NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Chronic Conditions (UK). Tuberculosis: Clinical Diagnosis and Management of Tuberculosis, and Measures for Its Prevention and Control. London: Royal College of Physicians (UK); 2006. (NICE Clinical Guidelines, No. 33.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Tuberculosis

Tuberculosis: Clinical Diagnosis and Management of Tuberculosis, and Measures for Its Prevention and Control.

Show details

7Management of non-respiratory tuberculosis

7.1. Meningeal tuberculosis

7.1.1. Clinical introduction

Tuberculous meningitis occurs when there is blood-borne spread of the TB bacteria to the brain. In the days before treatment was available this usually occurred within 12 months of the original (primary) infection.106 It is sometimes part of a more widespread blood-borne dissemination, with chest X-ray patterns typical of miliary tuberculosis.107 It can present with systemic features if due to miliary disease, or more local central nervous system signs if limited to the brain. Unlike acute bacterial meningitis with, for example, the meningococcus, the onset of TB meningitis is insidious over a few weeks. In infants there may be non-specific symptoms such as not feeding or a failure to thrive. There can be headache and vomiting, then increasing drowsiness, and localised neurological signs such as cranial nerve palsies or hemiparesis, progressing to coma.

Clinically, the meningitis is classified according to the following stages:

  • stage I: no clouding of consciousness or focal neurological signs
  • stage II: clouding of consciousness and/or focal neurological signs
  • stage III: coma.108

The diagnosis is supported by lumbar puncture suggesting CSF changes: a low glucose, raised protein and a lymphocyte dominant pattern of white blood cells. Diagnosis is confirmed by demonstrating M. tuberculosis on microscopy or culture of the CSF, or demonstrating M. tuberculosis DNA by PCR testing. TB meningitis may be accompanied by tuberculomas, inflammatory masses in the brain, which can either be present at diagnosis on CT brain scan or develop during treatment.109 Although only approximately 100 cases of TB meningitis occur in England and Wales each year, this form of TB has a high morbidity and mortality when compared to nearly all other forms of non-respiratory tuberculosis.110 Disability and death can still occur despite early diagnosis and appropriate treatment.

7.1.2. Methodological introduction: duration of treatment in adults

Studies were included where the majority of patients were adults (16 years of age and over) and where a modern drug treatment regimen was used to treat TB meningitis. Thus, treatment had to include at least isoniazid, rifampicin and pyrazinamide.

Two cohort studies performed in Turkey111 and Thailand112 were identified which compared different durations of treatment for TB meningitis. Two case series performed in Thailand113 and Ecuador114 and one treatment arm of a study performed in India115 were also considered. All of the studies were completed more than 15 years ago and were excluded due to methodological limitations.

There is a lack of high-level evidence in this area. There are no RCTs which compare different durations of treatment for TB meningitis and there are no good quality cohort studies. This seems to be due to the relative rarity of the condition (small patient numbers in studies) and the associated high mortality and morbidity. The studies that do exist are plagued by a number of methodological problems including small sample size, a lack of generalisability due to completion in developing countries, patients in variable stages of clinical severity, problems with definitive diagnosis of TB meningitis, concurrent use of glucocorticoid therapy and a lack of inferential statistics. Due to the low quality of the studies in this area, it was not possible to write evidence statements.

7.1.3. Methodological introduction: duration of treatment in children

One systematic review of case series studies116 was identified. This compared studies of six months treatment duration for TB meningitis with those of more than six months treatment duration. Nine studies were included, four of which were in the six months duration group113,114,117,118 and five in the more than six months duration group.111,119–123 Approximately 75% of the patients included were children. The review had several methodological limitations and due to these issues, the studies included in this review and performed in children were assessed separately. These were two studies performed in India,120,122 one in Thailand117 and another in South Africa;118 however all of these studies were excluded on the basis of methodological limitations.

Within the area of treatment duration for TB meningitis in children (as with adults) there is a lack of high-level evidence. Studies had similar methodological limitations to those in adult populations. Additionally, the issue of generalisability of results to the UK was even more marked as one study reported high levels of childhood malnutrition.122 Due to the low quality of the studies in this area, it was not possible to write evidence statements.

7.1.4. Methodological introduction: glucocorticoids as an adjunct to antituberculous drugs

A Cochrane systematic review124 compared the effects of glucocorticoids in combination with anti-TB treatment with anti-TB treatment alone in patients with TB meningitis. The review consisted of six RCTs125–130 and was methodologically sound and hence it could technically be given a grading of 1++/1+. However, the methodological limitations of individual studies contained within the review meant that there was insufficient robust data from which to derive evidence statements. The authors of the review concluded that

‘adjunctive steroids might be of benefit in patients with TB meningitis. However, existing studies are small, and poor allocation concealment and publication bias may account for the positive results found in this review’.

In the study steroids were associated with fewer deaths (RR 0.79, 95%CI 0.65 to 0.97) and a reduced incidence of death and severe residual disability (RR 0.58, 95%CI 0.38 to 0.88). Subgroup analysis suggested an effect on mortality in children (RR 0.77, 95%CI 0.62 to 0.96) but the results in a smaller number of adults were inconclusive (RR 0.96, 95%CI 0.50 to 1.84).

Another systematic review131 was also appraised; however this was excluded due to methodological limitations.

One further RCT was identified.132 This was a very high-quality study performed in Vietnam in adults and included patients who were HIV positive.

Studies were excluded where glucocorticoids were administered intrathecally as this rarely occurs due to the necessity of a lumbar puncture. This was the approach taken in the Cochrane systematic review.124

Due to the methodological issues associated with the studies in the Cochrane review124 there is no sound evidence available for the use of corticosteroids in children with TB meningitis. There is also no compelling evidence in this area for HIV-positive patients.

7.1.5. Evidence statements

Mortality and severe residual disability

In a RCT performed in Vietnam132 in TB meningitis patients over 14 years of age, adjunctive treatment with dexamethasone was associated with a reduced risk of death (RR 0.69, 95%CI 0.52 to 0.92, p=0.01). It was not however associated with a significant reduction in the proportion of severely disabled patients or in the proportion of patients who either died or were severely disabled after nine months.132 (1++)

Disease severity and HIV status

The treatment effect of adjunctive dexamethasone was consistent across subgroups that were defined by:

  • disease severity grade (stratified RR of death, 0.68, 95%CI 0.52 to 0.91, p=0.007)132
  • HIV status, although the reduction in the risk of death was not significant (the number of HIV-infected patients was too small to confirm or reject confidently a treatment effect).132 (1++)

Adverse effects

Significantly fewer serious adverse events occurred in the dexamethasone group than in the placebo group (26 of 274 patients vs. 45 of 271 patients, p=0.02). In particular eight severe cases of hepatitis (one fatal) occurred in the placebo group and none occurred in the dexamethasone group (p=0.004).132 (1++)

7.1.6. From evidence to recommendations

The evidence base in this area is hampered by the difficulty of recruiting patients for participation in studies. Mostly the existing studies included people following a presumptive diagnosis with few positive culture confirmations.

There is no evidence to support treatment durations of less than 12 months, but all the evidence on duration has some methodological limitations. Given the serious risk of disability and mortality, the advice given in the 1998 BTS guidelines68 remains appropriate.

There is also no evidence to inform the choice of drugs. Caution is required with ethambutol in unconscious patients, streptomycin should be avoided in pregnancy if at all possible (fetal 8th nerve damage) and there is potential teratogenicity with ethionamide and prothionamide.133

The important factor in drug choice was penetration into CSF. Ethionamide, isoniazid, prothionamide and pyrazinamide achieve best penetration. Rifampicin is less good in this regard, and ethambutol and streptomycin only penetrate into CSF if the meninges are inflamed.

Given the potential severe effects of neurological damage arising from TB meningitis, and the strong evidence in adults from the Vietnam study132 supporting additional glucocorticoids, this guideline recommends them. There is no reason to give a high-dose glucocorticoid to most patients, and the GDG reached a consensus on reviewing treatment response after 2–4 weeks with a view to starting to withdraw the glucocorticoid as soon as it is safe to do so.

RECOMMENDATIONS

R26.

Patients with active meningeal TB should be offered:

  • a treatment regimen, initially lasting for 12 months, comprising isoniazid, pyrazinamide, rifampicin and a fourth drug (for example, ethambutol) for the first two months, followed by isoniazid and rifampicin for the rest of the treatment period [D(GPP)]
  • a glucocorticoid at the normal dose range
    • – adults equivalent to prednisolone 20–40 mg if on rifampicin, otherwise 10–20 mg [A]
    • – children equivalent to prednisolone 1–2 mg/kg, maximum 40 mg [D(GPP)]

with gradual withdrawal of the glucocorticoid considered, starting within 2–3 weeks of initiation. [D(GPP)]

R27.

Clinicians prescribing treatment for active meningeal TB should consider as first choice:

  • a daily dosing schedule [B]
  • using combination tablets. [D]

Cross-referring:

For details of standard drug treatment, see section 6.1.

For details of managing drug-resistant TB, see chapter 9.

7.2. Peripheral lymph node tuberculosis

7.2.1. Clinical introduction

Lymph node tuberculosis is an important form of non-respiratory tuberculosis accounting for nearly half of all non-respiratory sites26,27 (see epidemiology in Appendix E). Since non-respiratory disease is found less commonly in white UK-born people than in others, who now make up nearly 70% of all cases in the UK, the number of cases of lymph node disease seen is rising.

Trials by the BTS and its predecessors with regimens of 18 months,134 nine months134,135 and six months duration,135–137 all showed a significant proportion of cases (up to 40%) to have residual nodes at the end of treatment, and up to 10% at 30 month follow-up. Sometimes new nodes and occasionally sinuses develop during treatment and/or during follow-up. Nearly all of these events are thought to be immunologically mediated responses to residual tuberculo-proteins, and not failure to respond to treatment or relapses. When cultured there is seldom evidence of bacteriological activity.

7.2.2. Methodological introduction

A meta-analysis138 of studies of varying designs compared six-month treatment regimens with nine month regimens in people with peripheral lymph node TB. However, this was excluded due to methodological limitations.

Two RCTs identified in the meta-analysis were assessed seperately.137 One UK trial comparing six months vs. nine months daily treatment was reported in two papers firstly as preliminary results136 and then follow-up results at 30 months.137 The other trial performed in Hong Kong139 compared six months and nine months thrice-weekly treatment, however this was excluded due to limitations in methodology.

There was a lack of high-quality comparative studies in this area, thus only one has been included as evidence.136,137

7.2.3. Evidence statements

A UK RCT136,137 of patients with peripheral lymph node TB compared two nine-month drug regimens (2HRE/7HR and 2HRZ/7HR) and one six-month regimen (2HRZ/4HR). Of those patients seen at 30 months (85%), there was no statistically significant difference between the groups in terms of reported residual measurable nodes, relapse, enlargement of existing nodes, development of new glands or sinuses or the need for new operative procedures. Aspiration after commencement of treatment was performed in eight patients: seven on the 2HRE/7HR regimen and the other on 2HRZ/4HR (2HRE/7HR versus all HRZ, p=0.005). (1+)

7.2.4. From evidence to recommendations

There was little evidence to guide the GDG in more practical issues, but it was felt that treatment should be stopped at the end of the regimen regardless of the appearance of new nodes, residual nodes or sinuses draining.

One study136,137 of six months vs. nine-months treatment duration shows equivalence for fully susceptible organisms. However, this trial used a three-drug initial phase (2RHZ), which may be inadequate in view of current drug resistance rates,140 and the isoniazid resistance rate of 12% in the trial.136,137 The standard six-month, four-drug regimen is therefore recommended.

Drug treatment is still required even if a gland has been surgically removed, because of the possibility of residual local and distal TB foci. Surgical excision biopsy for histology and culture is advised if pus cannot be aspirated from a gland. Fine needle aspiration does not give adequate samples for TB culture.

RECOMMENDATIONS

R28.

For patients with active peripheral lymph node tuberculosis, the first choice of treatment should:

  • be the standard recommended regimen (see section 6.1 for further details) [B]
  • use a daily dosing schedule [B]
  • include combination tablets. [D]
R29.

Patients with active peripheral lymph node TB who have had an affected gland surgically removed should still be treated with the standard recommended regimen. [D(GPP)]

R30.

Drug treatment of peripheral lymph node TB should normally be stopped after six months, regardless of the appearance of new nodes, residual nodes or sinuses draining during treatment. [D(GPP)]

Cross-referring:

For details of standard drug treatment, see section 6.1.

For details of managing drug-resistant TB, see chapter 9.

7.3. Bone and joint tuberculosis: drug treatment

7.3.1. Clinical introduction

Spinal tuberculosis accounts for approximately half of all the sites of bone and joint tuberculosis seen in England and Wales.22,26,27 As such it is an important subset of non-respiratory disease, and one which can sometimes have significant morbidity because of spinal cord compression from extradural abscess and/or vertebral collapse. For these reasons, the GDG considered the evidence base on the medical management of spinal tuberculosis as a proxy for the management of the many possible joint sites, in which separate drug trials have not been conducted.

7.3.2. Methodological introduction

Three RCTs were identified which compared different durations of treatment in those with TB of the spine.

A Hong Kong study141 with fourteen years of follow-up compared six, nine and eighteen months of treatment in those who had undergone radical anterior resection with bone grafting. The results of this trial (without the 18 month arm) were also reported at five years in a paper that presented the results of two further trials at five years in Madras and Korea,142 which both compared six months of treatment with nine months in patients who had not received surgery. The Madras trial was also reported with follow-up at ten years.143 The Korean trial142 was excluded due to a number of methodological limitations.

These trials were all originally commenced in the 1960s and 1970s by the British Medical Research Council (MRC) and although they subscribed to the methodological standards of the time, they do not include all patients in the analyses in the groups to which they were originally allocated (ie an intention to treat analysis). In line with NICE guidance in circumstances where an intention to treat analysis has not been used and there is little evidence available, these studies have been evaluated as if they were non-randomised cohort studies.

These studies did not use the standard, four-drug initial treatment regimens currently used in the UK and none of the studies reported blinding methods.

7.3.3. Evidence statements

In a Hong Kong study142 at five years follow-up, all analysed patients who had received radical anterior resection with bone grafting and a six- or nine-months treatment regimen of isoniazid, rifampicin and streptomycin (except one in each group) had favourable status at five years, and most had achieved favourable status by three years. (Favourable status was defined as full physical activity with radiographically quiescent disease, with neither sinuses nor clinically evident abscesses and with no myelopathy with functional impairment and no modification of the allocated regimen). (2+)

In the Hong Kong study141 at 14 years follow-up, clinical outcomes were similar in the six-, nine-and 18-month treatment regimen groups. One patient in the six months group had minor motor deficits whereas one patient in the 18 months group had partial unilateral sensory deficits. No patients had bladder or bowel disturbances at final follow-up and there was no recurrence or reactivation of tuberculosis in either group. Additionally there were no statistically significant differences in the change in mean angle of deformity between the groups and most side effects occurred early in treatment and were not related to duration of treatment. (2+)

In a study in Madras142 of patients who received treatment (isoniazid and rifampicin) without surgery for six or nine months, 91% in the six-month group and 98% in the nine-month group had a favourable status at five years (using the same definition as the Hong Kong study142). At ten years143 there was no significant difference in favourable status, or occurrence of complete bony fusion. The angle of kyphosis increased in both regimens with no significant difference between groups; however, in patients less than 15 years of age with angle of kyphosis >30°, the mean increase by ten years was 30°, compared with 10° in those >15 years (p=0.001). (2++)

7.3.4. From evidence to recommendations

A number of trials were conducted in association with the British MRC between the 1960s and 1980s in Korea, India and Hong Kong, designed according to the standards of the time. Whilst they did not use intention to treat analysis, these studies on six, nine and 18 months of treatment, with extensive follow-up of up to 10 years in some cases, show that six months duration of treatment performed just as well as longer regimens. The GDG agreed that these results are likely to be applicable to other forms of bone and joint tuberculosis, and accordingly recommended the standard six-month, four-drug regimen.

The GDG acknowledged the risk of CNS involvement via the spinal cord, and recommended scans to check for any patient with neurological signs or symptoms. The was no evidence to guide a choice of either CT or MR scanning.

RECOMMENDATIONS

R31.

The standard recommended regimen (see section 6.1 for details) should be planned and started in people with:

  • active spinal TB [B]
  • active TB at other bone and joint sites. [C]
R32.

Clinicians prescribing treatment for active bone and joint tuberculosis should consider as first choice:

  • a daily dosing schedule [B]
  • using combination tablets. [D]

See section 6.1 for details.

R33.

CT or MR scan should be performed on patients with active spinal TB who have neurological signs or symptoms. If there is direct spinal cord involvement (for example, a spinal cord tuberculoma), management should be as for meningeal TB (see section 7.1). [D(GPP)]

Cross-referring:

For details of managing drug-resistant TB, see chapter 9.

7.4. Bone and joint tuberculosis: routine therapeutic surgery

7.4.1. Clinical introduction

From before the age of anti-tuberculosis treatment, immobilisation and bed rest were thought to be important for bone and joint tuberculosis. This view continued after the development of anti-tuberculosis drugs and into the time when shorter durations of treatment with newer drugs were available. A series of studies by the MRC, commencing in 1965, showed the respective roles of anti-tuberculosis treatment and other routine management measures in spinal tuberculosis. Studies in Korea found no benefit from routine bed rest,144,145 or of a plaster jacket during therapy,145,146 and in Rhodesia no benefit from routine initial debridement of lesions.147 Prior to the introduction of rifampicin, trials of radical anterior fusion showed mixed results.142,148–151 The advent of rifampicin led to further trials on the use of anterior spinal fusion in conjunction with short-course treatment regimens.

7.4.2. Methodological introduction

Two RCTs were identified which compared surgery and drug treatment for those with TB of the spine with drug treatment alone.

A study in Rhodesia149 compared debridement and drug treatment with drug treatment alone but was excluded for methodological issues.

A Madras study, reporting at five142 and ten years,143 compared radical resection with bone grafting plus six months’ treatment with isoniazid and rifampicin with just six or nine months’ treatment with isoniazid and rifampicin.

The Madras trial, whilst in line with the methodological standards at the time it was commenced, did not include all patients in the analysis in the group to which they had been originally allocated (ie an intention to treat analysis). In line with NICE guidance in circumstances where an intention to treat analysis has not been used and there is little evidence available, these studies have been evaluated as if they were non-randomised cohort studies. Furthermore, it should be noted that a two-drug regimen would not now be used in the UK as standard therapy.

7.4.3. Evidence statements

At five years,142 radical resection with bone grafting in addition to six-months treatment regimen (with isoniazid and rifampicin) showed no benefit in status (favourable status was defined as no sinus or clinically evident abscess, no myelopathy and no modification of allocated regimen, no limitation of physical activity due to spinal lesion and radiologically quiescent disease) compared to six- or nine-months treatment regimen alone. (2++)

Whilst at ten years,143 the surgery and six-months treatment regimen was less effective in terms of favourable status than the nine-month treatment regimen alone (p=0.03), the difference being due to surgical complications. However, patients in the surgery and anti-tuberculosis drug treatment group had a faster resolution of sinuses and/or clinically evident abscesses (p<0.001 at two months) and a lower incidence (p=0.03) than those in the anti-tuberculosis drug treatment only groups. There was no significant differences found between the groups in terms of occurrence of complete bony fusion or angle of kyphosis. There were four deaths associated with spinal tuberculosis (all within the first six months and all in the surgery and anti-tuberculosis drug treatment group). Three died in the postoperative period and the other had complications of postoperative paraplegia. (2++)

7.4.4. From evidence to recommendations

Although the GDG concluded that the evidence showed no additional advantage of routinely carrying out anterior spinal fusion over standard chemotherapy, the recommendations for spinal surgery cannot be extrapolated to bone/joint tuberculosis at other sites.

Aspiration of paraspinal abscesses and/or biopsy from spinal sites may be needed for the diagnosis of TB, which is different from routine anterior fusion. Forms of surgery such as aspiration or arthroscopy of joints may be needed to obtain material for histology and culture by which to make the diagnosis of tuberculosis in bone/joint sites other than the spine.

RECOMMENDATIONS

R34.

In patients with spinal TB, anterior spinal fusion should not be performed routinely. [B]

R35.

In patients with spinal TB, anterior spinal fusion should be considered if there is spinal instability or evidence of spinal cord compression. [D(GPP)]

7.5. Pericardial tuberculosis

7.5.1. Clinical introduction

TB of the pericardium accounts for less than 4% of non-respiratory TB in England and Wales,140 but is potentially important because of the possibilities of cardiac tamponade and constrictive pericarditis, which have a mortality and morbidity higher than most other forms of extrapulmonary TB.

The presence of a pericardial effusion may require aspiration by pericardiocentesis for diagnosis, repeated during treatment. Similarly, considerable pericardial thickening, with or without fluid, may require surgery with pericardectomy or a pericardial window, which is a major invasive intervention. Additional glucocorticoids tailing from the equivalent of prednisolone 60 mg/day have been recommended in the UK,68 following studies in Transkei, South Africa, where this form of active tuberculosis was particularly common,152,153 which appeared to show reduced morbidity and mortality.

7.5.2. Methodological introduction

A Cochrane systematic review154 attempted to compare six-month anti-tuberculosis drug treatment regimens with regimens of nine months or more in people with tuberculous pericarditis. The Cochrane review search did not identify any RCTs which compared anti-tuberculosis drug regimens of these different durations.

No further studies were identified which compared six months of treatment with longer treatment durations, thus it was not possible to write evidence statements on the duration of treatment for TB pericarditis.

Two systematic reviews, which considered the effectiveness of glucocorticoids in addition to drug treatment in patients with TB pericarditis were identified. A Cochrane systematic review154 considered this issue in addition to a number of other treatment issues in TB pericarditis (treatment duration, pericardial drainage and pericardectomy) whilst a review by the same authors, published elsewhere, only considered the issue of additional glucocorticoids for TB pericarditis.155 The same four studies were included in both reviews152,153,156,157 and the results presented and the publication year were the same.

The two RCTs included in these reviews by Strang152,153 have since been reported at ten years.158 Results from this new report which now includes an intention to treat analysis, along with the two other RCTs identified in the systematic reviews, have thus been considered separately. One of these studies was excluded on methodological grounds.156 The other study included HIV-positive patients only.157

TB pericarditis is relatively rare and so it is difficult to find enough patients to study; furthermore, it is also difficult to diagnose. For example, the study in HIV patients157 was small (N=58) and the TB diagnosis was confirmed by culture in only 38% of the participants.

7.5.3. Evidence statements

The results of RCTs performed in Transkei, South Africa, comparing prednisolone to placebo in pericardial effusion and pericardial constriction patients with or without drainage are presented in the table below.158 Table 9 also includes the results of an RCT comparing prednisolone vs. placebo in HIV-positive pericardial effusion patients.157

Table 9. Summary of evidence for pericardial TB.

Table 9

Summary of evidence for pericardial TB.

7.5.4. From evidence to recommendations

The group were not aware of any further evidence on the treatment regimen and concluded that first-line treatment is with the standard six-month, four-drug regimen.

There are no comparative studies on which to base recommendations on the duration of treatment. Since this is a pauci-bacillary form of extrapulmonary disease by extrapolation from other forms of extrapulmonary disease with more evidence, a six-month duration of treatment is expected to be effective.

The GDG agreed that the RCT evidence157,158 strongly supported the use of glucocorticoids in adults with active pericardial tuberculosis and that they were also likely to be beneficial in children.

RECOMMENDATIONS

R36.

For patients with active pericardial TB, the first choice of treatment should:

  • be the standard recommended regimen (see section 6.1 for details) [B]
  • use a daily dosing schedule [B]
  • include combination tablets. [D]
R37.

In addition to anti-TB treatment, patients with active pericardial TB should be offered:

  • for adults, a glucocorticoid equivalent to prednisolone at 60 mg/day [A]
  • for children, a glucocorticoid equivalent to prednisolone 1 mg/kg/day (maximum 40 mg/day), with gradual withdrawal of the glucocorticoid considered, starting within two to three weeks of initiation. [D(GPP)]

Cross-referring:

For details of managing drug-resistant TB, see chapter 9.

7.6. Disseminated (including miliary) tuberculosis

7.6.1. Clinical introduction

In the 1997 guidance on notification, it was suggested that those with non-specific symptoms started on TB treatment should be described as having ‘cryptic disease’ with the term ‘cryptic miliary disease’ being reserved for those where the organism has been isolated from blood, from bone marrow or from multiple organ systems. In clinical texts there is usually a distinction between ‘classical miliary’ disease with the diffuse 1–2 mm uniform micronodular chest X-ray from acute haematogenous spread which may also involve other organs, including the CNS, and ‘cryptic miliary’ where the patient may have fever but few localising signs. The data collection form for enhanced TB surveillance gives possible sites of TB, including miliary and cryptic disseminated. Cryptic disseminated is defined as ‘systemic illness without localising features’.

These different labels for forms of what is essentially blood-borne spread of tuberculosis can cause confusion. Essentially, blood-borne spread may or may not be accompanied by chest X-ray or high-resolution CT changes. Such blood-borne spread often also causes significant liver function derangement because of diffuse liver involvement. This is a serious form of TB with a significant morbidity and mortality, so the risks of treating the disease with drugs which have a low incidence of hepatic side effects (3%), are much less than those of leaving the patient inadequately treated. The meninges are also not infrequently involved as part of the blood-borne spread, with up to 30% having clinical or lumbar puncture evidence of such involvement.140 The detection of CNS disease is important because of the longer duration of treatment required for CNS involvement.

7.6.2. Methodological introduction

One retrospective study159 where patients with disseminated TB received three different durations of treatment was identified, however this was excluded due to small sample size (N=6).

No other comparative studies were found, hence it was not possible to write evidence statements.

7.6.3. From evidence to recommendations

No data were found to inform recommendations. It is noted that all sites outside the CNS for which data exist show adequate response to a six-month, four-drug initial treatment regimen, but that six-month regimens have not been shown to be adequate for those with CNS involvement (see section 7.1).

Exclusion of CNS disease is important, by CT scan, MRI or lumbar puncture, so that the correct duration of treatment is applied.

Abnormal liver function should not prevent or delay the commencement of TB treatment, which usually causes improvement in liver function abnormalities due to the disease itself.

RECOMMENDATIONS

R38.

For patients with disseminated (including miliary) TB, the first choice of treatment should:

  • be the standard recommended regimen (see section 6.1 for details) [B]
  • use a daily dosing schedule [B]
  • include combination tablets. [D]
R39.

Treatment of disseminated (including miliary) TB should be started even if initial liver function tests are abnormal. If the patient’s liver function deteriorates significantly on drug treatment, advice on management options should be sought from clinicians with specialist experience of these circumstances. [D(GPP)]

R40.

Patients with disseminated (including miliary) TB should be tested for CNS involvement by:

  • brain scan (CT or MRI) and/or lumbar puncture for those with CNS signs or symptoms
  • lumbar puncture for those without CNS signs and symptoms.

If evidence of CNS involvement is detected, treatment should be the same as for meningeal TB (see section 7.1). [D(GPP)]

Cross-referring:

For details of managing drug-resistant TB, see chapter 9.

7.7. Other sites of infection

7.7.1. From evidence to recommendations

There is no evidence base available to derive recommendations for other sites of infection. However, as the pathogen and its drug susceptibility is the same, treatment has generally been given with the same regimen as is used for respiratory tuberculosis. The GDG’s clinical experience supported this and hence the recommendation below is extrapolated from the evidence base for respiratory tuberculosis, and other non-respiratory sites.

RECOMMENDATION

R41.

For patients with:

  • active genitourinary TB, or
  • active TB of any site other than:
    • – respiratory system
    • – CNS (typically meninges)
    • – peripheral lymph nodes
    • – bones and joints
    • – pericardium
    • – disseminated (including miliary) disease

the first choice of treatment should:

  • be the standard recommended regimen (see section 6.1 for details) [B]
  • use a daily dosing schedule [B]
  • include combination tablets. [D]

Cross-referring:

For details of managing drug-resistant TB, see chapter 9.

Copyright © 2006, Royal College of Physicians of London.

All rights reserved. No part of this publication may be reproduced in any form (including photocopying or storing it in any medium by electronic means and whether or not transiently or incidentally to some other use of this publication) without the written permission of the copyright owner. Applications for the copyright owner’s written permission to reproduce any part of this publication should be addressed to the publisher.

Bookshelf ID: NBK45806

Views

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...