[Downloaded free from http://www.neurologyindia.com on Monday, October 30, 2023, IP: 106.206.158.

76]

Review Article

Access this article online

Quick Response Code:
Three Decades of Vellore Grading
for Tuberculous Meningitis with
Hydrocephalus: A Reappraisal
Vedantam Rajshekhar
Website:
www.neurologyindia.com

DOI: Abstract:
10.4103/0028-3886.332251
Background: This review documents the evolution of the Vellore grading system for tuberculous meningitis
and hydrocephalus (TBMH), its evaluation by different authors, and analyzes the need for further modification
in light of the published literature.
Methods: Published literature was searched in PubMed and Google Scholar using the search terms,
“tuberculous meningitis hydrocephalus” and “Vellore grading.” The retrieved articles were reviewed by the
author and the appropriate ones were chosen for inclusion in the study.
Results: Vellore grade (1–4, with 1 being the best grade and 4 being the worst grade) was found to be the
sole statistically significant factor associated with outcome following VP shunt or ETV in several studies.
Additionally, Vellore grades also correlate with the likelihood of success following ETV. However, the use of
response to external ventricular drainage (EVD) in managing Vellore grade 4 patients has remained contentious
as a small but significant proportion of patients have a good outcome following shunt, irrespective of their
response to the EVD. The latter findings suggest that grade 4 patients might not constitute a homogenous
group. It is proposed that grade 4 be subdivided into grades 4a and 4b, which might help in prognostication
and in surgical management of the hydrocephalus in patients with TBMH.
Conclusions: Vellore grading has proved its utility as a prognostic tool and can aid surgical decision‑making.
However, management of patients in grade 4 might be better rationalized with its division into grades 4a
and 4b.
Key Words:
Endoscopic third ventriculostomy, outcome, prognosis, shunt surgery, tuberculous meningitis

Key Message:
Vellore grading for patients with TBMH has been found to be a reliable tool to prognosticate outcomes
following shunt or ETV. However, subcategorization of grade 4 into grades 4a and 4b might rationalize the
management of patients in this grade.

C linicians in India and several other

impoverished regions of the world continue
to see several patients with tuberculous
were made to correlate the preoperative grade
with the outcome following shunt surgery. The
variable response of patients with TBMH to
meningitis (TBM). One of the most common shunt surgery makes it important to grade the
complications of TBM is hydrocephalus (TBMH), severity of TBM to prognosticate the surgical
this complication being commoner in outcome.
Department of children.[1] Bhagwati, in 1971, first proposed the
Neurological Sciences, use of ventriculoatrial (VA) shunts to manage This article documents the history of grading
Christian Medical the hydrocephalus in patients with TBMH. [2] of patients with TBMH with a focus on the
College, Vellore, Although the reported outcomes following shunt development of Vellore grading and its
Tamil Nadu, India surgery in the series reported in the 1970s and utilization. The article also discusses the possible
1980s were variable, good outcomes were noted need for further changes to the Vellore grading
Address for in approximately 60% of patients on average.[2‑8] based on recent reports of outcomes in patients
correspondence: The authors of some of these series graded undergoing surgery for TBMH.
Dr. Vedantam Rajshekhar the severity of the TBM in their patients prior
Department of to shunt surgery, but no significant attempts
Neurological Sciences, How to cite this article: Rajshekhar V. Three
Christian Medical This is an open access journal, and articles are distributed under the terms Decades of Vellore Grading for Tuberculous Meningitis
College, Vellore, of the Creative Commons Attribution‑NonCommercial‑ShareAlike 4.0 with Hydrocephalus: A Reappraisal. Neurol India
Tamil Nadu ‑ 632 004, License, which allows others to remix, tweak, and build upon the work 2021;69:S569-74.
India. non‑commercially, as long as appropriate credit is given and the new
creations are licensed under the identical terms. Submitted: 05‑May‑2021   Revised: 09-Sep-2021
E‑mail: rajshekhar@ Accepted: 13-Sep-2021   Published: 11-Dec-2021
cmcvellore.ac.in For reprints contact: WKHLRPMedknow_reprints@wolterskluwer.com

© 2021 Neurology India, Neurological Society of India | Published by Wolters Kluwer - Medknow S569
[Downloaded free from http://www.neurologyindia.com on Monday, October 30, 2023, IP: 106.206.158.76]

Rajshekhar: Vellore grading for TBMH

Methods outside the authors’ institution. A modification of this grading

system was used to study the long‑term outcome in patients
PubMed and Google Scholar were searched using the terms with TBMH who had undergone shunt surgery. This study,
“tuberculous meningitis hydrocephalus” and “Vellore published in 1991, had the first version of what was later labeled
grading.” PubMed search yielded only five articles, whereas the Vellore grading system.[14] Some authors who used this
the Google Scholar search yielded 168 articles. The titles and grading system also referred to it as the “Palur” grading system
abstracts of these articles were read and the relevant articles after the first author of that publication. We prefer the term
were chosen for a detailed review. Several articles from the “Vellore grading system” as the framework for this grading
Google Scholar search were found to be irrelevant. Finally, system was taken from work done earlier in the department.
there were 13 articles that used Vellore grading while reporting
outcomes in patients with TBMH who had undergone CSF Vellore grading (1991)
diversion, and three review articles that reported on the use To address the lack of a validated grading system that had been
of Vellore grading in patients with TBMH. Other articles not used for prognostication in patients with TBMH, we evolved
revealed in the search of these two databases were obtained the Vellore grading system [Table 2]. The system was proposed
through cross‑references of the relevant articles or from the in a retrospective study of 114 patients with TBMH, all of whom
personal knowledge of the author. had undergone VA or VP shunts and who were followed
up for 6 months to 13 years (mean: 45.6 months).[14] The first
Early grading systems two grades were applied to patients with normal sensorium,
The grading system proposed by the Medical Research whereas the last two grades had patients with varying degrees
Council (MRC) of the United Kingdom in 1948 is the one of altered sensorium. The grading system was designed to
most frequently used to grade patients with TBM [Table 1].[9,10] be applied to patients in a retrospective study wherein the
Lincoln et al.,[11] in 1960, introduced a grading system for preoperative clinical data were gathered from hospital records.
TBM in children, which again had three grades but differed
considerably from the MRC grading [Table 1]. For several years, Based on the follow‑up outcome, we proposed a management
these were the only grading systems available for TBM. In 1977, algorithm for patients in different grades.[14] Essentially, the
Bhagwati and Singhal[12] proposed a grading system for TBM algorithm advised early shunting for patients in grades I and
that had five grades [Table 1]. However, this grading system II whereas those in grades III and IV should be chosen for
did not gain popularity and was only ever used in another shunt surgery on the basis of their response to two to three
article published by them in 1982.[5] days of external ventricular drainage (EVD) of CSF. Those who
improved should go on to have shunt surgery whereas those
Evolution of the Vellore grading system who did not improve should receive the best medical therapy.
Although the Vellore grading system for TBMH was formally
published in a peer‑reviewed publication in 1991; its origins date Modified Vellore grading (1997)
back to 1974. In 1974, Reddy et al.[13] from Vellore first proposed Although the Vellore grading of 1991 was found to be useful for
a simple grading system for TBMH with four grades [Table 1]. decision making and prognostication of patients with TBMH,
The grading was proposed only for patients with TBM with it suffered from the possibility of interobserver variability in
altered sensorium; those with normal sensorium were not the grading of patients as the patient assessment was subjective
graded by this system. In their study of 61 patients with TBMH to a large extent. To overcome this limitation, we modified
who underwent VA shunt or ventriculoperitoneal (VP) shunt, the grading by incorporating the Glasgow Coma Scale score
32 had normal sensorium and 29 had altered sensorium. Of in the grading [Table 2]. The modified grading system was
those with altered sensorium, 36.8% in grades I and II (better published as part of a prospective study of outcomes in
grades) died on follow‑up, whereas 70% of those in the worse poor‑grade (grades 3 and 4) TBMH patients.[15] Patients in
grades (III and IV) died. The grading system was not published grades 1 and 2 had a GCS score of 15, patients in grade 3 had a
in a peer‑reviewed journal and hence it remained unknown GCS score of 9–14, and those in grade 4 had a GCS score of 3–8.

Table 1: Commonly used grading/staging systems for patients with tuberculous meningitis (TBM) and
tuberculous meningitis with hydrocephalus (TBMH)
Grade/ Medical Research Lincoln et al., 1960 Reddy et al., 1974* Bhagwati and Singhal, 1975 Modified MRC
Stage Council (MRC), 1948 staging, 2005
1 Fully conscious, no Meningitis with no When the patient Only systemic symptoms and signs GCS 15
paresis neurological involvement was arousable and such as pyrexia, headache, etc., without
answering questions overt meningeal signs
2 Decreased level of Evidence of neurological Arousable, but not Presence of overt meningeal signs such GCS 10‑14 or 15
consciousness, localizing changes with no marked answering questions as neck stiffness etc. with neurological
pain changes in sensorium deficits
3 Deeply comatose±gross Marked neurological and Responding to pain Conscious but with a neurological GCS ≤10
paresis sensorial changes only deficit such as cranial nerve palsies and
hemiparesis
4 Comatose and not Stuporous to semiconscious, with or
responding to pain without focal neurological deficit.
5 Decerebrate or deeply comatose state.
*Grading was only used for patients with altered sensorium.

S570 Neurology India | Volume 69 | Supplement 2 | November-December 2021

[Downloaded free from http://www.neurologyindia.com on Monday, October 30, 2023, IP: 106.206.158.76]

Rajshekhar: Vellore grading for TBMH

Table 2: Vellore and modified Vellore grading grades, the modified Vellore grading system enables patients
systems with TBM and TBMH to be stratified into more homogeneous
Grading system Grading groups. This prevents too much variation in the prognosis of
Vellore grading, Grade I individual patients in a particular grade.
1991 (Palur 1. headache, vomiting, fever, and/or neck stiffness
et al.) 2. no neurological deficit Clinical utility of the Vellore grading system
3. normal sensorium The Vellore grading system was proposed both as a prognostic
tool and to decide on the benefit of shunt surgery in patients
Grade II
in different grades.
1. normal sensorium
2. neurological deficit present
i. Prognostic value
Grade III
In the initial 1991 retrospective study, mortality rates
1. altered sensorium but easily arousable for patients in the different Vellore grades at long‑term
2. dense neurological deficit may or may not be follow up were as follows: grade 1, 20%; grade 2, 34.7%;
present
grade 3, 51.9%; and grade 4, 100% (P < 0.001).[14] In the
Grade IV prospective study, using the modified Vellore grade, this
l. deeply comatose clear separation of mortality rates for the different grades
2. decerebrate or decorticate posturing was confirmed (P = 0.002), although the follow‑up was
Modified Vellore Grade 1: GCS 15; no deficits shorter than the retrospective study (mean: 23.1 months vs.
grading, 1998 Grade 2: GCS 15; neurological deficits present 45.6 months).[15] It should be noted that the prognostic value
(Mathew et al.) Grade 3: GCS 9‑4; neurological deficits may or of the grading was verified in patients with an average
may not be present follow‑up of over 2–4 years.
Grade 4: GCS 3‑8; neurological deficits may or Several studies from all over India and other countries have
may not be present validated the prognostic value of the Vellore and modified
Vellore grading in TBMH patients who underwent shunt
The use of GCS made the grading reliable and reproducible and surgery or ETV [Table 3].[16‑24]
minimized the chances of interobserver variability. The grading ii. Decision regarding surgery
The extremely poor outcome in patients in Vellore grade 4
also used Arabic numerals, in contrast to Roman numerals used
led us to question the policy of universal shunting in this
in the original Vellore grading. For the sake of uniformity, all
group of patients. It seemed that raised intracranial pressure
Vellore grades (original or modified) will be represented in
due to the hydrocephalus was not a significant cause of the
Arabic numerals from hereon.
altered sensorium in most of these patients. The altered
sensorium was possibly due to a combination of severe
Comparison with other grading systems
meningoencephalitis and ischemia of the basal ganglia,
All the grading systems for TBM prior to the introduction of the
diencephalic structures, and midbrain. We suggested
modified Vellore grading system were essentially subjective.
that patients in this group who would benefit from shunt
The Lincoln system was the most subjective of all, with no
surgery should be chosen based on their response to a
clear definitions of terms such as “neurological involvement,”
period of external ventricular drainage (EVD).[14] EVD was
“marked changes in sensorium,” or “sensorial changes.” The also suggested as an option for patients in Vellore grade III,
original Vellore grading also had similar shortcomings. “Easily but the prospective study[15] showed that response to EVD
arousable,” which was a feature of Vellore grade 3 patients, is did not correspond to outcome in patients in this grade;
open to different interpretations. The MRC grade I excludes hence, we do not recommend EVD for these patients except
patients with “paresis”; therefore, if a patient has no alteration if the patients are unfit for general anesthesia due to their
of sensorium but has paresis present, then it is not possible to nutritional status or other factors.
categorize such a patient using the MRC system. We suggest prompt surgery for patients in Vellore and
modified Vellore grades 1 and 2 and possibly 3 as the
The other major deficiency of the MRC and Lincoln grading prognosis on long‑term follow‑up is good in over 80%–
systems was that they had only three grades. Stage II of the 90% of patients in grade 1 and approximately 30% in
MRC system and Stages II and III of the Lincoln system grade 3 patients.[25,26]
encompass a wide range of clinical situations with varying iii. Predictor of success of ETV
degrees of alteration of sensorium. Understandably, there The ability to predict the success of ETV using the Vellore
would be significant heterogeneity in patients categorized in grade has been shown by some authors.[18,20,22] Goyal et al.[22]
these stages. Although the MRC staging was modified in 2005 and Yadav et al.[20] found that the ETV was successful in
to include the GCS score, it still has only three grades.[10] It 100% of patients in Grade 1, whereas the success rate was
addressed one of the problems with the original MRC staging, 33% to 50% in Vellore grade 4 patients. Possibly, the thicker
wherein patients with normal sensorium with neurological basal exudates and thicker third ventricular floor due to the
deficits could not be categorized. In the modified MRC staging, ependymitis, which are more likely to be associated with
patients with GCS 15/15 with neurological deficit were grade 4 disease, might be the factors that lead to higher
categorized as MRC stage 2.[10] However, Stage 3 of the MRC failure rates for ETV.
grading includes patients with GCS scores of 3–10. This is a Management of Vellore grade 4 patients
very wide range of GCS scores and therefore does not constitute Several authors have supported the use of response to EVD to
a homogeneous group for prognostication. By having four choose patients in Vellore grade 4 for shunt surgery.[14,15,17,19] In a
Neurology India | Volume 69 | Supplement 2 | November-December 2021 S571
[Downloaded free from http://www.neurologyindia.com on Monday, October 30, 2023, IP: 106.206.158.76]

Rajshekhar: Vellore grading for TBMH

Table 3: Validation of Vellore grading for TBMH

Authors, year, type of surgery Number of patients in different grades Improvement/Survival (%) Comments
Palur et al., 1991 114, C+A Grade 1: 80
Shunt Grade 1: 5 Grade 2: 65.3
Grade 2: 75 Grade 3: 48.1
Grade 3: 27 Grade 4: 0
Grade 4: 7
Singh and Kumar, 1996 140, C Grades 1 and 2: 100
Shunt Grade 1 and 2: 7 Grade 3: 60.3
Grade 3: 46 Grade 4: 34.5
Grade 4: 87
Mathew et al., 1998 30, C+A Grade 3: 66.7
Shunt Grade 3‑18 Grade 4: 8.3
Grade 4‑12
Agrawal et al., 2005 37, C Grade 2: 62
Shunt Grade 2: 16 Grade 3: 40
Grade 3: 15 Grade 4: 0
Grade 4: 6
Jha et al., 2007 14, A+C ETV failed in all grade 4 patients
ETV Grade 2: 2
Grade 3: 8
Grade 4: 4
Sil and Chatterjee, 2008 32, C Patients in Grade 2 had better outcome than those in Grade
Shunt Grade 2: 20 3.
Grade 3: 12
Yadav et al., 2011 59; 39 C, 20 A Grade 1: 95.2
ETV Grade 1:21 Grade 2: 88.9
Grade 2: 27 Grade 3: 57
Grade 3: 7 Grade 4: 50
Grade 4: 4 Vellore grade was the only significant prognostic factor for
survival on linear regression analysis (P=0.001)
Savardekar et al., 2013 26, C Grade 3: 71.4
Shunt Grade 3: 21 Grade 4: 20
Grade 4: 5
Goyal et al., 2014 24, C Grade 3: 70.9
Shunt/ETV ‑ RCT Mean grade: 3.1 Grade 4: 0
ETV success was related to the Vellore grade
Grade 1: 100%
Grade 4: 33.3%
Aranha et al., 2018 26, C Modified Vellore grade was the only factor associated
Shunt/ETV ‑ RCT Grades 1 and 2: 36 significantly with outcome in both shunt and ETV
Grades 3 and 4: 16 groups (Odds ratio: 4.2)
Harrichandpershad et al., 2019 15, A Grade 1: 100
HIV positive Grade 1: 3 Grade 3: 66.7
Shunt Grade 3: 12 Patients in grade 4 should undergo a trial of external
ventricular drainage and only those who improve should
undergo a definitive procedure
C, children; A, adults; ETV, endoscopic third ventriculostomy; RCT, randomized controlled trial.

recent publication from South Africa in 2019 on the outcome of studied the outcome in 95 patients with grade 4 TBMH.[27]
HIV‑positive patients with TBMH, the authors recommended They evaluated whether the response to EVD predicts the
that patients in grade 4 should only be shunted if they improve outcome following shunt surgery in grade 4 TBMH patients.
with a trial of EVD.[24] They conclude that this approach will Although the authors mention that 43 patients received an
allow for the optimal utilization of precious resources. EVD prior to shunt, the flow chart in their article suggests that
in some patients, only a ventricular tap was performed and
However, this approach has been challenged by others drainage was not done. It is obvious that sensorium, which
who suggest that a nihilistic approach to these patients is does not improve following a ventricular tap, might do so after
unwarranted and all patients with TBMH in grade 4 should 2–3 days of EVD. Besides this discrepancy, they lost 55 patients
undergo shunt surgery. The first such report published in 2009 to follow‑up. Of the 40 patients available for follow‑up for

S572 Neurology India | Volume 69 | Supplement 2 | November-December 2021

[Downloaded free from http://www.neurologyindia.com on Monday, October 30, 2023, IP: 106.206.158.76]

Rajshekhar: Vellore grading for TBMH

a median duration of 12 months, nine had died in hospital. Table 4: Suggested changes to modified Vellore
Therefore, follow‑up was only performed in 31 patients; 18 of grading
them had a good outcome or moderate disability (GOS 4 and 5). Grade Description
The authors did not comment on their finding that patients who Grade 1 GCS 15; no neurological deficit
improved following EVD had better outcome than those who Grade 2 GCS 15; neurological deficit is present
did not. EVD/ventricular tap was performed in 19 followed‑up Grade 3 GCS 9‑14; neurological deficit may or may not be present
patients. Of these, eight had improved to Grades 2 or 3 prior Grade 4a GCS 7, 8; no basal ganglia or brain stem infarcts on
to shunt and three (38%) had GOS 4 or 5 outcome. Of the 11 neuroimaging
who did not improve, only two (18%) had the same outcome Grade 4b GCS 7, 8; basal ganglia or brain stem infarcts present on
scores (GOS 4 or 5) at follow‑up. Although these numbers are neuroimaging Or GCS 3‑6
too small to analyze for statistical significance, they indicate
that improvement with EVD is probably associated with
better outcomes with shunt. Finally, of the 21 patients who altered sensorium. Even those who support universal shunt
underwent direct VP shunt without EVD, 13 (62%) improved surgery in grade 4 patients urge caution in performing shunts
to GOS 4 or 5 at follow‑up. Therefore, patients in the direct VP indiscriminately in this group of patients.[27] Based on the
shunt group seem to have had a significantly better outcome observations of these authors and our experience,[15,27,30‑32] we
than those who underwent EVD (P = 0.03; Fisher’s exact test). propose the creation of subgrades 4a and 4b [Table 4].
This difference was not analyzed or commented upon by the
authors. Furthermore, they did not include this as a prognostic Patients in grade 4a can be offered direct shunt surgery
factor in the multivariate analysis. Finally, a good outcome in without determining a response to EVD. In contrast, patients
62% of grade 4 TBMH patients following shunt surgery has in grade 4b are those in whom the prognosis is extremely poor
been reported in only one other publication from China that and response to EVD should be used to choose patients for
included 19 grade 4 patients.[28] shunt surgery. While it would be intuitive to use the duration of
altered sensorium (<2 weeks) as an additional factor to decide
Authors who oppose the use of response to EVD to select whether to operate a grade 4b patient, it has not been found to
patients for shunt surgery cite the following reasons: (i) be associated with outcome following shunt or ETV.[22]
improvement in sensorium following drainage of CSF might
not occur for several days and hence evaluating the response Conclusions
to EVD after 2 or 3 days of drainage can be misleading; (ii)
EVDs are prone to get infected and are therefore impractical; Since its introduction nearly three decades ago, Vellore grading
and (iii) some evidence suggests that response to EVD does not for patients with TBMH and its modified version have been
predict outcome following shunt.[21,27,28] Those who support the validated in many studies for their reliability, prognostic value
use of EVD argue that (i) universal shunting of these patients for outcome after shunt surgery, and utility in predicting the
is an inappropriate use of limited resources given the high success of ETV. With the available evidence that patients in a
mortality and morbidity in these patients; (ii) children with subgroup of grade 4 have significantly poorer outcomes than
low ventricular pressure do not have improvement following others in the same grade, it might be justified to have two
shunt; and (iii) use of response to EVD allows surgeons to select subcategories in this grade. Validation of this subcategorization
patients who are likely to benefit from shunt surgery.[14,15,17,19,24] is required in further studies.

Notwithstanding the shortcomings in the data presented by Financial support and sponsorship
those who argue for universal shunting, there seems to be some Nil.
merit in considering direct shunt in selected patients in grade 4.
Conflicts of interest
Subcategorization of Vellore grade 4 patients There are no conflicts of interest.
Patients in Vellore grade 4 have a GCS score that ranges from
3 to 8. The above discussion indicates these patients might References
not constitute a homogeneous group for prognostic and
management purposes and need further division. A recent 1. Tandon PN. Tuberculous meningitis (cranial and spinal). In:
systematic review on the outcome in patients with TBMH Vinken PJ, Bruyn GW, editors. Handbook of Clinical Neurology.
following VP shunt surgery also suggested the need for a Infections of the Nervous System. Vol 33. North‑Holland:
“predictive model” for choosing patients in Vellore grade 4 Amsterdam; 1978. p. 195‑262.
who might benefit from shunt surgery.[29] Subcategorization 2. Bhagwati SN. Ventriculoatrial shunt in tuberculous meningitiswith
of patients in grade 4 is an effort in that direction. hydrocephalus. J Neurosurg 1971;35:309‑12.
3. Singhal BS, Bhagwati SN, Syed AH, Laud GW. Raised intracranial
Several authors have shown that TBMH and TBM patients pressure in tuberculous meningitis. Neurol India 1975;23:32‑9.
with GCS of 3–5 or 3–6 and those with basal ganglia and brain 4. Bullock MR, Van Dellen JR. The role of cerebrospinal fluid shunting
stem infarcts have an extremely poor prognosis with high in tuberculous meningitis. Surg Neurol 1982;18:274‑7.
mortality rates.[27,30‑33] We reported that 8 of 16 (50%) patients 5. Chitale VR, Kasaliwal GT. Our experience of ventriculoatrial shunt
in Vellore grades 3 and 4 who died in our prospective study using Upadhyaya valve in cases of hydrocephalus associated with
of TBMH had brain stem infarcts.[15] Hydrocephalus in most tuberculous meningitis. Prog Pediatr Surg 1982;15:223‑36.
patients with GCS of 3–6 and in those with basal ganglia and 6. Upadhyaya P, Bhargava S, Sundaram KR, Mitra DK, George J,
brain stem infarcts might not contribute significantly to the Singh DC. Hydrocephalus caused by tuberculous meningitis:

Neurology India | Volume 69 | Supplement 2 | November-December 2021 S573

[Downloaded free from http://www.neurologyindia.com on Monday, October 30, 2023, IP: 106.206.158.76]

Rajshekhar: Vellore grading for TBMH

Clinical picture, CT findings and results of shunt surgery. in patients of tubercular meningitis with hydrocephalus in poor
Z Kinderchir 1983;38(Suppl 2):76‑9. neurological grade: A prospective study in the pediatric population
7. Roy TK, Sircar PK, Chandar V. Peritoneal ventriculo shunt and review of literature. Child Nerv Syst 2013;29:719‑25.
in the management of tuberculous meningitis. Indian Pediatr 22. Goyal P, Srivastava C, Ojha BK, Singh SK, Chandra A, Garg RK,
1979;16:1023‑7. et al. A randomized study of ventriculoperitoneal shunt versus
8. Gelabert M, Castro‑Gago M. Hydrocephalus and tubercular endoscopic third ventriculostomy for the management of
meningitis in children. Childs Nerv Syst 1988;4:268‑70. tubercular meningitis with hydrocephalus. Child Nerv Syst
9. Medical Research Council Streptomycin in tuberculosis trials 2014;30:851‑7.
committee. Streptomycin treatment of tuberculosis meningitis. 23. Aranha A, Choudhary A, Bhaskar S, Gupta LN. A randomized
Lancet 1948;2:582‑96. study comparing endoscopic third ventriculostomy versus
10. Thwaites GE, Tran TH. Tuberculous meningitis: Many questions, ventriculoperitoneal shunt in the management of hydrocephalus
due to tuberculous meningitis. Asian J Neurosurg 2018;13:1140‑7.
too few answers. Lancet Neurol 2005;4:160‑70.
24. Harrichandparsad R, Nadvi SS, Moosa M‑YS, van Dellen JR. Outcome
11. Lincoln EM, Sordillo SV, Davies PA. Tuberculous meningitis in
of ventriculoperitoneal shunt surgery in human immunodeficiency
children. J Pediatr 1960;57:807‑23.
virus positive patients on combination antiretroviral therapy with
12. Bhagwati SN, Singhal BS. Ventriculoatrial shunts in tuberculous tuberculosis meningitis and hydrocephalus. World Neurosurg
meningitis. In: Saksena DS, Purhoti A, editors. Proceedings of 2019;123:e574‑80.
Surgery in Tropics. Bombay, India: Bombay Hospital Trust; 1977.
25. Rajshekhar V. Management of hydrocephalus in patients with
p. 379‑80.
tuberculous meningitis. Neurol India 2009;57:368‑74.
13. Reddy PK, Abraham J, Chandy MJ, Mathai KV, Raman PT. Shunt
26. Rajshekhar V. Surgery for brain tuberculosis: A review. Acta
surgery in TB meningitis. Silver Jubilee Souvenir, Department of
Neurochir 2015;157:1665‑78.
Neurological Sciences, Christian Medical College Hospital, Vellore,
27. Srikantha U, Morab JV, Sastry S, Abraham R, Balasubramaniam A,
India. 1974. p. 123‑8.
Somanna S, et al. Outcome of ventriculoperitoneal shunt
14. Palur R, Rajshekhar V, Chandy MJ, Joseph T, Abraham J. Shunt
placement in Grade IV tubercular meningitis with hydrocephalus:
surgery for hydrocephalus in tubercular meningitis: A long‑term
A retrospective analysis in 95 patients. J Neurosurg Pediatr
follow up study. J Neurosurg 1991;74:64‑9. 2009;4:176‑83.
15. Mathew JM, Rajshekhar V, Chandy MJ. Shunt surgery for poor 28. Peng J, Deng X, He F, Omran A, Zhang C, Yin F, et al. Role of
grade patients with tuberculous meningitis and hydrocephalus: Effect ventriculoperitoneal shunt surgery in grade IV tubercular meningitis
of response to external ventricular drainage and other factors on with hydrocephalus. Childs Nerv Syst 2012;28:209‑15.
long‑term outcome. J Neurol Neurosurg Psychiatry 1998;65:115‑8.
29. Rizvi I, Garg RK, Malhotra HS, Kumar N, Sharma E, Srivastava C,
16. Singh D, Kumar S. Ventriculoperitoneal shunt in post tubercular et al. Ventriculo‑peritoneal shunt surgery for tuberculous meningitis:
hydrocephalus. Indian Pediatr 1996;33:854‑5. A systematic review. J Neurol Sci 2017;375:255‑63.
17. Agrawal D, Gupta A, Mehta VS. Role of shunt surgery in 30. Lamprecht D, Schoeman J, Donald P, Hartzenberg H.
pediatric tubercular meningitis with hydrocephalus. Indian Pediatr Ventriculoperitoneal shunting in childhood tuberculous meningitis.
2005;42:245‑50. Br J Neurosurg 2001;15:119‑25.
18. Jha DK, Mishra V, Choudhary A, Khatri P, Tiwari R, Sural A, et al. 31. Karande S, Gupta V, Kulkarni M, Joshi A. Prognostic clinical
Factors affecting the outcome of neuroendoscopy in patients with variables in childhood tuberculous meningitis: An experience from
tuberculous meningitis hydrocephalus: A preliminary study. Surg Mumbai, India. Neurol India 2005;53:191‑6.
Neurol 2007;68:35‑42. 32. Kalita J, Misra UK. Outcome of tuberculous meningitis at 6 and
19. Sil K, Chatterjee S. Shunting in tuberculous meningitis: 12 months: A multiple regression analysis. Int J Tuberc Lung Dis
A neurosurgeon’s nightmare. Childs Nerv Syst 2008;24:1029‑32. 1999;3:261‑5.
20. Yadav YR, Parihar V, Agrawal M, Bhatele PR. Endoscopic third 33. Morgado TC, Kinsky M, Carrara H, Rothemeyer S, Semple P.
ventriculostomy in tubercular meningitis with hydrocephalus. Prognostic value of computed tomography evident cerebral infarcts
Neurol India 2011;59:855‑60. in adult patients with tuberculous meningitis and hydrocephalus
21. Savardekar A, Chatterji D, Singh S, Mohindra S, Gupta S, treated with an external ventricular drain. World Neurosurg
Chhabra R. The role of ventriculoperitoneal shunt placement 2013;80:e255‑60.

S574 Neurology India | Volume 69 | Supplement 2 | November-December 2021