Although nontraumatic osteonecrosis (ON) is most

Transcription

1 Techniques in Shoulder & Elbow Surgery 9(1):23 30, 2008 T E C H N I Q U E Technique and Early Results of Intramedullary Pressure-Guided Core Decompression of the Humeral Head for Nontraumatic Osteonecrosis Milford H. Marchant Jr, MD Division of Orthopaedic Surgery Duke University Medical Center Durham, NC Allston J. Stubbs, MD North Carolina Sports Medicine Institute Winston-Salem, NC Carl J. Basamania, MD Triangle Orthopaedic Associates, P.A. Durham, NC Ó 2008 Lippincott Williams & Wilkins, Philadelphia ABSTRACT The purpose of this study was to describe a modified technique for humeral head core decompression using biplanar fluoroscopy and intraosseous pressure monitoring to identify osteonecrotic regions for targeted decompression. A bone marrow biopsy needle and arterial pressure monitoring equipment were used to locate the region of maximum pressure, and then acorn-style reamers decompressed the osteonecrotic bone. In 12 shoulders (11 patients) evaluated, the peak intramedullary lesion pressure was, on average, 6 times higher than the entry pressure. At 6 months, Visual Analog Scale pain scores had decreased in all shoulders by 70.3% (P G ) on average. Nine of 12 shoulders, with an average follow-up of 44.2 months (range, 12Y72.5 months), were evaluated for pain relief, disease progression, and need for arthroplasty. In that group, Visual Analog Scale pain scores had decreased by 56.8% (P G 0.008) on average. Only 1 patient, despite excellent initial results, had progression of symptoms and required arthroplasty at 12 months. Pressure-guided core decompression provides a technique for directly localizing the osteonecrotic bone and potentially improving decompression outcomes. Keywords: osteonecrosis, avascular necrosis, core decompression, intraosseous pressure monitoring, humeral head, humerus Reprints: Carl J. Basamania, MD, Triangle Orthopaedic Associates, P.A., 120 William Penn Plaza, Independence Park, Durham, NC ( cbasam@surgical.net). Although nontraumatic osteonecrosis (ON) is most frequently found in the femoral head, the humeral head is the second most common location. 1Y7 Initially described in the humeral head by Heimann and Freiberger 8 in 1960, most of our subsequent knowledge regarding the disease process has been extrapolated from femoral head disease. Although the 2 disease locations share the same etiologies and underlying pathology, humeral head ON is a different clinical entity. The functional capacity of the shoulder is more forgiving largely because of the fact that it is a nonweightbearing joint and is less constrained, and significant additional motion is afforded by the scapulothoracic articulation. 9 In addition, the shoulder is thought to have a greater vascular watershed. 6 Therefore, the treatment algorithm for humeral head ON is slightly different; often incorporating more conservative management. Core decompression for nontraumatic ON has been advocated for use in both the femoral and humeral heads. 3,5,9Y21 For accurate core decompression, several methods have been used to identify and confirm the area of necrotic bone. Preoperative evaluation uses imaging modalities including plain radiography, bone scans, and magnetic resonance imaging (MRI). Arthroscopy and/or biplanar fluoroscopy have been used intraoperatively to confirm lesion location. 11,19,22 Other studies have confirmed adequate decompression by using postoperative histology. 19,22 The goal of core decompression is to reduce the intraosseous/extravascular pressure to reestablish blood flow to the subchondral bone. Larsen 23 first observed Volume 9, Issue 1 23

2 Marchant et al TABLE 1. Demographic Data Shoulder Sex Age, y Diagnosis Etiology Shoulder Cruess stage Follow-up, mo Arthroplasty 1 F 45 Intra-articular injection CS L 1 2 NA 2 F 34 Intracerebral hemorrhage CS R 2 72 No 3 F 30 Sickle cell disease SCD L No 4 F 28 Sickle cell disease CS/SCD L 2 68 No 5 M 27 Autoimmune pericarditis CS L 3 60 No 6 F 43 Anaplastic ependymoma CS L 5 12 Yes 7* M 43 Leukemia CS R No 8* M 46 Leukemia CS L NA 9 F 44 Asthma CS L No 10 F 35 Hemolytic anemia CS R No 11 M 27 Meningioma CS L 2 44 No 12 M 42 Acoustic neuroma CS R 2 5 NA *Same patient. F indicates female; M, male; CS, corticosteroid use; SCD, sickle cell disease; L, left; R, right; NA, follow-up data less than 12 months. the relationship between bone ischemia and increased intramedullary pressure in Corticosteroid usey, alcohol abusey, Sickle cell diseasey, and Gaucher diseasey related ON have elevated intraosseous pressure as part of their proposed pathophysiologies. 2 Measurement of intramedullary pressure to identify ON has been demonstrated in the proximal femur but not in the humeral head. 9Y13,17 By identifying the region of maximum pressure within the bone, the accuracy of lesion identification and decompression could increase. The purpose of this study is to describe and evaluate the method of a modified technique for humeral head core decompression using combined biplanar fluoroscopy and intraosseous pressure monitoring to more accurately identify the regions of ON for targeted decompression. METHODS Patient records from November 2000 to October 2006 were retrospectively reviewed to identify patients who underwent pressure-guided core decompression for nontraumatic humeral head ON. Institutional review board approval was obtained to perform the chart and radiographic review. Twelve shoulders (11 patients) were identified as being treated with pressure-guided core decompression, and the intramedullary pressure data were available. Patient demographic data are summarized in Table 1. Evaluation All patients were evaluated preoperatively by plain radiography, and most patients had an MRI study. The most common reported risk factor for developing nontraumatic ON was corticosteroid use (11 of 12 shoulders). Core decompression had been advocated previously for patients with Cruess 1 modified stage 1, 2, and 3 disease of the humeral head. 2,3,5,19Y22 Stage 2 ON was the most common disease severity (7 shoulders) in our cohort, but all stages of disease were treated. The patient with stage 5 disease had multifocal ON with glenoid involvement but no collapse of the humeral head. FIGURE 1. Jamshidi bone marrow biopsy needle (top) and acorn-style reamer (bottom). FIGURE 2. Arterial pressure tubing connected to the bone biopsy needle. 24 Techniques in Shoulder & Elbow Surgery

3 Pressure-Guided Humeral Core Decompression FIGURE 3. Intramedullary pressure recordings. Entry pressure wave (A) and lesion pressure wave (B). Arrows mark the period of equilibrium. The primary chief complaint was pain. Patients were maximized with nonoperative treatment before consideration for core decompression. All patients were educated regarding risk factors. Patient symptoms were generally managed with a nonsteroidal anti-inflammatory and activity modification for 6 to 12 weeks. Because the superior central portion of the humeral head is the most common region affected, patients are advised to avoid overhead activities. Exercises to help patients maintain motion were used when necessary, with or without physical therapy assistance. Corticosteroid injections were generally avoided in patients with early stage disease to avoid the potential of disease progression. OPERATIVE TECHNIQUE The operative procedure was modified from a technique described by Ficat and Arlet 9 in the evaluation of femoral head ON and Downey et al 24 in a biomechanical study evaluating femoral head pressure during variable joint loading. Specific instruments used are shown in Figure 1; the operative sequence is shown in Figures 2 to 5. All patients received either general or regional anesthesia with sedation. Patients were placed in a beach-chair position. The fluoroscopy C-arm was positioned before preparing and draping to allow for intraoperative imaging. During positioning of the fluoroscopy machine, the C- arm was inverted, so that the ring was facing the ceiling, and the open portion was facing the floor. The shoulder was positioned within the radiographic field, so that the humeral head was easily visualized. Anteroposterior and lateral views of the humeral head could then be obtained simply by internally and externally rotating the arm. Once the shoulder and extremity were prepared in the standard sterile fashion, sterile U-drapes were placed on the shoulder, with the superior U-drape arranged over the top of the C-arm. Therefore, the entire case could proceed without having to move the patient or the C-arm machine to obtain fluoroscopic imaging. Before decompression, a diagnostic shoulder arthroscopy was performed to rule out significant intraarticular FIGURE 4. Guidance of the bone biopsy needle using fluoroscopic guidance. FIGURE 5. Acorn-style reamer used to decompress the lesion under fluoroscopic guidance. Volume 9, Issue 1 25

4 Marchant et al FIGURE 6. Preoperative MRI study showing region of ON. pathology including unstable cartilage flaps or articular collapse. No cartilage flaps or loose fragments were removed before proceeding with the decompression. Although the purpose of the arthroscopy was not for lesion identification, the cartilage overlying the suspected lesion was observed to be abnormal in several patients. Instead of its typical smooth appearance, the articular cartilage demonstrated macroscopic changes consistent with Outerbridge grade 1 chondromalacia. More specifically, it had a dimpled texture similar to peau d orange skin changes seen in thyroid disease and certain types of breast carcinoma. A 1-cm longitudinal incision was then made over the lateral humerus, 4 cm from the lateral edge of the acromion to avoid injury to the axillary nerve. With most lesions located on the superior aspect of the articular surface,thisstartingpointallowedforgoodaccesstothe lesion while staying within the safe area of the deltoid. 25 Through this incision, under fluoroscopic guidance, a Jamshidi bone marrow biopsy needle with a FIGURE 7. Postoperative MRI at 2 years, demonstrating resolution of ON and core decompression tract. stylus (Cardinal Health, Toronto, Ontario, Canada) was introduced through the lateral cortex. The stylus was removed, and an arterial pressure transducer was connected to the needle (Fig. 2). An intramedullary sinusoidal pressure wave was observed at the entry point and allowed to achieve equilibrium (Fig. 3A). Pressure readings were recorded from the arterial pressure monitor during the equilibrium period. The stylus was then reinserted. Under biplanar fluoroscopic guidance, a small mallet was used to slowly advance the needle to within 2 to 4 mm of the articular surface. Fluoroscopy was used to both assist with needle positioning within the osteonecrotic lesion and to monitor the advancement of the needle to avoid penetration of the articular surface. Several passes into the lesion were made with intermittent pressure monitoring at several sites within the subchondral bone (Fig. 4). Areas of increased intramedullary pressure within the lesion were located. On the pressure monitor, the intralesional pressure wave rose substantially and TABLE 2. Operative Data Shoulder No. Entry pressure, mm Hg Lesion pressure, mm Hg Pressure difference, mm Hg Lesion/entry ratio * * Average Maximum Minimum *Same patient. 26 Techniques in Shoulder & Elbow Surgery

5 Pressure-Guided Humeral Core Decompression FIGURE 8. Postoperative radiograph at 6 years demonstrating no articular collapse. remained high after achieving equilibrium (Fig. 3B). Once the area of peak pressure was identified, the arterial line was disconnected, and a guide wire was passed through the needle to the area of greatest pressure. Fluoroscopy was used additionally to ensure that the bone biopsy needle and the guide wire did not violate the glenohumeral joint. Subsequently, the bone biopsy needle was removed, and 2 passes with an acorn-style reamer (5Y7.5 mm) were performed to decompress the area of intramedullary hypertension (Fig. 5). The size of the reamer was determined based on the size of the lesion. Reamers greater than 7.5 mm were not used so as to avoid creating a large hole on the lateral cortex of the humerus. For that same reason, we avoided more than 1 entry site for the decompression. In 3 patients with large areas of ON, the needle and guide wire were redirected to another part of the lesion, and a second reaming occurred. This was easily performed through the initial lateral entry hole. The decision to make a second reaming was subjectively determined by the surgeon (C.J.B.). Generally, a second reaming was used if the size of the lesion was particularly large or if the first reamer selected was too small. Upon completion of the decompression, the instruments were then removed, and the wound was irrigated and closed in a standard layered fashion. Postoperative Care All patients were placed in a sling initially and instructed to wear it for 3 to 5 days postoperatively. Pendulum exercises were initiated on postoperative day 1 or 2. Patients were then instructed to progress slowly in terms of their range of motion with no overhead activity or strenuous activity for at least 6 weeks. At the 6-week mark, patients could transition to activity as tolerated. Analysis Entry and maximum intralesional intraosseous pressures were recorded for each patient. All patients were evaluated for pain relief, disease progression, and the need for arthroplasty. Preoperative and postoperative Visual Analog Scale (VAS) pain scores were recorded to evaluate pain relief. Progression to arthroplasty was considered to be the end point of the clinical course and conservative treatment strategies in previous studies evaluating core decompression 15,19,20,22 and subsequently was the final outcome measure. All 12 shoulders were evaluated at 6 months, and 9 of the 12 shoulders, received longterm follow-up for an average of 44.2 months (range, 12Y72.5 months) postoperative. Using a paired t test (GraphPad Software, San Diego, Calif), pressure differences and pain scores were compared preoperatively and postoperatively. A P G 0.05 was determined to be statistically significant. RESULTS There were no complications noted with the operative technique. Operative data, including pressure recordings, are displayed in Table 2. The average difference between thepeaklesionpressureandtheentrypressureswas mm Hg, with a 95% confidence interval (CI) from to mm Hg, and was found to be statistically significant (P G ). The average ratio of FIGURE 9. Postoperative MRI at 6 years confirming no articular incongruity. Volume 9, Issue 1 27

6 Marchant et al intramedullary lesion pressure to intramedullary entry pressure was 6.1, with a 95% CI ranging from 4.0 to 8.2. Patients generally reported good subjective clinical resultsinregardtopainrelief,oftenwithindaysofsurgery. The average preoperative and postoperative VAS pain scores at 6 months for all 12 patients were 7.4 (range, 5Y10; 95% CI, 6.5Y8.3) and 2.2 (range, 0Y8; 95% CI, 0.8Y3.6), respectively. At 6 months, pressureguided core decompression provided an average reduction in VAS pain score of 70.3% (P G ). The average postoperative pain score in 9 patients with follow-up of 1 year or more was 3.2 (range, 0Y10; 95% CI, 1.2Y6), which provided an average reduction of pain scores of 56.8% (P G 0.008). Of the 9 shoulders with long-term follow-up, only 2 patients have reported significant pain returning after surgery. One patient did very well for 4 years, with VAS pain scores remaining less than 4. At 5 years postsurgery, his radiographs did not demonstrate collapse but did show changes consistent with osteoarthritis. The other patient whose symptoms returned had complete pain relief for 11 months postsurgery. She eventually progressed to shoulder arthroplasty at exactly 1 year after the core decompression. The overall incidence of progression to arthroplasty was 11.1%, with 95% CI of to Although not all of the patients in our cohort received postoperative MRI studies, Figures 6 and 7 demonstrate the effect of targeted decompression. The preoperative lesion seen in Figure 6 has improved or, at the very least, has not progressed at the 2-year follow-up point shown in Figure 7. One can also identify the tracts from the core decompression hitting the osteonecrotic target site within the humeral head. Figures 8 and 9 show the radiographic and MRI image of another patient 6 years postsurgery. No appreciable collapse or articular damage can be identified. DISCUSSION The purpose of this report was to describe a means of identifying the region of ON during core decompression of the humeral head. Specifically, in regard to ON of the humeral head, localization of the osteonecrotic lesion has been previously accomplished via indirect methods such as preoperative MRI and intraoperative fluoroscopy. 19,22 Mont et al, 19 in 1993, first described the use of core decompression for nontraumatic humeral ON. The technique was performed using a modified deltopectoral approach through the anterior axillary fold using fluoroscopy as the primary means of lesion identification. Since that time, arthroscopy has been used to assist with lesion identification. 11,15 Arthroscopic debridement has also been described as an alternate treatment modality. 26 In review of the literature, this marks the first attempt at intraosseous pressure monitoring through a small lateral incision for humeral head during core decompression. The use of intramedullary pressure monitoring, however, is not new. Several authors have used varying techniques of pressure monitoring in regard to ON of the proximal femur. 10,12Y14 Authors including Ficat, 12 Ficat and Arlet, 9 and Hungerford 14 have demonstrated a high correlation between elevated intramedullary pressures and biopsy-confirmed ON. Only Ficat and Arlet 9 have evaluated the pressure differential between entry and intralesional intramedullary pressure as a means to identify the osteonecrotic lesion. Our technique takes advantage of radiographic diagnosis using plain radiograph, scintigraphy, and MRI for initial lesion localization and further exploits the pressure differential between normal and diseased bone for targeted decompression. Our data from 12 shoulders undergoing core decompression demonstrates a direct and reproducible technique for intraoperative localization of necrotic bone within the humeral head. A statistically significant increase in intraoperative intramedullary pressure can be appreciated between the diseased bone and the normal bone of the proximal humerus. Furthermore, our data suggest that the osteonecrotic lesion within the humeral head would be identified by an approximate 4-fold rise in the intramedullary pressure. These findings compare with the data collected by Ficat and Arlet, 9 which show an approximate 2-fold increase in intramedullary pressures between normal and osteonecrotic bone of the proximal femur. When evaluating outcomes from core decompression, similar to the results seen in ON of the hip, core decompression of the proximal humerus has had mixed results. Results from the initial study by Mont et al 19 found overall 73% good-to-excellent results in 30 shoulders at 5 or more years, with the best outcomes seen in patients with precollapse disease. Eight of the 30 patients required arthroplasty, with 5 of 8 requiring replacement within 1 year after core decompression. L Insalata et al 15 performed core decompressions on 5 stage 3 patients with progressive disease, and progression was not altered. Four of the 5 patients required arthroplasty, and the fifth was a candidate for the procedure at the time of publication. LaPorte et al 22 reported expanded results from the cohort of Mont et al 19 to include 63 shoulders (43 patients) with 2- to 18-year follow-up. A result pattern similar to the initial study was seen. Good-to-excellent University of CaliforniaYLos Angeles shoulder scores reported in 46 of 56 shoulders with stage 1 to 3 disease. Of the 16 patients that failed, the average time to arthroplasty was 24 months. In evaluating our cohort of patients, it is easy to see that the procedure afforded significant pain relief. It is 28 Techniques in Shoulder & Elbow Surgery

7 Pressure-Guided Humeral Core Decompression also worth noting that patients with stage 1 to 5 disease were all represented in our cohort, and only 1 patient with stage 5 disease went on to require arthroplasty. Furthermore, the 1 patient who did fail decompression was on chronic corticosteroid therapy for a malignant brain tumor and had multifocal ON with progressive glenoid disease. Therefore, it is difficult to determine the origin of her shoulder pain at 1 year and if she truly failed because of humeral pathology. Core decompression of humeral head ON has been shown to offer patients relief of symptoms, particularly those with precollapse disease. The benefits of a pressure-guided technique for localizing osteonecrotic lesions are greater confidence in decompressing diseased versus normal bone, thereby improving the decompression yields; small incisions; and a reduced need for additional indirect studies including the use of excessive fluoroscopy. It is still uncertain whether core decompression alters the natural history of ON, but this technique does not seem to negatively affect the clinical course or patient outcomes. Early clinical results from our study group suggest that this technique is successful in reducing pain symptoms at both short- and long-term follow-up. The incidence of conversion to arthroplasty was comparable to, if not lower than, other case series reported in the literature. Disadvantages of the procedure seem to be limited to the risks of surgery and the possible violation of the articular cartilage with the bone marrow biopsy needle. We did not encounter any of these complications in our study group. Further study will be needed to determine the accuracy and precision of this technique and whether pressureguided core decompression will delay or reduce the need for shoulder arthroplasty, resulting in significantly better clinical outcomes. ACKNOWLEDGMENT The authors thank Robin M. Queen, PhD, for assistance with our statistical evaluation. REFERENCES 1. Cruess RL. Experience with steroid-induced avascular necrosis of the shoulder and etiologic considerations regarding osteonecrosis of the hip. Clin Orthop Relat Res. 1978;130:86Y Cushner MA, Friedman RJ. Osteonecrosis of the humeral head. J Am Acad Orthop Surg. 1997;5:339Y Hasan SS, Romeo AA. Nontraumatic osteonecrosis of the humeral head. J Shoulder Elbow Surg. 2002;11:281Y Rutherford C, Cofield RH. Osteonecrosis of the shoulder. Orthop Trans. 1987;11: Sarris I, Weiser R, Sotereanos DG, et al. Pathogenesis and treatment of osteonecrosis of the shoulder. Orthop Clin North Am. 2004;35:397Y Usher BW Jr, Friedman RJ. Steroid-induced osteonecrosis of the humeral head. Orthopedics. 1995;18:47Y Hattrup SJ, Cofield RH. Osteonecrosis of the humeral head: relationship of disease stage, extent, and cause to natural history. J Shoulder Elbow Surg. 1999;8: 559Y Heimann WG, Freiberger RH. Avascular necrosis of the femoral and humeral heads after high-dosage corticosteroid therapy. N Engl J Med. 1960;263:672Y Ficat RP, Arlet J. Treatment of bone ischemia and necrosis. In: Hungerford DS, ed. Ischemia and Necrosis of Bone. Baltimore: Williams and Wilkins; 1980:171Y Camp JF, Colwell CW. Core decompression of the femoral head for osteonecrosis. J Bone Joint Surg Am. 1986;68A: 1313Y Chapman C, Mattern C, Levine WN. Arthroscopically assisted core decompression of the proximal humerus for avascular necrosis. Arthroscopy. 2004;20:1003Y Ficat RP. Idiopathic bone necrosis of the femoral head: early diagnosis and treatment. J Bone Joint Surg Br. 1985; 67:3Y Hopson CN, Silverhus SW. Ischemic necrosis of the femoral head: treatment by core decompression. J Bone Joint Surg Am. 1988;70:1048Y Hungerford DS. Bone marrow pressure, venography, and core decompression in ischemic necrosis of the femoral head. In: Sledge CB, ed. The Hip: Proceedings of the Seventh Open Scientific Meeting of the Hip Society. St. Louis: C.V. Mosby; 1979:218Y L Insalata JC, Pagnani MJ, Dines DM, et al. Humeral head osteonecrosis: clinical course and radiographic predictors of outcome. J Shoulder Elbow Surg. 1996;5:355Y Loebenberg MI, Plate AM, Zuckerman JD. Osteonecrosis of the humeral head. Instr Course Lect. 1999;48: 349Y Meyers MH. Osteonecrosis of the femoral head: pathogenesis and long-term results of treatment. Clin Orthop Relat Res. 1988;231:51Y Mont MA, Hungerford DS. Non-traumatic avascular necrosis of the femoral headvcurrent concepts review. J Bone Joint Surg Am. 1995;77:459Y Mont MA, Maar DC, Urquhart M, et al. Avascular necrosis of the humeral head treated by core decompressionva retrospective review. J Bone Joint Surg Br. 1993;75B: 785Y Mont MA, Jacque HP, Mohan V, et al. The results of core decompression for avascular necrosis of the humeral head. Specialty Society Day, American Shoulder and Elbow Society, American Academy of Orthopaedic Surgeons 1997 Annual Meeting; February 1997; San Francisco, CA. 21. Urquart MW, Mont MA, Maar DC, et al. Results of core Volume 9, Issue 1 29

8 Marchant et al decompression for avascular necrosis of the humeral head. Orthop Trans. 1992;16: LaPorte DM, Mont MA, Mohan V, et al. Osteonecrosis of the humeral head treated by core decompression. Clin Orthop Relat Res. 1998;355:254Y Larsen RM. Intramedullary pressure with particular reference to massive diaphyseal bone necrosis: experimental observations. Ann Surg. 1938;108: Downey DJ, Simkin PA, Taggart R. The effect of compressive loading on intraosseous pressure in the femoral head in vitro. J Bone Joint Surg Am. 1988;70A: 871Y Cetik O, Murad U, Halil IA, et al. Is there a safe area for the axillary nerve in the deltoid muscle? A cadaveric study. J Bone Joint Surg Am. 2006;88:2395Y Hardy P, Decrette E, Jeanrot C, et al. Arthroscopic treatment of bilateral humeral head osteonecrosis. Arthroscopy. 2000;16:332Y Techniques in Shoulder & Elbow Surgery