Author's Reply

Abstract

Many thanks to Dr. O'Driscoll for his interest and critical appraisal of our article.1Rapley J.H. Beavis R.C. Barber F.A. Glenohumeral chondrolysis after shoulder arthroscopy associated with continuous bupivacaine infusion.Arthroscopy. 2009; 25: 1367-1373Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar We certainly share his concern about the issue of mechanical damage to articular cartilage and the potentially disastrous implications for joint health. Indeed, we have published on this phenomenon related to sutures and suture anchors in the past.2Barber F.A. Biodegradable suture anchors have unique modes of failure.Arthroscopy. 2007; 23: 316-320Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar Although other etiologic agents such as gentian violet, chlorhexidine, and thermal treatment have been implicated with glenohumeral chondrolysis,3Good C.R. Shindle M.K. Kelly B.T. Wanich T. Warren R.F. Glenohumeral chondrolysis after shoulder arthroscopy with thermal capsulorrhaphy.Arthroscopy. 2007; 23: 797.e1-797.e5www.arthroscopyjournal.orgAbstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar, 4Lubowitz J.H. Poehling G.G. Glenohumeral thermal capsulorrhaphy is not recommended—Shoulder chondrolysis requires additional research.Arthroscopy. 2007; 23: 687Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar none of these was used in any of the cases in our report. It should also be noted that the subacromial space is apparently privileged,5Busfield B.T. Romero D.M. Pain pump use after shoulder arthroscopy as a cause of glenohumeral chondrolysis.Arthroscopy. 2009; 25: 647-652Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar because the subacromial use of a bupivacaine-instilling catheter is not associated with this condition. It remains our belief that the primary etiologic factor responsible for our cases of dramatic glenohumeral joint destruction was the chondrotoxic effect of local anesthetic eluted from high-flow intra-articular pumps for a time period greater than 48 hours. Other authors have made this association in basic science and clinical research.6Gomoll A.H. Kang R.W. Williams J.M. et al.Chondrolysis after continuous intra-articular bupivacaine infusion: An experimental model investigating chondrotoxicity in the rabbit shoulder.Arthroscopy. 2006; 22: 813-819Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar, 7Hansen B.P. Beck C.L. Beck E.P. et al.Postarthroscopic glenohumeral chondrolysis.Am J Sports Med. 2007; 35: 1628-1634Crossref PubMed Scopus (162) Google Scholar Our assertion that “[t]he pain pump and its catheter do not mechanically cause this problem” is supported by the absence of chondrolysis in the other patients who had intra-articular catheters but with low-flow rates and a 48-hour exposure. The current literature suggests that the acidity of the infusion fluid, the rate and volume of that infusion, and the duration of the articular cartilage exposure to this fluid have a synergistic effect and result in the development of articular cartilage damage. Additional reasons for our belief that an intra-articular catheter is not the cause of the observed global chondral damage are as follows:1Throughout the brief period during which the catheter remains intra-articular, the patient is fully immobilized and not allowed to move the shoulder. As a result, a limited (2- to 3-day) exposure in a fully immobilized joint cannot be compared with the prolonged exposure to proud suture anchors, loose sutures, or loose bodies that result in mechanical damage during rehabilitation and the return to full activity.2The customary placement of the catheter in the anterior shoulder would limit any hypothetical mechanical damage to the areas where that catheter could reach, and thus the damage would not be global as was observed.3The concept that a localized, mechanically induced articular cartilage lesion would initiate a cascade of events is not supported by prior clinical observations. Specific traumatic lesions are often seen along the anterior glenoid surface (glenoid articular rim divot or glenolabral articular disruption lesions), and sometimes articular cartilage damage occurs in association with instrument use. At long-term follow up (arthroscopic or radiographic), these mechanical lesions remain confined to the original sites and do not “spread” to a global destruction pattern. Furthermore, if a catheter could cause sufficient mechanical damage and begin a cascade of events leading to global glenohumeral chondrolysis, then the ultrahigh–molecular weight polyethylene material of routinely used high-strength nonabsorbable sutures would likely have the same effect. The volume of material associated with 3 intra-articular knots is similar to that of the multiport infusion catheter left within the joint. This suture material has the potential to cause abrasive damage, but these sutures have been used for years without causing significant global articular cartilage damage, even during range-of-motion and strengthening exercises.4The speed and aggressive destructiveness with which the global effect is observed (a matter of months) are unique to this condition and cannot be explained by a catheter placed in the area of the anterior glenohumeral ligaments, where it could not reach the posterior aspect of the humerus.5Finally, the observation that this response occurred exclusively in patients with the device in place in the glenohumeral joint for more than 48 hours (and not <48 hours) and only those receiving the infusion at 4.16 mL/h (and not those receiving 2.08 mL/h) cannot be explained by a mechanical effect alone, especially given that no evidence of any mechanical articular cartilage damage was present in any of the other cases in this series. In answer to the other issues raised by Dr. O'Driscoll: The patient whose radiograph was shown in Fig 2 (patient 2) has not undergone repeat arthroscopy, and arthroscopic images of the shoulder were not part of our report. The remark that “The difference between leaving a cannula in the joint for 5 days versus 2 days …” suggests that our cannulas were left in some of these joints for 5 days. To clarify, the small multiport catheters associated with the 100-mL containers were removed after 2 days, and the catheters associated with the 270-mL containers were removed when the infusion was exhausted (65 hours, or 2.7 days) or, at most, 3 days after surgery. Dr. O'Driscoll's expertise and excellent research in the area of cartilage research are both well known and highly respected. His point that a wide variety of ways exist for articular cartilage damage to develop is well supported. That these may result in a cascade of events leading to damage—ranging from mechanical damage to extensive cartilage damage—was recently outlined in great detail by Solomon et al.8Solomon D.J. Navaie M. Stedje-Larsen E.T. Smith J.C. Provencher M.T. Glenohumeral chondrolysis after arthroscopy: A systematic review of potential contributors and causal pathways.Arthroscopy. 2009; 25: 1329-1342Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar We appreciate Dr. O'Driscoll's comments and agree that many questions remain unanswered regarding the etiology and pathophysiology of postarthroscopy glenohumeral chondrolysis. The orthopaedic community and, ultimately, patients will certainly benefit from ongoing disclosure and discussion of this devastating complication. Glenohumeral Chondrolysis ArticleArthroscopyVol. 26Issue 4PreviewI read the article by Rapley et al.1 The authors present a consecutive series of patients in subgroups; in 3 of 16 patients an indwelling pain pump catheter was left in the glenohumeral joint for 65 hours. In the Discussion section, they state, “The pain pump and its catheter do not mechanically cause [articular cartilage damage].” I question the authors for their reasoning. The catheter has the potential to act as a “loose body,” and if it were to get into the space between the humeral head and glenoid, I see no reason why significant mechanical damage could not occur and initiate the cascade of events that we have discussed and presented evidence of in the literature. Full-Text PDF