Table 1.
Characteristics of Studies Meeting Inclusion Criteria
| First Author | Year | Study Design | Country | No. | Diagnostic Criteriaa | HIV, No. (%) |
Confirmed TBM, No. (%) |
Suspected TBM, No. (%)b |
MDR TB, No. (%) | INH Mono-R, No. (%) | Antituberculous Treatmentc | Steroidsc | Outcome(s) and Time Point Reported, mo |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Africa | |||||||||||||
| Luma [17] | 2013 | Case series | Cameroon | 54 | 2 a b |
54 (100) | 1 (2) | 53 (98) | 2RHZE/6-8RH | All received steroids: unspecified drug(s), dose, & duration | In-hospital mortality | ||
| Marais [18] | 2011 | Cohort | South Africa | 120 | 2 a b c d e f |
106 (88) | 47 (39) | 73 (61) | 3 (3) | 3 (3) | RHZE | Not specified | In-hospital and 6-mo mortality |
| Thinyane [19] | 2015 | Case series | Lesotho | 22 | 2 a b e f |
15 (68) | 0 (0) | 22 (100) | RHZE | Not given | Mortality at the end of follow-up | ||
| Cresswell [20] | 2018 | Cohort | Uganda | 195 | 2 a b c d |
106 (54) | 74 (38) | 93 (48) | 0 (0) | 0 (0) | RHZE | Not specified | In-hospital mortality |
| Raberahona [21] | 2017 | Case series | Madagascar | 75 | 1 | 3 (4) |
8 (11) | 44 (59) probable, 23 (31) possible | 2RHZE/6RH + S if prior TB (n = 2) | Not given | Mortality at 8 mo | ||
| South america | |||||||||||||
| Gonzalez-Duarte [22] | 2011 | Cohort | Mexico | 64 | 2 a c f |
14 (22) | 44 (69) | 20 (31) | 2RHZE/RH – mean time of therapy was 11.9 ± 7 mo | 57 (78%) received steroids, unspecified drug(s), dose, & duration | Mortality and neurological outcomes at 5 mo | ||
| Alarcon [23] | 2013 | Cohort | Ecuador | 310 | 2 a b d e f g h |
2 (1) |
140 (45) | 170 (55) | 2RHZ + E or S or quinolone/10RH (quinolone given to some) | Steroids given to patients with severe disease, unspecified drug(s), dose, & duration | Mortality and neurological outcomes at 12 mo | ||
| Asia | |||||||||||||
| Torok [24] | 2008 | Cohort | Vietnam | 58 | 2 b d e f |
58 (100) | 54 (93) | 4 (7) | 4 (7) | 3RHZE + S if prior TB/6RH | D (0.3–0.4 mg/kg) tapered over 6–8 wk | Mortality at 9 mo | |
| Torok [25] | 2011 | RCT | Vietnam | 253 | 2 a b d e f |
253 (100) | 158 (62) | 95 (38) | 4 (2) | 3RHZE + S if prior TB/6RH | D (0.3–0.4 mg/kg) tapered over 6–8 wk | Mortality and neurological outcomes at 9 and 12 mo | |
| Heemskerk [6] | 2016 | RCT | Vietnam | 817 | 1 | 349 (43) | 407 (50) | 214 (26) 174 (21) | 15 (2) | 86 (11) | 2RHZE/6RH + S if prior TB + L in 1 trial arm | D (0.3–0.4 mg/kg) for 6–8 wk | Mortality at 9 mo |
| Thwaites [26] | 2002 | Cohort | Vietnam | 56 | 2 a b d |
11 (20) | 56 (100) | 0 (0) | 0 (0) | 3 (5) | 3RHZE/6RHZ if HIV+ 3RHZS/6RHZ if HIV- | Not given | Mortality at 3 mo |
| Thwaites [27] | 2004 | RCT | Vietnam | 545 | 2 a b d e f |
98 (18) | 187 (34) | 358 (66) | 10 (2) | 3RHZS/6RHZ 3RHZE/6RHZ if HIV+ or prior history of TB |
D (0.3–0.4 mg/kg) tapered over 4 wk, then oral treatment (4 mg/d) tapered for 4 wk | Mortality and neurological outcomes at 9 mo | |
| van Laarhoven [8]d | 2017 | Cohort | Indonesia | 608 | 2 a b c d |
93 (15) | 336 (55) | 272 (45) | RHZE (n = 47: high-dose R; n = 25: M instead of E) | 91% received steroids; drug, dose, and duration not specified | Mortality at 12 mo | ||
| Singh [28] | 2016 | Cohort | India | 141 | 1 | 13 (9) | 54 (38) | 87 (62) | 2RHZS/7HE | D (0.3–0.4 mg/kg) tapered over 4 wk, then oral treatment (4 mg/d) tapered for 4 wk | Neurological outcomes at 9 mo | ||
| Tai [29] | 2016 | Cohort | Malaysia | 36 | 1 | 3 (8) | 23 (64) | 13 (36) | 2RHZE/10RH | Not specified | Neurological outcomes at 3 mo | ||
| Chen [30] | 2014 | Cohort | Taiwan | 38 | 2 b d f g |
2 (5) | Not reported | Not reported | 2RHZE/10-16RHE | D (12–16 mg), P (60–80 mg) tapered over 6–8 wk | Mortality and neurological outcomes at 18 mo | ||
| Kalita [31]e | 2014 | RCT | India | 60 | 2 a b c d e f |
3 (5) | 24 (40) | 36 (60) | RHZE | P (0.5 mg/kg/d) for 1 mo, tapered over 4 wk | Mortality and neurological outcomes at 6 mo | ||
| Sheu [32] | 2012 | Case series | Taiwan | 91 | 2 b c d e f g |
3 (3) | Not specified | Not specified | RHZE +/- S | Either D 12–16 mg/d or P 60–80 mg/d over 1.5–2 mo | In-hospital mortality and neurological outcomes | ||
| Wasay [33] | 2014 | Case series | Pakistan | 404 | 2 a b d e f g h |
1 (0.2) | 35 (9) | 369 (91) | 2 (0.5) | RHZE + 8% (n = 34) received S | Unspecified regimen given to all | Mortality and neurological outcomes at 2 mo | |
| Chotmongkol [34] | 1996 | RCT | Thailand | 59 | 2 a | 0 (0) | 6 (10) | 53 (90) | 2RHZS/4RH | 29 (52%) P 60 mg tapered over 5 wk | Mortality and neurological outcomes at 6 and 18 mo | ||
| Lu [35] | 2001 | Cohort | China | 36 | 2 a c d e f |
0 (0) | 23 (64) | 13 (36) | RHZE +/- C and/or S for drug toxicity | Unspecified steroid given to patients with clinical deterioration | Mortality and neurological outcomes at 3 and 6 mo | ||
| Wang [36] | 2002 | Cohort | China | 41 | 2 a d f g |
0 (0) | 22 (54) | 19 (46) | 0 (0) | 0 (0) | RHZE | Unspecified steroid given to 9 patients | Mortality at 6 mo |
| Chotmongkol [37]f | 2003 | Cohort | Thailand | 45 | 2 a b d |
0 (0) | 2 (4) | 42 (93) | 2RHZS/4RH | Not given | Mortality at 6 mo | ||
| Thwaites [38] | 2003 | Cohort | Vietnam | 21 | 2 a b d e f g |
0 (0) | 15 (71) | 6 (29) | 0 (0) | 3RHZS/6RHZ | Not given | Mortality and neurological outcomes at 9 mo | |
| Malhotra [39] | 2009 | RCT | India | 91 | 2 a b e f |
0 (0) | 18 (20) | 73 (80) | 2RHZE or S/7RH | D (0.3–0.4 mg/kg) tapered over 4 wk, then oral treatment (4 mg/d) tapered for 4 wk OR MP 5 d OD of either 1 g (weight >50 kg) or 20 mg/kg (<50 kg) | Mortality and neurological outcomes at 6 and 18 mo | ||
| Hsu [40] | 2010 | Case series | Taiwan | 108 | 2 a d c e f g h |
0 (0) | 46 (43) | 62 (57) | 1 (1) | 2 (2) | 6RHZ + S, C, or L in case of toxicity or side effects | P (minimum 20 mg) for >7 d given for 1 to >4 wk in n = 106 | Mortality at 9 mo |
| Sharma [41] | 2013 | Case series | India | 42 | 2 a e f g |
0 (0) | 4 (10) | 38 (90) | RHZE | 6 wk of steroids, unspecified drug(s) & dose | Mortality and neurological outcomes at 6 mo | ||
| Sun [42] | 2014 | Cohort | China | 33 | 2 a d e f h |
0 (0) | 7 (21) | 26 (79) | RHZE +/- PAS + L if in trial arm 2 | D 1.5–15 mg/d for 1.5–6 wk | In-hospital neurological outcomes | ||
| Kalita [43] | 2014 | Case series | India | 34 | 2 a b c d e f h |
0 (0) | 34 (34) | 0 (0) | 9RHZE/9RH | P (0.8 mg/kg, max 40 mg) for 1 mo | Mortality and neurological outcomes at 6 mo | ||
| Imam [44] | 2015 | Case series | Qatar | 80 | 2 a b c d e f g h |
0 (0) | 35 (44) | 45 (56) | RHZE + 4% received S, M, and A | D (med 21 mg/d) P (med 40 mg/d) over 3–9 wk | Mortality and neurological outcomes at 12 mo | ||
| Zhang [45] | 2016 | Cohort | China | 401 | 1 | 0 (0) | 131 (33) | 202 (50) | 4 (1) | 6 (1) | RHZE + L | Not specified | 5-y mortality |
| Kalita [46] | 2016 | RCT | India | 57 | 2 a b d e f h |
0 (0) | 18 (32) | 39 (68 | 6RHZE + L in trial arm/12RH for following year | P (0.5 mg/kg/d) for 1 mo tapered over 1 mo | Mortality and neurological outcomes at 3 and 6 mo | ||
| Li [47] | 2017 | Case series | China | 154 | 1 | 0 (0) | 18 (12) | 98 (61) probable, 42 (27) possible | 2-4RHZE/6-12RH | D (early treatment), unspecified dose & duration | Mortality and neurological outcomes at 8 mo | ||
| Mai [48] | 2018 | RCT | Vietnam | 120 | 1 | 0 (0) | 92 (77) | 26 (22) | 1 (0.1) | 10 (8) | 3RHZES/6RH | D (0.3–0.4 mg/kg) tapered over 4 wk, then oral treatment (4 mg/d) tapered for 4 wk | Mortality and neurological outcomes at 2 and 8 mo |
| Europe | |||||||||||||
| Cagatay [49] | 2004 | Cohort | Turkey | 42 | 2 a b d e f g h |
2 (5) | 10 (24) | 32 (76) | 3-6RHZE | D (8 mg) for 4–6 wk given to patients who were stage II or III | Mortality at 12 mo | ||
| Doganay [50] | 1995 | Cohort | Turkey | 72 | 2 a b d f |
0 (0) | 72 (100) | 51%: 2RHZS/6RH 49%: various combinations 12–16 mo R, H, Z, E, S | P or D 4–6 wk if MRC stage 3 diseases/signs of raised ICP | Mortality at 2 y | |||
| Sutlas [51] | 2003 | Cohort | Turkey | 61 | 2 b d e f g h |
0 (0) | 19 (31) | 42 (69) | 1RHZES/2-3RHZE/4-9RHZ (if no tuberculoma present)/10-12RH | P (1 mg/kg/d) for 1 mo, tapered for 4 mo | Mortality at 12 mo | ||
| Sengoz [52] | 2008 | Cohort | Turkey | 121 | 2 a b d e f g h |
0 (0) | 52 (43) | 69 (57) | 4 (3) | 2RHZ + E or S/7-10RH | 2D (16 mg/d) for those with neurological deficits | Mortality at the end of follow-up | |
| Miftode [53] | 2015 | Cohort | Romania | 127 | 1 | 0 (0) | 25 (20) | 35 (28) probable, 70 (55) possible | 2-3RHZE/7-9RH | All received: unspecified drug, dose, & duration | In-hospital mortality and neurological outcomes |
Abbreviations: TB treatment: Number of months placed in front of regimen code: A, amikacin; C, ciprofloxacin; D, dexamethasone; E, ethambutol; H, isoniazid; L, levofloxacin; M, moxifloxacin; P, Prednisolone; R, rifampicin; S, streptomycin; Z, pyrazinamide.
aDiagnostic criteria legend: 1= uniform case definition, 2 = other criteria used to diagnose and categorise patients, including a = suggestive CSF picture, b = microscopy, c = Xpert/PCR, d = culture, f = evidence of extraneural TB, g = response to treatment, h = other (history of TB or contact with a TB-infected individual, positive mantoux reaction, IGM AB in the CSF, biopsy, etc.)
bSome participants were considered “suspected” as well as “confirmed” TBM.
cTB treatment (given to all unless specified otherwise): number of months placed in front of regimen code: R = rifampicin, H = isoniazid, Z = pyrazinamide, E = ethambutol, S = streptomycon, L = levofloxacin, M = moxifloxacin, C = ciprofloxacin, A = amikacin, PAS = paraaminosalacylic acid, P = prednisolone, D = dexamethasone, MP = methylprenisolone. Where no duration of antituberculous therapy or steroids is stated, it means it was not clearly specified in the paper.
dvan Laarhoven et al. includes some data from 3 clinical trials in Indonesia [54–56]. The primary studies were excluded from the review to avoid duplication of data.
eOnly included participants who were treated with RHZE.
fTreatment information was taken from Chotmongkol [34], as they were from the same authors, hospital, and decade.