JBJS | Current Concepts Review - Tuberculosis of Bones and Joints*

The Journal of Bone & Joint Surgery, Volume 78, Issue 2

Current Concepts Review | February 01, 1996

Current Concepts Review - Tuberculosis of Bones and Joints*

HUGH G. WATTS, M.D.†, LOS ANGELES, CALIFORNIA; ROBERT M. LIFESO, M.D.‡, BUFFALO, NEW YORK

*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.

The Journal of Bone & Joint Surgery. 1996; 78:288-99

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

Article

References

text A A A

Introduction

The prevalence of tuberculosis¹⁵ in the United States has been rising since 1986, with morbidity increasing 14 per cent from 1985 through 1993. In 1986, the rate of tuberculosis in the United States was 9.3 new cases per 100,000 population. In 1991, the rate in New York State^12,13 had risen to 17.3 per 100,000, and the rate in central Harlem was 169 per 100,000, which is similar to rates reported in the eastern and central regions of Africa^7,12. Much of this increase has been in the rate of pulmonary tuberculosis, but it has been associated with a concomitant rise in the number of adults and children who have tubercular musculoskeletal involvement. In a prospective study of a cohort of intravenous drug abusers in New York City who tested positive with the tuberculin skin test, the risk of active tuberculosis was 14 per cent (seven of forty-nine patients) a year during a two-year follow-up interval⁷⁶. It is becoming increasingly likely that an orthopaedic surgeon who works in a developed region of the world (especially in a big city) will encounter a patient who has tuberculosis, a disease with which the surgeon may have little experience or training¹³.

Factors that have contributed to the increased rate of tuberculosis are the rise in the number of people who have suppression of the immune system, the development of drug-resistant strains of Mycobacterium, an aging population, and an increase in the number of health-care workers who are exposed to the disease. The human immunodeficiency virus remains the leading known risk factor for the reactivation of latent tuberculous infection^65,68,79, and patients with this virus who are exposed to Mycobacterium tuberculosum are more likely to have progression to an active disease state than non-infected patients.

Tuberculosis is an ancient disease. Spinal tuberculosis has existed for at least 5000 years³⁸, and mummified remains from northern Egypt dating from 3400 B.C. provide strong evidence of its presence³⁸. The first known description of tuberculous spondylitis was written in Sanskrit sometime between 1500 and 700 B.C.^38,90.

Tuberculosis, or "consumption," has paralleled the socioeconomic rise and fall of humankind. In the late eighteenth century, Pott noted the association between tuberculous involvement of the thoracic spine and paraplegia⁶¹. By the beginning of the twentieth century, tuberculosis was the leading cause of death in Western society^24,45. In the era before antitubercular drugs were available, patients for whom treatment in a sanitarium setting was considered to be successful still had a 60 per cent mortality rate within six years after discharge from the sanitarium⁶⁵.

The potential for control of tuberculosis with modern chemotherapeutic drugs has been shown among the Eskimo population in Northern Canada. In 1950, the rate of new infections was 25 per cent a year. By 1970, however, the incidence of new infections was so low as to be unmeasurable. This is a tribute to the effectiveness of medical treatment, public-health authorities, and a highly cooperative patient population^17,31.

Three related organisms—Mycobacterium tuberculosis, Mycobacterium africanum, and Mycobacterium bovis—are the causes of tuberculosis. Mycobacterium tuberculosis is by far the most common. Mycobacterium africanum is rarely found outside of Northwestern Africa, and disease due to Mycobacterium bovis is limited in developed countries by the widespread pasteurization of milk⁶⁵.

Mycobacterium tuberculosis is a thin rod with round ends, 2.0 to 2.5 micrometers long. It is non-motile, without a capsule, and is difficult to stain with use of the usual methods. If it is stained with the classic carbol fuchsin or Ziehl-Neelsen method, it resists decolorization with strong mineral acids and alcohol; hence, Mycobacterium tuberculosis is considered an acid-alcohol-fast or acid-fast bacillus. Its histological appearance is small curved or straight rods of red or pink. Mycobacterium tuberculosis does not grow on ordinary culture medium; it grows only on enriched medium containing an egg and potato base or serum (albumin) base. It has a slow rate of growth, and grossly visible colonies first appear at two to four weeks. Mycobacterium tuberculosis is a strict aerobe, and its rate of growth is highly dependent on oxygen tension. When tension is high, as in the tuberculosis cavity of the lung, Mycobacterium tuberculosis multiplies freely. When tension is much lower, as in the caseous foci of the lung, it multiplies slowly or not at all.

The World Health Organization has estimated that one-third of the global population is infected with Mycobacterium tuberculosis, and tuberculosis remains the most frequent cause of death and disability on a worldwide basis, accounting for close to three million deaths each year^10,43. An estimated ten million persons are presently infected with Mycobacterium tuberculosis in the United States, and 90 per cent of new activated cases come from this infected pool of individuals⁷⁶.

In non-Hispanic white people, the median age at diagnosis is sixty-one years⁶. Among American minority groups, tuberculosis is primarily a disease of younger people, with a median age at diagnosis of thirty-nine years⁶. Americans who are more than sixty-five years old account for 6.5 per cent of the United States population but account for 26 per cent of the total number of reported cases of tuberculosis^11,68.

When the prevalence of tuberculosis in a community is high, most of the population in that community has been infected by the age of twenty years. When the prevalence of tuberculosis in a community is low, almost all clinically infected patients are more than fifty years old and probably were infected years earlier and are showing signs of reactivation of the disease^30,31. Reported cases¹³ of tuberculosis in American children less than fifteen years old declined from 6036 in 1962 to 1177 in 1986. This number increased 36 per cent, to 1596 patients⁶, in 1987.

In the United States, one-fifth of diagnosed new cases of tuberculosis are associated with extrapulmonary disease¹². About one-third of patients with tuberculosis who are also infected with the human immunodeficiency virus will have extrapulmonary disease with or without a pulmonary component^58,67,75.

Tuberculosis has been reported in all bones of the body^6,36. In the United States⁶, the spine is involved in 50 per cent of patients (the thoracic spine in 50 per cent, the cervical spine in 25 per cent, and the lumbar spine in 25 per cent); the pelvis, in 12 per cent; the hip and femur, in 10 per cent; the knee and tibia, in 10 per cent; the ribs, in 7 per cent; the ankle or shoulder, in 2 per cent; the elbow or wrist, in 2 per cent; and multiple sites, in 3 per cent.

Extrapulmonary tuberculosis is more common in children than in adults, and about one-third of children who have tuberculosis have extrapulmonary manifestations, the most common being involvement of the superficial lymph nodes (scrofula)⁸⁰. Involvement of the cervical spine is rare in children^13,81. The infecting organism is usually Mycobacterium tuberculosis, but it may be Mycobacterium bovis or even Mycobacterium kansasii, Mycobacterium fortuitum, Mycobacterium marinum, or Mycobacterium intracellulare.

†Shriners Hospital for Crippled Children, 3160 Geneva Street, Los Angeles, California 90020.

‡Erie County Medical Center, 462 Grider Street, Buffalo, New York 14215.

†Shriners Hospital for Crippled Children, 3160 Geneva Street, Los Angeles, California 90020.

‡Erie County Medical Center, 462 Grider Street, Buffalo, New York 14215.

Diagnosis of Tuberculosis in Bones and Joints

Clinical

The clinical presentation of musculoskeletal tuberculosis includes localized pain associated with fever and weight loss. If the spine is involved, truncal rigidity, muscle spasm, and neurological signs may be present. A cold abscess (swelling without inflammation) is strongly suggestive of musculoskeletal tuberculosis. Although a skeletal lesion may be the presenting symptom complex, any individual who has a skeletal lesion that is suspected of being tuberculous must be evaluated for the possibility of other involved sites, including the lungs, intestinal tract, and kidneys⁷⁷. Only about one-third of patients who have tuberculosis of a bone or joint have a history of pulmonary disease¹⁸. The diagnosis is sometimes more difficult to make in elderly individuals in whom the presentation of the disease often involves non-localized constitutional symptoms.

Plain Radiographs

There are no specific radiographic features that are pathognomonic of tuberculosis of bones or joints. Common findings that should arouse suspicion of joint involvement include osteopenia, soft-tissue swelling with minimum periosteal reaction, narrowing of the joint space, cysts in bone adjacent to a joint, enlargement of the epiphysis in children, and subchondral erosions involving both sides of the joint^35,84,89. (These erosions cross the epiphysis in more than one-third of affected children.)

Features in the spine include rarefaction of the vertebral end plates, soft-tissue abscess, increasing loss of disc height, variable degrees of osseous destruction, and new-bone formation. Often, multiple bones are involved in the spine, and late fusion or collapse of bone is not uncommon. Radiographic findings in patients who have disease related to the human immunodeficiency virus may be caused by other diseases and should not be automatically ascribed to tuberculosis.

Bone Scan

There are no specific scintigraphic features of tuberculosis that are pathognomic. In one study of fifty-six patients who had a tuberculous lesion, technetium-99m bone scans showed diffuse changes in uptake similar to those seen with metastatic disease in thirty-five patients (63 per cent), and the scans proved to be negative, or so-called cold, in the presence of active disease in twenty-one (38 per cent)⁸⁹. The negative scans may represent avascular segments of bone due to the formation of an abscess. Indium-111-scanning is non-specific but may show decreased activity in bone marrow⁵⁵.

Skin Test

After his discovery of the tubercle bacillus, Koch found that a concentrated filtrate from culture specimens of Mycobacterium tuberculosis that had been killed with heat protected guinea pigs from experimental tuberculosis. He called this material tuberculin. In the 1930's, precipitants of tuberculin were developed, and this precipitated material was called purified protein derivative⁷⁵. Essentially, purified protein derivative contains a number of components that are antigenic to the tubercle bacillus. At the turn of the century, approximately 90 per cent of the population reacted in some way to the injection of tuberculin. Interpretation of the results of the tuberculin or purified-protein-derivative skin test depends on the prevalence of exposure to tuberculosis, immunosuppression, or the possibility of previous vaccination with bacille Calmette-Guérin. At present, the general adult population of the United States has a rate of infection of 5 to 10 per cent, and the positive predictive value of the tuberculin test is beginning to decrease when applied to this population. At least 20 per cent of debilitated or malnourished patients who have extensive disease have a false-negative skin test⁷⁰. Similarly, patients who are co-infected with the human immunodeficiency virus and tuberculosis are prone to anergy and negative skin-testing, especially in the later stage of acquired immunodeficiency syndrome^57,69.

In our experience in Saudi Arabia, the rate of conversion to a positive tuberculin test was 10 per cent per decade. In other words, at the age of seventy years, 70 per cent of the population whom we tested had conversion to a positive tuberculin test without overt manifestations of tuberculosis.

Other Tests

General tests of inflammation, such as measurement of the erythrocyte sedimentation rate, are neither specific nor completely reliable. Serological testing, especially the ELISA (enzyme-linked immunosorbent assay) test, has a reported sensitivity of 60 to 80 per cent^16,59, but these tests may be negative for patients who have advanced disease. New work in chromatography, nucleic acid probes, and polymerase chain reaction in addition to nucleic acid probes by systems using antibodies against Mycobacterium tuberculosis are gradually being introduced, but at this stage they have not been tested extensively and are not widely available^{59,66,71,72,74}.

In regions where brucellosis is endemic, a Brucella complement-fixation test should be performed as brucellosis can mimic tuberculosis clinically. (We are aware of one patient who had proved brucellosis and proved tuberculosis at different levels of the spine.)

Culture

The ultimate diagnosis of tuberculosis depends on the recognition of Mycobacterium tuberculosis on either histological study or culture, or, ideally, both. Because of the frequency of associated tuberculosis in the lungs and kidneys, culture of specimens of sputum and urine can be helpful⁹¹, but these tests are often not performed on orthopaedic patients.

Biopsy

In countries where tuberculosis is highly prevalent and where medical facilities are limited, patients who have clinical symptoms and radiographic findings suggestive of the disease may not need to have a biopsy of the lesion in order for the physician to make a diagnosis and to institute treatment. Ideally, the prevalences of resistant strains and of other diseases that can mimic tuberculosis are known and can be excluded with the use of other modalities (such as the complement-fixation test for brucellosis). In this setting, biopsy and possible operative treatment are reserved for patients who fail to respond to adequate chemotherapy, who have substantial neurological impairment, or in whom either resistant strains or other disease entities are suspected. In these areas of endemic disease, it is important that the prevalence of disease and of antibiotic resistance be known and documented.

In areas where the disease is not prevalent or where strains that are resistant to multiple drugs are common, biopsy is mandatory, both to make the diagnosis and to determine antibiotic sensitivity. We recommend the initiation of antituberculosis treatment at the time of the biopsy to decrease the likelihood of dissemination of the disease during that procedure. This is most important when the infection may involve the central nervous system and when there is a high index of suspicion, either because of a high prevalence of endemic disease or because active tuberculosis has been suggested on the basis of the clinical history or the findings on frozen section. In these clinical circumstances, we use triple or quadruple-drug therapy if multiple-drug-resistant tuberculosis is suspected. For biopsies performed in areas where the disease is not endemic and for those performed percutaneously and thus associated with only a slight risk of contamination, we consider empirical antituberculosis treatment on a patient-by-patient basis.

If a biopsy specimen is taken from a joint, tissues from adjacent cystic lesions and synovial tissue should also be obtained and sent for both histological study and culture. Simple aspiration of the joint is much less likely to lead to a definitive diagnosis^29,89. In the spine, computed tomography-guided needle biopsy usually yields sufficient material either from the spine itself or from an adjacent abscess. Open biopsy of the spine is usually reserved for occasions when either closed techniques have proved insufficient or other procedures, such as decompression and possibly arthrodesis, are contemplated. Stains for acid-fast bacilli, fluorochrome, and the traditional Ziehl-Neelsen stain are relatively rapid and inexpensive but may produce false-negative results. Stains to identify acid-fast bacilli reliably require the presence of at least 10⁴ acid-fast bacilli per milliliter of specimen⁷⁰, but cultures are more sensitive and can reliably identify mycobacteria in a concentration of 10³ organisms per milliliter of specimen²⁸. Drug-susceptibility testing is possible with cultures. The major drawbacks are that conventional culture techniques are slow, requiring weeks before a positive result for Mycobacterium tuberculosis can be identified, and at least a moderately well equipped laboratory is needed.

Limiting the Spread of Tuberculosis in Health-Care Facilities

As the prevalence of tuberculosis increases, the perceived or real threat to health-care workers requires consideration. Any patient who is suspected of having active infectious pulmonary tuberculosis should be placed in a tuberculosis isolation room with the appropriate ventilation facilities²¹. Ideally, operative intervention should be delayed until the patient is no longer infectious. Respiratory precautions should be applied, as should universal blood and body-substance precautions. The use of bacille Calmette-Guérin vaccination for health-care personnel working in areas with a particularly high prevalence of tuberculosis is still being debated. The consensus is that the advantage of possible protection does not outweigh the loss of the use of the purified-protein-derivative tuberculin skin test to indicate when a health-care worker has conversion from a negative to a possibly positive infectious status¹⁶.

Medical Treatment of Musculoskeletal Tuberculosis

The treatment of musculoskeletal tuberculosis is primarily medical. Operative intervention is an adjunct to appropriate antituberculosis chemotherapy. Each year, the United States Centers for Disease Control change their recommendations as to the exact combinations, doses, and durations of drug treatment depending on the incidence and resistance patterns of the locally predominant strains of Mycobacterium tuberculosis. As a result, medical treatment can only be summarized with generalities. If the need to provide tuberculosis therapy is infrequent, we recommend that the Centers for Disease Control be contacted for their most recently updated guidelines. We also recommend consultation with an infectious-disease specialist.

Successful medical treatment of tuberculosis requires the prolonged administration of a minimum of three drugs to which the organisms are susceptible, and at least one of these drugs must be bactericidal. Because of the spontaneous emergence of drug resistance in a small number of tubercle bacilli, monotherapy with even the most potent bactericidal drug (isoniazid) may result in the selection of a resistant bacterial population and lead to failure of the treatment and to acquired drug resistance. Therefore, a combination of drugs is necessary to treat tuberculosis effectively. Prolonged drug therapy is necessary to eliminate or sterilize so-called persistent bacilli, which are small populations of metabolically inactive organisms.

The following are considered the most effective drugs at this time for the treatment of tuberculosis^14,19,21,33.

Isoniazid is the most potent bactericidal drug available and is particularly effective against actively growing organisms. It is relatively non-toxic, easily administered, and inexpensive. The usual adult dosage of three to five milligrams per kilogram of body weight a day penetrates well into all bodily fluids and cavities. Hepatic toxicity is a major side effect. Patients who have a history of excessive alcohol consumption or hepatitis infection have an increased probability of isoniazid hepatotoxicity. Peripheral neuropathy caused by the interference of isoniazid with the metabolism of pyridoxine is uncommon at this dosage, although it may be a substantial problem in persons with conditions that predispose them to neuropathy, such as diabetes or alcoholism. Ten milligrams of pyridoxine a day should be given in conjunction with isoniazid.

Rifampin and pyrazinamide are the most effective sterilizing drugs, and they are specifically effective against bacilli that are dormant and undergo periodic bursts of activity. Rifampin is bactericidal for Mycobacterium tuberculosis at the usual adult dosage of ten milligrams per kilogram of body weight a day. The most common adverse reaction is gastrointestinal upset, but mild jaundice may also occur. Pyrazinamide is bactericidal for Mycobacterium tuberculosis as well. The dosage is generally twenty to twenty-five milligrams per kilogram of body weight a day.

Ethambutol is bacteriostatic. Retrobulbar neuritis is the most frequent and serious adverse effect, with symptoms including blurred vision, central scomata, and red-green color blindness. This complication is dosage-related, and red-green color discrimination and visual acuity tests should be administered before and during treatment. The dosage is fifteen to twenty-five milligrams per kilogram of body weight a day to a maximum of 1.2 grams a day, and administration is discontinued at two to three months.

Streptomycin is bactericidal but must be given parenterally. The usual adult dosage is fifteen to twenty milligrams per kilogram of body weight to a maximum of one gram per day. Almost all streptomycin is excreted by the kidneys and therefore it must be used with extreme caution in patients who have renal insufficiency. The most common serious adverse effect is ototoxicity, usually resulting in vertigo. The total cumulative dose should not be more than 120 grams. The potential for increased toxicity, especially in elderly individuals and in immunocompromised patients who are taking other chemotherapeutic agents, must always be kept in mind.

Other drugs that are especially useful in the face of multiple-drug-resistant organisms include ethionamide, cycloserine, kanamycin, capreomycin, and para-aminosalicylic acid. All of these drugs may be more toxic and less well tolerated than the first-line agents. Before they are used, we recommend consultation with a specialist in the treatment of tuberculosis.

The quinolones (ciprofloxacin and ofloxacin) have an in vitro activity against Mycobacterium tuberculosis, but resistance frequently develops and their use is limited. The macrolide antibiotics also demonstrate in vitro activity against Mycobacterium, especially against the Mycobacterium avium complex; however, they do not appear to have substantial activity against Mycobacterium tuberculosis.

The current recommendation for treatment of adults who have musculoskeletal tuberculosis, with or without infection with the human immunodeficiency virus, is 300 milligrams of isoniazid a day, 600 milligrams of rifampin a day, and twenty to thirty milligrams of pyrazinamide per kilogram of body weight a day (discontinued after two months)^5,9,19,20,65. Ethambutol (or streptomycin for children who are too young to be monitored for visual acuity) should be included in the initial regimen until the results of drug-susceptibility studies are available, unless there is little possibility of drug resistance, such as when there is less than 4 per cent primary resistance to isoniazid in the community and the patient has had no previous treatment with antituberculosis medications, is not from a country with a high prevalence of drug resistance, and has had no known exposure to drug-resistant tuberculosis^5,9,19,65. Ten milligrams of pyridoxine a day is given as prophylaxis against possible isoniazid-induced neuropathy.

The optimum duration of treatment has been an issue of considerable debate, and much of the information now available concerns the treatment of pulmonary disease. The short-course regimens (six or nine months) may not be applicable to extrapulmonary tuberculosis, specifically those with osseous involvement. We recommend that treatment be continued for a minimum of twelve months for osteoarticular involvement, extending to perhaps eighteen months for certain problems. If isoniazid resistance is demonstrated, rifampin and ethambutol should be continued for a minimum of twelve months.

Children should be managed essentially the same as adults, with the use of appropriately adjusted doses of the drugs⁹.

Tuberculosis that is resistant to multiple drugs—that is, resistant to at least isoniazid and rifampin—presents a difficult treatment problem. Treatment must be individualized, and consultation with an expert in tuberculosis is strongly recommended.

Osteoarticular Tuberculosis

Tuberculosis of bones and joints often presents as gradually worsening arthritis. This often involves a cold abscess, with or without drainage. Systemic and pulmonary symptoms are frequently absent, and the differential diagnosis must include other possible causes of septic osteoarticular disease, inflammatory arthritis, and possibly internal derangement of the joint. Osteoarticular tuberculosis rarely involves more than one joint, and this may help to differentiate tuberculous septic arthritis from other types of polyinflammatory disease. Generally, the radiographic changes of tuberculosis that affect large weight-bearing joints are slow to develop compared with those of pyogenic infections, and a reduction in the joint space is often a late occurrence³⁶. Any synovial space, bursa, or tendon sheath may be infected.

Magnetic resonance imaging generally shows large intra-articular effusions, periarticular osteoporosis, and gross thickening of the synovial membrane⁴. A differential diagnosis between tuberculosis and pyogenic arthritis is difficult, and an accurate diagnosis usually requires biopsy of synovial tissue and demonstration of the appropriate organisms on either histological study or culture. Aspiration of synovial fluid is often insufficient to make a diagnosis, and culture specimens from draining sinuses are usually contaminated with other organisms²⁷.

Tuberculous infections of the joints are usually more flagrant and clinically obvious in adults than in children. As a consequence, the diagnosis of osteoarticular tuberculosis in children is often more difficult and a delay in treatment is not uncommon.

Pathological fractures are rare but can develop in or adjacent to a tuberculous joint³⁷. They can be treated with standard techniques of fracture management, including internal fixation. We recommend appropriate and prolonged chemotherapy at the time of treatment of the fracture, combined with radical débridement and drainage of abscesses.

Operative Treatment

Knee

In the early stages of disease of the knee, before there is substantial loss of bone or cartilage, operative intervention is necessary only to drain large abscesses and to obtain synovial tissue for biopsy^2,39. Synovectomy is rarely indicated in the early stages, and a prolonged course of adequate chemotherapy generally sterilizes the joint. We use external immobilization only to correct fixed deformities, as we believe that early motion is necessary after adequate débridement. The long-term prognosis depends on the extent of the disease at the time of the initial presentation and the adequacy and duration of chemotherapy³⁹.

Arthrodesis and joint replacement have been advocated in the later stages, when there is loss of the joint space and osseous architecture^8,39,91. We recommend arthrodesis of the knee in young patients who have severe destruction of the joint, marked loss of soft-tissue stabilizers, and possible deficiencies in soft-tissue coverage. Because tuberculosis rarely involves multiple joints, the patient is usually able to compensate with motion of other joints. We recommend radical débridement of all avascular tissue and the juxtaposition of viable cancellous bone at the time of the arthrodesis.

Total knee arthroplasty after chemotherapeutic sterilization of a tuberculous knee joint is gaining in popularity^22,40. At a minimum, three months of adequate chemotherapy is required to sterilize a joint, and only then if all clinical and laboratory criteria suggest that the joint is adequately sterile, should an arthroplasty be considered. The arthroplasty must be followed by a prolonged period of antituberculous chemotherapy⁹². In a study of twenty-two patients who had had a total knee arthroplasty for tuberculous arthritis, Kim reported a reactivation of infection in three who had not had preoperative chemotherapy⁴⁰. He thought that a quiescent period of at least one year was necessary before proceeding with total knee arthroplasty. We concur that, in the appropriate patient who has active infection, a two or three-stage total knee arthroplasty, combined with adequate chemotherapy, is an excellent procedure if the facilities are available. Eskola et al.²² had similar findings and also stressed the importance of chemotherapy.

Hip

Tuberculosis of the hip in adults presents a wide spectrum of disease. In the early acute phase, with preservation of the joint space and osseous architecture, a biopsy is often required to make the diagnosis and to decompress the joint. If the articular cartilage and osseous architecture are preserved, these patients do well after a prolonged course of chemotherapy^1,38. Excision arthroplasty remains an acceptable alternative in some patients who have increased destruction of bone and loss of articular cartilage. Unfortunately, these patients have an altered gait pattern, marked pain, and shortening of the limb, but the disease can usually be eradicated when an arthroplasty is combined with adequate chemotherapy.

In patients who have evidence of destruction of the joint and loss of articular cartilage, arthrodesis remains an effective treatment. The indications depend on the patient's occupation, the cultural milieu, and the expectations of the patient and his or her family. For patients who have complete destruction of the femoral head, arthrodesis from the trochanter to the iliac wing can be performed with use of standard internal fixation devices.

Low-friction arthroplasty has been recommended as treatment for tuberculous septic arthritis of the hip^{21,41,42,44,47}. The longest duration of follow-up of which we are aware was in a report by Kim et al., who followed sixty patients for eight to thirteen years⁴². They commented on difficulty with exposure related to contracted scar tissue and difficulty with wound-healing. Of the sixty patients, three had reactivation of the tuberculosis, which was thought to be related to failure of the patients to comply with the postoperative chemotherapeutic regimen. Total hip replacement without cement has also been used with apparently excellent results²³, but once again active chemotherapy must be maintained for a prolonged period of time.

Our preferred method of treatment for active tuberculosis of the hip in which the joint space has been destroyed consists of a radical decompression, drainage of the abscess, and removal of all avascular tissue. This is followed by high-dose antituberculosis chemotherapy with multiple drugs until there is clinical and hematological evidence that the disease has been eradicated. The second stage of treatment involves either an arthrodesis of the hip in a young patient or a hip arthroplasty without cement in an older patient. Again, these are followed by a course of adequate chemotherapy.

Spinal Tuberculosis

Spinal tuberculosis presents a series of interrelated problems. The disease begins in the anteroinferior portion of the vertebral body and tends to spread beneath the anterior longitudinal ligament to involve adjacent vertebral bodies. Narrowing of the disc space occurs as a late phenomenon when destruction of cancellous bone on both sides of a disc allows the disc to herniate into the affected vertebral body or bodies. Osseous infarction and osteonecrosis may lead to a decrease in vertebral height and may be accompanied by paravertebral and possibly epidural formation of an abscess⁸⁹. Because the anterior portion of a vertebral body is involved and the posterior portion is rarely involved, a sharp kyphosis may occur, and even with resolution of the acute infectious process the kyphosis may continue to cause anterior compression of the cord and late neurological sequelae⁸⁹. Multiple sites in the spine may be involved simultaneously. In our experience, 5 per cent of patients who were seen with neurological impairment had no obvious discernible vertebral lesion; epidural abscess, severe arachnoiditis, and intradural tuberculomas accounted for these neurological lesions⁸⁹. Spinal tuberculosis has often been called the great imitator because its radiographic appearance may mimic other pathological conditions affecting the spine. Consequently, other diagnoses must be kept in mind, even in areas endemic for tuberculosis.

Computer-assisted tomography and magnetic resonance imaging disclose morphological abnormalities related to vertebral and disc infection. Computer-assisted tomography is of great value in the delineation of encroachment of the spinal canal by posterior extension of inflammatory tissue, bone, or disc material⁶⁰; in the guidance of biopsies; and in the planning of operative procedures.

Magnetic resonance imaging is most useful for delineating soft-tissue masses in both the sagittal and the coronal plane and for indicating the extent of disease and the spread of tuberculous debris under the anterior and posterior ligaments. The subligamentous spread of a paraspinal mass and the involvement of multiple contiguous bones strongly suggest infection, but there is no pathognomonic finding on magnetic resonance imaging that reliably differentiates tuberculosis from other spinal infections or from a possible neoplasm^3,48. Magnetic resonance imaging may be of greatest value in the evaluation of intramedullary lesions and isolated extradural disease. Intramedullary lesions include tuberculomas; spinal cord cavitation; spinal cord edema; and possibly unsuspected, non-contiguous lesions throughout the spine. Intraspinal lesions seem to be more common in tropical countries, specifically India, and are relatively rare in the West³². Magnetic resonance imaging also assists in the differentiation of compression of the spinal cord by granulation material from compression by hard material such as bone or disc.

Despite newer modalities of imaging of the spine and spinal cord, the cytological diagnosis of vertebral tuberculosis depends on the procurement of adequate tissue for histological examination and culture. In settings where the disease is rare or where there is a possibility of multiple-drug-resistant organisms, computer-guided fine-needle aspiration and biopsy should be the initial invasive diagnostic procedure⁵³. Aspiration of a paravertebral abscess is safe and very effective in rapidly establishing the diagnosis. Sputum, urine, and tissue from other involved sites should also be cultured⁷³. In one study, fine-needle aspiration biopsy was effective in the diagnosis of nine of eleven patients who had a positive culture⁵⁰. We have also found core needle biopsy to be useful in the diagnosis of vertebral tuberculosis.

Treatment

Our recommendations for the management of patients who have acute spinal tuberculosis depend on the availability of appropriate facilities and trained personnel. The absolute indications for operative intervention are a marked neurological deficit, especially if it is related to severe kyphosis or retropulsed bone or disc in the neural canal; large abscesses in a patient in whom respiratory obstruction has developed; a neurological deficit that has worsened despite adequate chemotherapy; and continuing progression of kyphosis or instability despite adequate chemotherapy. Relative indications for operative intervention are related to the inability to obtain adequate material for culture by other means, neurological deficits in patients for whom prolonged bed rest may lead to other problems, persistence of pain or spasticity caused by a demonstrable mechanical block, or pain related to spinal instability where spontaneous fusion has not occurred^{34,51,78,85-87}.

There is continuing debate regarding the necessity of operative intervention in the spine to decrease the occurrence of late kyphosis compared with the use of chemotherapy alone. The Medical Research Council of the United Kingdom, in multiple studies throughout the world, has shown that drug therapy alone can be effective treatment for tuberculosis of the spine, with an acceptable resolution of neurological sequelae and prevention of substantial progression of the kyphosis in most patients^25,52,54,78. The likelihood of progressive kyphosis developing depends on the site of the disease, the age of the patient, and the number of vertebrae involved^62-64,90. In areas of the spine where there is pre-existing kyphosis, such as the thoracic spine and the cervical-thoracic junction, the likelihood of a late increase in kyphosis is high. Conversely, in areas of the spine where there is pre-existing lordosis, such as the lumbar spine, late kyphosis is a less common and serious occurrence.

In children, tuberculosis of the spine generally involves the osseous tissue of the vertebrae and not the cartilaginous growth plate. After the disease has been controlled with medication, the end plates can continue to grow and, as a consequence, approximately 50 per cent of children have a reduction in the kyphotic deformity with time. This is particularly true for children who are less than five years old^63,64. Similarly, if only one or two vertebrae are involved, the probability of progressive kyphosis is low, but the likelihood increases if more than two vertebrae are involved and varies with the number of vertebrae involved^64,83.

In the Medical Research Council study in Hong Kong, proponents of operative intervention in tuberculous spondylitis showed that, at a mean of fifteen years, radical débridement and arthrodesis produced better results than radical débridement alone^51,78,86. There was no increase in kyphosis in the patients who had had débridement and arthrodesis, but there was a mean increase of approximately 12 degrees in both the patients who had had débridement alone and those who had had non-operative treatment. Similarly, children who had been followed for a mean of seventeen years in a study by the Medical Research Council in Hong Kong had a better end result after radical débridement and arthrodesis than after débridement alone^85,87. The children who had had débridement and arthrodesis had an over-all correction of kyphosis, whereas those who had had débridement alone had an increase in the kyphosis. Both procedures had similar results regarding recovery from neurological deficits and relief of pain. Consequently, if an operation is indicated, decompression should be combined with arthrodesis.

Spinal tuberculosis primarily involves the anterior vertebral structures and, therefore, anterior operative approaches are usually recommended. From an anterior approach, abscesses can be evacuated, all avascular material can be excised, and anterior decompression of the spinal cord can be performed safely. Tissue is easily obtained for histological study and culture, and the kyphosis can be corrected or at least stabilized with use of autogenous bone graft.

Indications for posterior operative approaches to the spine, although rare, include situations in which the posterior spinal elements are more involved than the anterior ones or those in which both the anterior and the posterior elements are involved and posterior stabilization is needed before anterior decompression and arthrodesis is performed⁸².

At specific anatomical sites, such as the occipito-cervical junction, we recommend a transoral biopsy to decompress and obtain tissue for culture, followed by a posterior stabilization procedure⁴⁶. A posterior operative approach is also indicated in patients with a stable spine who have slight deformity or involvement of the bone but who also have intramedullary or possibly extramedullary tuberculomas and an epidural abscess.

The costotransversectomy approach has somewhat limited applications but is useful for the drainage of a large abscess in the thoracic spine in a patient who is not medically fit to have a formal thoracotomy. It is also useful in patients with substantial thoracic kyphosis, in whom an anterior transthoracic approach is technically difficult. Often, a costotransversectomy allows sufficient exposure for removal of an anterior bone bar and for limited bone-grafting.

Posterior stabilization with various metallic implants does not appear to increase the risk of prolonged infection, and it may allow patients to be mobilized earlier with less need for postoperative immobilization⁵⁶. We do not have any personal experience with attempted closed reduction of kyphotic deformities, although the method appears to be simple and inexpensive and does not require special expertise in terms of anesthetic management. However, prolonged immobilization of the spine is required after the manipulation²⁶.

Late-onset paraplegia occurs in patients in whom a marked kyphotic deformity has developed and who have had prolonged anterior impingement on the cord by a sharp osseous kyphosis or possibly from constriction caused by fibrosis around the neural elements. Osseous and fibrous material anterior to the spinal cord must be removed very carefully and, because of the severe angulation of the osseous deformity, the operative procedure is technically difficult. Even with adequate decompression, recovery of normal neurological function is prolonged and usually incomplete. This is probably because of vascular insufficiency in addition to marked myelopathic changes in the cord³⁴.

Some authors⁴⁹ have recommended the use of anterior vascularized bone grafts followed by posterior osteotomy and arthrodesis with the placement of internal fixation devices to correct a fixed kyphosis. The concept involves elongation of the anterior column with a simultaneous shortening of the posterior column. This procedure is costly and requires longer hospitalization and a great deal of operative expertise; however, it may produce a more cosmetically acceptable and better balanced spine.

It has been reported that there is no substantial progression of kyphosis after a short anterior arthrodesis of the spine in children, even if the procedure is performed at an early age⁸⁸. Specifically, there does not appear to be posterior spinal overgrowth contributing to a recurrence of the deformity. Consequently, prophylactic posterior arthrodesis of the spine in a growing child is not indicated after an adequate anterior decompression and arthrodesis⁸⁸.

It has been our experience that adequate anterior decompression reliably leads to resolution of paraplegia if it is performed within nine months after the onset of acute paraplegia. Operative decompression can lead to substantial neurological improvement if it is performed between nine and eleven months after the onset of paraplegia, although spasticity will probably remain and the degree of resolution is usually incomplete. Ho and Leong found that operative intervention more than one year after the onset of paraplegia rarely eradicated neurological deficits completely, and an operation performed after two years rarely led to substantial recovery of spinal cord function³⁴.

Tuberculosis remains a major public-health problem in most of the world. It is axiomatic that, before the disease can be treated, it must be recognized and, before it can be recognized, it must be considered a diagnostic possibility.

References

Adjrad, A., and |and |Martini, N.: L'osteó-arthrite tuberculeuse de la hanche chez l'adulte. Internat. Orthop.,1: 227-233, 1987.1227 1987 [CrossRef]

Aguirre, N.; Bago, J.; and |and |Martin, N.: Tuberculosis of the knee. Surgical or conservative treatment?. Acta Orthop. Belgica,55: 22-25, 1989.5522 1989

Ahmadi, J.; Bajaj, A.; Destian, S.; Segall, H. D.; and |and |Zee, C. S.: Spinal tuberculosis: atypical observations at MR imaging. Radiology,189: 489-493, 1993.189489 1993 [PubMed]

Araki, Y.; Tsukaguchi, I.; Shino, K.; and |and |Nakamura, H.: Tuberculous arthritis of the knee: MR findings [letter]. AJR: Am J. Roentgenol.,160: 664, 1993.160664 1993

Bass, J. B., Jr.; Farer, L. S.; Hopewell, P. C.; O'Brien, R.; Jacobs, R. F.; Ruben, F.; Snider, D. E., Jr.; Thornton, G.; American Thoracic Society; and |and the |Centers for Disease Control and Prevention: Treatment of tuberculosis and tuberculosis infection in adults and children. Am. J. Respir. and Crit. Care Med.,149: 1359-1374, 1994.1491359 1994

Bloch, A. B.; Rieder, H. L.; Kelly, G. D.; Cauthen, G. M.; Hayden, C. H.; and |and |Snider, D. E.: The epidemiology of tuberculosis in the United States. Sem. Respir. Infect.,4: 157-170, 1989.4157 1989

Brudney, K., and |and |Dobkin, J.: Resurgent tuberculosis in New York City. Human immunodeficiency virus, homelessness, and the decline of tuberculosis control programs. Am. Rev. Respir. Dis.,144: 745-749, 1991.144745 1991 [PubMed]

Carnesale, P. G.: Arthrodesis of lower extremity and hip. In Campbell's Operative Orthopaedics, edited by A. H. Crenshaw. Ed. 8, pp. 317-352. St. Louis, Mosby-Year Book, 1992.

Centers for Disease Control: Tuberculosis and human immunodeficiency virus infection: recommendations of the Advisory Committee for the Elimination of Tuberculosis (ACET). Morbid. and Mortal. Weekly Rep.,38: 236-250, 1989.38236 1989

Centers for Disease Control: A strategic plan for the elimination of tuberculosis in the United States. Morbid. and Mortal. Weekly Rep.,38 (Supplement 3): 1-25, 1989.38 (Supplement 3)1 1989

Centers for Disease Control: Prevention and control of tuberculosis in facilities providing long-term care to the elderly. Recommendation of the Advisory Committee for Elimination of Tuberculosis (ACET). Morbid. and Mortal. Weekly Rep.,39(R-10): 7-13, 1990.39(R-10)7 1990

Centers for Disease Control: Summary of notifiable diseases, United States. 1991. Morbid. and Mortal. Weekly Rep.,41: 1-63, 1992.411 1992

Centers for Disease Control: Tuberculosis Statistics in the United States 1989. HHS Publication Number (CDC)91-8322. Atlanta, Public Health Service, 1991.

Centers for Disease Control: Guidelines for preventing the transmission of Mycobacterium tuberculosis in health-care facilities, 1994. Morbid. and Mortal. Weekly Rep.,43(RR-13): 4-68, 1994.43(RR-13)4 1994

Centers for Disease Control: Essential components of a tuberculosis prevention and control program. Recommendations of the Advisory Council for the Elimination of Tuberculosis. Morbid. and Mortal. Weekly Rep.,44(RR-11): 1-16, 1995.44(RR-11)1 1995

Colditz, G. A.; Brewer, T. F.; Berkey, C. S.; Wilson, M. E.; Burdick, E.; Fineberg, H. V.; and |and |Mosteller, F.: Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. J. Am. Med. Assn.,271: 698-702, 1994.271698 1994 [CrossRef]

Comstock, G. W.: Tuberculosis—a bridge to chronic disease epidemiology. Am. J. Epidemiol.,124: 1-16, 1986.1241 1986 [PubMed]

Daniel, T. M., and |and |DeBanne, S. M.: The serodiagnosis of tuberculosis and other mycobacterial diseases by enzyme-linked immunosorbent assay. Am. Rev. Respir. Dis.,135: 1137-1151, 1987.1351137 1987 [PubMed]

De Cock, K. M.; Grant, A.; and |and |Porter, J. D.: Preventive therapy for tuberculosis in HIV-infected persons: international recommendations, research, practice. Lancet,345: 833-836, 1995.345833 1995 [PubMed][CrossRef]

Department of Health and Human Services, Centers for Disease Control and Prevention: Dosage recommendations for the initial treatment of TB among children and adults [table]. Fed. Reg.,58(195): 52832, 1993.58(195)52832 1993

Department of Health and Human Services, Centers for Disease Control and Prevention: Guidelines for preventing the transmission of Mycobacterium tuberculosis in health-care facilities, 1994. Fed. Reg.,59(208): 54242-54303, 1994.59(208)54242 1994

Eskola, A.; Santavirta, S.; Konttinen, Y. T.; Tallroth, K.; and |and |Lindholm, S. T.: Arthroplasty for old tuberculosis of the knee. J. Bone and Joint Surg.,70-B(5): 767-769, 1988.70-B(5)767 1988

Eskola, A.; Santavirta, S.; Konttinen, Y. T.; Tallroth, K.; Hoikka, V.; and |and |Lindholm, S. T.: Cementless total replacement for old tuberculosis of the hip. J. Bone and Joint Surg.,70-B(4): 603-606, 1988.70-B(4)603 1988

Farer, L. S.; Lowell, A. M.; and |and |Meador, M. P.: Extrapulmonary tuberculosis in the United States. Am. J. Epidemiol.,109: 205-217, 1979.109205 1979 [PubMed]

Fifth Report of the Medical Research Council Working Party on Tuberculosis of the Spine: A five-year assessment of controlled trials of in-patient and out-patient treatment and of plaster-of-Paris jackets for tuberculosis of the spine in children on standard chemotherapy. Studies in Masan and Pusan, Korea. J. Bone and Joint Surg.,58-B(4): 399-411, 1976.58-B(4)399 1976

Galvagno, S.,, and |and |Meo, G.: Treatment of Pott's paraplegia in a rural African hospital. Eastern African Med. J..,68: 124-129, 1991.68124 1991

Gillespie, W. J.; Mayo, K. M.; and |and |Johnstone, V.: Skeletal tuberculosis in New Zealand since the introduction of chemotherapy. Australian and New Zealand J. Surg.,57: 727-732, 1987.57727 1987 [CrossRef]

Glassroth, J.: Diagnosis of tuberculosis. In Tuberculosis. A Comprehensive International Approach, pp. 149-165. Edited by L. B. Reichman and E. S. Hershfield. New York, Marcel Dekker, 1993.

Griffiths, D.: Orthopaedic tuberculosis. British J. Hosp. Med.,14: 146-157, 1975.14146 1975

Grzybowski, S.: Ontario studies on tuberculin sensitivity. Canadian J. Pub. Health,56: 181-192, 1965.56181 1965

Grzybowski, S.: Tuberculosis in the Third World [editorial]. Thorax,46: 689-691, 1991.46689 1991 [PubMed][CrossRef]

Gupta, R. K.; Gupta, S.; Kumar, S.; Kohli, A.; Misra, U. K.; and |and |Gujral, R. B.: MRI in intraspinal tuberculosis. Neuroradiology,36: 39-43, 1994.3639 1994 [PubMed][CrossRef]

Haas, D. W., and Des Prez, R. M.: Mycobacterium tuberculosis. In Mandell, Douglas and Bennett's Principles and Practice of Infectious Diseases, edited by G. L. Mandell, J. E. Bennett, and R. Dolin. Ed. 4, pp. 2213-2243. New York, Churchill Livingstone, 1995.

Ho, E. K., and Leong, J. C.: Tuberculosis of the spine. In The Pediatric Spine. Principles and Practice, pp. 837-849. Edited by S. L. Weinstein. New York, Raven Press, 1994.

Hoffman, E. B.; Crosier, J. H.; and |and |Cremin, B. J.: Imaging in children with spinal tuberculosis. A comparison of radiography, computed tomography and magnetic resonance imaging. J. Bone and Joint Surg.,75-B(2): 233-239, 1993.75-B(2)233 1993

Hsu, S. H.; Sun, J. S.; Chen, I. H.; and |and |Liu, T. K.: Reappraisal of skeletal tuberculosis: role of radiological imaging. J. Formosan Med. Assn.,92: 34-41, 1993.9234 1993

Jenyo, M. S., and |and |Komolafe, F.: Tuberculous pathological fracture of the femur in a 15-year-old boy. Pediat. Radiol.,16: 260-261, 1986.16260 1986 [PubMed][CrossRef]

Keers, R. Y.: Pulmonary Tuberculosis: a Journey down the Centuries. London, Bailliere Tindall, 1978.

Kerri, O., and |and |Martini, M.: Tuberculosis of the knee. Internat. Orthop.,9: 153-157, 1985.9153 1985 [CrossRef]

Kim, Y.-H.: Total knee arthroplasty for tuberculous arthritis. J. Bone and Joint Surg.,70-A: 1322-1330, Oct. 1988.70-A1322 1988

Kim, Y.-H.; Han, D.-Y.; and |and |Park, B.-M.: Total hip arthroplasty for tuberculous coxarthrosis. J. Bone and Joint Surg.,69-A: 718-727, June 1987.69-A718 1987

Kim, Y. Y.; Ko, C. U.; Ahn, J. Y.; Yoon, Y. S.; and |and |Kwak, B. M.: Charnley low friction arthroplasty in tuberculosis of the hip. An eight to 13-year follow-up. J. Bone and Joint Surg.,70-B(5): 756-760, 1988.70-B(5)756 1988

Kochi, A.: The global tuberculosis situation and the new control strategy of the World Health Organization [editorial]. Tubercle,72: 1-6, 1991.721 1991 [PubMed][CrossRef]

Laforgia, R.; Murphy, J. C. M.; and |and |Redfern, T. R.: Low friction arthroplasty for old quiescent infection of the hip. J. Bone and Joint Surg.,70-B(3): 373-376, 1988.70-B(3)373 1988

Leff, A.; Lester, T. W.; and |and |Addington, W. W.: Tuberculosis. A chemotherapeutic triumph but a persistent socioeconomic problem. Arch. Intern. Med.,139: 1375-1377, 1979.1391375 1979 [PubMed][CrossRef]

Lifeso, R.: Atlanto-axial tuberculosis in adults. J. Bone and Joint Surg.,69-B(2): 183-187, 1987.69-B(2)183 1987

Lin, E.; Oliver, S.; Caspi, I.; Ezra, E.; Bubis, J. J.; and |and |Nerubay, J.: Hip arthroplasty in quiescent mycobacterial infection of hip. Orthop. Rev.,15: 232-236, 1986.15232 1986 [PubMed]

Liu, G. C.; Chou, M. S.; Tsai, T. C.; Lin, S. Y.; and |and |Shen, Y. S.: MR evaluation of tuberculous spondylitis. Acta Radiol.,34: 554-558, 1993.34554 1993 [PubMed][CrossRef]

Louw, J. A.: Spinal tuberculosis with neurological deficit. Treatment with anterior vascularised rib grafts, posterior osteotomies and fusion. J. Bone and Joint Surg.,72-B(4): 686-693, 1990.72-B(4)686 1990

Masood, S.: Diagnosis of tuberculosis of bone and soft tissue by fine-needle aspiration biopsy. Diagn. Cytopathol.,8: 451-455, 1992.8451 1992 [PubMed][CrossRef]

Medical Research Council Working Party on Tuberculosis of the Spine: A controlled trial of anterior spinal fusion and débridement in the surgical management of tuberculosis of the spine in patients on standard chemotherapy. A study in Hong Kong. British J. Surg.,61: 853-866, 1974.61853 1974 [CrossRef]

Medical Research Council Working Party on Tuberculosis of the Spine: A controlled trial of debridement and ambulatory treatment in the management of tuberculosis of the spine in patients on standard chemotherapy. A study in Bulawayo; Rhodesia. J. Tropical Med. and Hygiene,77: 72-92, 1974.7772 1974

Mondal, A.: Cytological diagnosis of vertebral tuberculosis with fine-needle aspiration biopsy. J. Bone and Joint Surg.,76-A: 181-184, Feb. 1994.76-A181 1994

Ninth Report of the Medical Research Council Working Party on Tuberculosis of the Spine: A 10-year assessment of controlled trials of inpatient and outpatient treatment and of plaster-of-Paris jackets for tuberculosis of the spine in children on standard chemotherapy. Studies in Masan and Pusan, Korea. J. Bone and Joint Surg.,67-B(1): 103-110, 1985.67-B(1)103 1985

Nocera, R. M.; Sayle, B.; Rogers, C.; and |and |Wilkey, D.: Tc-99m MDP and indium-111 chloride scintigraphy in skeletal tuberculosis. Clin. Nucl. Med.,8: 418-420, 1983.8418 1983 [PubMed][CrossRef]

Oga, M.; Arizono, T.; Takasita, M.; and |and |Sugioka, Y.: Evaluation of the risk of instrumentation as a foreign body in spinal tuberculosis. Clinical and biologic study. Spine,18: 1890-1894, 1993.181890 1993

Okwera, A.; Eriki, P. P.; Guay, L. A.; Ball, P.; and |and |Daniel, T. M.: Tuberculin reactions in apparently healthy HIV-seropositive and HIV-seronegative women—Uganda. Morbid. and Mortal. Weekly Rep.,39: 638-646, 1990.39638 1990

Onorato, I. M., and |and |McCray, E.: Prevalence of human immunodeficiency virus infection among patients attending tuberculosis clinics in the United States. J. Infect. Dis.,165: 87-92, 1992.16587 1992 [PubMed][CrossRef]

Pandey, J., and |and |Talib, V. H.: Laboratory diagnosis of tuberculosis: use of ELISA and PCR. Indian J. Pathol. and Microbiol.,36: 512-518, 1993.36512 1993

Pantongrag-Brown, L., and |and |Suwanwela, N.: CT findings in tuberculous spondylitis. Australasian Radiol.,36: 4-7, 1992.364 1992 [CrossRef]

Pott, P.: Remarks on that kind of palsy of the lower limbs, which is frequently found to accompany a curvature of the spine, and is supposed to be caused by it, together with its method of cure. Med. Classics,1: 281-297, 1936-1937.1281 1936-1937

Pun, W. K.; Chow, S. P.; Luk, K. D. K.; Cheng, C. L.; Hsu, L. C. S.; and |and |Leong, J. C. Y.: Tuberculosis of the lumbosacral junction. Long-term follow-up of 26 cases. J. Bone and Joint Surg.,72-B(4): 675-678, 1990.72-B(4)675 1990

Rajasekaran, S., and |and |Shanmugasundaram, T. K.: Prediction of the angle of gibbus deformity in tuberculosis of the spine. J. Bone and Joint Surg.,69-A: 503-509, April 1987.69-A503 1987

Rajasekaran, S., and |and |Soundarapandian, S.: Progression of kyphosis in tuberculosis of the spine treated by anterior arthrodesis. J. Bone and Joint Surg.,71-A: 1314-1323, Oct. 1989.71-A1314 1989

Reichman, L. B., and Hershfield, E. S. [editors]: Tuberculosis. A Comprehensive International Approach. New York, Marcel Dekker, 1993.

Reimer, L. G.: Laboratory detection of mycobacteremia. Clin. Lab. Med.,14: 99-105, 1994.1499 1994 [PubMed]

Rieder, H. L.; Snider, D. E., Jr.; and |and |Cauthen, G. M.: Extra-pulmonary tuberculosis in the United States. Am. Rev. Respir. Dis.,141: 347-351, 1990.141347 1990 [PubMed]

Rieder, H. L.; Cauthen, G. M.; Comstock, G. W.; and |and |Snider, D. E., Jr.: Epidemiology of tuberculosis in the United States. Epidemiol. Rev.,11: 79-98, 1989.1179 1989 [PubMed]

Rieder, H. L.; Cauthen, G. M.; Bloch, A. B.; Cole, C. H.; Holtzman, D.; Snider, D. E., Jr.; Bigler, W. J.; and |and |Witte, J. J.: Tuberculosis and acquired immunodeficiency syndrome—Florida. Arch. Intern. Med.,149: 1268-1273, 1989.1491268 1989 [PubMed][CrossRef]

Rooney, J. J., Jr.; Crocco, J. A.; Kramer, S.; and |and |Lyons, H. A.: Further observations on tuberculin reactions in active tuberculosis. Am. J. Med.,60: 517-522, 1976.60517 1976 [PubMed][CrossRef]

Ryan, M. R., and |and |Murray, P. R.: Laboratory detection of anaerobic bacteremia. Clin. Lab. Med.,14: 107-117, 1994.14107 1994 [PubMed]

Salfinger, M., and |and |Morris, A. J.: The role of the microbiology laboratory in diagnosing mycobacterial diseases. Am. J. Clin. Pathol.,101 (4 Supplement 1): 6-S13, 1994.101 (4 Supplement 1)6 1994

Sant, M., and |and |Bajaj, H.: Role of histopathology in the diagnosis of tuberculous synovitis. J. Indian Med. Assn.,90: 263-264, 1992.90263 1992

Schluger, N. W., and |and |Rom, W. N.: Current approaches to the diagnosis of active pulmonary tuberculosis. Am. J. Respir. and Crit. Care Med.,149: 264-267, 1994.149264 1994

Seibert, F. B.: The isolation and properties of the purified protein derivative of tuberculin. Am. Rev. Tubercul.,30: 713-720, 1934.30713 1934

Selwyn, P. A.; Hartel, D.; Lewis, V. A.; Schoenbaum, E. E.; Vermund, S. H.; Klein, R. S.; Walker, A. T.; and |and |Friedland, G. H.: A prospective study of the risk of tuberculosis among intravenous drug users with human immunodeficiency virus infection. New England J. Med.,320: 545-550, 1989.320545 1989 [CrossRef]

Sharrard, W.: Infections of bones and joints. In Paediatric Orthopaedics and Fractures, edited by W. J. W. Sharrard. Ed. 3, pp. 1247-1284. Oxford, Blackwell Scientific, 1993.

Sixth Report of the Medical Research Council Working Party on Tuberculosis of the Spine: Five-year assessments of controlled trials of ambulatory treatment, debridement and anterior spinal fusion in the management of tuberculosis of the spine. Studies in Bulawayo (Rhodesia) and in Hong Kong. J. Bone and Joint Surg.,60-B(2): 163-177, 1978.60-B(2)163 1978

Small, P. M.; Schecter, G. F.; Goodman, P. C.; Sande, M. A.; Chaisson, R. E.; and |and |Hopewell, P. C.: Treatment of tuberculosis in patients with advanced human immunodeficiency virus infection. New England J. Med.,324: 289-294, 1991.324289 1991 [CrossRef]

Starke, J. R., and |and |Taylor-Watts, K. T.: Tuberculosis in the pediatric population of Houston, Texas. Pediatrics,84: 28-35, 1989.8428 1989 [PubMed]

Stuart, D.: Local osteo-articular tuberculosis complicating closed fractures. Report of two cases. J. Bone and Joint Surg.,58-B(2): 248-249, 1976.58-B(2)248 1976

Travlos, J., and |and |Du Toit, G.: Brief reports. Spinal tuberculosis: beware the posterior elements. J. Bone and Joint Surg.,72-B(4): 722-723, 1990.72-B(4)722 1990

Tuli, S.: Tuberculosis of the Skeletal System, p. 268. New Delhi, Jaypee Brothers Medical, 1991.

Upadhyay, S. S.; Saji, M. J.; Sell, P.; and |and |Yau, A. C. M.: The effect of age on the change in deformity after radical resection and anterior arthrodesis for tuberculosis of the spine. J. Bone and Joint Surg.,76-A: 701-708, May 1994.76-A701 1994

Upadhyay, S. S.; Saji, M. J.; Sell, P.; Sell, B.; and |and |Hsu, L. C. S.: Spinal deformity after childhood surgery for tuberculosis of the spine. A comparison of radical surgery and debridement. J. Bone and Joint Surg.,76-B(1): 91-98, 1994.76-B(1)91 1994

Upadhyay, S. S.; Saji, M. J.; Sell, P.; Sell, B.; and |and |Yau, A. C.: Longitudinal changes in spinal deformity after anterior spinal surgery for tuberculosis of the spine in adults. A comparative analysis between radical and debridement surgery. Spine,19: 542-549, 1994.19542 1994 [PubMed][CrossRef]

Upadhyay, S. S.; Sell, P.; Saji, M. J.; Sell, B.; Yau, A. C.; and |and |Leong, J. C.: 17-year prospective study of surgical management of spinal tuberculosis in children. Hong Kong operation compared with debridement surgery for short- and long-term outcome of deformity. Spine,18: 1704-1711, 1993.181704 1993 [PubMed][CrossRef]

Versfeld, G. A., and |and |Solomon, A.: A diagnostic approach to tuberculosis of bones and joints. J. Bone and Joint Surg.,64-B(4): 446-449, 1982.64-B(4)446 1982

Weaver, P., and |and |Lifeso, R. M.: The radiological diagnosis of tuberculosis of the adult spine. Skel. Radiol.,12: 178-186, 1984.12178 1984 [CrossRef]

Webb, G. B.: Tuberculosis, pp. 20-24. New York, Hoeber, 1936.

Wolfgang, G. L.: Tuberculosis joint infection. Clin. Orthop.,136: 257-263, 1978.136257 1978 [PubMed]

Wolfgang, G. L.: Tuberculosis joint infection following total knee arthroplasty. Clin. Orthop.,201: 162-166, 1985.201162 1985 [PubMed]

Topics