness rotator cuff tears is arthroscopic. No open operations are performed. With the arthroscopic technique, we can avoid deltoid origin detachment. Tendon tear size is not a factor. We have more difficulty with chronic, retracted tears regardless of size than we do repairing mobile large or massive tears. T h o u g h w e are unable to statistically d o c u m e n t our impressions, we believe that arthroscopic repair results in imp r o v e d cosmesis, decreased postoperative pain, and more rapid gains in motion w h e n c o m p a r e d with open operative treatment of patients with similar lesions. Arthroscopic repair has particular advantages in an athletic population. Typically the rotator cuff lesions are small and amenable to arthroscopic repair. Coexisting intra-articular lesions can be identified and corrected with arthroscopic techniques. Individuals with sports injuries are particularly knowledgeable about arthroscopic techniques and seem to appreciate all efforts that can speed their return to athletics.
REFERENCES 1. Altchek DW, Warren RF, WickiewiczTL, et al: Arthroscopic acromioplasty: Techniqueand results. J BoneJoint Surg Am 72:1198-1207,1990 2. Ellman H: Arthroscopic subacromial decompression:Analysis of oneto three-year results. Arthroscopy 3:173-181,1987 3. Gartsman GM: Arthroscopic acromioplasty for lesions of the rotator cuff.J BoneJoint Surg Am 72:169-180,1990 4. Ellman H: Diagnosis and treatment of incomplete rotator cuff tears. Clin Orthop 64-74, 1990 5. EschJC: Arthroscopic subacromial decompression: Results according to the degree of rotator cuff tear. Arthroscopy 4:241-249,1988 6. Gartsman GM, Mihle J: Partial articular surface tears of the rotator cuff.J Shoulder Elbow Surg 4:409-416,1995 7. Gartsman GM, Hammerman SM: Full-thickness tear: Arthroscopic repair. Orthop Clin North Am 28:83-98,1997 8. Gartsman GM, Hammerman SM: Arthroscopic repair of fullthickness rotator cuff tears: Operative technique. Op Tech Orthop 8:226-235, 1998 9. Gartsman GM, Khan M, Hammerman SM: Arthroscopic repair of full-thickness rotator cuff tears. J BoneJoint Surg Am 80:832-840,1998 10. Gartsman GM, Brinker MR, Khan M: Early effectiveness of arthroscopic repair for full-thickness tears of the rotator cuff--An outcome analysis. J BoneJoint Surg Am 80:33-40,1998 11. Gartsman G: Rotator cuff repair: Why I prefer arthroscopy. Sports Med Arth Rev 7:85-92,1999 12. Gazielly DF, Gleyze P, Montagnon C, et al: Arthroscopic repair of distal supraspinatus tears with Revo Screw and permanent mattress sutures: A preliminary report. Conference Proceeding, Amelia Island, FL, 1996 13. Snyder S: Partial thickness rotator cuff tears: Results of arthroscopic treatment. Arthroscopy 7:1-7, 1991 14. Ellman I-{,Gartsman GM (eds): Arthroscopic Shoulder Surgery and Related Procedures. Philadelphia, PA, Lea & Febiger, 1993 15. Gartsman GM: Arthroscopic assessment of rotator cuff reparability. Arthroscopy 12:546-549,1996 16. Miller C, Savoie FH: Glenohumeral abnormalities associated with full thickness tears of the rotator cuff. Orthop Rev 159-162,1994 17. Gartsman GM, Taverna E: The incidence of glenohumeral joint abnormalities associated with full-thickness, reparable rotator cuff tears. Arthroscopy 13:450-455,1997
GARY M. GARTSMAN
The arthroscopic operation for repair of full-thickness rotator cuff tears is successful and has the advantages of glenohumeral joint inspection, treatment of intra-articular lesions, smaller incisions, no deltoid detachment, less soft tissue dissection, less pain, and more rapid rehabilitation. However, these advantages must be balanced against the technical difficulty of the method, which limits its application to surgeons skilled in both open and arthroscopic shoulder operations. This article contains many technical pearls to, as much as is possible, simplify and improve all arthroscopic cuff repair. KEY WORDS: Arthroscopic rotator cuff repair
The role of the arthroscope in the management of rotator cuff lesions is evolving. In the last decade, shoulder arthroscopy had advanced from its use as a diagnostic tool to an effective treatment option for stage 2 impingement and acromioclavicular joint arthritis. >3Arthroscopic decompression has also been described in the management of more severe rotator cuff lesions such as partial-thickness tears. 4-6 Recently, repair of full-thickness rotator cuff tears by use of arthroscopic technique has been described. 7q3 The advantages of arthroscopic repair are that it allows smaller skin incisions, glenohumeral joint inspection, treatment of intra-articular lesions, no deltoid detachment, less soft tissue dissection, less pain, and more rapid rehabilitation. Arthroscopic repair of full-thickness rotator cuff tears has certain specific advantages for patients who present with athletic injur~Les. Commonly these individuals are younger, more athletic, present earlier, and have smaller tendon tears. These qualities make this patient population ideally suited for arthroscopic rotator cuff repair. Perhaps most importantly, arthroscopy allows the orthopedist to inspect and repair concomitant intra-articular lesions that occur frequently. Sometimes lesions in inaccessible locations, or in younger patients, for example, a patient with posterior cuff tear and an anterior Bankart lesion, would be difficult to manage through a single open incision.
INDICATIONS The indications for arthroscopic rotator cuff repair are identical to those for an open repair as discussed elsewhere in this section and should not in any way be altered or "broadened" in the mistaken view that arthroscopic repair is a lesser procedure. Although the skin incisions may be smaller and the deltoid is left attached, all elements of an open repair are performed arthroscopically, and patients From the Department of Orthopaedic Surgery, University of Texas
Houston Health Science Center, and the Fondren Orthopedic Group, Texas Orthopedic Hospital, Houston, TX. Address reprint requests to Gary M. Gartsman, MD, Fondren Orthopedic Group, Texas Orthopedic Hospital, Houston, TX. Copyright© 2000 by W.B. Saunders Company 1060~1872/00/0804-0006510.00/0 doi:10.1053/otsm.2000.17779
who are unable to tolerate either the surgery or the postoperative rehabilitation of an open procedure are not candidates for an arthroscopic rotator cuff repair. Surgery is performed for persistent pain that interferes with activities of daily living, work, or sports, and is unresponsive to a 6- to 12-month period of nonoperative care.
OPERATIVE
TECHNIQUE
Positioning I prefer the patient sitting, because easy access is afforded to the anterior, lateral, and posterior aspects of the shoulder. It is easier for most surgeons to convert to an open repair when the patient is in this position. Arm positioning during repair is greatly facilitated by the use of a McConnell arm holder (McConnell Co, Greenville, TX). Portals
Posterior. As the complexity of shoulder arthroscopy increases, precise portal placement becomes more critical. I prefer to mark the outlines of the acromion and lateral clavicle. Three portals are used (Fig 1). The posterior portal is 1.5 cm medial and 1.5 cm inferior to the posterolateral acromial border. This portal is made superior to the traditional point of entry and has an important advantage in that the arthroscope enters the subacromial space parallel to the acromial undersurface, allowing the surgeon to "look down" at the rotator cuff tear from a distance, improving perspective. Lateral. The lateral portal is made 10 to 15 m m posterior to the anterior acromial border and approximately 2 to 4 cm lateral to the acromion border. The axillary nerve can be in jeopardy if the lateral portal is placed more than 5 cm lateral to the acromial border (Fig 1). The lateral portal should allow the cannula to enter midway between the humeral head and the acromion. I identify the precise location of the lateral portal with a spinal needle bisecting the distance between the anterior and posterior edges of the tear.
Operative Techniques in Sports Medicine, Vol 8, No 4 (October), 2000: pp 293-300
293
\
\ \
/ t
Fig 1, (A) Portal sites. (B) Cannulas in position.
Anterior. The anterior portal is located approximately 2 to 3 cm anterior to the acromial border midway between the anterolateral border and the AC joint.
Inspection of Glenohumeral Joint The glenohumeral joint is entered and inspected. Because the most common location for partial-thickness tears is in the anterior portion of the supraspinatus, the lesion is well visualized. Shoulder rotation and abduction are then altered so that other areas of cuff tearing are clearly visualized. Frequently the tear will be overlooked unless the arm is placed in 30 ° to 40 ° abduction and 20 ° external rotation. Particular attention should be directed to the attachment and integrity of the glenohumeral ligaments and the biceps-labrum complex. The labrum should be carefully examined for signs of glenohumeral instability. If the posterior aspect of the shoulder cannot be seen clearly, then viewing should be performed from the anterior portal.
PARTIAL-THICKNESS ROTATOR CUFF TEARS With the advent and increased use of shoulder arthroscopy, partial articular surface tears have been frequently described. 5,6,13 The articular surface of the rotator cuff is extremely well visualized arthroscopically, and tears can be precisely measured. I use Ellman and Gartsman's
294
classification system. 14 They classified partial-thickness tears according to the depth of the lesion and the anatomic site. Tears are noted to be present on the articular surface or the bursal surface. Grade 1 lesions show definite disruption of the tendon fibers but are less than one fourth of the tendon (<3 mm) thickness. Grade 2 lesions are less than half of the tendon thickness and are 3 to 6 m m in depth. Grade 3 lesions are more than half the tendon thickness and are more than 6 m m in depth. If the margins and depth of the tear are not well visualized, then a small synovial resector is introduced through the anterior portal and the tear debrided. The decision process in the management of these lesions is complex, and the following are general guidelines. Lower-grade lesions and smaller rotator cuff tears may be treated with decompression, whereas more advanced lesions require repair. Certain bone shapes or changes are consistent with extrinsic tendon compression. Type 3 or hooked acromion, anterior acromial spurs, and inferior acromioclavicular joint osteophytes are radiographic findings that suggest that the patient will respond to operative decompression, whereas patients with benign bone anatomy more likely have intrinsic tendon lesions and require tendon repair. If tendon repair is needed, the area of partial tear is marked. With the tear in direct view, a spinal needle is introduced percutaneously into the center of the tear. A #1 PDS suture (Ethicon, Boston, MA) is passed down the GARY M. GARTSMAN
spinal needle into the joint. The needle is withdrawn; the arthroscope is removed from the joint and inserted into the subacromial space. The blue suture is identified and removed. A motorized shaver is used to convert the tear into a full-thickness lesion that is then repaired arthroscopically, as follows.
very thick and require resection with the electrocautery. This is particularly true in the area of the coracohumeral ligament. I have not found intra-articular division of the joint capsule to be helpful. Remember that tendon reparability is based not only on tendon mobility but also the integrity of the tissue to hold sutures and thus be used in the repair. 15
FULL-THICKNESS ROTATOR CUFF TEARS
Acromioplasty
Visualization
If the tear is reparable, I then evaluate the acromion. The purpose of acromioplasty is to create adequate space for the rotator cuff tendons. If the acromion compromises the space for the cuff or if there is clear evidence of impingement (coracoacromial ligament or acromial undersurface fraying), I perform an acromioplasty. Because the thickness and shape of the acromion varj6 the amount of bone removed during acromioplasty also varies. My goal is to achieve a flat acromial undersurface. If the acromion is flat and there is adequate space for the cuff, then acromioplasty is not performed.
After inspection of the glenohumeral joint, the arthroscope is removed and redirected into the subacromial space. ! establish a lateral portal and use a motorized shaver to remove any bursal tissue that impedes a clear view of the tendon tear. Surprisingly, the subacromial space is easily visualized in full-thickness tears. The thick subacromial bursitis that exists in stage 2 impingement is usually absent.
Tear Classification The arthroscope is then rotated so that it is pointed directly down at the rotator cuff tear. With small to medium tears, the size and geomefry of the tear is easily appreciated. The length of the tear from anterior to posterior as well as the amount of medial retraction is noted. Straight medial retraction or retraction in an elliptical shape are the most common findings. As tear size increases, the surgeon's ability to appreciate tear geometry becomes more difficult. Reverse "U' tears with a longitudinal component along the rotator interval will allow the tear to rotate posteriorly. "U'-shaped tears have a longitudinal limb posteriorly, often at the junction of the supraspinatus and infraspinatus in addition to the lateral detachment at the greater tuberosity. Longitudinal tears may be present in the area of the rotator interval and occasionally within a substance of the supraspinatus. "V"-shaped tears have the longitudinal component in addition to a lateral detachment. Only when tear geometry is appreciated can an effective repair be performed. I use a grasper and pull on the tear edge, attempting to determine its anatomic location. I will pull on the tendon edge while varying elevation and rotation until a "best fit" is obtained. The McConnell is then secured to maintain arm position.
Cuff Mobilization Adhesions may have formed within the subacromial space between cuff and acromion or cuff and deltoid, interfering with tendon mobilization. Adhesions to the coracoid or coracohumeral ligament contracture may falsely give the impression of irreparability. The adhesions must be released to enable mobilization. Posterior adhesions usually are not dense, and often I can release them by inserting a metal trocar and cannula through the lateral portal, placing it superior to the anterior tear edge and sweeping posteriorly directly beneath the arthroscope. Occasionally electrocautery is used to divide adhesions if they are particularly thick. I have found it unwise to attempt to remove the adhesions with a power shaver, because bleeding often will result. Anterior adhesions to the coracoid are usually ALL ARTHROSCOPIC ROTATOR CUFF REPAIRS
Acromioclavicular Joint Inferior acromioclavicular joint osteophytes are removed if necessary, as determined by preoperative radiograph or inspection at the time of surgery. The acromioclavicular joint is removed only if preoperative symptoms and signs are positive.
Repair Site Preparation The next step is preparation of the bone surface at the repair site (Fig 2). A 4-mm round bur is used to prepare a cancellous bed for the tendon. One to two millimeters of bone is removed until the cancellous bone is visible. I do not place the tendon in a trough. The site of bone preparation is based on tendon mobility. If an anatomic repair is possible, then the bone is prepared from the articular surface margin to the greater tuberosity. The length of the bone preparation site is determined by tendon tear length. The width is 1 to 2 cm. If anatomic repair is not possible without excessive tendon tension, then the tendon is repaired medially, and I adjust the bone decortication site accordingly. Anchor
Selection
No currently available suture anchor is ideal. I believe the best choices as of this writing are the ROC 5-mm anchor (Orthopedic Biosystems Ltd, Scottsdale, AZ) and the 5-mm Corkscrew anchor (Arthrex, Naples, FL). Each anchor offers relative advantages and disadvantages when compared with the others, and the surgeon should select the anchor based on personal preference. I use the ROC 5 anchor for rotator cuff repair. Suture
Selection
I use #2 braided, nonabsorbable suture of different colors for rotator cuff repair. The ROC 5 anchor has 2 sutures on each anchor, 1 dark green and the other white. This allows the surgeon to identify at a glance where each suture belongs. 295
/!
Fig 2. (A) Repair site abrasion, superior view. (B) Repair site abrasion, coronal view.
Anchor Placement The number of anchors is dependent on the length of the rotator cuff tear. For all but the smallest tears, I advise using 2 anchors. Tears involving 2 or more tendons generally require 3 to 4 anchors. 1 place the anchors lateral to the greater tuberosity, for the following reasons: (1) The anchor is placed in bone with an intact cortical surface as compared with the prepared bed of the repair site. (2) Lateral placement allows the anchors to be positioned so that the line of tendon pull is 90 ° to the direction of the anchor, which minimizes anchor pullout. (3) The tendon is repaired anatomically. As the anchors are moved medially onto the superior surface of the tuberosity, the tendon healing site also is moved medially. The anchors are placed proportionally along the length of the greater tuberosity corresponding to the length of the torn tendon, beginning anteriorly approximately 1 cm posterior to the bicipital groove and proceeding posteriorly (Fig 3). The anchors are placed through the lateral cannula and their sutures withdrawn out the anterior cannula with a crochet hook. The arm is gradually internally rotated to allow anchor insertion posteriorly. All anchors are placed in bone before any tendon suturing starts. Suture Placement and Passing Once suture anchor placement has been completed, the braided sutures must be passed through the torn tendon (Fig 4). The soft tissue grasper is passed through the lateral cannula, and the precise location for the tendon repair as well as the location and spacing of each suture is estimated. The sutures should be evenly spaced from the anterior tendon edge to its posterior margin. The sutures are placed 296
approximately 5 mm from the tendon edge. I begin suturing anteriorly and work posteriorly. The Caspari suture punch will accept only monofilament suture. I currently employ a 2-0 nylon suture, which I fold in half. The 2 free ends are passed into the suture punch, and the loop end exits from the handle. The tendon is grasped and punctured and the 2 ends advanced into the subacromial space (Fig 4). A grasper hook is used to retrieve the free ends out the anterior cannula and a hemostat applied to the suture ends. The suture punch jaws are opened and the suture punch is withdrawn out the lateral cannula. The crochet hook then is used to retrieve 1 of the sutures from the anterior cannula to thelateral cannula (Fig 5). The suture retriever is passed through the lateral cannula into the subacromial space and the nylon sutures grasped. The retriever is removed from the cannula to assure that the nylon loop and the braided suture have not become entangled. This is a critical step and should be repeated with each suture. The end of the lateral braided suture is placed within the loop of the 2-0 nylon. Traction is placed on the 2 ends of the nylon suture anteriorly, and the braided suture is pulled from the lateral cannula, into the subacromial space, through the tendon and out the anterior cannula (Fig 6A). At this point, a simple suture has been placed through the rotator cuff. This process is repeated until all of the sutures have been placed (Fig 6B). Many surgeons find suture passing with a nylon loop cumbersome and prefer to use instruments that allow the surgeon to pass the braided sutures directly through the cuff tendon. The Cuff Stitch (Orthopedic Biosystems Ltd., Scottsdale, AZ) is such a device. After the suture anchors are placed, the sutures are drawn out the lateral cannula. GARY M. GARTSMAN
Fig 3. (A) Anterior sutures withdrawn through anterior cannula. (B) Anchor angle of insertion.
//
Fig 4. Suture punch piercing anterior tendon. ALL ARTHROSCOPIC ROTATOR CUFF REPAIRS
297
Fig 5. Crochet hook pulling nylon sutures through cannula.
Fig 6. (A) Nylon loop pulling anterior suture through tendon. (B) Multiple sutures through tendon.
298
GARY M. GARTSMAN
One suture limb from the most anterior anchor is pulled back out through the anterior cannula and loaded on the Cuff Stitch. The instrument is introduced into the subacromial space from the anterior cannula and used to pierce the tendon. The suture is withdrawn out the lateral carmula with a loop forceps or crochet hook. This process is repeated until all of the sutures are passed through the tendon edge. Knot Tying
Arthroscopic knot tying can be a frustrating experience that is only overcome with practice. However, a few fundamentals may ease the process. Always make certain that no tangles exist. The suture retrieval forceps should be passed through the knot-tying cannula and used to grasp 1 suture. The suture retrieval forceps is then withdrawn out the cannula. This will separate any suture wrapping. I grasp the suture that exits from the bursal cuff surface. This bursal suture is then passed underneath the other suture limb and a surgeon's knot created with 2 half-hitches in the same direction. The bursal suture is then placed in the knot pusher, which is passed into the subacromial space. If the knot pusher end is placed 1 cm away from the knot, a 180 ° pull can be obtained and a true surgeon's knot tied. Another half-hitch is placed in the same direction, and then a half-hitch is placed in the opposite direction. The suture limb that did not pass through the tendon is then placed in the knot pusher, reversing the suture post. A half-hitch is then placed, followed by a half-hitch in the opposite direction. The knot is then cut. I prefer to use simple sutures for the rotator cuff repair. The knots are tied on the bursal surface (Fig 7), and I have not been aware of any
problems with this approach. Other surgeons may be more comfortable with mattress sutures or buried knots. My operative technique is described in more detail. 7
instruments A number of specialized arthroscopic instruments are useful in this procedure. My preferences include the Modified Caspari suture punch (Linvatec, Naples, FL), crochet hook, suture retrieval forceps, suture grasper, knot pusher, and Cuff Stitch (Orthopedic Biosystems Ltd, Scottsdale, AZ).
POSTOPERATIVE MANAGEMENT A sling maintains the arm in 15 ° of abduction, and an ice pack wrap decreases shoulder swelling and pain. Passive motions in elevation and external rotation are started the afternoon of operation and continued at home for a period of 6 weeks. Patients go home the morning after the operation. At 6 weeks, they begin active range of motion, which is continued until week 12, at which point strengthening is begun. Muscles that are strengthened with rubber tubing include the deltoid, infraspinatus, supraspinatus, scapular rotators, and the biceps. Patients continue range of motion and strengthening exercises for i year. Individuals who have a SLAP repair in addition to a rotator cuff repair follow a modified program. Because external rotation beyond 40 ° causes the SLAP repair to separate, passive external rotation beyond 40 ° is not allowed for 6 weeks. Patients with a rotator cuff repair and a Bankart repair must have their postoperative program individualized. The "safe" zone of movement that does not disrupt the
// \
Fig 7. (A) Posterior knot tied. (B) Completed repair, superior view. (C) Completed repair, coronal view. ALL ARTHROSCOPIC ROTATOR CUFF REPAIRS
299
glenohumeral reconstruction should be d o c u m e n t e d at operation. C o m m o n l y m y postoperative p r o g r a m allows passive m o t i o n 90 ° of elevation and 20 ° of external rotation immediately after operation. After week 2, this is a d v a n c e d to 120 ° of elevation and 40 ° of external rotation. At the start of w e e k 6, unlimited active elevation and external rotation are allowed.
RESULTS Our results 8,9 from arthroscopic repair of full-thickness rotator cuff tears is satisfactory and certainly equals the results reported from open repair. To s u m m a r i z e our experience: 1. At final evaluation, 90% of patients rated satisfaction with the shoulder as good to excellent. 2. Postoperative pain relief was self-assessed b y patients as good to excellent in 78%. 3. Shoulder function was i m p r o v e d in all 10 of the American Shoulder and Elbow Surgeons' activities of daily living functional evaluation, and 90% assessed shoulder function as good to excellent. Patients were particularly pleased with the i m p r o v e d ability to sleep comfortably. 4. Active range of m o t i o n was i m p r o v e d in all m e a s u r e d planes with a preoperative m e a n Constant m o t i o n score of 27.2 (out of a possible 40) and a postoperative score of 37.9. 5. Strength i m p r o v e d from 7.5 p o u n d s preoperatively to 14.0 p o u n d s postoperatively. 6. Patient self-assessed levels of general health were imp r o v e d in all scales and s u m m a r y measures of the SF-36 H e a l t h Survey, with the exception of the General Health and the Mental C o m p o n e n t Summary.
DISCUSSION Miller and Savoie 16 d o c u m e n t e d a 76% incidence of intraarticular p a t h o l o g y in patients u n d e r g o i n g mini-open repair and concluded that g l e n o h u m e r a l arthroscopic examination is useful. However, the significance of the lesions described in that report is unclear. M y experience has been that although g l e n o h u m e r a l joint abnormalities are common, occuring in 68% (137 of 200 patients) of patients with complete rotator cuff tears, most of the abnormalities are minor, and in only 12.5% (25 of 200 patients) were the findings of significance. 17 We used 3 criteria to define a major intra-articular abnormality. The lesion f o u n d (1) required surgical treatment, (2) changed the postoperative rehabilitation, or (3) altered the expected goals of the procedure. Does treatment of the intra-articular lesion make a difference? Yes! W h e n patients with full-thickness rotator cuff tears u n d e r g o i n g arthroscopic repair were analyzed, m e a n preoperative UCLA scores were 23.7 for the normal group (without intra-articular lesions) and 10.9 for the group with a major glenohumeral lesion. Postoperative UCLA scores were 31.2 for the n o r m a l group and 29.9 for the group with a major glenohumeral lesion, differences that are not statistically significant (P = .23). The compari-
300
son indicates that identification and treatment of intraarticular lesions resulted in these patients achieving results comparable to those of patients w i t h o u t intra-articular lesions. 9 Our operative treatment of full-thickness rotator cuff tears is arthroscopic. No open operations are performed. With the arthroscopic technique, we can avoid deltoid origin detachment. Tendon tear size is not a factor. We have more difficulty with chronic, retracted tears regardless of size than we do repairing mobile large or massive tears. T h o u g h w e are unable to statistically d o c u m e n t our impressions, we believe that arthroscopic repair results in imp r o v e d cosmesis, decreased postoperative pain, and more rapid gains in motion w h e n c o m p a r e d with open operative treatment of patients with similar lesions. Arthroscopic repair has particular advantages in an athletic population. Typically the rotator cuff lesions are small and amenable to arthroscopic repair. Coexisting intra-articular lesions can be identified and corrected with arthroscopic techniques. Individuals with sports injuries are particularly knowledgeable about arthroscopic techniques and seem to appreciate all efforts that can speed their return to athletics.
REFERENCES 1. Altchek DW, Warren RF, WickiewiczTL, et al: Arthroscopic acromioplasty: Techniqueand results. J BoneJoint Surg Am 72:1198-1207,1990 2. Ellman H: Arthroscopic subacromial decompression:Analysis of oneto three-year results. Arthroscopy 3:173-181,1987 3. Gartsman GM: Arthroscopic acromioplasty for lesions of the rotator cuff.J BoneJoint Surg Am 72:169-180,1990 4. Ellman H: Diagnosis and treatment of incomplete rotator cuff tears. Clin Orthop 64-74, 1990 5. EschJC: Arthroscopic subacromial decompression: Results according to the degree of rotator cuff tear. Arthroscopy 4:241-249,1988 6. Gartsman GM, Mihle J: Partial articular surface tears of the rotator cuff.J Shoulder Elbow Surg 4:409-416,1995 7. Gartsman GM, Hammerman SM: Full-thickness tear: Arthroscopic repair. Orthop Clin North Am 28:83-98,1997 8. Gartsman GM, Hammerman SM: Arthroscopic repair of fullthickness rotator cuff tears: Operative technique. Op Tech Orthop 8:226-235, 1998 9. Gartsman GM, Khan M, Hammerman SM: Arthroscopic repair of full-thickness rotator cuff tears. J BoneJoint Surg Am 80:832-840,1998 10. Gartsman GM, Brinker MR, Khan M: Early effectiveness of arthroscopic repair for full-thickness tears of the rotator cuff--An outcome analysis. J BoneJoint Surg Am 80:33-40,1998 11. Gartsman G: Rotator cuff repair: Why I prefer arthroscopy. Sports Med Arth Rev 7:85-92,1999 12. Gazielly DF, Gleyze P, Montagnon C, et al: Arthroscopic repair of distal supraspinatus tears with Revo Screw and permanent mattress sutures: A preliminary report. Conference Proceeding, Amelia Island, FL, 1996 13. Snyder S: Partial thickness rotator cuff tears: Results of arthroscopic treatment. Arthroscopy 7:1-7, 1991 14. Ellman I-{,Gartsman GM (eds): Arthroscopic Shoulder Surgery and Related Procedures. Philadelphia, PA, Lea & Febiger, 1993 15. Gartsman GM: Arthroscopic assessment of rotator cuff reparability. Arthroscopy 12:546-549,1996 16. Miller C, Savoie FH: Glenohumeral abnormalities associated with full thickness tears of the rotator cuff. Orthop Rev 159-162,1994 17. Gartsman GM, Taverna E: The incidence of glenohumeral joint abnormalities associated with full-thickness, reparable rotator cuff tears. Arthroscopy 13:450-455,1997
GARY M. GARTSMAN