The Journal of Arthroplasty Vol. 27 No. 7 2012
Extensor Mechanism Repair Failure With Use of Bidirectional Barbed Suture in Total Knee Arthroplasty Russell C. Wright, BS,* Craig T. Gillis, BS,*y Stephan V. Yacoubian, MD,*z Raymond B. Raven III, MD,*z Yuri Falkinstein, MD,*z and Shahan V. Yacoubian, MD*z
Abstract: Total knee arthroplasty (TKA) continues to advance as innovative devices become available. #2 PDO Quill SRS (Angiotech, Reading, Pa) bidirectional barbed suture was used for 161 primary TKAs at our facility. We report on 3 separate cases of extensor mechanism repair failure after primary TKA in which a barbed suture was used for extensor mechanism closure. Before the implementation of this device, there were no reported failures in 385 primary TKAs. We recommend that surgeons who use this device for extensor mechanism repair of a medial parapatellar arthrotomy in TKA exercise caution when operating on patients with morbid obesity, diabetes, and rheumatoid arthritis. We have discontinued use of the bidirectional barbed suture until more definitive large orthopedic studies establish its efficacy and safety. Keywords: bidirectional barbed suture, knotless suture, quill suture, total knee arthroplasty, arthrotomy closure, extensor mechanism failure. © 2012 Elsevier Inc. All rights reserved.
Orthopedic surgeons continually strive to refine all aspects of total joint arthroplasty to improve patient outcomes and increase efficiency in the operating room. The bidirectional barbed suture provides soft tissue approximation in a continuous manner eliminating the need for knot tying. This suture has a needle attached to each end and is characterized by 2 segments of unidirectional barbs radiating outward in opposite directions on either side of the transitional, unbarbed, center segment. The design allows the suture to glide as it is introduced into the soft tissue and the barbs engage the soft tissue when tension is created in a retrograde direction, preventing unraveling. The proposed advantages of the knotless suture include increased speed, efficiency, and strength with the ability to
distribute tension uniformly across the closure with each pass of the suture. There has been an increased interest in the application of knotless sutures in arthroplasty, and some debate over the device's ability to function in a high stress, dynamic subdermal environment. Although the efficacy of knotless sutures has been verified in the urologic , obstetric , laparoscopic gynecologic , and plastic surgery literature [4-6], there is a paucity of reports supporting the use of this suture in arthroplasty [7-8]. We report 3 cases of extensor mechanism repair failure after primary total knee arthroplasty (TKA) in which a barbed suture was used for extensor mechanism closure.
Case 1 From the *Department of Orthopaedic Surgery Research, Orthopaedic Surgery Specialists, Burbank, California; yWestern University of Health Sciences, College of Osteopathic Medicine of the Pacific, Pomona, California; and zDepartment of Orthopaedic Surgery, Providence Saint Joseph Medical Center, Burbank, California. Submitted May 2, 2011; accepted August 14, 2011. The Conflict of Interest statement associated with this article can be found at doi:10.1016/j.arth.2011.08.013. Reprint requests: Russell C. Wright, BS, Orthopaedic Surgery Specialists, 2625 W Alameda Suite 116, Burbank, CA 91505. © 2012 Elsevier Inc. All rights reserved. 0883-5403/2707-0026$36.00/0 doi:10.1016/j.arth.2011.08.013
A 56-year-old woman with severe osteoarthritis of the right knee elected to proceed with a TKA after failing conservative measures. Preoperatively, the patient had bilateral varus deformity of the knees with a 5° flexion contracture and flexion to 110°. She was 5 ft 3 in tall, weighed 294 lb, with a body mass index of 54.1 kg/m 2, and her medical history included type II diabetes mellitus, obesity, and hypertension. The TKA was performed through a standard medial parapatellar arthrotomy. The arthrotomy repair was performed with the knee in slight flexion using an
1413.e2 The Journal of Arthroplasty Vol. 27 No. 7 August 2012 absorbable #2 PDO Quill SRS (Angiotech, Reading, Pa). In accordance with the manufacturer recommendations, the repair was begun at the center of the arthrotomy. The barbed suture was inserted through the full thickness of the extensor mechanism (quadriceps tendon and retinaculum) and advanced past the unbarbed midsegment until the opposing barbs were engaged. One arm of the suture was advanced to the proximal end, and the other was advanced to the distal end of the arthrotomy. At the ends of the arthrotomy, the suture was passed back along the arthrotomy toward the center in a J-loop fashion for a minimum of 3 throws. The suture was then advanced through the tissue lateral to the wound edge and cut flush with the tissue without tying knots. The knee was then taken through multiple cycles of motion under direct visualization to confirm that the patella was tracking centrally and the repair was sound. The subcutaneous tissues were repaired using an interrupted technique with 20 Vicryl (Ethicon, Inc, Somerville, NJ). The skin was then closed using a running technique with 4-0 Monocryl (Ethicon, Inc). Postoperatively, the patient was started on a continuous passive range of motion machine on the day of surgery. She was discharged to home 4 days postoperatively in stable condition. Approximately 3 weeks after surgery, the patient returned complaining of right knee pain and a sensation of instability. Physical examination did not reveal any drainage or clinical signs of infection. There were no objective signs of gross instability; however, the patella subluxed laterally in the flexed position, and it reduced centrally in extension. The patient was able to perform a straight leg raise with the leg in the extended position. Radiographs demonstrated a laterally displaced patella. These findings suggested that the patient had a disruption of the medial retinacular repair. She was placed in a knee immobilizer and scheduled for surgery. At the time of surgery, a tear of the proximal two thirds of the medial retinaculum extending in to the quadriceps was identified along the prior arthrotomy site. The distal one third of the repair inferior to the inferior pole of the patella was intact. Component rotation was assessed intraoperatively and determined to be appropriate. Intraoperative fluid collection and cultures did not reveal any signs of infection. The knee was lavaged, and the extensor mechanism was repaired with figureof-eight interrupted stitches using #2 FiberWire (Arthrex, Inc, Naples, Fla). The knee was taken through multiple cycles of motion to confirm that the patella was tracking centrally and the repair was sound. The subcutaneous tissues were closed with 2-0 Vicryl, and the skin was closed with staples. Postoperatively, the patient was immobilized in extension for 2 weeks, after which range of motion exercises were initiated. The patient is currently pleased with her results and has good range of motion and a centrally tracking patella.
Case 2 A 69-year-old man with severe osteoarthritis of the left knee elected to proceed with a TKA after failing conservative measures. Preoperatively, the patient had bilateral varus deformity of the knees with a 5° flexion contracture and flexion to 110°. He was 6 ft tall, weighed 325 lb, with a body mass index of 44.5 kg/m 2, and his medical history included type II diabetes mellitus, hyperlipidemia, gout, and hypertension. The TKA was performed using the same technique as described in the aforementioned case. The same postoperative protocol was followed, and the patient was discharged home 3 days postoperatively in stable condition. The patient returned 10 weeks postoperatively with a small area of serosanguineous spotting from the incision. This raised concern for an infection, and the patient was taken back to the operating room for an irrigation and debridement. At the time of surgery, the dissection was taken down to the extensor mechanism, and a direct communication through the arthrotomy site into the joint was visualized. The middle one third of the arthrotomy site had failed to heal at the superomedial pole of the patella. The distal extent of the repair inferior to the inferior pole of the patella was intact. Component rotation was assessed intraoperatively and determined to be appropriate. Intraoperative fluid collection revealed a white blood cell count of 31 300 cells/mL; therefore, the knee was lavaged, and a new liner was inserted. The extensor mechanism was repair using the same technique as described in case 1. The knee was placed in an immobilizer and locked in full extension. Postoperatively, all cultures were determined to be negative; however, the patient was placed on vancomycin due to the intraoperative elevated white blood cell count and appearance of tissues. He is currently doing well and has completed his course of antibiotics with resolution of all symptoms.
Case 3 A 71-year-old man with severe osteoarthritis of the right knee elected to proceed with a TKA after failing conservative measures. Preoperatively, the patient had bilateral varus deformity of the knees with a 10° flexion contracture and flexion to 95°. He was 5 ft 10 in tall, weighed 210 lb, with a body mass index of 32.95 kg/m 2. He smoked half of a pack of cigarettes daily, and his medical history included hypertension and rheumatoid arthritis. The TKA was performed using the same technique as described in case 1. The same postoperative protocol was followed, and the patient was discharged home 5 days postoperatively in stable condition. At his 1-month postoperative visit, the patient had new onset serous drainage from the knee. Physical examination did not reveal any other clinical signs of
Extensor Mechanism Repair Failure
Wright et al 1413.e3
Fig. 1. Intraoperative photo demonstrating failure of medial retinacular and quadriceps tendon repair. Common to all 3 cases, the site of failure spanned from the inferior aspect of the patella (bottom arrow) to approximately 3 cm proximal to the superior aspect of the patella (top arrow).
infection. He had no pain with range of motion and no fever or chills. The patient was taken back to the operating room, and the knee was explored (Fig. 1). The joint was exposed, and there were no obvious signs of infection. A retinacular and quadriceps tendon tear was identified along the prior arthrotomy site spanning from the inferior aspect of the patella to 3 cm proximal to the superior aspect of the patella. As in the 2 aforementioned cases, the distal extent of the repair inferior to the inferior pole of the patella was intact. Component rotation was assessed intraoperatively and determined to be appropriate. The knee was lavaged, and the extensor mechanism was repair using the same technique as described in case 1. Postoperatively, all cultures results were confirmed as negative. One month later, the patient again developed drainage about the incision site and was taken back to the operating room to treat a possible infection. At the time of surgery, there was a small pinpoint wound dehiscence that needed to be excised. There was again a disruption of the extensor mechanism about the superomedial pole of the patella. Significant seropurulent fluid was observed, and the synovial fluid white blood cell count exceeded 100 000 cells/mL. Consequently, an explantation was performed with placement of an antibiotic-impregnated cement spacer. Postoperatively, the cultures were determined to be positive, and the patient was treated with appropriate antibiotics and is currently awaiting replacement.
Extensor mechanism repair failure after primary TKA is a rare complication. In our previous 385 primary knee arthroplasties using interrupted sutures for extensor mechanism repair, there have been no extensor mechanism repair ruptures to our knowledge. Although we were early adopters of this new application of an existing technology, we have now stopped use until more definitive orthopedic studies become available. Widespread adoption of the bidirectional barbed suture in arthroplasty surgery may not be without consequence. There were certainly confounding factors in our 3 cases. Two of our patients were morbidly obese, one had rheumatoid arthritis, one developed a definite infection, and another was treated as though he were infected. However, in both cases 2 and 3, the extensor mechanism clinically appeared to be in poorer condition than is often seen in patients with an acute infection. In the present study, the subjective degree of failure of the extensor mechanism was magnified in the cases in which the barbed suture was used. We remain concerned that an initial extensor mechanism disruption with persistent drainage of joint fluid may have set the stage for an ensuing infection as opposed to an infection, which lead to an extensor mechanism repair disruption. Although it is plausible that an infection led to the failure of the extensor mechanism in 2 of our 3 cases, we feel as though the degree of disruption was significantly greater in these patients than is commonly seen in similar patients with acute/subacute infections after knee arthroplasty. It is imperative to note that surgical approach may also contribute to the success of bidirectional barbed suture. All of our cases were performed through a medial parapatellar approach. The use of other approaches such as the midvastus and subvastus approaches, which do not require proximal tendon repair, may be more amenable to the use of these types of sutures. We feel this may be the case because, in all 3 of our failures, the distal one third of the repair was found to be healing, whereas the repair proximal to the superior pole of the patella demonstrated the most significant zone of failure. In light of our experience, we advocate that surgeons who continue to feel confident in using the bidirectional barbed suture exercise caution in morbidly obese patients, in patients who may have tissue compromise such as those with autoimmune disorders and diabetes and when a classic medial parapatellar approach is used. This potential complication may be avoided by using a more straight, longitudinal incision in the superior patellar zone and avoiding a horizontal portion of the medial parapatellar approach. Larger studies need to be performed before it is widely accepted as a truly reliable orthopedic surgical device.
1413.e4 The Journal of Arthroplasty Vol. 27 No. 7 August 2012
References 1. Moran ME, Marsh C, Perrotti M. Bidirectional-barbed sutured knotless running anastomosis v classic Van Velthoven suturing in a model system. J Endourol 2007; 21:1175. 2. Murtha AP, Kaplan AL, Paglia MJ, et al. Evaluation of a novel technique for wound closure using a barbed suture. Plast Reconstr Surg 2006;117:1769. 3. Greenberg JA, Einarsson JI. The use of bidirectional barbed suture in laparoscopic myomectomy and total laparoscopic hysterectomy. J Minim Invasive Gynecol 2008;15:621 [Epub 2008 Jul 10].
4. Warner JP, Gutowski KA. Abdominoplasty with progressive tension closure using a barbed suture technique. Aesthet Surg J 2009;29:221. 5. Ruff G. Technique and uses for absorbable barbed sutures. Aesthet Surg J 2006;26:620. 6. Villa MT, White LE, Alam M, et al. Barbed sutures: a review of the literature. Plast Reconstr Surg 2008;121:102e. 7. Vakil JJ, O'Reilly MP, Sutter EG, et al. Knee arthrotomy repair with a continuous barbed suture a biomechanical study. J Arthroplasty 2011;26:710. 8. Eickmann T, Quane E. Total knee arthroplasty closure with barbed sutures. J Knee Surg 2010;23:163.