158 ABSTRACTS

throscope. Two patients had loose bodies in the gle- nohumeral joint. Five patients had full-thickness ro- tator cuff tears that were either unsuspected (nega- tive arthrogram) or larger than suspected by arthrography. Twenty-six patients demonstrated la- brum fraying or tears, five of which we termed “si- lent subluxators.” Most of the shoulders with inter- articular pathology would have been undiagnosed by open subacromial decompression. The strong correlation between shoulder instability and im- pingement syndrome was unexpected. In reviewing our results, it appears that arthroscopic subacro- mial decompression can be a successful alternative to open decompression. The key to success for closed decompression is related to (a) accurate di- agnosis, (b) selective treatment, (c) adequate bone resection when required, and (d) repair of full- thickness cuff tears. Postoperative rehabilitation is critical.

Ankle Arthroscopy Complications. F. Alan Barber, James Click, and Bradley T. Britt. Plano, Texas, U.S.A.

In a recent AANA survey of almost 400,000 ar- throscopies, almost 4,500 were of the ankle (>I%). In this same study, 2,215 complications were re- ported, but only one of these (an infection) was re- lated to the ankle. No one reported nerve injury, vascular injury, or reflex sympathetic dystrophy as- sociated with ankle arthroscopy. Ankle arthroscopy is a challenging technique. The anatomy of the an- kle is complex and many structures can be damaged by imprecise portal placement. A review of our an- kle arthroscopy experience found 4 complications in 30 ankle arthroscopies, representing a complica- tion rate of 13%. There were two dorsal sensory nerve injuries, one wound infection, and one syno- vial fistula. A literature review found reports of other complications including instrument breakage, synovitis, and fibula fracture. Although ankle ar- throscopy is a very useful technique, the relation- ship of the five common ankle portals to adjacent anatomical structures must be well understood. Ad- ditionally, the significant potential for and true in- cidence of complications associated with this tech- nique is not presently reflected in the literature.

Caveat Arthroscopes (Revisited). Kenneth M. Yaw, Henry J. Mankin, and John B. McGinty. Charles- ton, South Carolina, U.S.A.

In 1983 Joyce and Mankin reported on 12 patients during a 4-year period who had undergone diagnos-

Arthroscopy, Vol. 5, No. 2, 1989

tic or therapeutic knee arthroscopy before recogni- tion that their pathology was extraarticular. Since that time, the Massachusetts General Hospital Or- thopaedic Oncology Unit has seen an additional 21 patients who have had arthroscopy before referral for an extraarticular problem because their lesions were either not detected or were not appreciated as the source of their clinical symptoms. This paper is a retrospective review of all 33 cases for the pur- pose of analyzing a potential pitfall of arthroscopy so that such occurrences can be avoided. These cases include 27 malignant tumors, 5 benign tu- mors, and 1 infection. Two of the malignant tumors and none of the benign tumors were of soft-tissue origin. Errors that occurred were failure to obtain adequate (or in some cases any) radiographs before arthroscopy, failure to detect lesions visible on pre- operative radiographs, failure to appreciate recog- nized extraarticular lesions as the source of articu- lar symptoms, and transsynovial arthroscopic bi- opsy of extraarticular lesions, thereby potentially seeding the knee joint with tumor. Arthroscopy is an invasive surgical procedure with potentially se- rious (albeit unlikely) complications. When it is un- dertaken without adequate prior history, examina- tion, radiographs, and laboratory studies, there is an additional danger of delaying and possibly com- promising future treatment of a condition for which arthroscopy is inappropriate.

Arthroscopic Anterior Cruciate Ligament Recon- struction-comparison of Results Following Semi- tendinosus and Gracilis Vs. Patellar Tendon Au- tografts. Angelo L. Otero, Robert H. Schmidt, and Kathy L. McDermott. Forth Worth, Texas, U.S.A.

Reconstruction of the anterior cruciate ligament (ACL) is being performed by many surgeons using many different techniques and a variety of grafts. The most important factors in the success of an ACL reconstruction is the selection of a graft with enough strength to withstand joint stresses, isomet- ric placement of the graft, and secure fixation of the graft after it has been placed. This study was con- ducted to analyze the success of arthroscopic ACL reconstruction using one of three grafts: single- strand semitendinosus and gracilis (ST&G) au- tograft (n = 10, follow-up time = 34 months), dou- bled ST&G autograft (n = 41, follow-up time = 25 months), and central third patellar tendon (PAT) autograft (n = 24, follow-up time = 14 months). Due to the difference in type of grafts used, fixation

ABSTRACTS 159

was different. ST&G (single strand and doubled) were secured with suture from a Krachow stitch at each end of the graft tied to a screw. The PAT grafts each had a 2-cm bone block at each end and were fixed with a Kurosaka screw. Rehabilitation was also slightly altered between these two types of grafts. ST&G grafts were immobilized for 8 weeks with passive range-of-motion exercises at 4 weeks. The patellar tendon grafts are also immobilized for 6 weeks, but passive range-of-motion exercises out- side of the brace are instituted at 1 week because the method of fixation is more secure for these grafts. PAT graft patients must work harder to get full extension, but they also tend to be much more stable than those with the ST&G grafts. The single- strand ST&G grafts have a 60% rate of success, the doubled ST&G grafts have a higher success rate (93%), but also tend to stretch out. Although follow- up is short, only one PAT graft has failed (96%). Based on our experience, we believe the PAT au- tograft is the gold standard in ACL reconstruction today.

The Medial Approach for Elbow Arthroscopy. Thom- as N. Lindenfeld. Cincinatti, Ohio, U.S.A.

We undertook a cadaveric dissection study to confirm the hypothesis that beginning with the an- terior medial portal in elbow arthroscopy is safer than starting with the anterior radial portal.

In six cadaveric elbows, we distended the capsule with saline. Both anterior medial and anterior radial approaches were made with the elbow flexed 90”. Four and one-half millimeter arthroscopic sheaths with inserted oblurators were then left in place while the saline was drained; expanding polyure- thane foam was used to distend the capsule. We allowed the polyurethane foam to harden, then dis- sected all elbows, with special attention given to exposure of the radial and medial nerve, and bra- chial artery.

The hardened foam allowed for continued capsu- lar distention during these dissections and recreated normal distances from instrument portals to neuro- vascular portals. We then measured the minimum distances from the path of the arthroscopic sheath to large neurovascular structures. The distance from the medial portal to the nearest neurovascular structure (median nerve, brachial artery) averaged 23 mm. The distance from the radial portal to the nearest neurovascular structure (radial nerve) aver-

aged 3 mm. The ulnar nerve averaged a 25mm clearance from the medial portal. Even when the medial portal was made by an incorrect method, the minimum clearance to the median nerve averaged 11 mm.

The most frequently recommended current stan- dard technique for elbow arthroscopy involves be- ginning with an anterior radial portal. Based on our findings, we suggest, however, that an anterior me- dial portal is a superior starting point. The medial portal allows good visualization of the joint, and helps with safe and accurate placement of the radial portal. Most importantly, the medial portal may be placed a safe distance from important neurovascu- lar bundles.

Intramuscular Compartment Pressures During Shoulder Arthroscopy. Lester Cohn, Yu Fon Lee, and S. Michael Tooke. Sepulveda, California, U.S.A.

A large amount of fluid extravasation into the sur- rounding muscle and soft tissue may occur during arthroscopy of the shoulder and particularly of the subacromial space. The tense swollen appearance of the shoulder has caused concern that the intra- muscular compartment pressures may become ele- vated to dangerous levels. Therefore, the intramus- cular shoulder compartment pressures were moni- tored.

The intramuscular compartment pressures were continuously measured in 24 consecutive arthro- scopic procedures of the shoulder or subacromial space. A slit catheter was placed in the deltoid mus- cle compartment.

Shoulders with only an intraarticular procedure had minimal elevation of the intramuscular com- partment pressures. There was a transient elevation of the pressures (mean 38 mm Hg) when both a shoulder arthroscopy and subacromial decompres- sion were done. Additionally, the pressures were higher with the use of an infusion pump (mean 71 mm Hg). The intramuscular compartment pressures returned to baseline levels within 10-30 minutes af- ter the termination of the surgical procedure al- though the tense clinical swelling remained for up to 4 h postoperatively.

Intramuscular compartment pressures became transiently elevated during arthroscopy of the shoulder and especially the subacromial space. However, the pressure elevations are sustained only as long as fluid infusion occurs.

Arthroscopy, Vol. S, No. 2. 1989