Physical Therapist Management of Tuberculous Arthritis of the Elbow.

Tuberculosis (TB) was a common and deadly disease in the first half of the 20th century until antibiotics led to the decline of all forms of TB in industrialized countries.[1] Tuberculosis cases in the United States decreased until the mid-1980s when a resurgence occurred, reaching its height in 1992.[2] Two important factors for this resurgence were the human immunodeficiency virus (HIV) epidemic and the emergence of multidrug-resistant TB.[2] Recently though, the Centers for Disease Control and Prevention reported that 1998 marked the sixth consecutive year in which cases of TB in the United States decreased.[3] Conversely, TB cases among residents born outside the United States continue to increase, and rates remain 4 to 6 times higher than for US-born people.[3] The risk of developing TB has been reported to relate to a longer lifetime experience abroad, and, even after entry into the United States, the incidence of TB remains high for several years after arrival among people originating from high-prevalence countries.[4] In 1998, foreign-born people whose birth countries were Mexico, Philippines, and Vietnam had the highest number of cases of TB.[3]

Tuberculous arthritis occurs in approximately 1% to 5% of all patients with TB.[5] It can involve any of the bones or joints of the body but is usually confined to one location, with 10% of tuberculous arthritis in the upper extremity[6] and up to 8% in the elbow.[7] The sites most frequently affected are the spine, sacroiliac, hip, and knee.[8] Because weight-bearing joints are the most frequently involved, some authors[5] suspect that trauma plays a role in the pathogenesis of bone and joint TB.

Tuberculous arthritis is usually secondary to hematogenous dissemination of tubercle bacilli from a primary pulmonary lesion.[1,8] Less commonly, it can occur by spreading through the lymphatic system or into adjacent tissue.[8] Joints can become infected by activation of dormant lymphatic or blood stream areas of morbidity.[9] In the long bones, TB originates in the epiphysis in response to mycobacteria and causes tubercle formation in the marrow, with secondary infection of the trabeculae.[8] The joint synovium responds to the mycobacteria by developing an inflammatory reaction, followed by formation of granulation tissue. The pannus of granulation tissue formed then begins to erode and destroy cartilage and eventually bone, leading to demineralization.[5] Because TB is not a pyogenic infection, proteolytic enzymes, which destroy peripheral cartilage, are not produced. The joint space, therefore, is preserved for a considerable time. If allowed to progress without treatment, however, abscesses may develop in the surrounding tissue.[5] Asaka et al[10] described an abscess around the elbow joint and between the biceps brachii and brachioradialis muscles in a patient with tuberculous arthritis.

In the United States, the most common early symptoms of tuberculous arthritis are insidious onset of local pain and swelling around the joint. In advanced cases, which occur primarily in countries where TB is more common and often is allowed to progress, sinuses and joint deformities may develop.[8] The granulomatous process eventually imparts a "boggy" or "doughy" feeling to the joint and periarticular structures.[9] Localized pain may precede other symptoms of inflammation or radiograph changes by weeks or even months.[9] Other symptoms include joint stiffness, reduced range of motion, fever, night sweats, or weight loss.[8,11] Because of the rarity of tubercular infections of joints and because the usual signs of inflammation (eg, erythema, heat) do not occur, diagnosis of tuberculous arthritis affecting peripheral joints is often delayed.[8,11] When diagnosis is not timely, joint contractures and limited functional improvement after treatment are more likely to occur, especially if bone and articular cartilage are destroyed.[12] Authors have reported diagnoses of olecranon bursitis,[13,14] tennis elbow,[15] and pyogenic arthritis, osteomyelitis, neopathic articular disease, and neoplasm before an eventual diagnosis of tuberculous arthritis.[16]

The purpose of this case report is to describe a case of tuberculous arthritis of the elbow. The patient described in this report had numerous previous diagnoses for chronic elbow pain and was ultimately referred for physical therapy evaluation and intervention.

Case Description

Patient

The patient was a 36-year-old, foreign-born (Mexico), right-hand-dominant man who reported experiencing intermittent sharp pain with insidious onset and swelling in his left elbow 10 months previously. He reported that his symptoms were aggravated with movements of the elbow and eased with rest. There was no known history of left elbow or arm injury. The patient did not report any recent fever or weight loss, and he said that he was healthy except for the elbow pain. He stated that he had been an intravenous (IV) drug user for 5 years, during which he used his left arm for injections, but he said he had not used any IV drugs for 2 years prior to the physical therapist examination and evaluation. The patient was not working at the time of the examination. His goal was to play handball pain-free.

The patient had a 10-month history of evaluations for left elbow pain, swelling, and decreased range of motion. The patient had been diagnosed with lateral epicondylitis, degenerative joint disease, synovitis, and tenosynovitis by 3 different physician assistants at 3 different facilities, and he had been treated with nonsteroidal anti-inflammatory drugs. After 10 months, an orthopedic surgeon examined the patient. The physician referred the patient to the physical therapist for examination, evaluation, and intervention for chronic elbow pain and ordered electromyography (EMG) and nerve conduction studies (NCS).

Three series of elbow radiographs were taken prior to the physical therapy evaluation. Each of the 3 series of elbow radiographs was taken at a different facility, with the last series being taken at my facility. The first series, taken 10 months previously, showed no noticeable abnormalities. Two months later, a second series was negative for fracture, but there were cyst-like structures and mild exostotic bone formation in the region of the lateral epicondyle, and there was another cyst-like structure in the proximal shaft of the ulna (Fig. 1). The lateral view showed exostotic bone formation at the anterior distal humerus, which the radiologist stated may have been indicative of an old injury. The third radiographic series 4 months before the physical therapy evaluation revealed a posterior fat-pad sign, which the radiologist suggested may have been created by joint effusion or an occult fracture (Fig. 2). Normally, the posterior fat pad, which lies deep in the olecranon fossa, is not visible on the lateral view. It can be displaced out of the fossa by blood or synovial fluid within the joint, thus becoming visible.[17] The radiologist who interpreted the third series recommended further evaluation if the patient's complaints continued.

[ILLUSTRATIONS OMITTED]

Nerve conduction studies of motor and sensory components of the left median, ulnar, and radial nerves completed just prior to the physical therapy evaluation were within normal limits. Electromyograms of the middle deltoid, biceps brachii, brachioradialis, pronator teres, abductor pollicis brevis, and first dorsal interosseus muscles also were within normal limits. The patient had positive purified protein derivative (PPD) tests since the previous year. A standard posteroanterior chest radiograph for patients with a positive PPD test was normal. A normal chest radiograph shows no pleurisy with effusion. Pleurisy with effusion results when the pleural space is seeded with Mycobacterium tuberculosis.[18]

Examination

The patient held his left elbow in a flexed position and apparently was guarding the elbow against his body. He had diffuse left elbow effusion, with the left elbow joint girth 1.5 cm greater than the right elbow joint girth measured at the elbow flexion crease. There was no ecchymosis at the time of examination, but wasting of the biceps and triceps muscles was noticeable. The patient had elbow active and passive range of motion of 30 to 110 degrees, with pain at both flexion and extension end ranges. Wrist range of motion was normal, but the patient did have a sharp pain at the lateral and medial condyles during end ranges of pronation and supination, respectively.[19] The shoulder was cleared for pathology using overpressure during active flexion, abduction, and while the patient was reaching behind his back. I performed overpressure by applying a force to the patient's end range at the point where his active range of motion stopped. The wrist was cleared when ,overpressure was performed during active flexion and extension. Because both procedures failed to reproduce the patient's elbow pain, I considered the shoulder and wrist cleared as the source of his pathology. I tested light touch sensation by moving my index fingers along the patient's C4-T2 dermatomes and upper-extremity nerve fields bilaterally. Sensation was recorded as intact and symmetrical. Muscle stretch reflexes were not tested.

Manual muscle tests of the upper-extremity musculature were performed during the examination as described by Kendall and McCreary.[19] The trapezius, middle deltoid, wrist flexor, dorsal and palmar interosseus, and extensor pollicis longus muscles were painless and rated normal bilaterally. The patient said that he was unable to hold the left biceps brachii, triceps brachii, and wrist extensor muscles in the test position against resistance because he said that it reproduced his pain. Because pain limited the patient's effort during these muscle tests, grading was not done.

Palpation revealed a mild increase in warmth around the left elbow compared with the right elbow. Palpation at the olecranon and both lateral and medial epicondyles caused a sharp pain that did not radiate. Palpation of the patient's entire anterior forearm also reproduced his elbow pain.

Evaluation

A posterior fat-pad sign has been reported to be a possible sign of interarticular fracture or swelling.[17] Due to local tenderness, swelling, and a documented fat-pad sign on this patient's radiographic report, I chose to rule out systemic pathology or a fracture before initiating aggressive stretching or joint immobilization intervention. The patient began a light physical therapy regimen of active range of motion exercises for 10 to 15 minutes 3 times a week on an upper-body cycle(*) to maintain his present range of motion, followed by ice massage for 10 minutes. The patient was instructed to use ice bags for 10 to 15 minutes on his own throughout the day. He was also instructed to stop playing handball. I discussed the case with a physician assistant, who subsequently ordered follow-up radiographs, including an oblique view to rule out an interarticular fracture as was originally advised in the most recent radiologist's report.

Re-evaluation and Intervention

The new radiographs showed a smaller posterior fat-pad sign but no fractures or evidence of other pathologies in osseous structures. Therefore, the patient continued his physical therapy program and was re-evaluated 2 weeks after the initial evaluation. During the week 2 follow-up, the patient reported that the pain had lessened and that his elbow was tender to palpation only at the olecranon. Both active and passive ranges of motion were unchanged, as was the elbow flexion crease girth. Resistive exercises were added because the patient expressed concern about the atrophy in his biceps and triceps muscles. Because he was reporting less elbow pain with palpation and range of motion end ranges, I decided to allow the patient to perform seated biceps muscle curls and supine triceps muscle extension exercises in a pain-free range. The patient performed 3 sets of 10 repetitions, 3 times a week, in the clinic under my supervision.

During the week 4 follow-up evaluation, the patient reported increased pain in the area of the medial and lateral epicondyles. Examination of elbow girth, active and passive ranges of motion, and palpation revealed no other changes. Based on the patient's continued pain and swelling, the physician assistant and I agreed that a magnetic resonance image (MRI) could be informational. At the same time, the physician assistant referred the patient back to the orthopedic surgeon for re-evaluation following the MRI. Physical therapy was discontinued until the MRI and orthopedic evaluations were completed. The MRI showed a large joint effusion and increased marrow signal within the radial neck (Fig. 3). Signal intensity refers to the strength of the radio wave that a tissue emits following excitation. The strength of the radio wave determines the degree of brightness of the imaged structures. A bright (white) area in any image is said to demonstrate a high signal intensity, and a dark (black) area demonstrates a decreased intensity.[17] Hematopoietic marrow normally displays a low to intermediate signal intensity, whereas fluid displays a higher signal intensity on T2 weighted MRI.[17] The radiologist suspected infection and recommended aspiration of synovial fluid and a biopsy.

[ILLUSTRATION OMITTED]

During the second orthopedic evaluation, 2 months after the MRI, the surgeon aspirated the elbow and ordered a bone scan. A culture of the aspirated fluid was negative for growth, but the bone scan image was consistent with possible septic arthritis and osteomyelitis. At the orthopedic follow-up 3 months later, the surgeon ordered an open debridement and biopsy based on the bone scan reports and performed an arthrotomy of the left elbow with open debridement of synovium and biopsy of the capitellum and radial head the next day. The culture was positive for acid-fast bacilli, which was later identified as Mycobacteria tuberculosis. Following identification of TB, a physician specializing in infectious diseases evaluated the patient. The bacterium was sensitive to ethambutol, pyrazinamide, isoniazid, and rifampin, and the patient began a 4-drug anti-TB regimen for no less than 1 year.

Outcomes

Four months after initiating the drug regimen, the patient reported that he was pain-free, and he was discharged from the orthopedic surgeon's care. I attended a weekly orthopedic clinic at my facility during which patients were evaluated by an orthopedic surgeon. At this particular evaluation by the orthopedic surgeon, I was not following the patient in this case report in the physical therapy clinic. Six months later, I wanted to see how the patient was doing, so I scheduled him for a visit to the physical therapy clinic. At 12 months after the diagnosis of TB, the patient had recovered normal elbow range of motion, and manual muscle tests of the biceps brachii, triceps brachii, and wrist extensor muscles were normal and painless.[19] He said that he was working and playing handball without pain. The patient performed janitorial work, which consisted of cleaning floors, walls, and bathroom fixtures.

Discussion

Tuberculous arthritis usually occurs in an insidious manner, with pain and swelling of the affected joint. It is rare among people born in the United States and is more often found in people born in other countries or those with a compromised immune system. The patient in this case report had chronic elbow pain and swelling without signs of infection. Lack of signs of infections is consistent with other cases of tuberculous arthritis described.[15,16] Although he did not have compromised immunity, he had immigrated to the United States from Mexico, where TB is endemic. He also reported a history of IV drug use, which, along with direct joint trauma, interarticular steroid injections, and systemic illness, has been found to be a predisposing factor for tuberculous arthritis.[16] These factors and this patient's history suggest an onset of TB that is consistent with reports of other patients who developed tuberculous arthritis.

Joint effusion, such as that seen in this patient, often occurs with tuberculous arthritis and has been shown to affect muscles and nerves around the elbow.[20,21] Chen and Eng[20] noted compression of the posterior interosseous nerve at the region of the arcade of Frohse. Prem et al[21] noted wasting of muscles around the upper limbs and shoulder girdle along with obliteration of bony landmarks due to swelling around an elbow infected with tuberculous arthritis. Yao and Sartoris[1] also stated that weakness and muscle wasting could be present around involved joints. The patient in this case report did not have sensory deficits, but he did have noticeable wasting of his biceps and triceps muscles. Persistent effusion in the knee affects afferent activity of intracapsular receptors and can cause reflex inhibition of the quadriceps femoris muscle.[22-24] A similar mechanism may have occurred in this patient, causing wasting of the biceps and triceps muscles due to capsular distention and intracapsular pressures. An alternative hypothesis might also attribute the muscle wasting to disuse secondary to pain during elbow motion.

Radiographs can be powerful diagnostic tools, but they are not always beneficial during evaluation of a patient with tuberculous arthritis. Some authors have described normal chest radiographs in patients with tuberculous arthritis[20,25] and old or active pulmonary disease evident in only 50% of chest radiographs in patients with tuberculous arthritis.[8,16] Elbow radiographs can also be negative, even when the disease is present.[15] Unlike pyogenic organisms that produce rapid destruction of bone, TB has a gradual progression of symptoms.[26] It has been reported to begin in the distal end of the humerus, olecranon, or synovium of the elbow joint.[13,25] The first radiograph report of the patient's elbow was normal. The second series of radiographs identified a cyst-like structure and mild exostotic bone formation that was not identified on the first and final radiographs. Munk and Lee[26] contended that a normal appearance on imaging is the rule with TB infections because the underlying bone reacts (by forming cysts and producing sclerotic borders at the margins of the infected lesion) in an attempt to wall off the infectious process. Thus, a cyst-like appearance in the involved bone is not uncommon. The third set of radiographs revealed no abnormalities in bone or joint space, with the exception of a positive fat-pad sign. Greenspan[17] reported that a positive fat-pad sign could be indicative of interarticular swelling or a fracture. The fourth set of radiographs eliminated the possibility of a fracture that had not been diagnosed, but they revealed a smaller fat-pad sign, which most likely appeared because of interarticular swelling. When radiographs are normal, an MRI may be beneficial by revealing early changes such as edema that are not visible on radiographs.[27] The patient's MRI identified the complex effusion in his elbow, but a biopsy that was needed for the definitive diagnosis.

Biopsy is the most definitive test for tuberculous arthritis.[6,9,13,15] Some authors have reported that synovial fluid or tissue cultures establish a diagnosis in 90% of the cases of tuberculous arthritis.[11] Material for the culture may be obtained from aspiration of joint fluid, but this may be inconclusive, as it was in this patient's case. Laboratory tests such as sedimentation rate, granulocyte count, and lymphocyte count are not thought to be helpful.[7] This patient's prior tuberculin skin tests were positive, which is consistent with researchers' findings for patients with tuberculous arthritis.[6,10,20,25] However, as was described in cases involving a 66-year-old woman[15] and a 76-year-old man[16] with tuberculous arthritis of the elbow, a negative TB skin test does not exclude diagnosis of tuberculous arthritis. Repeated negative tuberculin tests, however, practically eliminate TB as a possible etiology.[7]

Before the advent of anti-TB chemotherapy, the classic treatment in adults consisted of excision or arthrodesis of the elbow joint.[28] The disadvantage of arthrodesis was loss of motion, and the risk of excision was an unstable elbow.[28] Anti-TB agents are effective in halting the destructive process and treating the infection. However, they cannot repair the anatomical defects that can occur in later stages.[8] During these stages, fibrous tissue can result in ankylosis of the joint. Similarly, the untreated cases can evolve to bony ankylosis.[16]

The literature provides few specifics for the physical therapist management of TB. Investigators[29] have reported using prolonged immobilization for an average of 18 months. With the introduction of TB drugs, this is no longer necessary.[12] Some authors[6,28] advocated immobilizing the elbow for 1 to 2 months at 90 degrees to relieve pain and, in the event of fusion, to achieve a functional position. After removing the cast, rehabilitation proceeded daily for 3 to 6 months, with a back splint used between therapy sessions to prevent extension deformity and help the elbow flexors regain power.[6] No specific descriptions of the splint or interventions were reported.

Surgery may be necessary in certain cases when the disease does not respond to drugs or to correct deformities or improve joint function.[8] Vohra and Kang[25] treated 6 cases of elbow TB, ranging from the disease being restricted to within the synovial membrane to extensive articular cartilage involvement. Patients were treated with 3 to 6 weeks of immobilization after surgery followed by encouraging active movements and using night splints for 2 to 5 months. No other intervention specifics were given. Other authors[30] reported that using a hinged long arm brace for a month after surgically removing granulation tissue returned the patient's elbow to being pain-free with full range of motion.

Chen et al[12] reported that a continuous passive movement (CPM) device improved functional results after synovectomy and intra-articular debridement. Following surgery, the arc of movement was set at 30 to 90 degrees and then increased to a level that the patients were able to tolerate. Patients used the CPM device for 2 to 4 weeks until movement exceeded 120 degrees. The average flexion deformity in a group of 8 patients who used the CPM device was 24 degrees versus 34 degrees in a group of 8 patients who were treated with active and passive movement. Active and passive movement was not defined.

The patient in this report responded well to antibiotics and regained full elbow function without immobilization or surgery. This improvement could have been due, in part, to the location of the disease in the joint. Vohra and Kang[25] stated that prognosis is excellent in synovial and extra-articular lesions, whereas involvement of articular cartilage reduces the chances of maintaining good range of motion. In addition, this patient's improvement could have been due to diagnosing tuberculous arthritis early and administering anti-TB treatment before severe destruction occurred. Chen et al[12] noted that joints with severe intra- and extra-articular destruction usually become stiff with fibrosis and adhesions. Martini and Gottesman[28] hypothesized that, unlike the lower-limb joints, the elbow is non-weight bearing and therefore more able to recover a normal, painless range of motion, as this patient was able to do.

Conclusion

Patients with tuberculous arthritis are not often examined or treated by physical therapists in the United States due to the relative rarity of TB infections of joints. Because of its often slow progression, tuberculous arthritis is a frequently misdiagnosed condition, which delays treatment and can lead deformities and functional deficits. This patient's disease was identified as a result of diagnostic tests and communication between a physical therapist and other health care providers. Physical therapists and other health care providers can learn from this case to consider tuberculous arthritis in the differential diagnosis of unexplained musculoskeletal complaints, especially in patients with compromised immunity or a history of immigration to the United States from an area where TB is endemic.

(*) Biodex Medical Systems, Inc, Brookhaven C & D Plaza, 20 Ramsay Rd, Box 702, Shirley, NY, 11967.

References

[1] Yao DC, Sartoris DJ. Musculoskeletal tuberculosis. Radiol Clin North Am. 1995;33:679-689.

[2] Centers for Disease Control and Prevention. Tuberculosis morbidity--United States, 1997. MMWR Morb Mortal Wkly Rep. 1998;47: 253-275.

[3] Centers for Disease Control and Prevention. Progress toward the elimination of tuberculosis--United States, 1998. MMWR Morb Mortal Wkly Rep. 1999;48:732-736.

[4] Zuber PL, McKenna MT, Binkin NJ, et al. Long-term risk of tuberculosis among foreign-born persons in the United States. JAMA. 1997;278:304-307.

[5] Davidson PT, Horowitz I. Skeletal tuberculosis: a review with patient presentations and discussion. Am J Med. 1970;48:77-84.

[6] Martini M, Benkeddache Y, Medjani Y, Gottesman H. Tuberculosis of the upper limb joints. Int Orthop. 1986;10:17-23.

[7] Martini M, Ouahes M. Bone and joint tuberculosis: a review of 652 cases. Orthopedics. 1988;11:861-866.

[8] Wright T, Sundaram M, McDonald D. Radiologic case study: tuberculous osteomyelitis and arthritis. Orthopedics. 1996;19:699-702.

[9] Rotrosen D. Infectious arthritis. In: Wilson JD, Braunwald E, Issel-bacher KJ, et al, eds. Harrison's Principles of Internal Medicine. 12th ed. New York, NY: McGraw-Hill; 1991:544-548.

[10] Asaka T, Takizawa Y, Kariya T, et al. Tuberculous tenosynovitis in the elbow joint. Intern Med. 1996;35:162-165.

[11] Naides SJ. Infectious arthritis: viral and less common agents. In: Schumacher HR, Klippel JH, Koopman WJ, et al, eds. Primer on the Rheumatic Diseases. 10th ed. Atlanta, Ga: Arthritis Foundation; 1993: 199-200.

[12] Chen WS, Wang CJ, Eng HL. Tuberculous arthritis of the elbow. Int Orthop. 1997;21:367-370.

[13] Parkinson RW, Hodgson SP, Noble J. Tuberculosis of the elbow: a report of five cases. J Bone Joint Surg Br. 1990;72:523-524.

[14] Holder SF, Hopson CN, Vonkuster LC. Tuberculous arthritis of the elbow presenting as chronic bursitis of the olecranon. J Bone Joint Surg Am. 1985;67:1127-1129.

[15] Patel S, Collins DA, Bourke BE. Don't forget tuberculosis. Ann Rheum Dis. 1995;54:174-175.

[16] George JC, Buckwalter KA, Braunstein EM. Case report 824: tuberculosis presenting as a soft tissue forearm mass in a patient with a negative tuberculin skin test. Skeletal Radiol. 1994;23:79-81.

[17] Greenspan A. Orthopedic Radiology: A Practical Approach. 2nd ed. Philadelphia, Pa: Lippincott-Raven; 1997.

[18] Daniel TM. Tuberculosis. In: Wilson JD, Braunwald E, Isselbacher KJ, et al, eds. Harrison's Principles of Internal Medicine. 12th ed. New York, NY: McGraw-Hill; 1991:637-645.

[19] Kendall FP, McCreary EK. Muscles: Testing and Function. 3rd ed. Baltimore, Md: William & Wilkins; 1983:18-293.

[20] Chen WS, Eng HL. Posterior interosseous neuropathy associated with tuberculous arthritis of the elbow joint: report of two cases. J Hand Surg [Am]. 1994;19:611-613.

[21] Prem H, Babu NV, Chittaranjan BS, et al. Tuberculosis of the elbow: an unusual presentation. Tuber Lung Dis. 1994;75:157-158.

[22] Fahrer H, Rentsch HU, Gerber NJ, et al. Knee effusion and reflex inhibition of the quadriceps: a bar to effective retraining. J Bone Joint Surg Br. 1988;70:635-638.

[23] Spencer JD, Hayes KC, Alexander IJ. Knee joint effusion and quadriceps reflex inhibition in man. Arch Phys Med Rehabil. 1984;65: 171-177.

[24] Stratford P. Electromyography of the quadriceps femoris muscles in subjects with normal knees and acutely effused knees. Phys Ther. 1982;62:279-283.

[25] Vohra R, Kang HS. Tuberculosis of the elbow: a report of 10 cases. Acta Orthop Scand. 1995;66:57-58.

[26] Munk PL, Lee MJ. Musculoskeletal case 3: musculoskeletal tuberculosis. Can J Surg. 1999;42:120-121.

[27] Gordon AC, Friedman L, White PG. Pictorial review: magnetic resonance imaging of the paediatric elbow. Clin Radiol. 1997;52: 582-588.

[28] Martini M, Gottesman H. Results of conservative treatment in tuberculosis of the elbow. Int Orthop. 1980;4:83-86.

[29] Wilson JN. Tuberculosis of the elbow: a study of thirty-one cases. J Bone Joint Surg Br. 1953;35:551-560.

[30] Yip KH, Lin J, Leung PC. Cystic tuberculosis of the bone mimicking osteogenic sarcoma. Tuber Lung Dis. 1996;77:566-568.

CS Dahl, LT, PT, MPT, is Staff Physical Therapist, Federal Medical Center, E Center St, Rochester, MN 55903 (USA) ([e-mail]).

LT Dahl provided concept/project design, writing, data collection and analysis, and project management. David Nestor, Eric D Payne, Jessie W Lief, and Matt Walsworth provided consultation (including review of the manuscript before submission), and Greg Larsen provided expert technical assistance.

The opinions or assertions contained herein are the private views of the author and are not to be construed as official or as reflecting the views of the US Public Health Service or Federal Bureau of Prisons.

This article was submitted March 1, 2000, and was accepted October 25, 2000.