Hagan and Nathani Critical Care 2013, 17:240

http://ccforum.com/content/17/5/240

REVIEW

Clinical review: Tuberculosis on the intensive care

unit
Guy Hagan and Nazim Nathani*

(HDU)/ICU setting for a variety of reasons. Mortality for

Abstract patients admitted with active TB and respiratory failure
Rates of tuberculosis (TB) are increasing in most west requiring mechanical ventilation is poor, with reported
European nations. Patients with TB can be admitted in-hospital mortalities of 33 to 67% [3-5]. It is a treatable
to an ICU for a variety of reasons, including respiratory disease and a proactive approach with timely intervention
failure, multiorgan failure and decreased consciousness is required in the treatment of critically ill TB patients, as
associated with central nervous system disease. TB delays to starting therapy can be associated with worse
is a treatable disease but the mortality for patients survival [6]. TB patients on ICU present special challen-
admitted with TB to an ICU remains high. Management ges, including obtaining microbiological confirmation,
challenges exist in establishing a prompt diagnosis providing effective anti-tuberculosis treatment (ATT)
and administering effective treatment on the ICU with with poor absorption and high rates of organ dysfunction,
potentially poor gastric absorption and high rates of and apparent deterioration of TB during appropriate
organ dysfunction and drug toxicity. In this review treatment (paradoxical reactions). This review describes
reasons for ICU admission, methods of achieving the reasons patients with TB may be on a HDU/ICU and
a confident diagnosis through direct and inferred aims to discuss management with particular relevance to
methods, anti-tuberculosis treatment (including the intensivist and the ICU environment.
steroid and other adjuvant therapies) and specific
management problems with particular relevance to Reasons and outcome for ICU admission in
the intensivist are discussed. The role of therapeutic tuberculosis patients
drug monitoring, judicious use of alternative regimes in TB usually affects the lungs but may present acutely in
the context of toxicity or organ dysfunction and when almost any organ system and mimic other infectious or
to suspect paradoxical tuberculosis reactions are also non-infectious processes [7]. Most studies of TB patients
covered. Diagnostic and therapeutic algorithms are on ICU involve patients with pulmonary TB [6,8].
proposed to guide ICU doctors in the management of Common reasons for admission are acute respiratory
this sometimes complicated disease. failure [3,4], and development of multi-organ failure
(MOF) [9]; high rates of acute respiratory distress syn-
drome (ARDS) are seen [10], although post mortem
Introduction studies suggest that confluent tuberculous bronchopneu-
Tuberculosis (TB) remains a significant public health monia may mimic ARDS [3]. Neurological deterioration
problem worldwide, with an estimated 8.7  million cases due to tuberculosis meningitis (TBM) is a rarer but
and 1.4  million deaths from it in 2011 [1]. Immigration important reason for ICU admission. Presentations of TB
patterns, the HIV pandemic and iatrogenic immuno- are myriad and more unusual reasons for ICU admission
suppression have made TB a more common disease in exist (some of these are listed in Table 1).
western European nations; TB rates in the UK have
increased over the past 2  years [2], with a rate of Respiratory failure due to pulmonary tuberculosis
14.4  cases/100,000 population. Patients with suspected Advanced pulmonary TB can cause respiratory failure;
or actual tuberculosis may be on a high dependency unit the incidence of respiratory failure in hospitalised pul-
monary TB patients is about 1.5% [3]. The majority of
*Correspondence: n.nathani@nhs.net
patients will have abnormal chest X-rays (CXRs) includ-
Department of Respiratory Medicine, Birmingham City Hospital, Dudley Road, ing cavitatory lesions and bilateral infiltrates [8] (Figure 1).
Birmingham B18 7QH, West Midlands, England, United Kingdom Contributing factors such as bacterial pneumonia,
chronic obstructive pulmonary disease and malignancy
© 2010 BioMed Central Ltd © 2013 BioMed Central Ltd may be present in about 72% of cases [11].
Hagan and Nathani Critical Care 2013, 17:240 Page 2 of 10
http://ccforum.com/content/17/5/240

Table 1. Other potential reasons for admission of a patient

with tuberculosis to an ICU
Presentation Possible cause
Massive haemoptysis Rasmussen aneurysm
Cardiogenic shock Massive pericardial effusion
Known tuberculosis patient Post-thoracic surgery
electively admitted to ICU
Liver failure Drug reaction
Renal failure Drug reaction (usually rifampicin)
Disseminated intravascular coagulation Miliary TB
Pituitary apoplexy/stroke mimic Cerebral tuberculoma
Airway obstruction Laryngeal/retropharyngeal TB
TB, tuberculosis.

Patients with TB requiring ICU care may have high

rates of co-morbidities and ICU related complications. In
one German study [8] 65.5% of patients had deranged
liver function, 12.1% chronic pancreatitis, 8.6% chronic
renal failure and 6.9% HIV co-infection. ICU related
complications were also common, with nosocomial
pneumonia in 67.2% patients, pneumothorax in 13.8%,
ARDS in 12.1%, acute renal failure in 12.1% and MOF in
Figure 1. Male patient who presented with type I respiratory
3.4%. Rates of co-existing extra-pulmonary TB can be up failure. Sputum grew fully sensitive Mycobacterium tuberculosis.
to 19 to 22% [6,8].
Delay in appropriate treatment due to lack of early
recognition of TB may result in progressive multi-organ CXR may be initially normal for the first few weeks of the
involvement and ICU admission [6] and an important disease. Mortality is about 25% overall [16] but assumed
opportunity between hospital admission and ICU pre- near 100% if untreated. Patients with miliary TB may be
sentation may exist. Given the high ARDS rates in more likely to develop ARDS than patients with isolated
ventilated patients with TB, standard mechanical ventila- pulmonary TB and MOF may account for most of the
tion strategies to reduce ARDS may be appropriate, mortality of patients with miliary TB on ICU [10]. A
including lower tidal volumes and a conservative fluid retrospective study by Silva and colleagues [4], which
strategy [12]. included high rates of extrapulmonary TB (about 63%)
Mortality is high for patients with active TB and and HIV seropositivity, described MOF in over 80% of
respiratory failure; a Canadian study found a significantly patients.
higher in-hospital mortality of 69% for patients requiring Disseminated TB can rarely lead to a septic shock with
mechanical ventilation for TB in comparison to ARDS of MOF presentation, sometimes described as Landouzy
any cause (56%) and nontuberculous pneumonia (36%) septicaemia after the original report [18]. This is usually
requiring mechanical ventilation [13]. Risk factors for described in association with HIV infection [19], and
mortality include older age, nosocomial pneumonia, TB recently associated with monoclonal antibodies used in
destroyed lung, MOF, a duration of symptoms of more the treatment of rheumatological disease [17], but can
than 4 weeks, and an APACHE-II score >20 [14,15]. also occur in patients with no obvious risk factors [20].
Disseminated TB may also cause adrenal insufficiency
Miliary tuberculosis [21], which should be considered in the context of refrac-
Miliary TB is a form of TB where there is haematological tory hypotension or hyponatraemia.
dissemination from focal infection into the blood, leading
to seeding of multiple organs with TB bacilli. A minority Tuberculosis meningitis and other central nervous system
of patients may present with symptoms of less than four tuberculosis
weeks duration. An underlying predisposing condition Tuberculosis of the central nervous system (CNS) occurs
such as diabetes or steroid use will be present in about a in approximately 1% of TB cases, and includes TBM and
third of patients [16] and immunosuppression due to cerebral tuberculomas. It carries a high mortality and is
HIV or iatrogenic reasons may also contribute [17]. The probably fatal if untreated. It is the reason for about 6 to
Hagan and Nathani Critical Care 2013, 17:240 Page 3 of 10
http://ccforum.com/content/17/5/240

18% [4,6] of TB related ICU admissions. In a French case pulmonary TB submit at least two sputum specimens for
series of TBM admitted to ICU [22], nearly all patients microscopic examination [32]. Culture not only confirms
were admitted due to a falling conscious level, 75% the diagnosis but also provides drug susceptibility testing.
required mechanical ventilation and 33% underwent a In the UK, 8.4% of isolates were resistant to any first line
neurosurgical procedure; 1  year mortality was 65%. drug, and 1.6% of isolates were multi-drug resistant TB
Clinical symptoms range from an acute illness mimicking (MDR TB) [2]; rates of MDR TB are much higher in some
bacterial meningitis to a non-specific illness of fever and countries.
headache, with a smaller proportion having cranial nerve New liquid culture techniques should give a culture
palsies [22]. The mean duration of symptoms ranges from result in 2 to 4 weeks [31]. An initial targeted sample for
12 to 29 days in most series and approximately one-third auramine or Ziehl-Neelsen stain is important as this
will have symptoms lasting less than a week [7]. The would help confirm TB in the appropriate clinical
majority of the literature on CNS TB comes from the situation. There is little evidence specific to diagnosis in
paediatric age group. Cerebral tuberculomas can present an ICU setting. TB can present acutely and a high index
as seizures (focal and generalised), as well as with focal of suspicion should be had in the majority of ill patients
neurological signs; occasionally a tuberculous brain with an abnormal CXR, particularly if risk factors for TB
abscess can form that may require neurosurgical inter- are present. An incidental finding of acid fast bacilli
vention as well as ATT [23]. (AFB) in sputum in a patient not suspected to have TB
As well as a high mortality, TBM poses two additional may be due to non-tuberculous mycobacteria (NTM) and
particular challenges for the intensivist relevant to correlation with other diagnostic data such as radiology
neurocritical care - hydrocephalus and hyponatraemia. and immune status is advised; 35% of AFB seen in
Hydrocephalus is common and develops radiologically in sputum was due to NTM in one study [33]. NTM may
about 77% [24] of cases. It is usually due to communi- cause pulmonary disease especially in the immuno-
cating hydrocephalus associated with tuberculous exu- suppressed or may not be significant. Conversely, an
dates in the basal cisterns, but may be non-communi- AFB-positive specimen in a patient with a suggestive
cating in a smaller (17 to 25%) proportion of cases - for history, clinical features and radiology should be assumed
example, due to obstruction at the outlet foramen of the to be TB unless proved otherwise; TB PCR may have a
fourth ventricle or the cerebral aqueduct by oedema or a role if there is doubt.
tuberculoma [25]. Hydrocephalus may increase the intra- In patients who are expectorating, serial sputum samp-
cranial pressure, leading to reduced cerebral perfusion ling probably provides a yield similar to bronchoscopy
and ischaemia, and in more advanced cases cause brain [34]. Microbiological sampling in the non-expectorating
herniation. Non-communicating hydrocephalus usually patient may be carried out via bronchoscopy if the
requires a neurosurgical procedure such as a shunt patient is intubated. Transbronchial biopsy may provide
procedure or an endoscopic third ventriculostomy [26]. histology to enable rapid diagnosis and may increase the
Cases of communicating hydrocephalus should also be diagnostic yield [35]; this can carry a risk of complications
discussed with a neurosurgical centre. There may be a in the mechanically ventilated patient (mainly bleeding
role for external ventricular drainage in patients with a and pneumothorax [36]) but in selected patients the risk/
low Glasgow coma score and TBM [27]. Hyponatraemia benefit ratio may be favourable, particularly in patients
is common in TBM patients and is independently with diffuse lung shadowing or miliary disease. Special-
associated with a worse outcome [26]. It is multifactorial, ised procedures such as bronchoscopy through a non-
and the syndrome of inappropriate anti-diuretic hormone invasive ventilation mask may facilitate sampling in a
and cerebral salt wasting probably both play a role [28]. hypoxic non-intubated patient and this is carried out in
The best approach to TBM-associated hyponatraemia is our institution. Where facilities exist, induced sputum
uncertain, and hypertonic saline, fluid restriction, fludro- may have a role in some patients and has been shown to
cortisone and demeclocycline may all have a role have similar or better sensitivities to bronchoscopy (73
depending on the fluid state of the patient [26]. Other and 87% sensitivity, respectively, in smear-negative
aspects of neurocritical care may be relevant for TBM. patients [37]). No data on the utility of respiratory
These additional interventions may include intra-cranial secretions obtained at suction for diagnosis of TB on
pressure monitoring [25], a higher transfusion threshold ICU/HDU exist but transtracheal aspirates have shown a
[29], and control of fever [30]. high sensitivity (88%) in smear-negative patients [38]. If
there are signs and symptoms reinforced by appropriate
Diagnosis of tuberculosis investigations consistent with a TB diagnosis, treatment
Culture confirmation of tuberculosis should be obtained should be started without waiting for culture results, and
where possible [31]; the World Health Organisation continued even if subsequent culture results are negative
(WHO) recommends all patients suspected to have [31].
Hagan and Nathani Critical Care 2013, 17:240 Page 4 of 10
http://ccforum.com/content/17/5/240

Non-pulmonary samples yielded positive results in the majority of patients with

Extrapulmonary TB is common in ICU patients with miliary TB when it was carried out [16], although the
pulmonary TB and in the context of advanced immuno- place of bone marrow examination in the absence of
suppression associated with HIV [39]; in the latter, haematological abnormalities requires clarification.
visualised AFB may be due to NTM. Whilst culture is
central to confirmation of the mycobacterium, PCR for Interferon gamma release assays/nucleic acid amplification
TB may have a role in this patient population to help tests
establish a diagnosis [40]. Most forms of extrapulmonary The past decade has seen the development of interferon
TB have a lower bacterial load than pulmonary disease gamma release assays (IGRAs) in the diagnosis of latent
and histology/cytology such as pleural biopsies and TB. IGRAs work on the principle of measuring the
lymph node aspiration play a more prominent role in cytokine interferon gamma released from T cells in res-
diagnosis. A lymphocytic pleural effusion, caseating ponse to synthetic antigens that are also found in MTB
granulomas or granulomas with Langhan’s giant cells are and have an approximate 90% sensitivity and 99%
suggestive of TB [31], although other conditions may also specificity for diagnosis of latent TB [48]. These antigens
cause these histological changes; lymphoma is a main are absent from Mycobacterium bovis and most NTM so
differential for a lymphocytic effusion. A lymphocytic are not affected by prior Bacillus Calmette-Guérin
effusion in an ICU setting with a high index of suspicion vaccination or most NTM exposure.
for TB should trigger ATT. A small proportion (6.7%) of IGRAs cannot distinguish between latent and active TB
patients, especially if the duration of illness is short, may and therefore may be positive in an individual who has
have a predominately polymorphonuclear leucocytic latent TB but another cause for ICU admission. It is
pleural effusion [41]. Adenosine deaminase levels may recommended that IGRAs should not be used as a
have a role in conjuction with other pleural fluid results, routine diagnostic tool in active TB, although there may
although this is a controversial area [42]. Urine culture be a role in using IGRA with other complementary tests
for TB may be useful in miliary disease, with positive when TB is suspected, especially if samples are difficult
cultures in about 25 to 69% of cases [43,44]. Urine to obtain [49]. A negative test does not exclude active TB.
microscopy for AFB is not usually carried out due to high ICU admission has also been associated with a false
rates of NTM. negative IGRA [50] and so the role of IGRA in intensive
Gastric aspirates may be easy to obtain on an ICU. care patients is less clear. There is ongoing research into
There is little evidence for their utility in diagnosing TB next generation IGRAs and T-cell-based diagnostic
in adults, and none from an ICU setting. The presence of platforms that may overcome some of the current
AFB in a gastric aspirate may be due to atypical myco- limitations of IGRAs [51].
bacteria leading to a low specificity in some studies [45] Nucleic acid amplification tests (NAATs) tend not to be
but not others [46]. A culture rate of about 5% for Myco- routinely used, and the sensitivity and specificity can be
bacterium tuberculosis (MTB) has been described in highly variable compared with culture results [40]. A
gastric aspirates in patients with pulmonary TB [46]. negative result does not exclude TB and a positive result
The diagnosis of TBM is based on typical cerebrospinal does not give drug sensitivities. One exception is the
fluid (CSF) findings (high protein, low glucose and Xpert MTB/RIF probe, which has shown a 98.2% and
predominately lymphocytic CSF) and confirmation if 72.5% sensitivity for diagnosing TB in smear-positive and
possible of TB elsewhere in the body. There may be a role smear-negative patients, respectively [52]. The probe also
for repeat lumbar puncture in 48  hours in atypical CSF identifies mutations associated with rifampicin (R) resis-
findings. Culture of TB in CSF may occur in up to 80% of tance and may have a role where MDR TB is suspected;
cases if larger volumes of CSF (>6  ml) are submitted. for this reason WHO suggests this test as a follow on test
Repeated lumbar punctures may increase diagnostic yield for patients with smear-negative samples. British recom-
[26]. TBM is probably fatal if not treated and there should mendations for NAATs currently restrict their role to the
be a low threshold for empirical therapy unless a clear case where rapid confirmation of a TB diagnosis in a
alternative diagnosis is made. smear-positive patient would alter management (for
Blood cultures for mycobacteria are particularly useful instance, if an NTM was suspected). There are few
in the diagnosis of disseminated TB and NTM (especially recommendations on the role of NAATs on non-
mycobacterium avium intracellulare) in HIV affected respiratory samples. Data from meta-analysis gives a 56%
individuals with a low CD4 count, with mycobacteraemia sensitivity and 98% specificity for CSF, which is similar to
being proportional to the CD4 count [47]. The yield microscopy [26]. There may be a role for NAAT on CSF
drops to zero with CD4 counts above 300  cells/μl. The when ATT treatment has been started without a
utility of mycobacterial blood cultures in non-HIV immuno- confirmed diagnosis, as mycobacterial DNA may remain
suppresion is unknown. Bone marrow examination detectable for a month after the start of treatment [26].
Hagan and Nathani Critical Care 2013, 17:240 Page 5 of 10
http://ccforum.com/content/17/5/240

Radiology
Upper lobe disease on a CXR was shown to increase the
odds ratio of TB by 14.6 [53]. There are fewer data
specific to ICU patients; small nodular or cavitary
patterns on a CXR as well as a duration of illness of more
than 2  weeks may be predictive of TB in some studies
[54], although other studies fail to identify radiological
changes specific for TB on an ICU [55]. In HIV co-
infected patients, radiological appearances can vary and
cavitation becomes less common as immunosuppression
advances. An American study described a normal or near
normal CXR in 19% of pulmonary TB patients with a
CD4 count less than 200  cells/μl [56]. Computed
tomography (CT) scanning may have a role in identifying
active TB and allowing differentiation from old fibrotic
lesions, with centrilobular nodules and a ‘tree in bud’
pattern often seen in active disease. Mediastinal
lymphadenopathy and cavitation may also raise the
suspicion of TB (Figure 2), and miliary shadowing may be
present on CT even with a normal chest X-ray [57]. CT Figure 2. CT scan of thorax of female patient intubated due
chest may help in gathering diagnostic information in an to respiratory failure. Tuberculosis subsequently cultured from
bronchoscopy specimens. Note the cavitation mainly in the right
intubated patient where TB is suspected but not
upper lobe.
confirmed, and also allow targeting of bronchoscopy.
A proposed diagnostic algorithm for respiratory failure
is presented in Figure 3.
High rates of hepatic and renal dysfunction in ICU
patients with TB provide specific challenges. Patients on
Anti tuberculosis treatment and adjuvant the ICU may have uncertain enteral absorption [58].
therapies (including paradoxical reactions) Subtherapeutic levels of ATT have been associated with a
The standard treatment for non-MDR TB involves slow clinical response, treatment failure and drug
combination therapy of more than three drugs; R and resistance [59]. The incidence of subtherapeutic levels of
isoniazid (H) provide a crucial backbone to ATT regimes, anti-TB drugs in ICU patients is not known but it is
allowing shorter courses of 6 to 9 months (1 year for CNS reasonable to have a low threshold for therapeutic drug
TB) to be efficacious due to their bacteriocidal action. R monitoring in TB patients on the ICU. Our practice is to
and H are associated with the potential serious side prefer parenteral therapy in severely ill patients for the
effects of hepatotoxicity. Two other first line ATT, pyra- initial 72 hours.
zinamide (Z) and ethambutol (E), are renally excreted Additional bacterial infection can complicate TB-
and E may be associated with optic nerve toxicity. Other related ICU admissions [8] and a low threshold for
drugs used to treat TB include fluoroquinolones (for additional anti-bactericidal therapy should be present.
example, moxifloxacin), aminoglycosides (for example,
streptomycin or amikacin) and a range of other second Hepatotoxicity
line anti-TB drugs such as cycloserine or prothionamide. Hepatotoxicity may be associated with older age, mal-
Not all ATT is available parenterally; parenteral prepara- nutrition, alcoholism, HIV or viral hepatitis co-infection
tions are available for R, H, fluoroquinolones and amino- [60]. In miliary TB it is important to ensure that the cause
glycosides. The management of HIV co-infection is of the deranged liver function is not due to TB itself (by
complicated and detailed discussion is beyond the scope imaging or consideration of liver biopsy), as the
of this article; guidelines exist for management outside of management of deranged liver function in this context
the HDU/ICU [40]. Principles of HIV/TB co-infection would be management of the TB. International guidelines
management include awareness of the immune reconsti- differ slightly but suggest stopping TB medication if
tution inflammatory syndrome (IRIS), and when highly transaminases are more than three to five times the
active antiretroviral therapy (HAART) should be started; upper limit of normal or there is a bilirubin rise [60,61]. If
this depends on the CD4 count but is recommended as it is crucial to continue ATT in the short term (for
soon as is practical in the very immunosuppressed (CD4 instance in TBM where treatment interruptions are an
count <100 cells/μl) [40]. independent risk factor for death [26]), a combination of
Hagan and Nathani Critical Care 2013, 17:240 Page 6 of 10
http://ccforum.com/content/17/5/240

Figure 3. Proposed diagnostic algorithm for respiratory failure due to suspected tuberculosis (TB). AFB, acid fast bacilli; CSF, cerebrospinal
fluid; ET, endotracheal; IGRA, interferon gamma release assay; NIV, non-invasive ventilation.

relatively non-hepatotoxic drugs, such as an aminoglyco- Nephrotoxicity

side, E and a fluoroquinolone, could be given [60]. Dose adjustments are required with Z and E with a
Ethionamide and prothionamide may be an alternative to glomerular filtration rate of less than 30 ml/minute/1.73 m2.
E as they penetrate the meninges well and do not have Aminoglycosides and cycloserine may require similar
the concern of optic nerve toxicity [62]. dose adjustments [63]. No data exist for patients on
As R and H are important for TB treatment, they are continuous renal replacement therapy, and collaboration
usually sequentially reintroduced once liver function between intensivists, pharmacists, TB physicians, renal
tests improve with close monitoring, and standardised physicians and monitoring of drug levels is appropriate.
re-challenge protocols are provided [60,61]. In cases of
severe or prolonged hepatotoxicity who have had R and Corticosteroids as adjuvant therapy
H re-introduced, it may be reasonable not to re-challenge Corticosteroids inhibit release of inflammatory cyto-
with Z and extend treatment to 9  months [60]. The kines, which may help lessen tissue damage and consti-
management of patients with decompensated liver tutional symptoms. There have been many studies of
disease and TB is not clear, and the standard regime with corticosteroids in TB, including advanced pulmonary TB.
close monitoring, or 18 to 24  months of E, a fluoro- The only indications for steroids recognised in most
quinolone and an aminoglycoside is suggested [60]. international guidelines are a) TBM, where corticosteroids
Hagan and Nathani Critical Care 2013, 17:240 Page 7 of 10
http://ccforum.com/content/17/5/240

Figure 4. Proposed algorithm for treatment of suspected or actual pulmonary tuberculosis (TB) on the ICU. MDR-TB, multi-drug resistant
tuberculosis; TDM, therapeutic drug monitoring.

help reduce risk of death or disability [26,31] and meningitis) for a total of 6 to 8  weeks [65]; British
b) tuberculous pericardial effusion where corticosteroids guidelines suggest prednisolone (or equivalent) 20 to
decrease the amount and rate of re-accumulation of 40 mg/day with gradual withdrawal. There are no direct
tuberculous pericardial effusion [31,64]. The largest trial comparisons of the dose or type of steroid in TBM. For
of steroids in TBM used a reducing dose of dexa- tuberculous pleural effusion there is some evidence for
methasone, initially given intravenously at a dose of 0.3 to adjuvant corticosteroids resulting in a faster resolution of
0.4  mg/kg/day (depending on the severity of the pleural effusion at 4 weeks, but no difference in residual
Hagan and Nathani Critical Care 2013, 17:240 Page 8 of 10
http://ccforum.com/content/17/5/240

fluid at 8 weeks or death rates between corticosteroid and evidence base for diagnosis and management of TB
non-corticosteroid groups [66]. Many trials of cortico- specific to an ICU setting is sparse and what may be true
steroids for pulmonary TB were carried out in the 1950s for stable TB patients may not translate to an ICU. These
to 1960s and suggested a more rapid clinical and patients have a high mortality and high rates of organ
radiological improvement compared to control patients, dysfunction. Diagnosis of TB if not confirmed prior to
particularly in severe disease [67], but an absence of ICU admission may be challenging but culture of
longer term effects on survival or risk of long-term lung mycobacterium TB should be attempted, with radiology,
damage. The implications of this for patients with histology and possibly IGRAs contributing to the clinical
advanced TB and respiratory failure on an ICU are picture. Treatment is complicated by drug toxicity, erratic
unclear. A recent meta-analysis suggests a non-significant absorption and organ dysfunction, and therapeutic drug
trend towards benefit of steroids in pulmonary TB [68], monitoring and judicious use of alternative regimes may
and a 2008 study [14] suggested a lower mortality rate in help this problem. Corticosteroids have a role in TBM
patients with pulmonary TB who received corticosteroids and pericardial TB and may have a role in far advanced
but firm conclusions cannot be drawn due to the pulmonary TB on a case by case basis. The clinician
retrospective nature of the study. should be alert to paradoxical reactions as a cause of
apparent treatment failure or disease progression.
Paradoxical reactions/immune reconstitution
inflammatory syndrome Abbreviations
AFB, acid fast bacilli; ARDS, acute respiratory distress syndrome; ATT, anti-
A paradoxical reaction in TB is defined as a clinical or tuberculosis treatment; CNS, central nervous system; CSF, cerebrospinal fluid;
radiological worsening of pre-existing tuberculous CT, computed tomography; CXR, chest X-ray; E, ethambutol; H, isoniazid;
lesions or the development of new lesions in patients HAART, highly active antiretroviral therapy; HDU, high dependency unit; IGRA,
interferon gamma release assay; IRIS, immune reconstitution inflammatory
receiving ATT in the absence of an alternative explana- syndrome I; MDR TB, multi-drug resistant tuberculosis; NAAT, nucleic acid
tion such as drug resistance, ineffective drug delivery or a amplification test; NTM, non-tuberculous mycobacterium; PCR, polymerase
secondary diagnosis [69]. Patients with HIV are more chain reaction; R, rifampicin; TB, multi-drug resistant tuberculosis; MOF, multi-
organ failure; TB, tuberculosis; TBM, tuberculosis meningitis; WHO, World
likely to develop these reactions, which are referred to in Health Organisation; Z, pyrazinamide.
the context of HIV co-infection as IRIS [40]. The
aetiology of these reactions is unknown, but in HIV may Competing interests
The authors have no conflicts of interest to declare.
relate to HAART causing a reconstitution of immunity
leading to an immune response to dead bacilli. The Acknowledgements
incidence in HIV-negative patients is probably between 2 The authors are grateful to Mark McBreen, Head Graphic Designer,
Department of Medical Illustration, City Hospital Birmingham, for assistance
and 23% [70], and about 32 and 36% in HIV-positive with layout of the diagnostic algorithms.
patients, and may be more in patients with advanced TB.
Paradoxical reactions are usually mild and may manifest Published: 27 September 2013
as recurrent fever, deterioration in radiological appear- References
ances or lymph node inflammation [70] but may cause 1. WHO: Global tuberculosis report 2012
significant morbidity, including airway obstruction, [http://www.who.int/tb/publications/global_report/en]
2. Pedrazzoli D, Fulton N, Anderson L, Lalor M, Abubakar A, Zenna D:
splenic rupture, or worsening neurology due to new or Tuberculosis in the UK: 2012 report [http://www.hpa.org.uk/Publications/
enlarging intracranial tuberculomas [69,71]. The median InfectiousDiseases/Tuberculosis/1206TBintheUK2012report/]
time of onset is about 26  days after treatment [70], but 3. Levy H, Kallenbach JM, Feldman C, Thorburn JR, Abramowitz JA: Acute
respiratory failure in active tuberculosis. Crit Care Med 1987, 15:221-225.
can occur months into treatment. Risk factors for IRIS 4. Silva DR, Menegotto DM, Schulz LF, Gazzana MB, Dalcin PT: Mortality among
include a low baseline CD4 count, rapid recovery in CD4 patients with tuberculosis requiring intensive care:a retrospective cohort
numbers and HAART started within 2  months of study. BMC Infect Dis 2010, 10:54.
5. Lin SM, Wang TY, Liu WT: Predictive factors for mortality among non-HIV-
diagnosis [40]. There are no firm recommendations on infected patients with pulmonary tuberculosis and respiratory failure. Int J
how to treat paradoxical reactions but a tapering dose of Tuberc Lung Dis 2009, 13:335-340.
corticosteroids is reasonable [69]. Thalidomide may have 6. Zahar JR, Azoulay E, Klement E, De Lassence, Lucet JC, Regnier B, Schlemmer
B, Bedos JP: Delayed treatment contributes to mortality in ICU patients
a role in severe CNS TB paradoxical reactions un-
with severe active pulmonary tuberculosis and acute respiratory failure.
responsive to corticosteroids [72], and montelukast has Intensive Care Med 2001, 27:513-520.
been used in IRIS. An algorithm for treatment of 7. Jacob JT, Mehta AK, Leonard MK: Acute forms of tuberculosis in adults. Am J
pulmonary TB on the ICU based on the above review is Med 2009, 122:12-17.
8. Erbes R, Oettel K, Raffenberg M, Mauch H, Schmidt-Ioanas, Lode H:
presented in Figure 4. Characteristics and outcome of patients with active pulmonary
tuberculosis requiring intensive care. Eur Resp J 2006, 27:1223-1228.
Conclusion 9. Sydow M, Schauer A, Crozier TA, Burchardi H: Multiple organ failure in
generalised disseminated tuberculosis. Respir Med 1992, 86:517-519.
Respiratory failure and miliary TB are common reasons 10. Piqueras AR, Marruecos L, Artigas A, Rodriguez C: Miliary tuberculosis and
for admission of patients with TB to an ICU. The adult respiratory distress syndrome. Intensive Care Med 1987, 13:175-182.
Hagan and Nathani Critical Care 2013, 17:240 Page 9 of 10
http://ccforum.com/content/17/5/240

11. Frame RN, Johnson MC, Eichenhorn MS, Bower GC, Popovich J: Active Intensive Care Med 2008, 34:2194-2201.
tuberculosis in the medical intensive care unit:a 15-year retrospective 34. Mase S, Ramsay A, Ng N, Henry M, Hopewell PC, Cunningham J, Urbanczik R,
analysis. Crit Care Med 1987, 15:1012-1014. Perkins M, Aziz MA, Pai M, Yield of serial sputum specimen examinations in
12. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, Sevransky the diagnosis of pulmonary tuberculosis: a systematic review. Int J Tuberc
JE, Sprung CL, Douglas IS, Jaeschke R, Osborn TM, Nunnally ME, Townsend SR, Lung Dis 2007, 11:485-495.
Reinhart K, Kleinpell RM, Angus DC, Deutschman CS, Machado FR, Rubenfeld 35. Jacomelli M, Silva PR, Rodrigues AJ, Demarzo SE, Seicento M, Figueiredo VR:
GD, Webb S, Beale RJ, Vincent JL, Moreno R: Surviving Sepsis Bronchoscopy for the diagnosis of pulmonary tuberculosis in patients
Campaign:International guidelines for management of severe sepsis and with negative sputum smear microscopy results. J Bras Pneumol 2012,
septic shock:2012. Crit Care Med 2013, 41:580-637. 38:167-173.
13. Penner C, Roberts D, Kunimoto D, Manfreda J, Long R: Tuberculosis as a 36. O’Brien JD, Ettinger NA, Shevlin D, Kollef MH: Safety and yield of
primary cause of respiratory failure requiring mechanical ventilation. Am J transbronchial biopsy in mechanically ventilated patients. Crit Care Med
Respir Crit Care Med 1995, 151:867-872. 1997, 25:440-446.
14. Kim YJ, Pack KM, Jeong E, Na JO, Oh YM, Lee SD, Kim WS, Kim DS, Kim WD, 37. Anderson C, Inhaber N, Menzies D: Comparison of sputum induction with
Shim TS: Pulmonary tuberculosis with acute respiratory failure. Eur Resp J fibre-optic bronchoscopy in the diagnosis of tuberculosis. Am J Resp Crit
2008, 32:1625-1630. Care Med 1995, 152:1570-1574.
15. Ryu YJ, Koh WJ, Kang EH, Suh GY, Chung MP, Kim H, Kwon OJ: Prognostic 38. Thadepalli H, Rambhatla K, Niden AH: Transtracheal aspiration in diagnosis
factors in pulmonary tuberculosis requiring mechanical ventilation for of sputum-smear negative tuberculosis. JAMA 1977, 238:1037-1040.
acute respiratory failure. Respirology 2007, 12:406-411. 39. Jones BE, Young SM, Antoniskis D, Davidson PT, Kramer F, Barnes PF:
16. Maartens G, Willcox PA, Benatar SR: Miliary tuberculosis:rapid diagnosis, Relationship of the manifestations of tuberculosis to CD4 cell counts in
haematologic abnormalities, and outcome in 109 treated adults. Am J Med patients with human immunodeficiency virus infection. Am Rev Respir Dis
1990, 89:291-296. 1993, 148:1292-1297.
17. Hess S, Hospach T, Nossal R, Dannecker G, Magdorf K, Uhlemann F: Life- 40. Pozniak AL, Coyne KM, Miller RF, Lipman MC, Freedman AR, Omerod LP,
threatening disseminated tuberculosis as a complication of TNF-α Johnson MA, Collins S, Lucas SB: British HIV Association guidelines for the
blockade in an adolescent. Eur J Pediatr 2011, 170:1337-1342. treatment of TB/HIV coinfection. HIV Med 2011, 12:517-524.
18. Landouzy L: A note of la typho-bacillose. Lancet 1908, 172:1440-1441. 41. Light RW: Update on pleural effusion. Respirology 2010, 15:451-458.
19. Barber TW, Craven DE, McCabe WR: Bacteremia due to mycobacterium 42. McGrath EE, Anderson PB: Diagnostic tests for tuberculous pleural effusion.
tuberculosis in patients with human immunodeficiency virus infection, Eur J Clin Microbiol Infect Dis 2010, 29:1187-1193.
A report of 9 cases and a review of the literature. Medicine (Baltimore) 1990, 43. Alsoub H, Al Alousi FS: Miliary tuberculosis in Qatar: a review of 32 adult
69:375-383. cases. Ann Saudi Med 2001, 21:16-20.
20. Lim KH, Chong KL: Multiple organ failure and septic shock in disseminated 44. Kim JH, Langston AA, Gallis HA: Miliary tuberculosis: epidemiology, clinical
tuberculosis. Singapore Med J 1999, 40:176-178. manifestations, diagnosis and outcome. Rev Infect Dis 1990, 12:583-590.
21. Al-Mamari A, Balkhair A, Gujjar A, Ben Abid F, Al-Farqani A, Al-Hamadani A, 45. Chierakul N, Anantasetagoon T, Chaiprasert A, Tingtoy N: Diagnostic value of
Jain R: A case of disseminated tuberculosis with adrenal insufficiency. gastric aspirate smear and polymerase chain reaction in smear-negative
Sultant Qaboos Univ Med J 2009, 9:324-327. pulmonary tuberculosis. Respirology 2003, 8:492-496.
22. Verdon R, Chevret S, Laissy JP, Wolff M: Tuberculous meningitis in adults: 46. Bahammam A, Choudhri S, Long R: The validity of acid-fast smears of
review of 48 cases. Clin Infect Dis 1996, 22:982-988. gastric aspirates as an indicator of pulmonary tuberculosis. Int J Tuberc
23. Kumar R, Pandey CK, Bose N, Sahay S: Tuberculous brain abscess:clinical Lung Dis 1999, 3:62-67.
presentation, pathophysiology and treatment (in children). Childs Nerv Syst 47. Jones BE, Young SM, Antoniskis D, Davidson PT, Kramer F, Barnes PF:
2002 18:118-123. Relationship of the manifestations of tuberculosis to CD4 cell counts in
24. Thwaites GE, Macmullen-Price J, Tran TH, Pham PM, Nguyen TD, Simmons CP, patients with human immunodeficiency virus infection. Am Rev Respir Dis
White NJ, Tran TH, Summers D, Farrar JJ: Serial MRI to determine the effect 1993, 148:1292-1297.
of dexamethasone on the cerebral pathology of tuberculous meningitis: 48. Pai M, Zwerling A, Menzies D: Systemic review: T-cell-based assays for the
an observational study. Lancet Neurol 2007, 6:230-236. diagnosis of latent tuberculosis infection: an update. Ann Intern Med 2008,
25. Figaji AA, Fieggen AG: The neurosurgical and acute care management of 149:177-184.
tuberculous meningitis:Evidence and current practice. Tuberculosis 2010, 49. Health Protection Agency: Health Protection Agency position statement
90:393-400. on the use of Interferon Gamma Release Assay (IGRA) tests for
26. Thwaites G, Fisher M, Hemingway C, Scott G, Solomon T, Innes J: British Tuberculosis (TB) [http://www.hpa.org.uk/webc/HPAwebFile/
Infection Society guidelies for the diagnosis and treatment of tuberculosis HPAweb_C/1204186168242]
of the central nervous system in adults and childen. J Infection 2009, 50. Cho K, Cho E, Kwon S, Im S, Sohn I, Song S, Kim H, Kim S: Factors associated
59:167-187. with indeterminate and false negative results of QuantiFERON-TB Gold
27. Mathew JM, Rajshekhar V, Chandy MJ: Shunt surgery in poor grade patients In-Tube Test in active tuberculosis. Tuberc Respir Dis (Seoul) 2012, 72:416-425.
with tuberculous meningitis and hydrocephalus:effects of response to 51. Abubakar I, Stagg HR, Whitworth H, Lalvani A: How should I interpret an
external ventricular drainage and other variables on long term outcome. interferon gamma release assay result for tuberculosis infection? Thorax
J Neurol Neurosurg Psychiatry 1998, 65:115-118. 2013, 68:298-301.
28. Moller K, Larsen FS, Bie P, Skinhoj P: The syndrome of inappropriate 52. Boehme CC, Nabeta P, Hillemann D, Nicol MP, Shenai S, Krapp F, Allen J, Tahirli
secretion of antidiuretic hormone and fluid restriction in meningitis-how R, Blakemore R, Rustomjee R, Milovic A, Jones M, O’Brien SM, Persing DH,
strong is the evidence? Scand J Infect Dis 2001, 33:13-26. Ruesch-Gerdes S, Gotuzzo E, Rodrigues C, Alland D, Perkins MD: Rapid
29. Figaji AA, Zwane E, Kogels M, Fieggen AG, Argent AC, Le Roux PD, Peter JC: molecular detection of tuberculosis and rifampicin resistance. N Engl J Med
The effect of blood transfusion on brain oxygenation in children with 2010 363:1005-1015.
severe traumatic brain injury. Pediatr Crit Care Med 2010, 11:325-331. 53. Wisnivesky JP, Kaplan J, Henschke C, McGinn TG, Crystal RG: Evaluation of
30. Axelrod YK, Diringer MN: Temperature management in acute neurologic clinical parameters to predict mycobacterium tuberculosis in inpatients.
disorders. Neurol Clin 2008, 26:585-603. Arch Int Med 2000, 160:2471-2476.
31. National Institue for Health and Clinical Excellence: Tuberculosis: Clinical 54. Hui C, Wu CL, Chan MC, Kuo IT, Chiang CD: Features of severe pneumonia in
diagnosis and management of tuberculosis, and measures for its patients with undiagnosed pulmonary tuberculosis in an intensive care
prevention and control [http://guidance.nice.org.uk/CG117] unit. J Formos Med Assoc 2003, 102:563-569.
32. World Health Organisation: Treatment of Tuberculosis: guidelines for 55. Wu JY, Ku SC, Shu CC, Fan JY, Chen HY, Chen YC, Yu CJ: The role of chest
national programmes; WHO/HTM/TB/2009.420 [http://www.who.int/tb/ radiography in the suspicion for and diagnosis of pulmonary tuberculosis
publications/tb_treatmentguidelines/en/index.html] in intensive care units. Int J Tuberc Lung Dis 2009, 13:1380-1386.
33. Shu CC, Lee CH, Wang JY, Jerng JS, Yu CJ, Hsueh PR, Lee LN, Yang PC, 56. Greenberg SD, Frager D, Suster B, Walker S, Stavropoulos C, Rothpearl A:
Nontuberculous mycobacteria pulmonary infection in medical intensive Active pulmonary tuberculosis in patients with AIDS: spectrum of
care unit: the incidence, patient characteristics and clinical significance. radiographic findings (including a normal appearance). Radiology 1994,
Hagan and Nathani Critical Care 2013, 17:240 Page 10 of 10
http://ccforum.com/content/17/5/240

193:115-119. Dexamethasone for the treatment of tuberculous meningitis in

57. Lee KS, Im JG: CT in adults with tuberculosis of the chest: characteristic adolescents and adults. N Engl J Med 2004, 351:1741-1751.
findings and role in management. Am J Roentgenol 1995, 164:1361-1367. 66. Engel ME, Matchaba PT, Volmink J: Corticosteroids for tuberculous pleurisy.
58. Thompson JS: The intestinal response to critical illness. Am J Gastroenterol Cochrane Database of Syst Rev 2007:CD001876.
1995, 90:190-200. 67. Dooley DP, Carpenter JL, Rademacher S: Adjunctive corticosteroid therapy
59. Kimberling ME, Phillips P, Patterson P, Hall M, Robinson CA, Dunlap NE: Low for tuberculosis:a critical reappraisal of the literature. Clin Infect Dis 1997,
serum antimycobacterial drug levels in non-HIV-infected tuberculosis 25:872- 887.
patients. Chest 1998, 113:1178-1183. 68. Critchley JA, Young F, Orton L, Garner P: Corticosteroids for prevention of
60. Saukkonon JJ, Cohn DL, Jasmer RM, Schenker S, Jereb JA, Nolan CM, Peloquin mortality in people with tuberculosis:a systemic review and meta-analysis.
CA, Gordin FM, Nunes D, Stader DB, Bernardo J, Venkataramanan R, Sterling Lancet Infect Dis 2013, 13:223-237.
TR: An Official ATS Statement: Hepatotoxicity of Antituberculosis Therapy. 69. Bloch S, Wickremasinghe M, Wright A, Rice A, Thompson M, Kon OM:
Am J Resp Crit Care Med 2006, 174:935-952. Paradoxical reactions in non-HIV tuberculosis presenting as
61. Joint Tuberculosis Committee of the British Thoracic Society: Chemotherapy endobronchial obstruction. Eur Resp J 2009, 18:295-299.
and management of tuberculosis in the United 70. Cheng SL, Wang HC, Yang PC: Paradoxical response during anti-
Kingdom:recommendations. Thorax 1998, 53:536-548. tuberculosis treatment in HIV negative patients with pulmonary
62. Donald PR, Seifart HI: Cerebrospinal fluid concentrations of ethionamide in tuberculosis. Int J Tuberc Lung Dis 2007, 11:1290-1295.
children with tuberculous meningitis. J Pediatr 1989, 115:483-486. 71. Nicolls DJ, King M, Holland D, Bala J, del Rio C: Intracranial tuberculomas
63. Milburn H, Ashman N, Davies P, Doffman S, Drobniewski F, Khoo S, Ormerod P, developing while on therapy for pulmonary tuberculosis. Lancet Infect Dis
Ostermann M, Snelson C: Guidelines for the prevention and management 2005, 5:795-801.
of Mycobacterium tuberculosis infection and disease in adult patients 72. Schoeman JF, Fieggen G, Seller N, Mandelson M, Hartzenberg B: Intractable
with chronic kidney disease. Thorax 2010, 65:559-570. intracranial tuberculous infection responsive to thalidomide:report of four
64. Strang JI, Kakaza HH, Gibson DG, Girling DJ, Nunn AJ, Fox W: Controlled trial cases. J Child Neurol 2006, 21:301-308.
of prednisolone as adjuvant in treatment of tuberculous constrictive
pericarditis in Transke. Lancet 1987, 19:1418-1422.
doi:10.1186/cc12760
65. Thwaites GE, Nguyen DB, Nguyen HD, Hoang TQ, Do TT, Nguyen TC, Nguyen
Cite this article as: Hagan G, Nathani N: Clinical review: Tuberculosis on the
QH, Nguyen TT, Nguyen NH, Nguyen TN, Nguyen NL, Nguyen HD, Vu NT, Cao
intensive care unit. Critical Care 2013, 17:240.
HH, Tran TH, Pram PM, Nguyen TD, Stepniewska K, White NJ, Tran TH, Farrar JJ: