Abstract
Background: Osteoid osteoma is a painful benign neoplasm that is rarely found in the elbow region.Methods: The study included fourteen patients, and we believe that this is the largest reported series of patients with osteoid osteoma of the elbow evaluated at one institution. Most of the patients had had symptoms for a prolonged period and had had multiple invasive procedures before an accurate diagnosis was made. Although findings on physical examination generally are nonspecific and are not always accurate in localizing the lesion, plain tomograms and computed tomography scans were most helpful in identifying the nidus in the present study. Thirteen of the patients had limited motion of the elbow before the definitive diagnosis was made, and ten of these thirteen had a mean flexion contracture of 38 degrees.Results: Removal of the nidus resulted in relief of pain and improvement in the range of motion of the elbow in all fourteen patients. A persistent postoperative flexion contracture was more common in the patients who had had a previous arthrotomy of the elbow than in those who had not had that procedure.Conclusions: It is important to recognize this uncommon entity to avoid the morbidity associated with a prolonged delay in diagnosis. Because the symptoms resolve after excision of the lesion, the surgeon can avoid unnecessary soft-tissue dissection and release of the contracture.
Osteoid osteoma is a painful bone lesion of obscure etiology5,7. This clearly demarcated lesion is usually one centimeter or less in diameter and generally involves the diaphysis of a long bone. The most conspicuous clinical symptom is unrelenting pain in the region of the nidus of the lesion. The pain is classically relieved with aspirin or other nonsteroidal anti-inflammatory drugs15,20. On plain radiographs, an osteoid osteoma appears as a well circumscribed round or oval lesion with a radiolucent nidus. Extensive reactive sclerosis around cortical lesions may obscure the central nidus17. When plain radiographs are inconclusive, technetium bone scans and thin-cut computed tomography scans are the most helpful modalities used to make the diagnosis1,16 (Figs. 1-A, 1-B, 1-C, 1-D). The radiographic differential diagnosis includes stress fracture, infection, a bone island, histiocytosis X, osteoblastoma, and Ewing sarcoma. These entities can usually be differentiated by their clinical presentation and histological characteristics.
The elbow is a rare location in which to find osteoid osteoma. Brugera and Newman3 identified only four osteoid osteomas among all primary tumors of the elbow in the Leeds Regional Bone Tumour Registry. Bednar et al.2 reviewed their experience with osteoid osteomas of the upper extremity and identified ten about the elbow. There have also been isolated case reports of lesions in the olecranon or the distal aspect of the humerus6,9,11. When these lesions occur about the elbow, they present a challenge with regard to diagnosis10. Orthopaedic surgeons do not always include this neoplasm in their differential diagnosis when a patient has pain in the elbow. Frequently, the patients have long-standing symptoms and synovitis of the elbow joint, which leads to a flexion contracture4. The contracture can become a serious functional problem when the dominant arm is involved. The diagnosis is particularly difficult when loss of motion, not pain, is the major presenting problem. The loss of motion is often managed with injections, casts, and even an arthrotomy of the elbow before osteoid osteoma is diagnosed.
We reviewed our experience with osteoid osteoma of the elbow at our institution. Our goals were (1) to highlight the difficulty in making this diagnosis and to identify features that allow a more timely discovery of the lesion, (2) to document the residual functional deficits in patients who have a delay in diagnosis, and (3) to determine whether the flexion contracture resolves with time or should be released when the lesion is excised.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†Section of Orthopaedics, Department of Surgical Oncology, University Boulevard, Box 106, Houston, Texas 77030. E-mail address: kwever@mdanderson.org.
‡Department of Orthopaedic Surgery, Mayo Clinic, 200 First Street S.W., Rochester, Minnesota 55905.
Fourteen patients who had osteoid osteoma of the elbow were identified in a review of the tumor registry at the Mayo Clinic. The approval of the institutional review board was obtained before the chart reviews and telephone interviews of the patients were performed. The patients had been evaluated by orthopaedic surgeons at our institution between August 1976 and March 1996. All of the lesions were located within three centimeters of the elbow joint. Each diagnosis of osteoid osteoma was confirmed histologically by a trained musculoskeletal pathologist.
The median duration of symptoms before the diagnosis was nineteen months (range, four to seventy-two months) (Table I). Most of the patients had extensive diagnostic or therapeutic intervention before the diagnosis was made at our institution (Table I). Only three did not have some type of invasive procedure during their assessment elsewhere. In addition to the invasive procedures, common treatments included application of a cast for treatment of flexion contracture, extensive rheumatological evaluation for presumed arthritis, and use of a transcutaneous electrical nerve stimulator unit.
Physical examination revealed positive findings in all of the patients (Table I). If a full range of motion is considered to be 0 to 140 degrees of flexion and 70 degrees each of pronation and supination, eleven patients had limited flexion-extension and two had limited pronation-supination. Only one patient had a full range of motion before a definitive operative excision of the osteoid osteoma. The mean preoperative range of motion was 91 degrees of flexion-extension (range, 5 to 140 degrees). Ten patients had a mean preoperative flexion contracture of 38 degrees (range, 10 to 90 degrees). One patient had a pronation contracture of 30 degrees and a supination contracture of 10 degrees, and another had a supination contracture of 20 degrees.
None of the lesions were located in the epiphysis of a skeletally immature patient (Table I). Radiographic imaging studies included plain radiographs of thirteen patients, plain tomograms of ten patients, technetium bone scans of nine patients, computed tomography scans of five patients, and magnetic resonance images of six patients.
Most of the studies were brought with the patient to our institution, where the most recent evaluation was performed. Osteoid osteoma was suspected in all fourteen patients before the definitive operative excision; however, the differential diagnosis also included osteomyelitis and stress fracture. Plain radiographs demonstrated sclerosis and cortical thickening and, occasionally, marked periosteal reaction. One patient had mild degenerative changes of the ulnohumeral joint. All bone scans demonstrated diffusely increased uptake of the tracer in the elbow region. The plain tomograms and computed tomography evaluations were the most accurate studies for identifying the nidus. In only one patient was a magnetic resonance image instrumental in the diagnosis.
Excision of the lesion was recommended for all fourteen patients. Thirteen of the patients had the excision at our institution, where the diagnosis was made, and the remaining patient had it elsewhere. The site of the lesion was determined by matching the position of the lesion on radiographic studies with superficial osseous landmarks. The operative approach varied depending on the location of the lesion but included medial, lateral, and posterior approaches to the elbow. Four patients had a synovectomy in addition to removal of the lesion. Only two patients had documented release of the anterior part of the elbow capsule; one of the two had a preoperative flexion contracture of 90 degrees and only 5 degrees of total flexion-extension. In general, the lesion was excised without the removal of much additional bone, but the osseous defect in two patients was large enough to require filling with particulate femoral allograft. Specimens from each patient were sent for aerobic and anaerobic culture. Specimens were also evaluated by trained musculoskeletal pathologists, and the diagnosis of osteoid osteoma was confirmed in all of the patients.
The most recent follow-up data were obtained by examination of seven patients at our institution and by telephone interview of three patients. The range of motion was determined on the basis of elbow tracings sent by the patients if they were unable to return.
Three patients were lost to follow-up, and one died of unrelated causes after the initial postoperative examination. The remaining ten patients had a median follow-up period of 117 months (mean, 121 months; range, fourteen to 240 months).
All fourteen patients had complete relief of pain after operative excision of the osteoid osteoma by the third postoperative day (Table I). Two patients had a return of pain at the time of the most recent follow-up, and it is noteworthy that both had had two arthrotomies of the elbow before the diagnosis and excision of the osteoid osteoma. Their preoperative range of flexion-extension had been 30 to 95 degrees and 40 to 120 degrees. One patient had a mild decrease in strength postoperatively and he had had a previous elbow arthrotomy because of an incorrect diagnosis before the eventual excision of the osteoid osteoma. Five patients had a mean postoperative flexion contracture of 20 degrees, one had a pronation contracture of 30 degrees, and one had a supination contracture of 20 degrees. Two of these patients had mild limitations secondary to the decreased range of motion. One had difficulty receiving change after a cash purchase because of a 30-degree pronation contracture, and the other had difficulty with prolonged waterskiing because of a 20-degree flexion contracture. All of the patients were employed outside of the home or were performing regular, strenuous house and yard work, and they stated that their functional status was greatly improved compared with their preoperative status.
As the diagnosis of osteoid osteoma had been confirmed histologically and all of the patients had had initial postoperative relief of pain, routine follow-up radiographs were not made for all of the patients. Radiographs of five patients revealed residual sclerosis in two and demineralization of the bones of the elbow in one. In the latter patient, a postoperative magnetic resonance image showed osteonecrosis of the capitellum. This patient had a chronic pain syndrome. The three patients who had had the osteoid osteoma excised before skeletal maturity had postoperative radiographs that demonstrated normal bone and physeal growth without deformity.
There were no perioperative complications, and no patient had an infection. The patient with a pain syndrome after excision of the osteoid osteoma had additional resection of scar tissue three months post-operatively.
This series of patients is too small for meaningful statistical analysis. However, of the nine patients who had had an arthrotomy or a biopsy as part of the initial assessment for the pain in the elbow, five had a residual postoperative flexion contracture. All five of these patients had had one, two, or three arthrotomies of the elbow or negative biopsies before an accurate diagnosis was made. The mean preoperative flexion contracture of these five patients had been 42 degrees (range, 20 to 90 degrees), and the total preoperative flexion had been 63 degrees (range, 5 to 90 degrees).
CASE 1. An eighteen-year-old right-hand dominant man had pain in the left elbow that had begun insidiously without previous trauma. The pain, which was not related to activity, increased and was most bothersome at night. The pain was almost completely relieved by aspirin. Five months later, plain radiographs, a bone scan, and a computed tomography scan were reported as revealing negative findings, but motion of the elbow was limited, with a flexion contracture of 20 degrees and flexion to 125 degrees. The patient was managed with serial casts for approximately two months. The pain was not relieved, and a subsequent complete rheumatological evaluation was negative. Three months later, repeat radiographic studies were suggestive of osteoid osteoma and the patient was managed with an operative excision of the suspicious area. He had only transient relief of pain and was referred to our institution four months later (fourteen months after the start of the symptoms). At that time, the range of flexion-extension was 20 to 110 degrees. A review of the histological specimens from the previous excision showed bone fragments but no evidence of osteoid osteoma. Repeat radiographs revealed a lucent area near the area of previous excision (Figs. 2-A, 2-B, and 2-C). Computed tomography confirmed the nidus, and the patient had an en bloc excision of the area and synovectomy. The gross and histological specimens were consistent with an eight-millimeter osteoid osteoma. At 168 months postoperatively, the patient had no pain and no limitation of activities. He had a residual flexion contracture of 10 degrees and flexion to 110 degrees. The patient had a desk job as a financial consultant and was very satisfied with the result of the procedure.
CASE 2. A thirty-one-year-old ambidextrous man had pain in the right elbow without a history of trauma. He had previously been managed with two injections of steroids without relief of pain. Aspirin and other nonsteroidal anti-inflammatory drugs relieved most of the pain, but this treatment was discontinued after the patient began to have elevated levels of liver enzymes. A magnetic resonance image made four months after the onset of symptoms revealed normal findings, and six months later the patient had an exploratory arthrotomy through a lateral approach that revealed nonspecific synovitis and normal-appearing articular cartilage. A full rheumatological evaluation was negative. Nine months later, plain radiographs revealed normal findings but a bone scan showed increased uptake of the tracer in the distal aspect of the humerus. An aspiration of the elbow was negative for infection. The pain increased, and the patient began to have considerable stiffness of the elbow. Thirty-six months after the symptoms began, the patient was referred to our institution, where physical examination revealed tenderness and swelling about the elbow with mild atrophy of the triceps. Flexion-extension was only 5 degrees, with an arc from 90 to 95 degrees, even though the patient had full pronation and supination. Marked periosteal reaction of the distal aspect of the humerus was noted on plain radiographs, but no definite nidus was seen (Figs. 3-A, 3-B, 3-C, 3-D). Plain tomograms showed evidence of a cortical osteoid osteoma, and a magnetic resonance image confirmed the presence of a round, partially ossified density in the olecranon fossa. During the operation, the triceps was split to gain exposure of the olecranon fossa. The olecranon fossa was removed en bloc with a dowel cutting device from posterior to anterior, and the tissue was consistent with osteoid osteoma. Intraoperative cultures were negative. An anterior and posterior capsular release left the patient with a flexion contracture of 10 degrees. A catheter was placed around the axillary nerve for postoperative relief of pain with instillation of 0.25 percent bupivacaine. Range-of-motion exercises were started immediately, and the patient was discharged with the extremity in an extension brace; the hospitalization was followed by an intensive program of physical therapy. At the time of the most recent follow-up (thirty months postoperatively), the patient had no pain or limitations. He had full strength but had a residual flexion contracture of 30 degrees and had flexion to 110 degrees. He was pleased with the clinical result, worked as an industrial arts teacher, and played on a softball team.
We reported on this group of patients for two reasons: (1) to highlight the difficulty involved in the diagnosis of osteoid osteoma of the elbow, and (2) to document the residual functional deficits in patients who have a delay in diagnosis. The fact that osteoid osteomas are rare in the elbow, cause joint dysfunction similar to that resulting from other conditions, and are not always obvious on plain radiographs explains why these lesions go unrecognized for so long.
This group of patients had a median delay in diagnosis of nineteen months, during which time eleven had extensive workups. The working diagnoses included chronic inflammatory arthritis, osteomyelitis, and overuse syndromes commonly found in the elbow. Multiple medications, injections, aspirations, joint explorations, and nondiagnostic open biopsies were performed before accurate localization of the osteoid osteoma. Previous arthrotomy and the delay in diagnosis, during which the lesion caused a secondary inflammatory response, contributed to the morbidity that was noted when the patients were first seen for definitive treatment. The patients who had the lesion in the distal aspect of the humerus or the proximal aspect of the ulna had loss of flexion and extension, whereas those who had the lesion in the proximal aspect of the radius had loss of pronation and supination. Other findings such as swelling and weakness were nonspecific, demonstrating that physical examination alone frequently fails to localize the lesion accurately. The patients who had had symptoms for a prolonged period and multiple invasive procedures were more likely to have a preoperative restriction of motion. This group was also more likely to have a residual postoperative flexion contracture.
After a prolonged period of symptoms, the plain radiographs often showed extensive periosteal reaction and cortical thickening, which made it difficult to discern the small nidus. The bone scans showed increased uptake of the tracer in the region about the elbow but were nonspecific. Magnetic resonance images were often misleading, especially when the patient had had previous invasive diagnostic procedures. Plain tomograms or thin-cut computed tomography scans were the most reliable studies for delineating the actual nidus.
One treatment option for osteoid osteoma described in the literature is management with anti-inflammatory drugs for as long as three years, at which point the symptoms may resolve8. However, given the tendency for synovitis and loss of motion with a prolonged period of symptoms from osteoid osteoma of the elbow, we strongly argue against medical management in this location. Operative excision is the recommended treatment for osteoid osteoma of the elbow. The lesions in our series could be localized by comparison of the preoperative imaging studies with the superficial osseous landmarks about the elbow, but alternative methods of localization used in several other series include radioisotope scanning and tetracycline-fluorescence labeling5,18. After definitive diagnosis and excision, all fourteen patients in the present series had initial relief of pain and improvement in the range of motion.
Other recent advances in the treatment of osteoid osteoma include computed tomography-guided percutaneous excision or radioablation of the lesion12-14, 19. Results with these techniques are promising, with excellent relief of pain, low rates of recurrence of the lesion, and fewer complications than with operative excision. The nidus can be removed without a large amount of additional osseous resection. This may be a good treatment option for lesions that are discovered early in the course of a patient's symptoms. However, operative excision allows the surgeon the option of performing a synovectomy or capsular release if the patient has a flexion contracture or has had multiple previous procedures, or both. Only two patients in the present series had a formal capsular release at the time that the osteoid osteoma was excised. In general, if a patient has a small flexion contracture due to pain and synovitis, removal of the lesion without capsular release will allow resolution of the inflammation and a return of nearly normal motion of the elbow. However, in patients who have a severe flexion contracture, a capsular release is recommended to improve extension.
In summary, it is important to recognize the possibility of osteoid osteoma when a patient has pain in the elbow that is refractory to nonoperative treatment and loss of motion of unknown etiology. Most of the patients in the present series had classic symptoms of pain at rest or at night that was relieved with aspirin or other nonsteroidal anti-inflammatory drugs. A prolonged workup with invasive diagnostic or failed therapeutic procedures can lead to severe morbidity in terms of the range of motion of the elbow. The lesion may be seen on a number of different imaging modalities; however, plain tomograms or computed tomography scans were the most helpful in identifying the nidus in the present series. Excision of the lesion leads to resolution of symptoms and patient satisfaction. Residual deficits in strength and the range of motion seem to be associated with a prolonged, invasive preoperative workup.
Assoun, J.; Richardi, G.; Railhac, J. J.; Baunin, C.; Fajadet, P.; Giron, J.; Maquin, P.; Haddad, J.; and Bonnevialle, P.: Osteoid osteoma: MR imaging versus CT. Radiology,191: 217-223, 1994.191217Â
1994Â
[PubMed] Â
Bednar, M. S.; McCormack, R. R., Jr.; Glasser, D.; and Weiland, A. J.: Osteoid osteoma of the upper extremity. J. Hand Surg.,18A: 1019-1025, 1993.18A1019Â
1993Â
Â
Brugera, J. A., and Newman, R. J.: Primary tumors of the elbow: a review of the Leeds Regional Bone Tumour Registry. Orthopedics,21: 551-553, 1998.21551Â
1998Â
[PubMed] Â
Cronemeyer, R. L.; Kirchmer, N. A.; De Smet, A. A.; and Neff, J. R.: Intra-articular osteoid-osteoma of the humerus simulating synovitis of the elbow. A case report. J. Bone and Joint Surg.,63-A: 1172-1174, Sept. 1981.63-A1172Â
1981Â
Â
Greenspan, A.: Benign bone-forming lesions: osteoma, osteoid osteoma, and osteoblastoma. Clinical, imaging, pathologic, and differential considerations. Skel. Radiol.,22: 485-500, 1993.22485Â
1993Â
Â
Heybeli, N., and Babacan, M.: Intraarticular osteoid osteoma of the distal humerus. J. Shoulder and Elbow Surg.,6: 311-313, 1997.6311Â
1997Â
Â
Klein, M. H., and Shankman, S.: Osteoid osteoma: radiologic and pathologic correlation. Skel. Radiol.,21: 23-31, 1992.2123Â
1992Â
Â
Kneisl, J. S., and Simon, M. A.: Medical management compared with operative treatment for osteoid- osteoma. J. Bone and Joint Surg.,74-A: 179-185, Feb. 1992.74-A179Â
1992Â
Â
Marcove, R. C., and Freiberger, R. H.: Osteoid osteoma of the elbow—a diagnostic problem. Report of four cases. J. Bone and Joint Surg.,48-A: 1185-1190, Sept. 1966.48-A1185Â
1966Â
Â
Radcliffe, S. N.; Walsh, H. J.; and Carty, H.: Osteoid osteoma: the difficult diagnosis. European J. Radiol.,28: 67-79, 1998.2867Â
1998Â
Â
Rettig, A., and Goris, J. E.: Osteoid osteoma of the olecranon. Orthopedics,19: 977-979, 1996.19977Â
1996Â
[PubMed] Â
Roger, B.; Bellin, M.-F.; Wioland, M.; and Grenier, P.: Osteoid osteoma: CT-guided percutaneous excision confirmed with immediate follow-up scintigraphy in 16 outpatients. Radiology,201: 239-242, 1996.201239Â
1996Â
[PubMed] Â
Rosenthal, D. I.; Springfield D. S.; Gebhardt, M. C.; Rosenberg, A. E.; and Mankin, H. J.: Osteoid osteoma; percutaneous radiofrequency ablation. Radiology,197: 451-454, 1995.197451Â
1995Â
[PubMed] Â
Rosenthal, D. I.; Hornicek, F. J.; Wolfe M. W.; Jennings, L. C.; Gebhardt, M. C.; and Mankin, H. J.: Percutaneous radiofrequency coagulation of osteoid osteoma compared with operative treatment. J. Bone and Joint Surg.,80-A: 815-821, June 1998.80-A815Â
1998Â
Â
Schulman, L., and Dorfman, H. D.: Nerve fibers in osteoid osteoma. J. Bone and Joint Surg.,52-A: 1351-1356, Oct. 1970.52-A1351Â
1970Â
Â
Smith, F. W., and Gilday, D. L.: Scintigraphic appearances of osteoid osteoma. Radiology,137: 191-195, 1980.137191Â
1980Â
[PubMed] Â
Swee, R. G.; McLeod, R. A.; and Beabout, J. W.: Osteoid osteoma. Detection, diagnosis, and localization. Radiology,130: 117-123, 1979.130117Â
1979Â
[PubMed] Â
Vigorita, V. J., and Ghelman, B.: Localization of osteoid osteomas—use of radionuclide scanning and autoimaging in identifying the nidus. Am. J. Clin. Pathol.,79: 223-225, 1983.79223Â
1983Â
[PubMed] Â
Voto, S. J.; Cook, A. J.; Weiner, D. S.; Ewing, J. W.; and Arrington, L. E.: Treatment of osteoid osteoma by computed tomography guided excision in the pediatric patient. J. Pediat. Orthop.,10: 510-513, 1990.10510Â
1990Â
Â
Wold, L. E.; Pritchard, D. J.; Bergert, J.; and Wilson, D. M.: Prostaglandin synthesis by osteoid osteoma and osteoblastoma. Mod. Pathol.,1: 129-131, 1988.1129Â
1988Â
[PubMed] Â