Hip Arthroscopy for Femoroacetabular Impingement: The Changing Nature and Severity of Associated Complications Over Time Myung-Sik Park, M.D., Sun-Jung Yoon, M.D., Yong-Jin Kim, M.D., and Woo-Chul Chung, M.D.

Purpose: The aim of this study was to assess complications related to arthroscopy for femoroacetabular impingement (FAI) and how these complications have changed as we have gained more experience with this procedure. Methods: The authors reviewed 200 hips (197 patients). The average patient age was 44.64 years and the mean follow-up time was 28.2 months. All patients underwent hip arthroscopy in the supine position. Clinically, Modified Harris Hip Scores (MHHS) and patient satisfaction with outcome were used. We divided complications into 3 groups: Group 1 related to traction, group 2 related to surgical technique or implant failure, and group 3 related to outcomes. Results: Clinically, the MHHS improved from 69.96 (6.10) to 80.45 (7.00), and patient satisfaction with the achieved outcome increased to 8.87 (0.76). The overall complication rate was 15% (30 of 200 hips). Group 1 consisted of 4 patients with pudendal neuropraxia and 2 patients with ankle joint pain (P ¼ .013). Group 2 consisted of 2 patients with lateral femoral cutaneous neuropraxia, 2 patients with iatrogenic labral perforations, one patient with a labral tear, and 4 patients with femoral head scuffs. There were 4 incidents of instrument breakage. Furthermore, 3 suture anchors failed, a second-degree burn occurred in one patient, and there was incomplete reshaping in 5 hips (P ¼ .045). Group 3 included one patient with a snapping sound and heterotopic ossification. Second-look arthroscopy was performed for 5 hips. All the complications outlined in groups 1 and 2 are related to the learning curve and have statistical significance (P < .05). Conclusions: Complications relating to hip arthroscopy took different forms during the early learning period, but overall complication rates decreased along the learning curve. Surgical techniqueerelated complications such as problems with suture anchors and the reshaping of cam impingements were also considered during the later stage. Level of Evidence: Level IV, therapeutic case series.

F

emoroacetabular impingement (FAI) is increasingly recognized as an important cause of labral tears, chondral lesions, and hip pain in young adults, as well as a leading cause of early osteoarthritis of the hip.1,2 Hip arthroscopy has emerged as a promising modality in the treatment of FAI. Furthermore, arthroscopic

From the Department of Orthopedic Surgery (M-S.P., S-J.Y., Y-J.K.), Chonbuk National University Hospital, Research Institute of Clinical Medicine, Jeonju and Department of Orthopedic Surgery (W-C.C.), Carollo Hospital, Sunchun, Korea. Supported by the Research Institute of Clinical Medicine, Chonbuk National University Hospital, Korea. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received August 20, 2013; accepted March 19, 2014. Address correspondence to Myung-Sik Park, M.D., Department of Orthopedic Surgery, Chonbuk National University Hospital, Research Institute of Clinical Medicine, Jeonju, 561-712, Korea. E-mail: [email protected] Ó 2014 by the Arthroscopy Association of North America 0749-8063/13597/$36.00 http://dx.doi.org/10.1016/j.arthro.2014.03.017

procedures have approached results similar to those with an open technique.3-5 Although hip arthroscopy is now used more widely, it nevertheless gives rise to a unique set of challenges that makes it technically more demanding than arthroscopy of other joints. However, there are few reported complications and those that are reported generally refer to soft tissue injuries caused by traction or portal creation procedures.6,7 Reported complication rates associated with hip arthroscopy range from 0.5% to 6.4%.8-10 Most complications are related to the required distraction of the joint.6 Transient neuropraxia of the pudendal, sciatic, and peroneal nerves is the most common injury,8-10 but permanently disabling cases have also occurred,11-16 as have major complications such as avascular necrosis,15 femoral neck fracture,13 and cardiac arrest16 after intra-abdominal fluid extravasation in an acetabular fracture. Recently, hypothermia (body temperature 10 cm in diameter) to protect the genitals and pudendal nerve as described by Byrd.21 In this series, all patients underwent hip arthroscopy while in a supine position with 2 or 3 portals (anterolateral, anterior, and midanterior). All surgery was performed by the same surgeon. During the arthroscopic procedure, capsulotomy was usually performed using a beaver blade and radiofrequency (RF) device through the working portal. The traction times, procedures, and intraoperative complications were recorded for each operation. Postoperatively, we checked the 3 pelvic views again and compared them with the preoperative radiographs. We encouraged early ambulation and 3 weeks of walking

with crutches with partial weight bearing. No impact sport activities were permitted for a period of 6 months after the operation. Patients visited our clinic at 6-week, 3-month, 6-month, and 1-year intervals. The first 100 hips were regarded as part of the learning curve. All patients were clinically assessed using Modified Harris Hip Scores (MHHS),22 and patient satisfaction with outcome was also recorded postoperatively (1 ¼ unsatisfied; 10 ¼ very satisfied).5 All information concerning perioperative complications was recorded by a research fellow. We divided complications into 3 groups; the types of complications were broadly linked to group 1 (traction related), group 2 (surgical technique related), and group 3 (outcome related) (Table 1). Statistical Analysis The statistical analysis pertained to 2 data sets: the first 100 hips of the early stage and the remaining hips. The adjusted differences were calculated as part of a multivariate analysis to account for differences in the 2 groups. The statistical comparison of the first 100 hips and the remaining hips was performed using a nonparametric test. We set statistical significance to P ¼ .05 or less.

Results Clinically, the MHHS scores improved to 80.45 (7.00) at final follow-up from 68.96 (6.10), and the Table 1. Complications During the Early and Later Stages Complication Group 1 (traction-related complications) Neuropraxia Pudendal nerve Sciatic nerve Femoral nerve Ankle pain Group 2 (surgical or implantrelated complications) LFC nerve Soft tissue and cartilage damage Labral perforation Labral tearing Femoral head scuffing Instrument breakage Ratchet tip Flexible RF tip Beaver blade Suture anchor problem Incomplete reshaping Second-degree burn Group 3 (outcome-related complications) Snapping sounds Heterotopic ossification

Early Stage (n ¼ 100)

Later Stage (n ¼ 100)

4

0

2

0

P Value .013

.043

.045 2

0

2 1 2

0 0 2

1 1 1 2 4 1

0 1 0 1 1 0

1 0

0 1

.157 .251

.044

LFC, lateral femoral cutaneous; RF, radiofrequency.

.246 .082

.319 .319

HIP ARTHROSCOPY FOR FEMOROACETABULAR IMPINGEMENT

3

Fig 1. The broken blade of an arthroscopic knife.

patient satisfaction outcome score was 8.87 (0.76). One total hip arthroplasty was excluded from the clinical score. The MHHS of the early stage improved to 79.71 (6.72) from 68.52 (5.54) and the patient satisfaction score was 8.4 (0.7). The MHHS of the later stage improved to 83.15 (6.79) from 70.06 (8.35), and the patient satisfaction score was 9.2 (0.41). The following procedures were used for the 200 FAI lesions: pincer type for 53 hips (26.5%), cam type for 7 (3.5%) hips, and mixed type for 140 hips (70%). Analysis of our 200 hip arthroscopy cases showed that 30 (15%) perioperative complications occurred. The complication rate for the early stage was 24% and 6% for the later stage. Second-look arthroscopy was performed in 5 hips. The average traction time was 127 minutes (range, 62 to 150 minutes) during the early stage and 112 minutes (range, 65 to 125 minutes) during the later stage. Group 1 (Traction-Related Complications) The types of nerve injuries were pudendal nerve neuropraxia in 4 hips. Only one male patient reported a dull sensation in the scrotal area (similar to pudendal

Fig 2. The broken tip of a flexible radiofrequency probe.

nerve neuropraxia) for more than 1 month. Ankle joint pain was reported in 2 patients. The traction-related complications showed some statistical significance between the early and late stages (P ¼ .013). Group 2 (Surgical TechniqueeRelated Complications) Instrument breakage or failures occurred in 4 hips. A ratchet tip of a grasper (Elite suture grasper; Smith & Nephew, Andover, MA) broke during one operation, and the blade of a double-edged knife (model 28146m, Storz, Germany) broke during capsulotomy (Fig 1). A motorized blade (Dyonics power shaver blades, Smith & Nephew) and 2 tips of a flexible RF probe (Vulcan ablator, Smith & Nephew) broke (Fig 2). All broken surgical instruments were successfully removed, except a tip from a broken RF probe. However, this patient did not show any pain or mechanical symptoms attributable to the broken tip nor was the piece visible on radiography (P ¼ .045). Three suture anchors failed. During the early stage, one bioabsorbable anchor broke during insertion and

4

M-S. PARK ET AL.

Fig 3. (A) Anchor being pulled out (arrow) (PEEK BIORAPTOR 2.9 AB). (B) Inadequate fixation of an anchor (arrow) (Peek Bioraptor 2.3 PK).

another one was inserted in the wrong place. Furthermore, this anchor was placed too close to the articular cartilage, causing the acetabular cartilage to bulge along the anchor thread (Fig 3). The third anchor complication was the wrinkling and breakage of a suture strand. This happened because the suture strand could not slide freely along the anchor eyelet hole and Arthro-Pierce (Smith & Nephew) hole. We removed this anchor using a bur and straight curette. In the early stage, there was a second-degree burn along the anterior portal in one patient, which is believed to have been caused by protracted use of an RF ablator (Dyonics ablation probe; Smith & Nephew) during mixed-type FAI. Blister formation was observed along the line of fluid extravasation from the anterior working portal during the operation, which extended postoperatively. The burn wound took 3 months to resolve completely, leaving severe scar formation over the hip (Fig 4). Group 3 (Outcome-Related Complications) Postoperatively, one female patient complained about an occasional snapping sound in the inguinal region, but lesions were not identified using radiography and ultrasonography. We suspected that the snapping sound was caused by the iliopsoas tendon directly

Fig 4. Second-degree burn injuries around the lateral thigh and trochanteric area.

touching the superior pubic ramus after removal of the cushioning covering the bone of the hip joint capsule during capsulotomy or capsulectomy. However, this patient did not complain of pain in the inguinal region. The snapping sounds had ceased at her follow-up visit 6 months later. Heterotopic bone formation occurred in one hip along the gluteus medius like a comet’s tail, but the hip showed no limited range of motion or increase in size at last follow-up. Second-look arthroscopy was performed in 5 hips because of recurring inguinal pain. Four hips required reshaping of the cam lesion and labral trimming, and one hip required a repeated repair of the adjacent labral tears and labral adhesiolysis from the capsule (4 early cases and 1 later case). One patient underwent total hip arthroplasty because of progressive osteoarthritis. This patient had a dysplastic hip in which preoperative radiography showed a lateral center-edge angle of 20 .

Discussion Hip arthroscopic surgery is being used increasingly to investigate and treat disorders of the hip, especially FAI. We diagnosed the FAI with conventional pelvic radiographs and also using indirect MRA,19 which involves the intravenous injection of a gadolinium contrast agent (dose, 0.1 mmol/kg), followed by a variable delay or physical activity, or both, before magnetic resonance imaging is performed. The gadolinium contrast agent distributes within the joint space, diffusely enhancing the synovial fluid to provide greater contrast resolution between the joint fluid and the labrum, cartilage, and capsule. It has a sensitivity of 88% and an accuracy of 90% in detecting labral pathologic conditions compared with a direct MRA sensitivity of 63% (specificity, 44%).19 Harris et al.23 reported performing a systemic meta-analysis that revealed a major complication rate of 0.58% and a minor complication rate of 7.5%. In this study, we analyzed and divided complications into 3 groups.

HIP ARTHROSCOPY FOR FEMOROACETABULAR IMPINGEMENT

Group 1 Traction-related complications can be reduced along the surgical learning curve. However, we also experienced these complications in a stiff joint during the later stage of our experience. Traction is not directly related to nerve injuries. The perineal post compresses the perineum, and sustained traction can result in transient ankle pain. Making portals however, is related to nerve injuries (lateral femoral cutaneous nerve). Two of our patients complained of ankle joint pain. However, subjective paresthesia and a cold sensation resolved after 2 months. Group 2 During the early stage, surgical injuries were probably caused by poor technique, whereas later-stage occurrences were caused by insufficient traction or difficult access to stiff joints. Mild arthroscopic trauma was observed in 9 hips: labral perforation in 3 hips (caused by nitinol wire or spinal needles) and femoral head scuffing in the remainder. These iatrogenic arthroscopic traumas occur more frequently in association with central access portals. Injury to the labrum occurred in patients with stiff hip joints. In these patients, we were unable to create an anterior portal of adequate width because of insufficient traction. Femoral head scuffing was caused by a guidewire or metal cannula in 4 hips during the early stage and in 2 hips at the later stage. Badylak and Keene24 reported that iatrogenic labral punctures do not affect clinical results.25 In this study, femoral head scuffing was more common than iatrogenic labral punctures. These injuries did not affect clinical results at short-term follow-up. Instrument breakage or failure usually occurred during the early stage of our experience. This may be because of a lack of skill or old instruments. However, these complications decreased along the learning curve. A multitude of studies have discussed potential complications of fluid heating from RF.26-28 To our knowledge, the second-degree burn complication mentioned earlier has not been reported previously during hip arthroscopy, but second-degree burns of the shoulder girdle have been previously reported.27 According to Zurich et al.,28 a temperature of greater than 50 C occurred in 12% of cases in a no-flow setting after only 5 seconds and in 22% of cases after 10 seconds. This temperature threshold is high enough to cause damage to chondrocytes according to the findings of a fresh human cadaver shoulder study.22,23 The time needed to cool to a safe temperature was significantly longer in the no-flow state (average, 140.5 seconds) than it was in the 50% flow state (average, 12.5 seconds) or the 100% flow state (average, 8.5 seconds).27 We believe that temperature-checking devices are very important for surgeons to control the flow rate and the pressure of fluids as well as for RF system research.

5

Anchor-related complications included anchor breakage and penetration into the acetabular fossa when using hard anchors (Peek Bioraptor 2.9 AB; Smith & Nephew). Recently introduced soft anchors (JuggerKnot, Biomet, Warsaw, IN) may overcome hard anchor failures but need further study. Ilizalituri20 stated that incomplete reshaping is probably under-reported and is more frequent for arthroscopy. Philippon et al.29 reported that this was the reason for repeated arthroscopic reshaping in 92% of their revision cases. In another revision series by Heyworth et al.,30 repeated arthroscopic reshaping was the reason for revision arthroscopy in 79% of cases. To avoid a bad cam resection, a multiplane examination of the femoral head-neck junction with an image intensifier is necessary to fully visualize the 3-dimensional pattern of the deformity and to ensure a complete circumferential resection. Adequate exposure of the deformity is mandatory. Second-look hip arthroscopy was performed for 5 patients with recurrent pain (4 insufficient cam lesions and one pincer lesion). They were re-evaluated using radiography of the pelvis. Surgical complications also occurred with respect to femoroplasty. We identified 7 hips with insufficient femoral osteoplasty, but we excluded 2 patients because one hip was affected by Legg-Calve-Perthes disease, and the other patient had a history of poliomyelitis sequelae. Two hips showed asymmetry during preoperative radiography. In 3 further hips we observed a furrow at the head-neck junction in frog-leg view. We refer to a furrow such as this as a “gouge sign,” which is caused by a bur procedure at the head-neck junction that does not extend to the neck base (Fig 5). Patients with gouge signs often complain of inguinal pain but do not always require surgery. Eight months after hip arthroscopy, one middle-aged male patient complained of recurrent groin pain despite medication and a local injection. Indirect MRA showed a labral detachment and swelling. Second-look arthroscopy was also performed for labral repair and adhesiolysis. A gouge sign seen in any view indicates inadequate reshaping. We presumed that an inferior furrow may again have damaged the labrum, which showed labral fraying and capsular adhesions during second-look arthroscopy (Fig 6). There is little information available in published reports about complications relating to the reshaping of cam lesions of FAI.20 Generally, surgical or implant-related complications are decreased along the learning curve. Group 3 The complications in group 3 were outcome related. One patient complained about a mild snapping sound in the inguinal region while standing up, although we did not find any abnormalities during an ultrasonographic examination. There are no reports about this kind of

6

M-S. PARK ET AL.

Fig 5. Insufficient reshaping procedure. (A) Preoperative and (B) postoperative radiographs showing a hemispheric defect (gouge sign) in frog-leg lateral view.

complication. The snapping sounds were asymptomatic and resolved with rehabilitation (strengthening of the iliopsoas). We did not use any kind of prophylaxis for heterotopic ossification. We did observe it, however, in one patient during the later stage of our experience. The patient had no complaints and its size did not increase during the follow-up period. One female patient’s condition progressed to osteoarthritis. She had to undergo total hip arthroplasty 2.5 years after hip arthroscopy. We believe that hip arthroscopy should be subject to stricter indications in the case of dysplastic hips. We presume that microinstability after hip arthroscopy may accelerate osteoarthritis in dysplastic hips.

Generally, all complications related to hip arthroscopy decreased along the learning curve as our skill and experience in hip arthroscopy increased. There was a high frequency of minor traumas such as soft tissue injuries during the early stage (but no burn injuries). However, during the later stage (more than 100 cases), new complications relating to surgical techniques, such as suture anchors and reshaping of cam impingements, arose as more operations were performed. However, we did not encounter major complications such as fluid extravasation, femoral head subluxation, deep venous thrombosis, or suture site infection as experienced by many other surgeons performing hip arthroscopy.8-11,16

Fig 6. Second-look arthroscopic views. (A) Primary labral repair site, (B) showing synovitis and fraying of labrum, (C) arthroscopic view showing a gouge sign on femoral head neck junction, and (D) reshaping with burr.

HIP ARTHROSCOPY FOR FEMOROACETABULAR IMPINGEMENT

Limitations Limitations of this study are its relatively small scope and short follow-up period. Furthermore, the study was carried out retrospectively in the form of an observational report by a single surgeon, thereby precluding other significant statistical analyses of other surgeons’ experiences during the learning period. Nonetheless, this study revealed unreported complications and analyzed specific divisions as traction-related, surgical techniqueerelated, and outcome-related complications.

Conclusions Complications relating to hip arthroscopy took different forms during the early learning period, but overall complication rates decreased along the learning curve. Additional complications relating to surgical technique, such as suture anchors and femoral osteoplasty, arose as more complex procedures were performed.

References 1. Myers SR, Eijer H, Ganz R. Anterior femoroacetabular impingement after periacetabular osteotomy. Clin Orthop Relat Res 1999;363:81-92. 2. Leunig M, Ganz R. Femoroacetabular impingement: A common cause of hip complaints leading to arthrosis. Unfallchirug 2005;108:9-17. 3. Domb BG, Stake CE, Bobster IB, Jackson TJ. Surgical dislocation of the hip versus arthroscopic treatment of femoroacetabular impingement: A prospective matchedpair study with average 2-year follow-up. Arthroscopy 2013;29:1506-1513. 4. Byrd JWT, Jones KS. Arthroscopic management of femoroacetabular impingement: minimum 2-year follow-up. Arthroscopy 2011;27:1379-1388. 5. Philippon MJ, Souza BGS, Briggs KK. Hip arthroscopy for femoroacetabular impingement in patients aged 50 years or older. Arthroscopy 2012;28:59-65. 6. Byrd JWT, Pappas JN, Pedley MJ. Hip arthroscopy: An anatomic study of portal placement and relationship to the extraarticular structures. Arthroscopy 1995;11:418-423. 7. Smart LR, Oetgen M, Noonan B, Medevecky M. Beginning hip arthroscopy: Indication, position, portals, basic techniques, and complications. Arthroscopy 2007;23:1348-1353. 8. Clarke MT, Villar RN. Hip arthroscopy: Complications in 1054 cases. Clin Orthop Relat Res 2003;406:84-88. 9. McCarthy JC, Lee JA. Hip arthroscopy: Indications, outcomes, and complications. Instr Course Lect 2006;55:301-308. 10. Sampson TG. Complications of hip arthroscopy. Clin Sports Med 2001;20:831-835. 11. Committee on Complications of the Arthroscopy Association of North America. Complications in arthroscopy. The knee and other joints. Arthroscopy 1986;2:253-256. 12. Espinosa N, Beck M, Rothenfluh DA, Ganz R, Leunig M. Treatment of femoroacetabular impingement: Preliminary results of labral refixation. J Bone Joint Surg Am 2007;89: 36-53.

7

13. Aveni OR, Bedi A, Lorich DG, Kelly BT. Femoral neck fracture after arthroscopic management of femoroacetabular impingement: A case report. J Bone Joint Surg Am 2011;93:e47. 14. Byrd JWT. Complications associated with hip arthroscopy. In: Byrd JWT, ed. Operative hip arthroscopy. New York: Thieme 1998;171-176. 15. Tannast M, Goricki D, Beck M, Murphy SB, Siebenrock KA. Hip damage occurs at the zone of femoroacetabular impingement. Clin Orthop Relat Res 2008;466: 273-280. 16. Bartlett CS, DeFelice GS, Buly RL, Quinn TJ, Green DST, Helfet DL. Cardiac arrest as a result of intraabdominal extravasation of fluid during arthroscopic removal of a loose body from the hip joint of a patient with an acetabular fracture. J Orthop Trauma 1998;12:294-299. 17. Parodi D, Tobar C, Valderrama J, et al. Hip arthroscopy and hypothermia. Arthroscopy 2012;28:924-928. 18. Vaughn ZD, Safran MR. Arthroscopic femoral osteoplasty/ cheilectomy for cam-type femoroacetabular impingement in the athlete. Sport Med Arthrop Rev 2010;18:90-99. 19. Rakhra KS. Magnetic resonance imaging of acetabular labral tears. J Bone Joint Surg Am 2011;93:28-34. 20. Ilizalituri VM Jr. Complications of arthroscopic femoroacetabular impingement treatment. A review. Clin Orthop Relat Res 2009;467:760-768. 21. Byrd JWT. Hip arthroscopy utilizing the supine position. Arthroscopy 1994;10:275-280. 22. Byrd JWT, Jones KS. Prospective analysis of hip arthroscopy with 2-year follow-up. Arthroscopy 2000;16: 578-587. 23. Harris JD, McCormick FM, Abrams GD, et al. Complications and reoperations during and after hip arthroscopy: A systemic review of 92 studies and more than 6,000 patients. Arthroscopy 2013;29:589-595. 24. Badylak JS, Keene JS. Do iatrogenic punctures of the labrum affect the clinical results of hip arthroscopy. J Arthro Relat Surg 2011;27:761-767. 25. Villar RN. Arthroscopic debridement of the hip. J Bone Joint Surg Br 1991;73:170-171. 26. Ho E, Cofield RH, Balm MR, et al. Neurologic complications of surgery for anterior shoulder instability. J Shoulder Elbow Surg 1999;8:266-270. 27. Kouk SN, Zoric B, Stetson WB. Complication of the use of a radiofrequency device in arthroscopic shoulder surgery: Second-degree burn of the shoulder girdle. J Arthrosc Relat Surg 2001;27:136-141. 28. Zurich BB, Horn N, Braun S, et al. Factor influencing intra-articular fluid temperature profiles with radiofrequency ablation. J Bone Joint Surg Am 2009;91: 2448-2454. 29. Philippon MJ, Stubb AJ, Shenker ML, Maxwell RB, Ganz R, Leunig M. Arthroscopic management of femoroacetabular impingement: Osteoplastic technique and literature review. Am J Sports Med 2007;35:1571-1580. 30. Heyworth BE, Shindle MK, Voos JE, Rudski JR, Kelly BT. Radiologic and intraoperative findings in revision hip arthroscopy. Arthroscopy 2007;23:1295-1302.

Hip arthroscopy for femoroacetabular impingement: the changing nature and severity of associated complications over time.

The aim of this study was to assess complications related to arthroscopy for femoroacetabular impingement (FAI) and how these complications have chang...
1MB Sizes 2 Downloads 3 Views