- Email: [email protected]
Tuberculous cerebral vasculitis: Retrospective study of 10 cases
European Journal of Internal Medicine 22 (2011) e99–e104
Contents lists available at ScienceDirect
European Journal of Internal Medicine j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e j i m
Original article
Tuberculous cerebral vasculitis: Retrospective study of 10 cases Nicolas Javaud a,b,⁎, Rita Da Silva Certal b, Jérôme Stirnemann b, Anne-Sophie Morin b, Jean-Marie Chamouard b, Alexandre Augier c, Olivier Bouchaud d, Antoine Carpentier e, Robin Dhote f, Jean-Luc Dumas c, Bruno Fantin g, Olivier Fain b a
Service des Urgences, Hôpital Jean-Verdier, Assistance Publique-Hôpitaux de Paris, Université Paris 13, avenue du 14 Juillet, 93140 Bondy Cedex, France Service de Médecine Interne, Hôpital Jean-Verdier, Assistance Publique-Hôpitaux de Paris, Université Paris 13, avenue du 14 Juillet, 93140 Bondy Cedex, France Service de Radiologie, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France d Service de Maladies Infectieuses et Tropicales, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France e Service de Neurologie et Rééducation Neurologique, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France f Service de Médecine Interne, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France g Service de Médecine Interne, Hôpital Beaujon, Assistance Publique-Hôpitaux de Paris, Université Paris 7, 100, bd du Général-Leclerc, 92110 Clichy Cedex, France b c
a r t i c l e
i n f o
Article history: Received 22 February 2011 Received in revised form 12 April 2011 Accepted 14 April 2011 Available online 6 May 2011 Keywords: Cerebral vasculitis Tuberculosis Tuberculous meningitis
a b s t r a c t Background: Tuberculous cerebral vasculitis is a complication of tuberculous meningitis. This study was undertaken to determine the epidemiological characteristics, context, diagnostic means and outcomes under treatment of tuberculous cerebral vasculitides. Methods: All consecutive patients diagnosed with tuberculous cerebral vasculitis were identified from the databases of three Internal Medicine, one Neurology and one Infectious Disease Departments in three suburban Parisian hospitals. Results: We describe 10 cases: five men and five women (median age 33.5 [range: 27–55] years). Two were infected with the human immunodeficiency virus. Nine patients had tuberculous meningitis, eight with extraneurological involvement. The following manifestations led to the diagnosis: motor deficit, acute confusional state, headaches, involvement, coma and/or seizures. The cerebral vasculitis revealed tuberculosis in three patients, but tuberculosis was already known when vasculitis was diagnosed for the seven others. The cerebral computed-tomography scan showed cerebral infarctions in five patients, hydrocephalus and tuberculomas in four, while magnetic resonance imaging detected infarctions and leptomeningitis in nine patients, pachymeningitis in one, hydrocephalus and tuberculomas in seven. Therapy combined antituberculous agents with oral corticosteroids for all patients, preceded by a methylprednisolone pulse for five patients. Outcome was favorable for nine patients. Conclusion: We described the non-negligible frequency of tuberculous cerebral vasculitides, their clinical manifestations and their potential severity, and the diagnostic and monitoring contributions of magnetic resonance imaging and magnetic resonance angiography. © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Cerebral vasculitis corresponds to inflammation of cerebral blood vessel walls [1,2]. Primary and secondary cerebral vasculitides are distinguished, with tuberculous cerebral vasculitis belonging among the latter [2–6]. Tuberculous cerebral vasculitis represented 12.5% (3/24) of secondary cases, with other etiologies of secondary central nervous system vasculitides including systemic vasculitides, infections, hematological malignancies, toxic vasculitis and miscellaneous origins [3].
Tuberculous cerebral vasculitis is a complication of tuberculous meningitis, which occurred in 1.5–5% of declared tuberculosis cases in France [7,8]. Cerebral vasculitis-induced strokes had occurred in 41% of the autopsy series of 100 patients with tuberculous meningitis [9]. Depending on the imaging technique used, their frequency ranged from 20.5 to 47.8% with computed tomography (CT) scans [10–12] or 54–66% with magnetic resonance imaging (MRI) [13,14]. The objective of this study was to determine the epidemiological characteristics, context, diagnostic means and outcome under treatment of tuberculous cerebral vasculitides. 2. Materials and methods
⁎ Corresponding author at: Service des Urgences, Hôpital Jean-Verdier, avenue du 14 Juillet, 93143 Bondy Cedex, France. Tel.: + 33 1 48 02 65 34; fax: + 33 1 48 02 63 61. E-mail address: [email protected] (N. Javaud).
We conducted a retrospective study on consecutive patients diagnosed with tuberculous cerebral vasculitis, between January 1995
0953-6205/$ – see front matter © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.ejim.2011.04.004
e100
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
and September 2007, identified from the databases of three Internal Medicine, one Neurology and one Infectious Disease Departments in three suburban Parisian hospitals. The following inclusion criteria were used. Tuberculosis was diagnosed based on at least one of the following: Mycobacterium tuberculosis growth in any cultured sample, epithelioid and giant-cell granuloma with/without caseous necrosis seen in any biopsy, M. tuberculosis-DNA detection by polymerase chain reaction (PCR, bacillus of Koch (BK)-positive); clinical history compatible with tuberculosis; and anti-tuberculous therapy effective at 3 months for all patients. Cerebral vasculitis was diagnosed based on two or more of the following [4,5]: MRI-confirmed infarcts in a patient with a neurological deficit and no cardiovascular risk factors; irregular calibers of one or more cerebral arteries visualized by magnetic resonance angiography (MRA); increased cerebrospinal fluid (CSF) cellularity. Non-inclusion criteria were normal CSF cellularity with normal MRI. The following information was also retrieved from each patient's medical file: age, sex, clinical presentation, laboratory findings and treatments. The same neuroradiologist reread all imaging documents, unblinded for MRA and MRI. 3. Results Thirteen cases of tuberculous cerebral vasculitis were identified between January 1995 and September 2007, representing 21% of the 62 patients with tuberculous meningitis during the same period in the same hospital departments. After excluding three patients because of insufficient data, 10 were analyzed. Ten patients, five men and five women, with a median age of 33.5 [range: 27–55] years, were included (Table 1). Six originated from sub-Saharan Africa, two from Pakistan, one from North Africa and the last from Eastern Europe. Patients 6 and 10 were human immunodeficiency virus (HIV)-infected and their respective CD4+ T-cell counts were 5/mm3 and 42/mm3; the other eight were HIVseronegative. Median follow-up was 18 [4–60] months. Tuberculous cerebral vasculitis revealed tuberculosis infection in three patients. Tuberculosis was known at the time of vasculitis for the other seven patients. For HIV-infected patients 6 and 10, the tuberculosis was extraneurological and had already been treated for 6 months, and their antituberculous drugs had been discontinued for 1 month (pulmonary tuberculosis) or 5 months (lymph-node tuberculosis), respectively. Vasculitis developed in the other five during the course of progressive tuberculosis treated with antituberculous agents for a median of 51 [24–139] days. For patients 2, 3 and 4, it occurred despite ongoing prednisone (1 mg/kg/day), for a median of 35 [24–51] days, and, for patient 9, 1 month after stopping corticosteroids that had been taken for 2 months.
Table 2 Cerebrospinal fluid parameters at the time tuberculous cerebral vasculitis was diagnosed in 9 patients. CSF parameter
n (%)
Median [range]
Protein concentration N 0.5 g/L Glucose concentration/glycemia b 0.5 Chloride concentration b 115 mmol/L Cellularity N5 elements/mm3 Lymphocyte predominance Neutrophil predominance Not determined
8 (89) 7 (78) 8 (89) 8 (89) 4 (50) 3 (38) 1 (13)
2.5 [0.49–4.5] g/L 0.31 [0.17–0.76] mmol/L 108 [94–123] mmol/L 56 [1–630] elements/mm3 51 [23–99] 47 [1–77] –
Data for Patient 6 are not reported.
The neurological manifestations at the time of vasculitis diagnosis were: motor deficit (n = 7), acute confusional state (n = 7), meningeal syndrome (n = 6), headaches (n = 5), cranial nerve involvement (n = 7), coma (n = 4) and/or convulsions (n = 3). Eight patients had extraneurological tuberculosis affecting the lungs in seven, lymph nodes in six and spleen in two. Nine patients underwent lumbar puncture (Table 2) at the time of neurological deterioration. Their CSF had elevated protein levels, low glucose concentrations and increased cellularity in eight. M. tuberculosis was detected in the CSF cultures of five patients, whose PCR was also BK-positive. Initially, seven patients had CT scans. The first scan contributed to the diagnosis of cerebral vasculitis for patients 3, 7 and 10. The abnormalities seen were: cerebral infarcts (n = 5), meningeal contrast-medium uptake (n = 5), hydrocephalus (n = 4) and/or tuberculomas (n = 4). All 10 patients underwent brain MRI (Fig. 1A and B), which contributed to the diagnosis for nine of them. For the other case, MRI showed no signs indicative of cerebral infarcts. For half of the patients, when the CT scan had not been informative, brain MRIrevealed the following abnormalities: infarctions (n = 9); white matter hypersignals (n = 3); leptomeningitis, an inflammation of the pia mater and the arachnoid membrane in contact with cerebral structures (n = 9); pachymeningitis, an inflammation of the dura mater adhering to the skull (n = 1); hydrocephalus (n = 7); and/or tuberculomas (n = 7). In all cases, one or several cerebral infarcts were seen in the carotid area, localized in the deep or superficial sylvian region in six and four patients, respectively. Brain ischemia was also observed in the vertebrobasilar territory in two patients with carotid region involvement. Five patients had several strokes. MRA examination (Fig. 1C) of the circle of Willis found narrowing of the middle cerebral arteries in all eight patients examined, associated with the anterior cerebral arteries in three and the internal carotid artery in three. Bilateral involvement was observed in three patients.
Table 1 Main demographic, clinical, biological and radiological characteristics of the 10 patients at the time of tuberculous cerebral vasculitis diagnosis. Patient
1 2 3 4 5 6 7 8 9 10
Sex/age (yr)
ATD
M/55 F/28 M/35 M/50 F/51 M/26 M/32 F/29 F/28 F/45
No Yes Yes Yes No No Yes No Yes No
Taking prednisone
Focal deficit
Syndrome
Infected sites
Stroke
Meningeal
Inflammatory
CSF Anomalies
BK
Lung
Other
1st CT
1st MRI
No Yes Yes Yes No No No No Yes No
Yes Yes Yes No Yes No No Yes Yes Yes
Yes Yes Yes No Yes No No No Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes No Yes
Yes Yes Yes Yes Yes No Yes Yes Yes Yes
No Yes No No Yes No No Yes Yes Yes
Yes Yes No Yes Yes Yes Yes Yes No No
LN LN, spleen No No No LN, spleen LN LN LN No
– No Yes – No No Yes No – Yes
Yes Yes Yes Yes Yes Yes Yes No Yes Yes
MRA-detected stenosis Yes Yes – Yes Yes – Yes Yes Yes Yes
Abbreviations: ATD = antituberculous drug; CT = computed tomography; CSF = cerebrospinal fluid; BK = bacillus of Koch; MRI = magnetic resonance imaging; MRA = magnetic resonance angiography; LN = lymph node.
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
e101
Fig. 1. Brain imaging. (A) Magnetic resonance imaging T1-weighted sequence after gadolinium injection: 1: basal leptomeningitis enveloping the middle cerebral arteries; 2: level of the sylvian valleys. (B) Magnetic resonance imaging T2-weighted sequence showing ischemic lesions: 1: a hypersignal in the right temporal pole; 2: the cerebral trunk. (C) Magnetic resonance angiography of the circle of Willis: note the irregular, stringy appearance of the middle cerebral arteries.
The therapeutic regimen combined antituberculous drugs and corticosteroids. The former, which lasted a median of 12.5 [range: 4– 32] months, initially given to 9/10 patients, consisted of rifampicin, isoniazid, ethambutol and pyrazinamide. Pregnant patient 2 did not receive pyrazinamide. Rifabutin was prescribed for patient 10, instead of rifampicin, after the reintroduction of antiviral therapy for HIV infection. Initial oral prednisone was given to all patients at the median dose of 1.5 [range: 1.2–2] mg/kg/day, preceded by methylprednisolone pulses, for five patients. The median total duration of corticosteroid treatment was 9.5 [range: 2–27] months. Three patients received antiplateletaggregating drugs and three were prescribed low-weight heparin-type anticoagulants at effective doses. At the end of the 18-month median follow-up, nine patients had favorable outcomes, with persistent
sequelae in five: motor deficit in two, seizures in two and psychomotor retardation in one.
4. Discussion Tuberculous cerebral vasculitis is a complication of tuberculous meningitis. We identified 13 tuberculous cerebral vasculitis cases, which represented 21% of the 62 tuberculous meningitis episodes admitted during the same period to the same hospital departments. We did not find any study on tuberculous cerebral vasculitis but three case series of patients with tuberculous meningitis reported complication rates with cerebral infarcts of 6–47% [9,10,12,15–18].
e102
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
The younger age of our patients and those of Koh et al. [17] can explain the absence of cardiovascular risk factors, as opposed to those followed by Chan et al. [16], who included patients at risk. Because the incidence of stroke increases exponentially with age [19], elderly patients were excluded. Our patients 6 and 10 had HIV coinfections. The results of several studies showed no significant difference concerning the occurrence of CT-detected focal neurological signs or cerebral infarcts during tuberculous meningitis in patients coinfected with HIV or not [15,20,21]. Tuberculous cerebral vasculitis revealed tuberculosis in three of our patients. In the report by Chan et al., only 2/12 (16.7%) patients had cerebral infarct(s) at the time meningitis was diagnosed, but cerebral infarcts complicated the outcomes of their 10 other patients [16]. Indeed, tuberculous cerebral vasculitis can develop despite treatment with antituberculous drugs and corticosteroids. For our patients, the mean time to the appearance of tuberculous cerebral vasculitis was 40 [range: 0–139] days after the onset of antituberculous therapy, and it developed despite corticotherapy given to four patients. This observation agrees with the findings of Chan et al., who found a 40 [range: 0–128]-day interval for 5/12 of the patients who were given corticosteroids [16,22]. Tuberculous cerebral vasculitis should be included in the differential diagnosis of any neurological deterioration arising during the course of tuberculosis, particularly tuberculous meningitis. The clinical neurological signs seen in our patients (focal neurological deficit, acute confusional state, meningeal syndrome, headaches, cranial nerve paralysis, coma and seizures) are in accordance with those found in patients with tuberculous meningitis complicated by stroke. These signs are not specific to cerebral vasculitis [16–18]. In addition, vasculitis-associated strokes were asymptomatic in 2/9 of our patients, which is close to the 3/12 reported by Chan et al. [16]. Lumbar puncture led to the diagnosis of meningitis for 8/9 of our patients, but was unable to do so for HIV-infected Patient 6 because of her severe immunodepression (5 CD4+ T cells/mm3). M. tuberculosis was found more often in the CSF of HIV-infected than non-infected patients [23], whereas CSF cellularity was lower in HIV-infected patients [24]. In a prospective, double-blinded, randomized study on the use of dexamethasone to treat tuberculous meningitis in 96 HIVinfected and 432 non-infected patients [21], M. tuberculosis was found in the CSF of 41.7% versus 29.6% (p = 0.029), respectively, with corresponding CSF cellularity values of 99 versus 120 leukocytes/mm3 (p = 0.069). A retrospective analysis of 53 tuberculous meningitis patients showed that CSF protein concentrations (1.24 and 1.75 g/L) and cellularity values (146/mm3 and 206/mm3) were significantly lower for the 22 HIV-infected than the 31 non-infected, respectively [20]. The CSF PCR was BK-positive for five of our patients, whose CSF cultures grew M. tuberculosis. These findings are in agreement with a reported 56% sensitivity and 98% specificity of CSF PCR for BK [25]. The sensitivity of this technique is not better than classical bacteriology, because 81% of the 132 tuberculous meningitis patients had bacteriological diagnoses: 58% had multidrug-resistant acid-fast bacteria on direct microscope examination and 71% had positive cultures [25]. Cerebral CT scan can contribute to diagnosing tuberculous cerebral vasculitis [5] and yielded the diagnosis for 3/8 of our patients. However, cerebral vasculitis is not directly visible on CT scans. The indirect signs of arteritis are those of cerebral infarcts, appearing as hypodense areas with variable and irregular uptake of contrast medium [1,24]. The CT scan-demonstrated frequency of cerebral ischemia ranged from 20.5 to 38% [24]. Brain MRI contributed to the diagnosis of tuberculous cerebral vasculitis for nine of our 10 patients. It was sensitive (90–100%) but poorly specific for the detection of vasculitis-associated parenchymatous anomalies in published series [3,5,26]. The results of several studies showed the superiority of MRI, compared to CT scans, for the
diagnosis of tuberculous meningitis and its complications, such as stroke [14,26,27]. The MRI-detected abnormalities suggestive of cerebral vasculitis are infarctions [11,28] and white matter hypersignals. However, these signs are not specific to tuberculous cerebral vasculitis [3,5,6,26]. Signs of tuberculosis neuromeningitis orienting the etiological diagnosis can be present: leptomeningitis [27], pachymeningitis, hydrocephalus and tuberculomas [13,14,29]. The localization of strokes associated with tuberculous cerebral vasculitis is a factor orienting the etiological diagnosis. These strokes were found, in 6/10 cases, in the deep sylvian region, which agrees with previous studies [13,18,30] and can be explained by the presence of tuberculous meningitis exudates at the level of the basal cisterna, leading to vasculitis in the vessels crossing those exudates. The smallcaliber arteries, such as the lenticulostriate arteries, are obstructed more easily, thereby provoking strokes in the central gray nucleus and internal capsule, corresponding to the deep sylvian region [13,18,30]. Furthermore, these strokes are frequently multiple and bilateral [30]. Notably, half our patients had multiple but unilateral strokes. Depending on the series, MRA sensitivity for the diagnosis of cerebral vasculitis ranged from 59 to 90% [5]. All of our patients who underwent MRA had lesions suggestive of cerebral vasculitis. MRA specificity for the etiology is low [3], as brain MRA can be abnormal during, for example, systemic vasculitides (Takayasu arteritis, polyarteritis nodosa, giant-cell arteritis…), systemic autoimmune diseases (systemic lupus erythematosus, Sjögren's syndrome, antiphospholipid syndrome…), cancer (lymphomas…), drug-induced vasculitides (narcotics, amphetamines…), infectious vasculitides (syphilis, HIV…) and reversible brain angiopathy. MRA classically visualizes segmental and arterial narrowing, parietal irregularities and, sometimes, obstructions. Although the cerebral vasculature can be visualized by MRA, direct angiography remains the gold standard for imaging the vascular lumen [4,5]. The arteries most frequently involved by tuberculous cerebral vasculitis are those at the base of the skull, particularly, the middle cerebral artery [5,13,18], which was affected in all our patients. Patient 8's MRA showed narrowing of the right middle cerebral artery, with no visible infarcts on the brain MRI. Gupta et al. found the same frequency (10%) [13]. Moreover, tuberculous cerebral vasculitis was bilateral in three of our patients but no bilateral cerebral infarcts were seen. Thus, the absence of brain MRI-detected infarcts does not exclude tuberculous cerebral vasculitis. Finally, some authors described strokes during the course of tuberculous meningitides, with normal MRA images (other stroke causes had been excluded) [31]. Gupta et al. reported 20% strokes [13]. Other authors did not eliminate tuberculous cerebral vasculitis, thinking that the infarct-affected vessels were of smaller caliber at the limit of MRA resolution [36]. Although all our patients with stroke had abnormal MRA images, MRA had not been done for two of our patients. Treatment of tuberculous cerebral vasculitis is based on corticosteroids, by analogy with other cerebral vasculitides, and their efficacy against tuberculous meningitis, even though no controlled study has demonstrated it [16,22,23,26,32–36]. The pathophysiological mechanisms implicated in cerebral vasculitis are diverse and can include: a direct pathogenic effect of the infectious agent on the vessels, immunological involvement via the induction of antigen expression on endothelial cells and the formation of immune complexes. The non-infectious immunological mechanisms can play an important role in infectious vasculitides, justifying the use of immunosuppressants [4,37]. The corticosteroids evaluated for the treatment of tuberculous meningitis are dexamethasone and prednisolone [22,23,32,35,36]. Dexamethasone use prolonged survival. In a prospective, double-blinded, randomized study on dexamethasone treatment of tuberculous meningitis, 87/274 (32%) of treated versus 112/271 (41%) of untreated patients died (p= 0.01) [36]. Patients received intravenous dexamethasone as follows, for weeks 1–4, respectively: 0.4 mg/kg/day, 0.3 mg/kg/day,
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
0.2 mg/kg/day and 0.1 mg/kg/day. Then, 4 weeks of oral treatment, starting at 4 mg/day week 1, and declining 1 mg/week thereafter. The frequency of MRI-detected strokes was lower with the same dexamethasone regimens but not significantly so. In a prospective, randomized study [23], even though infarcts were rare at presentation (2/22 patients had an initial CT scan), 11/27 patients had infarcts after 60 days of followup. The patients treated with dexamethasone had a lower stroke frequency than those given the placebo (27% versus 58%, respectively; p= 0.13). Corticosteroids should be prescribed in combination with antituberculous drugs to treat tuberculous meningitis. This regimen lowers the risk of death and neurological sequelae in survivors [22]. The steroid dose should be adapted to the enzyme-inducer effect of rifampicin [32]. In 1993, McAllister et al. showed that rifampicin diminished the area under the prednisolone time–concentration curve by 66% [38]. The duration of corticosteroid administration should also be evaluated, as in our study, the median time to the onset of tuberculous cerebral vasculitis after starting antituberculous therapy was 51 [range: 24–139] days. In light of the sometimes late occurrence of cerebral vasculitis during treatment, one can wonder about the potential usefulness of prolonging corticotherapy, especially at full dose. Randomized–controlled trials are needed to assess the optimal duration of steroid administration (6 versus 12 weeks), their dose and type of glucocorticoid (dexamethasone, prednisolone or prednisone). Treatment must also address the causal disease, hence antituberculous agents. For 90% of the patients, the initial regimen comprised four molecules, combining isoniazid, rifampicin, ethambutol and pyrazinamide. This combination is also the currently recommended one [24,25]. Patient 2's initial therapy combined three of those drugs, without pyrazinamide, which is contraindicated during pregnancy. Rifabutin, which has a weaker enzyme-inducer effect, replaced rifampicin for our HIV-infected patient 10, as it had been shown to be well-tolerated and to have efficacy comparable to that of rifampicin for the treatment of tuberculosis in HIV-infected patients [39]. The median treatment duration for our patients was 12.5 months, which is consistent with current guidelines recommending 9–12 months [24,25]. Although the duration of anti-tuberculous therapy for tuberculous meningitis is controversial, the evidence base for shortterm therapy for tuberculous meningitis is weak [40]. Adding an antiplatelet-aggregating agent is an option, in light of their efficacy during the acute phase of ischemic strokes, but they have not been assessed in cerebral vasculitides [41]. In a recent study, aspirin obtained non-significantly fewer strokes and significantly lower 3-month mortality of tuberculous meningitis patients [42]. Development of tuberculous cerebral vasculitis is a poor-prognosis factor. None of our patients died but 12–24% mortality has been reported [12,16–18]; however, our patient population was small. In agreement with previous reports, 50% of our patients had persistent neurological sequelae [12,16,17]. We described 10 patients who developed tuberculous cerebral vasculitis, diagnosed based on clinical symptoms, biological analyses and brain MRI and MRA findings, while the gold standard for its diagnosis is direct angiography and biopsy [4,5]. 5. Conclusion Herein, we described the non-negligible frequency of tuberculous cerebral vasculitides, their clinical manifestations and potential severity, and the diagnostic and monitoring inputs of MRI and MRA. Further investigations are needed to determine: the contributions of MRI and MRA for all tuberculous meningitis patients to detect tuberculous cerebral vasculitis early and increase ongoing corticotherapy; the modalities of the latter, including the type of glucocorticoid to be prescribed, its dose and treatment duration; and the potential role of antiplatelet-aggregating agents.
e103
Learning points • Tuberculous cerebral vasculitis is a complication of tuberculous meningitis. • Tuberculous cerebral vasculitis can reveal tuberculosis but can develop despite treatment with antituberculous drug and corticosteroids. • Further investigations are needed to determine the contributions of MRI and MRA for all tuberculous meningitis patients to detect tuberculous cerebral vasculitis early. Funding source None. Conflicts of interest statement It is certified that there is no actual or potential conflict of interest in relation to this article. References [1] Sarwar M, Falkoff G, Naseem M. Radiologic techniques in the diagnosis of CNS infections. Neurol Clin 1986;4:41–61. [2] Siva A. Vasculitis of the nervous system. J Neurol 2001;248:451–68. [3] Appenzeller S, Faria AV, Zanardi VA, Fernandes SR, Costallat LT, Cendes F. Vascular involvement of the central nervous system and systemic diseases: etiologies and MRI findings. Rheumatol Int 2008;28:1229–37. [4] Calabrese LH. Primary angiitis of the central nervous system: reflections on 20 years of investigation. Clin Exp Rheumatol 2009;27:S3–4. [5] Neel A, Pagnoux C. Primary angiitis of the central nervous system. Clin Exp Rheumatol 2009;27:S3–4. [6] Younger DS. Vasculitis of the nervous system. Curr Opin Neurol 2004;17:317–36. [7] Antoine D, Che D. Les cas de tuberculose déclarés en France en 2007. Bull Epidémiol Hebd 2009;12–13:106–9. [8] Fain O, Lortholary O, Lascaux V, Amoura I, Badinet P, Beaudreuil J, et al. Extrapulmonary tuberculosis in the northeasten suburbs of Paris: 141 cases. Eur J Intern Med 2000;11:145–50. [9] Dastur DK, Lalitha VS, Udani PM, Parekh U. The brain and meninges in tuberculous meningitis. Gross pathology in 100 cases and pathogenesis. Neurology (Bombay) 1970;18:86–100. [10] Barghava S, Grupta AK, Tandon PN. Tuberculous meningitis — a CT study. Br J Radiol 1982;55:189–96. [11] Jinkins JR. Computed tomography of intracranial tuberculosis. Neuroradiology 1991;33:126–35. [12] Lan SH, Chang WN, Lu CH, Lui CC, Chang HW. Cerebral infarction in chronic meningitis: a comparison of tuberculous meningitis and cryptococcal meningitis. Q J Med 2001;94:247–53. [13] Gupta RK, Gupta S, Singh D, Sharma B, Kohli A, Gujral RB. MR imaging and angiography in tuberculous meningitis. Neuroradiology 1994;36:87–92. [14] Offenbacher H, Fazekas F, Schmidt R, Kleinert R, Payer F, Kleinert G, et al. MRI in tuberculous meningoencephalitis: report of four cases and review of the neuroimaging literature. J Neurol 1991;238:340–4. [15] Berenguer J, Moreno S, Laguna F, Vicente T, Andrados M, Ortega A, et al. Tuberculous meningitis in patients infected with the human immunodeficiency virus. N Engl J Med 1992;326:668–72. [16] Chan KH, Cheung RTF, Lee R, Mak W, Ho SL. Cerebral infarcts complicating tuberculous meningitis. Cerebrovasc Dis 2005;19:391–5. [17] Koh SB, Kim BJ, Ho Park M, Yu SW, Park KW, Lee DH. Clinical and laboratory characteristics of cerebral infarction in tuberculous meningitis: a comparative study. J Clin Neurosci 2007;14:1073–7. [18] Leiguarda R, Berthier M, Starkstein S, Nogués M, Lylyk P. Ischemic infarction in 25 children with tuberculous meningitis. Stroke 1988;19:200–4. [19] Feigin VL, Lawes CM, Bennett DA, Anderson CS. Stroke epidemiology: a review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2003;2:43–53. [20] Katrak SM, Shembalkar PK, Bijwe SR, Bhandarkar LD. The clinical, radiological and pathological profile of tuberculous meningitis in patients with and without human immunodeficiency virus infection. J Neurol Sci 2000;181:118–26. [21] Thwaites GE, Duc Bang N, Huy Dung N, Thi Quy H, Thi Tuong Oanh D, Thi Cam Thoa N, et al. The influence of HIV infection on presentation, response to treatment and outcome in adults with tuberculous meningitis. J Infect Dis 2005;192:2135–40. [22] Prasad K, Singh MB. Corticosteroids for managing tuberculous meningitis. Cochrane Database Syst Rev 2008;23:CD002244. [23] Thwaites GE, Macmullen-Price J, Thi Hong Chau T, Phuong Mai P, Thi Dung N, Simmons CP, et al. Serial MRI to determine the effect of dexamethasone on the cerebral pathology of tuberculous meningitis: an observational study. Lancet Neurol 2007;6:230–6. [24] Katti MK. Pathogenesis, diagnosis, treatment and outcome aspects of cerebral tuberculosis. Med Sci Monit 2004;10:215–29.
e104
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
[25] Thwaites GE, Hien TT. Tuberculous meningitis: many questions, too few answers. Lancet Neurol 2005;4:160–70. [26] Hajj-Ali RA, Calabrese LH. Central nervous system vasculitis. Curr Opin Rheumatol 2009;21:10–8. [27] Semlali S, El Kharras A, Mahi M, Hsaini Y, Benameur M, Aziz N, et al. Imaging features of CNS tuberculosis. J Radiol 2008;89:209–20. [28] Schoeman J, Hewlett R, Donald P. MR of childhood tuberculosis meningitis. Neuroradiology 1988;30:473–7. [29] Dastur DK, Dave UP. Ultrastructural basis of the vasculopathy in and around brain tuberculomas. Am J Pathol 1977;89:35–50. [30] Hsieh FY, Chia LG, Shen WC. Locations of cerebral infarctions in tuberculous meningitis. Neuroradiology 1992;34:197–9. [31] Belorgey L, Lalani I, Zakaria A. Ischemic stroke in the setting of tuberculous meningitis. J Neuroimaging 2006;16:364–6. [32] Blumberg HM, Burman WJ, Chaisson RE, Daley CL, Etkind SC, Friedman LN, et al. American Thoracic Society, Centers for Disease Control and Prevention and the Infectious Diseases Society. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis. Am J Respir Crit Care Med 2003;167:603–62. [33] Schoeman JF, Van Zyl LE, Laubscher JA, Donald PR. Effect of corticosteroids on intracranial pressure, computed tomography findings and clinical outcome in young children with tuberculous meningitis. Pediatrics 1997;99:226–31.
[34] Simmons CP, Thwaites GE, Quyen NT, Chau TT, Mai PP, Dung NT, et al. The clinical benefit of adjunctive dexamethasone in tuberculous meningitis is not associated with measurable attenuation of peripheral or local immune responses. J Immunol 2005;175:579–90. [35] Thwaites G, Fisher M, Hemingway C, Scott G, Solomon T, Innes J. British Infection Society guidelines for the diagnosis and treatment of tuberculosis of the central nervous system in adults and children. J Infect 2009;59:167–87. [36] Thwaites GE, Nguyen DB, Nguyen HD, Hoang TQ, Do TT, Nguyen TC, et al. Dexamethasone for the treatment of tuberculous meningitis in adolescents and adults. N Engl J Med 2004;351:1741–51. [37] Donald PR, Schoeman JF. Tuberculous meningitis. N Engl J Med 2004;351:1719–20. [38] McAllister WAC, Thompson PJ, Al-Habet SM, Rogers HJ. Rifampicin reduces effectiveness and bioavailability of prednisolone. Br Med J 1983;286:923–5. [39] Havlir DV, Barnes PF. Tuberculosis in patients with human immunodeficiency virus infection. N Engl J Med 1999;340:367–73. [40] Prasad K, Sahu JK. Duration of anti-tubercular treatment in tuberculous meningitis: challenges and opportunity. Neurol India 2010;58:723–6. [41] van der Worp HB, van Gijn J. Acute ischemic stroke. N Engl J Med 2007;357:572–9. [42] Misra UK, Kalita J, Nair PP. Role of aspirin in tuberculous meningitis: a randomized open label placebo controlled trial. J Neurol Sci 2010;293:12–7.
Contents lists available at ScienceDirect
European Journal of Internal Medicine j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e j i m
Original article
Tuberculous cerebral vasculitis: Retrospective study of 10 cases Nicolas Javaud a,b,⁎, Rita Da Silva Certal b, Jérôme Stirnemann b, Anne-Sophie Morin b, Jean-Marie Chamouard b, Alexandre Augier c, Olivier Bouchaud d, Antoine Carpentier e, Robin Dhote f, Jean-Luc Dumas c, Bruno Fantin g, Olivier Fain b a
Service des Urgences, Hôpital Jean-Verdier, Assistance Publique-Hôpitaux de Paris, Université Paris 13, avenue du 14 Juillet, 93140 Bondy Cedex, France Service de Médecine Interne, Hôpital Jean-Verdier, Assistance Publique-Hôpitaux de Paris, Université Paris 13, avenue du 14 Juillet, 93140 Bondy Cedex, France Service de Radiologie, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France d Service de Maladies Infectieuses et Tropicales, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France e Service de Neurologie et Rééducation Neurologique, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France f Service de Médecine Interne, Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris, Université Paris 13, 125, route de Stalingrad, 93009 Bobigny Cedex, France g Service de Médecine Interne, Hôpital Beaujon, Assistance Publique-Hôpitaux de Paris, Université Paris 7, 100, bd du Général-Leclerc, 92110 Clichy Cedex, France b c
a r t i c l e
i n f o
Article history: Received 22 February 2011 Received in revised form 12 April 2011 Accepted 14 April 2011 Available online 6 May 2011 Keywords: Cerebral vasculitis Tuberculosis Tuberculous meningitis
a b s t r a c t Background: Tuberculous cerebral vasculitis is a complication of tuberculous meningitis. This study was undertaken to determine the epidemiological characteristics, context, diagnostic means and outcomes under treatment of tuberculous cerebral vasculitides. Methods: All consecutive patients diagnosed with tuberculous cerebral vasculitis were identified from the databases of three Internal Medicine, one Neurology and one Infectious Disease Departments in three suburban Parisian hospitals. Results: We describe 10 cases: five men and five women (median age 33.5 [range: 27–55] years). Two were infected with the human immunodeficiency virus. Nine patients had tuberculous meningitis, eight with extraneurological involvement. The following manifestations led to the diagnosis: motor deficit, acute confusional state, headaches, involvement, coma and/or seizures. The cerebral vasculitis revealed tuberculosis in three patients, but tuberculosis was already known when vasculitis was diagnosed for the seven others. The cerebral computed-tomography scan showed cerebral infarctions in five patients, hydrocephalus and tuberculomas in four, while magnetic resonance imaging detected infarctions and leptomeningitis in nine patients, pachymeningitis in one, hydrocephalus and tuberculomas in seven. Therapy combined antituberculous agents with oral corticosteroids for all patients, preceded by a methylprednisolone pulse for five patients. Outcome was favorable for nine patients. Conclusion: We described the non-negligible frequency of tuberculous cerebral vasculitides, their clinical manifestations and their potential severity, and the diagnostic and monitoring contributions of magnetic resonance imaging and magnetic resonance angiography. © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Cerebral vasculitis corresponds to inflammation of cerebral blood vessel walls [1,2]. Primary and secondary cerebral vasculitides are distinguished, with tuberculous cerebral vasculitis belonging among the latter [2–6]. Tuberculous cerebral vasculitis represented 12.5% (3/24) of secondary cases, with other etiologies of secondary central nervous system vasculitides including systemic vasculitides, infections, hematological malignancies, toxic vasculitis and miscellaneous origins [3].
Tuberculous cerebral vasculitis is a complication of tuberculous meningitis, which occurred in 1.5–5% of declared tuberculosis cases in France [7,8]. Cerebral vasculitis-induced strokes had occurred in 41% of the autopsy series of 100 patients with tuberculous meningitis [9]. Depending on the imaging technique used, their frequency ranged from 20.5 to 47.8% with computed tomography (CT) scans [10–12] or 54–66% with magnetic resonance imaging (MRI) [13,14]. The objective of this study was to determine the epidemiological characteristics, context, diagnostic means and outcome under treatment of tuberculous cerebral vasculitides. 2. Materials and methods
⁎ Corresponding author at: Service des Urgences, Hôpital Jean-Verdier, avenue du 14 Juillet, 93143 Bondy Cedex, France. Tel.: + 33 1 48 02 65 34; fax: + 33 1 48 02 63 61. E-mail address: [email protected] (N. Javaud).
We conducted a retrospective study on consecutive patients diagnosed with tuberculous cerebral vasculitis, between January 1995
0953-6205/$ – see front matter © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.ejim.2011.04.004
e100
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
and September 2007, identified from the databases of three Internal Medicine, one Neurology and one Infectious Disease Departments in three suburban Parisian hospitals. The following inclusion criteria were used. Tuberculosis was diagnosed based on at least one of the following: Mycobacterium tuberculosis growth in any cultured sample, epithelioid and giant-cell granuloma with/without caseous necrosis seen in any biopsy, M. tuberculosis-DNA detection by polymerase chain reaction (PCR, bacillus of Koch (BK)-positive); clinical history compatible with tuberculosis; and anti-tuberculous therapy effective at 3 months for all patients. Cerebral vasculitis was diagnosed based on two or more of the following [4,5]: MRI-confirmed infarcts in a patient with a neurological deficit and no cardiovascular risk factors; irregular calibers of one or more cerebral arteries visualized by magnetic resonance angiography (MRA); increased cerebrospinal fluid (CSF) cellularity. Non-inclusion criteria were normal CSF cellularity with normal MRI. The following information was also retrieved from each patient's medical file: age, sex, clinical presentation, laboratory findings and treatments. The same neuroradiologist reread all imaging documents, unblinded for MRA and MRI. 3. Results Thirteen cases of tuberculous cerebral vasculitis were identified between January 1995 and September 2007, representing 21% of the 62 patients with tuberculous meningitis during the same period in the same hospital departments. After excluding three patients because of insufficient data, 10 were analyzed. Ten patients, five men and five women, with a median age of 33.5 [range: 27–55] years, were included (Table 1). Six originated from sub-Saharan Africa, two from Pakistan, one from North Africa and the last from Eastern Europe. Patients 6 and 10 were human immunodeficiency virus (HIV)-infected and their respective CD4+ T-cell counts were 5/mm3 and 42/mm3; the other eight were HIVseronegative. Median follow-up was 18 [4–60] months. Tuberculous cerebral vasculitis revealed tuberculosis infection in three patients. Tuberculosis was known at the time of vasculitis for the other seven patients. For HIV-infected patients 6 and 10, the tuberculosis was extraneurological and had already been treated for 6 months, and their antituberculous drugs had been discontinued for 1 month (pulmonary tuberculosis) or 5 months (lymph-node tuberculosis), respectively. Vasculitis developed in the other five during the course of progressive tuberculosis treated with antituberculous agents for a median of 51 [24–139] days. For patients 2, 3 and 4, it occurred despite ongoing prednisone (1 mg/kg/day), for a median of 35 [24–51] days, and, for patient 9, 1 month after stopping corticosteroids that had been taken for 2 months.
Table 2 Cerebrospinal fluid parameters at the time tuberculous cerebral vasculitis was diagnosed in 9 patients. CSF parameter
n (%)
Median [range]
Protein concentration N 0.5 g/L Glucose concentration/glycemia b 0.5 Chloride concentration b 115 mmol/L Cellularity N5 elements/mm3 Lymphocyte predominance Neutrophil predominance Not determined
8 (89) 7 (78) 8 (89) 8 (89) 4 (50) 3 (38) 1 (13)
2.5 [0.49–4.5] g/L 0.31 [0.17–0.76] mmol/L 108 [94–123] mmol/L 56 [1–630] elements/mm3 51 [23–99] 47 [1–77] –
Data for Patient 6 are not reported.
The neurological manifestations at the time of vasculitis diagnosis were: motor deficit (n = 7), acute confusional state (n = 7), meningeal syndrome (n = 6), headaches (n = 5), cranial nerve involvement (n = 7), coma (n = 4) and/or convulsions (n = 3). Eight patients had extraneurological tuberculosis affecting the lungs in seven, lymph nodes in six and spleen in two. Nine patients underwent lumbar puncture (Table 2) at the time of neurological deterioration. Their CSF had elevated protein levels, low glucose concentrations and increased cellularity in eight. M. tuberculosis was detected in the CSF cultures of five patients, whose PCR was also BK-positive. Initially, seven patients had CT scans. The first scan contributed to the diagnosis of cerebral vasculitis for patients 3, 7 and 10. The abnormalities seen were: cerebral infarcts (n = 5), meningeal contrast-medium uptake (n = 5), hydrocephalus (n = 4) and/or tuberculomas (n = 4). All 10 patients underwent brain MRI (Fig. 1A and B), which contributed to the diagnosis for nine of them. For the other case, MRI showed no signs indicative of cerebral infarcts. For half of the patients, when the CT scan had not been informative, brain MRIrevealed the following abnormalities: infarctions (n = 9); white matter hypersignals (n = 3); leptomeningitis, an inflammation of the pia mater and the arachnoid membrane in contact with cerebral structures (n = 9); pachymeningitis, an inflammation of the dura mater adhering to the skull (n = 1); hydrocephalus (n = 7); and/or tuberculomas (n = 7). In all cases, one or several cerebral infarcts were seen in the carotid area, localized in the deep or superficial sylvian region in six and four patients, respectively. Brain ischemia was also observed in the vertebrobasilar territory in two patients with carotid region involvement. Five patients had several strokes. MRA examination (Fig. 1C) of the circle of Willis found narrowing of the middle cerebral arteries in all eight patients examined, associated with the anterior cerebral arteries in three and the internal carotid artery in three. Bilateral involvement was observed in three patients.
Table 1 Main demographic, clinical, biological and radiological characteristics of the 10 patients at the time of tuberculous cerebral vasculitis diagnosis. Patient
1 2 3 4 5 6 7 8 9 10
Sex/age (yr)
ATD
M/55 F/28 M/35 M/50 F/51 M/26 M/32 F/29 F/28 F/45
No Yes Yes Yes No No Yes No Yes No
Taking prednisone
Focal deficit
Syndrome
Infected sites
Stroke
Meningeal
Inflammatory
CSF Anomalies
BK
Lung
Other
1st CT
1st MRI
No Yes Yes Yes No No No No Yes No
Yes Yes Yes No Yes No No Yes Yes Yes
Yes Yes Yes No Yes No No No Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes No Yes
Yes Yes Yes Yes Yes No Yes Yes Yes Yes
No Yes No No Yes No No Yes Yes Yes
Yes Yes No Yes Yes Yes Yes Yes No No
LN LN, spleen No No No LN, spleen LN LN LN No
– No Yes – No No Yes No – Yes
Yes Yes Yes Yes Yes Yes Yes No Yes Yes
MRA-detected stenosis Yes Yes – Yes Yes – Yes Yes Yes Yes
Abbreviations: ATD = antituberculous drug; CT = computed tomography; CSF = cerebrospinal fluid; BK = bacillus of Koch; MRI = magnetic resonance imaging; MRA = magnetic resonance angiography; LN = lymph node.
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
e101
Fig. 1. Brain imaging. (A) Magnetic resonance imaging T1-weighted sequence after gadolinium injection: 1: basal leptomeningitis enveloping the middle cerebral arteries; 2: level of the sylvian valleys. (B) Magnetic resonance imaging T2-weighted sequence showing ischemic lesions: 1: a hypersignal in the right temporal pole; 2: the cerebral trunk. (C) Magnetic resonance angiography of the circle of Willis: note the irregular, stringy appearance of the middle cerebral arteries.
The therapeutic regimen combined antituberculous drugs and corticosteroids. The former, which lasted a median of 12.5 [range: 4– 32] months, initially given to 9/10 patients, consisted of rifampicin, isoniazid, ethambutol and pyrazinamide. Pregnant patient 2 did not receive pyrazinamide. Rifabutin was prescribed for patient 10, instead of rifampicin, after the reintroduction of antiviral therapy for HIV infection. Initial oral prednisone was given to all patients at the median dose of 1.5 [range: 1.2–2] mg/kg/day, preceded by methylprednisolone pulses, for five patients. The median total duration of corticosteroid treatment was 9.5 [range: 2–27] months. Three patients received antiplateletaggregating drugs and three were prescribed low-weight heparin-type anticoagulants at effective doses. At the end of the 18-month median follow-up, nine patients had favorable outcomes, with persistent
sequelae in five: motor deficit in two, seizures in two and psychomotor retardation in one.
4. Discussion Tuberculous cerebral vasculitis is a complication of tuberculous meningitis. We identified 13 tuberculous cerebral vasculitis cases, which represented 21% of the 62 tuberculous meningitis episodes admitted during the same period to the same hospital departments. We did not find any study on tuberculous cerebral vasculitis but three case series of patients with tuberculous meningitis reported complication rates with cerebral infarcts of 6–47% [9,10,12,15–18].
e102
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
The younger age of our patients and those of Koh et al. [17] can explain the absence of cardiovascular risk factors, as opposed to those followed by Chan et al. [16], who included patients at risk. Because the incidence of stroke increases exponentially with age [19], elderly patients were excluded. Our patients 6 and 10 had HIV coinfections. The results of several studies showed no significant difference concerning the occurrence of CT-detected focal neurological signs or cerebral infarcts during tuberculous meningitis in patients coinfected with HIV or not [15,20,21]. Tuberculous cerebral vasculitis revealed tuberculosis in three of our patients. In the report by Chan et al., only 2/12 (16.7%) patients had cerebral infarct(s) at the time meningitis was diagnosed, but cerebral infarcts complicated the outcomes of their 10 other patients [16]. Indeed, tuberculous cerebral vasculitis can develop despite treatment with antituberculous drugs and corticosteroids. For our patients, the mean time to the appearance of tuberculous cerebral vasculitis was 40 [range: 0–139] days after the onset of antituberculous therapy, and it developed despite corticotherapy given to four patients. This observation agrees with the findings of Chan et al., who found a 40 [range: 0–128]-day interval for 5/12 of the patients who were given corticosteroids [16,22]. Tuberculous cerebral vasculitis should be included in the differential diagnosis of any neurological deterioration arising during the course of tuberculosis, particularly tuberculous meningitis. The clinical neurological signs seen in our patients (focal neurological deficit, acute confusional state, meningeal syndrome, headaches, cranial nerve paralysis, coma and seizures) are in accordance with those found in patients with tuberculous meningitis complicated by stroke. These signs are not specific to cerebral vasculitis [16–18]. In addition, vasculitis-associated strokes were asymptomatic in 2/9 of our patients, which is close to the 3/12 reported by Chan et al. [16]. Lumbar puncture led to the diagnosis of meningitis for 8/9 of our patients, but was unable to do so for HIV-infected Patient 6 because of her severe immunodepression (5 CD4+ T cells/mm3). M. tuberculosis was found more often in the CSF of HIV-infected than non-infected patients [23], whereas CSF cellularity was lower in HIV-infected patients [24]. In a prospective, double-blinded, randomized study on the use of dexamethasone to treat tuberculous meningitis in 96 HIVinfected and 432 non-infected patients [21], M. tuberculosis was found in the CSF of 41.7% versus 29.6% (p = 0.029), respectively, with corresponding CSF cellularity values of 99 versus 120 leukocytes/mm3 (p = 0.069). A retrospective analysis of 53 tuberculous meningitis patients showed that CSF protein concentrations (1.24 and 1.75 g/L) and cellularity values (146/mm3 and 206/mm3) were significantly lower for the 22 HIV-infected than the 31 non-infected, respectively [20]. The CSF PCR was BK-positive for five of our patients, whose CSF cultures grew M. tuberculosis. These findings are in agreement with a reported 56% sensitivity and 98% specificity of CSF PCR for BK [25]. The sensitivity of this technique is not better than classical bacteriology, because 81% of the 132 tuberculous meningitis patients had bacteriological diagnoses: 58% had multidrug-resistant acid-fast bacteria on direct microscope examination and 71% had positive cultures [25]. Cerebral CT scan can contribute to diagnosing tuberculous cerebral vasculitis [5] and yielded the diagnosis for 3/8 of our patients. However, cerebral vasculitis is not directly visible on CT scans. The indirect signs of arteritis are those of cerebral infarcts, appearing as hypodense areas with variable and irregular uptake of contrast medium [1,24]. The CT scan-demonstrated frequency of cerebral ischemia ranged from 20.5 to 38% [24]. Brain MRI contributed to the diagnosis of tuberculous cerebral vasculitis for nine of our 10 patients. It was sensitive (90–100%) but poorly specific for the detection of vasculitis-associated parenchymatous anomalies in published series [3,5,26]. The results of several studies showed the superiority of MRI, compared to CT scans, for the
diagnosis of tuberculous meningitis and its complications, such as stroke [14,26,27]. The MRI-detected abnormalities suggestive of cerebral vasculitis are infarctions [11,28] and white matter hypersignals. However, these signs are not specific to tuberculous cerebral vasculitis [3,5,6,26]. Signs of tuberculosis neuromeningitis orienting the etiological diagnosis can be present: leptomeningitis [27], pachymeningitis, hydrocephalus and tuberculomas [13,14,29]. The localization of strokes associated with tuberculous cerebral vasculitis is a factor orienting the etiological diagnosis. These strokes were found, in 6/10 cases, in the deep sylvian region, which agrees with previous studies [13,18,30] and can be explained by the presence of tuberculous meningitis exudates at the level of the basal cisterna, leading to vasculitis in the vessels crossing those exudates. The smallcaliber arteries, such as the lenticulostriate arteries, are obstructed more easily, thereby provoking strokes in the central gray nucleus and internal capsule, corresponding to the deep sylvian region [13,18,30]. Furthermore, these strokes are frequently multiple and bilateral [30]. Notably, half our patients had multiple but unilateral strokes. Depending on the series, MRA sensitivity for the diagnosis of cerebral vasculitis ranged from 59 to 90% [5]. All of our patients who underwent MRA had lesions suggestive of cerebral vasculitis. MRA specificity for the etiology is low [3], as brain MRA can be abnormal during, for example, systemic vasculitides (Takayasu arteritis, polyarteritis nodosa, giant-cell arteritis…), systemic autoimmune diseases (systemic lupus erythematosus, Sjögren's syndrome, antiphospholipid syndrome…), cancer (lymphomas…), drug-induced vasculitides (narcotics, amphetamines…), infectious vasculitides (syphilis, HIV…) and reversible brain angiopathy. MRA classically visualizes segmental and arterial narrowing, parietal irregularities and, sometimes, obstructions. Although the cerebral vasculature can be visualized by MRA, direct angiography remains the gold standard for imaging the vascular lumen [4,5]. The arteries most frequently involved by tuberculous cerebral vasculitis are those at the base of the skull, particularly, the middle cerebral artery [5,13,18], which was affected in all our patients. Patient 8's MRA showed narrowing of the right middle cerebral artery, with no visible infarcts on the brain MRI. Gupta et al. found the same frequency (10%) [13]. Moreover, tuberculous cerebral vasculitis was bilateral in three of our patients but no bilateral cerebral infarcts were seen. Thus, the absence of brain MRI-detected infarcts does not exclude tuberculous cerebral vasculitis. Finally, some authors described strokes during the course of tuberculous meningitides, with normal MRA images (other stroke causes had been excluded) [31]. Gupta et al. reported 20% strokes [13]. Other authors did not eliminate tuberculous cerebral vasculitis, thinking that the infarct-affected vessels were of smaller caliber at the limit of MRA resolution [36]. Although all our patients with stroke had abnormal MRA images, MRA had not been done for two of our patients. Treatment of tuberculous cerebral vasculitis is based on corticosteroids, by analogy with other cerebral vasculitides, and their efficacy against tuberculous meningitis, even though no controlled study has demonstrated it [16,22,23,26,32–36]. The pathophysiological mechanisms implicated in cerebral vasculitis are diverse and can include: a direct pathogenic effect of the infectious agent on the vessels, immunological involvement via the induction of antigen expression on endothelial cells and the formation of immune complexes. The non-infectious immunological mechanisms can play an important role in infectious vasculitides, justifying the use of immunosuppressants [4,37]. The corticosteroids evaluated for the treatment of tuberculous meningitis are dexamethasone and prednisolone [22,23,32,35,36]. Dexamethasone use prolonged survival. In a prospective, double-blinded, randomized study on dexamethasone treatment of tuberculous meningitis, 87/274 (32%) of treated versus 112/271 (41%) of untreated patients died (p= 0.01) [36]. Patients received intravenous dexamethasone as follows, for weeks 1–4, respectively: 0.4 mg/kg/day, 0.3 mg/kg/day,
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
0.2 mg/kg/day and 0.1 mg/kg/day. Then, 4 weeks of oral treatment, starting at 4 mg/day week 1, and declining 1 mg/week thereafter. The frequency of MRI-detected strokes was lower with the same dexamethasone regimens but not significantly so. In a prospective, randomized study [23], even though infarcts were rare at presentation (2/22 patients had an initial CT scan), 11/27 patients had infarcts after 60 days of followup. The patients treated with dexamethasone had a lower stroke frequency than those given the placebo (27% versus 58%, respectively; p= 0.13). Corticosteroids should be prescribed in combination with antituberculous drugs to treat tuberculous meningitis. This regimen lowers the risk of death and neurological sequelae in survivors [22]. The steroid dose should be adapted to the enzyme-inducer effect of rifampicin [32]. In 1993, McAllister et al. showed that rifampicin diminished the area under the prednisolone time–concentration curve by 66% [38]. The duration of corticosteroid administration should also be evaluated, as in our study, the median time to the onset of tuberculous cerebral vasculitis after starting antituberculous therapy was 51 [range: 24–139] days. In light of the sometimes late occurrence of cerebral vasculitis during treatment, one can wonder about the potential usefulness of prolonging corticotherapy, especially at full dose. Randomized–controlled trials are needed to assess the optimal duration of steroid administration (6 versus 12 weeks), their dose and type of glucocorticoid (dexamethasone, prednisolone or prednisone). Treatment must also address the causal disease, hence antituberculous agents. For 90% of the patients, the initial regimen comprised four molecules, combining isoniazid, rifampicin, ethambutol and pyrazinamide. This combination is also the currently recommended one [24,25]. Patient 2's initial therapy combined three of those drugs, without pyrazinamide, which is contraindicated during pregnancy. Rifabutin, which has a weaker enzyme-inducer effect, replaced rifampicin for our HIV-infected patient 10, as it had been shown to be well-tolerated and to have efficacy comparable to that of rifampicin for the treatment of tuberculosis in HIV-infected patients [39]. The median treatment duration for our patients was 12.5 months, which is consistent with current guidelines recommending 9–12 months [24,25]. Although the duration of anti-tuberculous therapy for tuberculous meningitis is controversial, the evidence base for shortterm therapy for tuberculous meningitis is weak [40]. Adding an antiplatelet-aggregating agent is an option, in light of their efficacy during the acute phase of ischemic strokes, but they have not been assessed in cerebral vasculitides [41]. In a recent study, aspirin obtained non-significantly fewer strokes and significantly lower 3-month mortality of tuberculous meningitis patients [42]. Development of tuberculous cerebral vasculitis is a poor-prognosis factor. None of our patients died but 12–24% mortality has been reported [12,16–18]; however, our patient population was small. In agreement with previous reports, 50% of our patients had persistent neurological sequelae [12,16,17]. We described 10 patients who developed tuberculous cerebral vasculitis, diagnosed based on clinical symptoms, biological analyses and brain MRI and MRA findings, while the gold standard for its diagnosis is direct angiography and biopsy [4,5]. 5. Conclusion Herein, we described the non-negligible frequency of tuberculous cerebral vasculitides, their clinical manifestations and potential severity, and the diagnostic and monitoring inputs of MRI and MRA. Further investigations are needed to determine: the contributions of MRI and MRA for all tuberculous meningitis patients to detect tuberculous cerebral vasculitis early and increase ongoing corticotherapy; the modalities of the latter, including the type of glucocorticoid to be prescribed, its dose and treatment duration; and the potential role of antiplatelet-aggregating agents.
e103
Learning points • Tuberculous cerebral vasculitis is a complication of tuberculous meningitis. • Tuberculous cerebral vasculitis can reveal tuberculosis but can develop despite treatment with antituberculous drug and corticosteroids. • Further investigations are needed to determine the contributions of MRI and MRA for all tuberculous meningitis patients to detect tuberculous cerebral vasculitis early. Funding source None. Conflicts of interest statement It is certified that there is no actual or potential conflict of interest in relation to this article. References [1] Sarwar M, Falkoff G, Naseem M. Radiologic techniques in the diagnosis of CNS infections. Neurol Clin 1986;4:41–61. [2] Siva A. Vasculitis of the nervous system. J Neurol 2001;248:451–68. [3] Appenzeller S, Faria AV, Zanardi VA, Fernandes SR, Costallat LT, Cendes F. Vascular involvement of the central nervous system and systemic diseases: etiologies and MRI findings. Rheumatol Int 2008;28:1229–37. [4] Calabrese LH. Primary angiitis of the central nervous system: reflections on 20 years of investigation. Clin Exp Rheumatol 2009;27:S3–4. [5] Neel A, Pagnoux C. Primary angiitis of the central nervous system. Clin Exp Rheumatol 2009;27:S3–4. [6] Younger DS. Vasculitis of the nervous system. Curr Opin Neurol 2004;17:317–36. [7] Antoine D, Che D. Les cas de tuberculose déclarés en France en 2007. Bull Epidémiol Hebd 2009;12–13:106–9. [8] Fain O, Lortholary O, Lascaux V, Amoura I, Badinet P, Beaudreuil J, et al. Extrapulmonary tuberculosis in the northeasten suburbs of Paris: 141 cases. Eur J Intern Med 2000;11:145–50. [9] Dastur DK, Lalitha VS, Udani PM, Parekh U. The brain and meninges in tuberculous meningitis. Gross pathology in 100 cases and pathogenesis. Neurology (Bombay) 1970;18:86–100. [10] Barghava S, Grupta AK, Tandon PN. Tuberculous meningitis — a CT study. Br J Radiol 1982;55:189–96. [11] Jinkins JR. Computed tomography of intracranial tuberculosis. Neuroradiology 1991;33:126–35. [12] Lan SH, Chang WN, Lu CH, Lui CC, Chang HW. Cerebral infarction in chronic meningitis: a comparison of tuberculous meningitis and cryptococcal meningitis. Q J Med 2001;94:247–53. [13] Gupta RK, Gupta S, Singh D, Sharma B, Kohli A, Gujral RB. MR imaging and angiography in tuberculous meningitis. Neuroradiology 1994;36:87–92. [14] Offenbacher H, Fazekas F, Schmidt R, Kleinert R, Payer F, Kleinert G, et al. MRI in tuberculous meningoencephalitis: report of four cases and review of the neuroimaging literature. J Neurol 1991;238:340–4. [15] Berenguer J, Moreno S, Laguna F, Vicente T, Andrados M, Ortega A, et al. Tuberculous meningitis in patients infected with the human immunodeficiency virus. N Engl J Med 1992;326:668–72. [16] Chan KH, Cheung RTF, Lee R, Mak W, Ho SL. Cerebral infarcts complicating tuberculous meningitis. Cerebrovasc Dis 2005;19:391–5. [17] Koh SB, Kim BJ, Ho Park M, Yu SW, Park KW, Lee DH. Clinical and laboratory characteristics of cerebral infarction in tuberculous meningitis: a comparative study. J Clin Neurosci 2007;14:1073–7. [18] Leiguarda R, Berthier M, Starkstein S, Nogués M, Lylyk P. Ischemic infarction in 25 children with tuberculous meningitis. Stroke 1988;19:200–4. [19] Feigin VL, Lawes CM, Bennett DA, Anderson CS. Stroke epidemiology: a review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2003;2:43–53. [20] Katrak SM, Shembalkar PK, Bijwe SR, Bhandarkar LD. The clinical, radiological and pathological profile of tuberculous meningitis in patients with and without human immunodeficiency virus infection. J Neurol Sci 2000;181:118–26. [21] Thwaites GE, Duc Bang N, Huy Dung N, Thi Quy H, Thi Tuong Oanh D, Thi Cam Thoa N, et al. The influence of HIV infection on presentation, response to treatment and outcome in adults with tuberculous meningitis. J Infect Dis 2005;192:2135–40. [22] Prasad K, Singh MB. Corticosteroids for managing tuberculous meningitis. Cochrane Database Syst Rev 2008;23:CD002244. [23] Thwaites GE, Macmullen-Price J, Thi Hong Chau T, Phuong Mai P, Thi Dung N, Simmons CP, et al. Serial MRI to determine the effect of dexamethasone on the cerebral pathology of tuberculous meningitis: an observational study. Lancet Neurol 2007;6:230–6. [24] Katti MK. Pathogenesis, diagnosis, treatment and outcome aspects of cerebral tuberculosis. Med Sci Monit 2004;10:215–29.
e104
N. Javaud et al. / European Journal of Internal Medicine 22 (2011) e99–e104
[25] Thwaites GE, Hien TT. Tuberculous meningitis: many questions, too few answers. Lancet Neurol 2005;4:160–70. [26] Hajj-Ali RA, Calabrese LH. Central nervous system vasculitis. Curr Opin Rheumatol 2009;21:10–8. [27] Semlali S, El Kharras A, Mahi M, Hsaini Y, Benameur M, Aziz N, et al. Imaging features of CNS tuberculosis. J Radiol 2008;89:209–20. [28] Schoeman J, Hewlett R, Donald P. MR of childhood tuberculosis meningitis. Neuroradiology 1988;30:473–7. [29] Dastur DK, Dave UP. Ultrastructural basis of the vasculopathy in and around brain tuberculomas. Am J Pathol 1977;89:35–50. [30] Hsieh FY, Chia LG, Shen WC. Locations of cerebral infarctions in tuberculous meningitis. Neuroradiology 1992;34:197–9. [31] Belorgey L, Lalani I, Zakaria A. Ischemic stroke in the setting of tuberculous meningitis. J Neuroimaging 2006;16:364–6. [32] Blumberg HM, Burman WJ, Chaisson RE, Daley CL, Etkind SC, Friedman LN, et al. American Thoracic Society, Centers for Disease Control and Prevention and the Infectious Diseases Society. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis. Am J Respir Crit Care Med 2003;167:603–62. [33] Schoeman JF, Van Zyl LE, Laubscher JA, Donald PR. Effect of corticosteroids on intracranial pressure, computed tomography findings and clinical outcome in young children with tuberculous meningitis. Pediatrics 1997;99:226–31.
[34] Simmons CP, Thwaites GE, Quyen NT, Chau TT, Mai PP, Dung NT, et al. The clinical benefit of adjunctive dexamethasone in tuberculous meningitis is not associated with measurable attenuation of peripheral or local immune responses. J Immunol 2005;175:579–90. [35] Thwaites G, Fisher M, Hemingway C, Scott G, Solomon T, Innes J. British Infection Society guidelines for the diagnosis and treatment of tuberculosis of the central nervous system in adults and children. J Infect 2009;59:167–87. [36] Thwaites GE, Nguyen DB, Nguyen HD, Hoang TQ, Do TT, Nguyen TC, et al. Dexamethasone for the treatment of tuberculous meningitis in adolescents and adults. N Engl J Med 2004;351:1741–51. [37] Donald PR, Schoeman JF. Tuberculous meningitis. N Engl J Med 2004;351:1719–20. [38] McAllister WAC, Thompson PJ, Al-Habet SM, Rogers HJ. Rifampicin reduces effectiveness and bioavailability of prednisolone. Br Med J 1983;286:923–5. [39] Havlir DV, Barnes PF. Tuberculosis in patients with human immunodeficiency virus infection. N Engl J Med 1999;340:367–73. [40] Prasad K, Sahu JK. Duration of anti-tubercular treatment in tuberculous meningitis: challenges and opportunity. Neurol India 2010;58:723–6. [41] van der Worp HB, van Gijn J. Acute ischemic stroke. N Engl J Med 2007;357:572–9. [42] Misra UK, Kalita J, Nair PP. Role of aspirin in tuberculous meningitis: a randomized open label placebo controlled trial. J Neurol Sci 2010;293:12–7.