The role of fluid restriction in acute bacterial meningitis: A systematic review

Abstract

Background

Acute bacterial meningitis (ABM) is a neurological emergency that requires prompt and appropriate antibiotics and corticosteroids. Previously fluid restriction was recommended in the treatment of ABM in view of studies reporting hyponatremia which was due to elevated antidiuretic hormone (ADH) levels.

Aims

The primary objective was to study the effects of adjuvant fluid restriction in patients of ABM on mortality, neurological and audiological parameters.

Methods

Electronic search was performed on the PubMed, Medical Literature Analysis and Retrieval System Online (MEDLINE), Excerpta Medica Database (EMBASE), Epistemonikos, Web of science, Cochrane Central Register of Controlled Trials (CENTRAL) and Clinical trial registry for studies published from 1946 to 6th Feb 2024. The systematic review included all studies that were Randomized controlled trials (RCT) wherein fluid restriction was employed in patients with ABM. Data were uploaded into the rayyan tool, and the review authors independently screened the articles. Quality of the studies was assessed using the Cochrane ‘Risk of bias 2’ (ROB2) tool. Risk ratios (RR) with 95% confidence intervals (CI) were used to measure treatment effect and the GRADE framework was used to evaluate the certainty of evidence for each outcome.

Results

Two articles with 407 participants were included in the meta-analyses and both of them had children less than 7 years old. Fluid restriction probably had little or no difference on mortality at admission compared with normal fluid maintenance (RR:1.19; 95% CI 0.77 to 1.85) and also on neurological sequalae (RR 1.31, 95% CI 0.74 to 2.30). A single study reported on mortality and neurological sequelae based on hyponatremia status and there was no difference in between the fluid restricted group and standard maintenance groups.

Conclusions

There are only two studies on the efficacy of Fluid restriction as an adjuv-ant therapy in the treatment of acute bacterial meningitis. Fluid restriction does not appear to significantly alter the mortality or neurological sequelae compared to maintenance therapy.

Trial registration

Prospero id: CRD42024533368.

Background

Acute bacterial meningitis (ABM) is a severe and potentially life-threatening condition, marked by high mortality and morbidity rates, that demands prompt diagnosis and treatment [1–3]. Common etiological agents vary with age and immune status, but typically include Streptococcus pneumoniae, Neisseria meningitidis, Group B Streptococcus (GBS), Haemophilus influenzae, Listeria monocytogenes, and some Gram-negative bacteria [4]. Early diagnosis and treatment is a critical point in the management of ABM as it can be fatal when there is delayed treatment. In order to reduce the incidence of bacterial meningitis, prevent epidemics, and reduce the associated mortality and morbidity burden, the World Health Organization (WHO), together with global partners and experts, coordinated the development of the Defeating meningitis by 2030: a global road map which was approved by the Seventy-third World Health Assembly in November 2020. As part of the Roadmap objectives, WHO has developed new clinical guidelines on meningitis diagnosis, treatment and care, published in April 2025 [5, 6].

Globally, the prevalence of meningitis approximately was of 10.5 million cases in 2017.

and resulted in 288,000 deaths [4]. The mortality of ABM ranges widely from 7 to 23% in high-income countries [7, 8] with prompt antibiotic treatment, but exceeds more than 50% in low-income countries [9, 10]. Survivors of acute bacterial meningitis often develop disabling morbidity in terms of hearing impairment, neuropsychiatric disturbances, cognitive impairment, seizures and learning disability in children [10].

Adjunctive therapies can be helpful in preventing the morbidity; among them corticosteroids have been proven to be helpful in the treatment of acute bacterial meningitis caused by specific pathogens [10]. In addition, several adjuvant therapies have been investigated, including fluid restriction as one such approach. Managing fluid and electrolyte balance is crucial, as both over-hydration and under-hydration can worsen outcomes. In the past, fluid restriction was advised in the management of children with meningitis [11, 12]. This recommendation was primarily based on frequent observations of hyponatremia, which is often associated with increased levels of antidiuretic hormone (ADH). Over half of children with meningitis may present with hyponatremia at admission [12–14]. Studies have linked the extent of hyponatremia with seizure occurrence, disease severity, and adverse neurodevelopmental outcomes [12]. A strong association between cerebral oedema and fatal outcomes in bacterial meningitis has led some researchers to suggest that fluid restriction might help reduce cerebral oedema severity and improve neurological prognosis [12–14]. The primary objective of this review was to study the effects of fluid restriction on mortality, neurological and audiological sequelae in patients with community acquired acute bacterial meningitis. This systematic review was conducted as part of the evidence review process that informed the development of the WHO guidelines in meningitis diagnosis, treatment and care.

Methods

Research question

Among suspected, probable or confirmed cases of acute bacterial meningitis, should fluid restriction be recommended as a way of decreasing morbidity and mortality outcomes?

Type of studies included

The systematic review involved all studies (in English language) categorized as randomized controlled trials (RCT) wherein fluid restriction was included in addition to antibiotics in patients with acute bacterial meningitis. Non-randomized studies without a comparator (e.g., case reports, case series, letters, etc.) were excluded. Additionally, any on-going trials and studies, with no evaluable outcome data were excluded.

Inclusion criteria

The study population included suspected, probable or confirmed cases of acute bacterial meningitis as defined by the study authors, based on clinical features, cerebrospinal fluid (CSF) characteristics and microbiological investigations. Subgroup analysis was performed based on the following parameters: (i) etiological pathogens: Neisseria meningitidis, Streptococcus pneumoniae, Haemophilus influenzae type b, and Group B Streptococcus; (ii) age group (i.e., child, adult); (iii) World bank income classification (i.e., HICs, LMICs) [15].

Exclusion criteria

The following disease categories were excluded: (i) hospital-acquired, nosocomial, and healthcare-associated meningitis; (ii) subacute and chronic meningitis, including tuberculous, cryptococcal and eosinophilic meningitis; (iii) meningitis in newborns (< 28 days); (iv) non-infectious meningitis (e.g., drugs, malignancy, autoimmune diseases).

Outcome measures

Following GRADE methodology the Guideline Development Group (GDG) identified critical and important outcomes. The primary outcome measures were mortality, neurological sequelae, audiological involvement and adverse effects. Neurological sequelae included hearing loss, speech and/or language impairment, seizures, cognitive impairment, psychological (stress, depression, behaviour), hydrocephalus, motor deficits, vision impairment. Secondary outcome included the impact of fluid restriction on disease course (i.e. time to resolution of symptoms, persistent fever).

Search strategy

Electronic searches were conducted in several databases, including Medline via Pubmed, EMBASE via Elsevier, the Cochrane Library and the ClinicalTrials.gov registry. The search was run on the 6th of February 2024. Additionally, reference lists of relevant publications were reviewed to identify any unlisted trials.

The search results were uploaded into the Rayyan tool [16] for screening. The title/abstract, full text screening and data extraction was carried out by two review authors independently. Each study was carefully reviewed to ensure that there were no duplicates and that multiple publications by the same researcher were included only once (Authors M.N., T.A.S.). Any disagreements during the selection process were resolved through discussion. Titles and abstracts were reviewed on the basis of the inclusion and exclusion criteria. The total number of studies retrieved, along with the reasons for their inclusion or exclusion, is presented in the PRISMA flowchart (see Fig. 1).

Fig. 1.

Risk of bias summary: review authors’ judgements about each risk of bias item for each included study

The methodological quality of the studies included was evaluated using the Version 2 of the Cochrane risk of bias’ (RoB 2) tool (Figs. 2 and 3) [17]. Each of the studies included was assessed on the basis of a number of pre-defined parameters, that include the following: analysis of the randomization process for selection bias, detection of any deviations from the protocol (performance bias), assessment of attrition bias, reporting bias, detection bias, and any other additional sources of bias. The results from the RoB 2 analysis informed the GRADE assessment of outcomes. Treatment effects were measured using risk ratios (RR) with 95% confidence intervals (CI). Funnel plots were visually inspected to assess for publication bias. Data were analyzed using Review Manager Web software (version 5.4) [18].

Fig. 2.

(a) Risk of bias graph: review authors’ judgements about each risk of bias item presented as percentages across all included studies. b Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies

Fig. 3.

Effect of fluid restriction vs. fluid maintenance on mortality

Assessment of certainty of evidence (GRADE evidence profiles)

The GRADE (Grading of Recommendations Assessment, Development and Evaluation) framework was used to evaluate the certainty of the evidence for each outcome, as developed by the GRADE Working Group [19]. The evidence was categorized as follows: (i) High: indicates a high level of confidence in the effect estimate, suggesting that the true effect is likely to be very close to the estimated value. (ii) Moderate: reflects a moderate level of confidence, implying that the true effect is likely near the estimated value, but there remains a chance of significant deviation. (iii) Low: signifies limited confidence in the estimate, indicating that the true effect could differ substantially from the estimated value. (iv) Very low: suggests minimal confidence in the estimate, indicating that the true effect is likely to differ significantly from the estimated value.

Heterogeneity was assessed by means of visual inspection of the forest plots to determine point estimate alignment and CI (confidence interval) overlap. We applied the Chi-square test with a P-value threshold of 0.10 to denote statistical significance and utilized the I² statistic to quantify heterogeneity. According to the Cochrane handbook for systematic reviews of interventions, the following ranges were used to interpret the I² values: 0% to 40% indicates negligible heterogeneity; 30% to 60% suggests moderate heterogeneity; 60% to 90% signifies substantial heterogeneity; and 75% to 100% indicates considerable heterogeneity [19]. The magnitude and direction of effects were considered, as were the strength of the evidence for heterogeneity (e.g. P-value from the Chi-square test when determining the importance of the observed I² value).

Results

Studies included in the review and the GRADE evidence profiles

A total of 4738 records were screened, of which 1176 duplicates were removed. Subsequently, 3562 articles were screened (Fig. 1). Out of these, 3562 were excluded, for the following reasons: 1852 articles were based on the wrong disease, 477 assessed parameters that were not relevant to the scope of this review, and 90 were based on unsuitable articles lacking real-world patient data, e.g. case reports, case series, pathogenicity studies, animal studies, editorials and correspondence. Finally, two RCTs [20, 21] were considered for meta-analysis (Table 1).

Table 1.

Characteristics of the studies included in the GRADE evidence profiles

Lead author (Year) Country of conduct	Study design	Overall Risk of Bias (study level)	Population; Age group	Intervention	Comparator	Outcome domains with available data (synthesis method/metric)	Specific outcomes measure	Time points of measurement
Duke 2002; Papua New Guines (20)	RCT	Low	Children up to 12 years of age;	Breast milk by nasogastric tube at 60% of normal maintenance volumes (n = 172)	Intravenous half-normal saline and 5% dextrose at 100% of normal maintenance volumes (n = 174) for the first 48 h of treatment.	Mortality	– Neurological sequelae (seizures, motor weakness), visual impairment, hydrocephalus, hearing impairment hypoglycemia, hyponatremia, pulmonary edema	Baseline, 14 days, 3 months
Singhi, 1995; India (21)	RCT	High	Children up to 7 years of age;	65% calculated maintenance fluid requirement, given as intravenous 1/5th normal saline in 5% dextrose for 24 h, followed by “a gradual liberalisation at a rate of 10 ml/kg per 8 h after 24 hours of hospital stay if serum sodium and plasma osmolality had returned to normal and if there were no clinical signs of dehydration (N = 28)	Maintenance fluid requirements (110 ml/kg for first 10 kg, 50 ml/kg for next 10 kg and 25 ml/kg for subsequent weight) given intravenously and comprising 1/5th normal saline in 5% dextrose” as long as they required intravenous fluids (n = 22)	Mortality	hypoglycemia, change in osmolality	Baseline and after 48 h

RCT: randomized controlled trial

Studies excluded from the review

Eight studies were excluded from the review (Table 2). Six articles were excluded because they were review articles [22–27], while the remaining two studies [28, 29] were excluded due to interventions not relevant to the scope of the review. One study [25] was excluded because it focused on indicators specific to bacterial meningitis.

Table 2.

Studies excluded from the systematic review

Lead author (Year)	Study type	Population	Intervention	Comparator	Outcome	Reasons for exclusion
Brown (1994) (22)	Review article	NA	NA	NA	NA	Not an RCT
Duke (1998) (23)	Review article	NA	NA	NA	NA	Not an RCT
Floret(1999) (24)	Review article	NA	NA	NA	NA	Not an RCT
Berkley (2004) (25)	Retrospective	Patients with bacterial meningitis	NA	NA	NA	Study of indicators of bacterial meningitis; not an RCT
Pelkonen (2011) (26)	RCT	Children 2 months to 13 years of age with acute bacterial meningitis	Cefotaxime infusion/bolus with paracetamol	Cefotaxime infusion/bolus with placebo	Mortality	Intervention is not relevant to this review
Maitland (2013) (27)	RCT	Children, aged 60 days to 12 years, with severe febrile illness	Albumin bolus/saline bolus	No bolus	Mortality	Intervention not relevant to the review; study included children with severe infections (not limited to acute bacterial meningitis)
Roine (2014) (28)	Post hoc analysis of Pelkonen et al.	Children 2 months to 13 years age with acute bacterial meningitis	NA	NA	NA	Not an RCT
van Paridon (2015) (29)	Retrospective study	Patients with sepsis	NA	NA	NA	Not an RCT

NA: not applicable; RCT: randomized controlled trial

Risk of bias

Among the included studies, the one by Singhi et al. [21] showed a high risk of bias, while the other by Duke et al. had a low risk of bias (Duke et al., 2002) [20]. Duke et al. [20] used computer-generated, numbered opaque sealed envelopes to ensure random allocation. The study by Singhi et al., [21] in contrast, used random numbers table but did not describe the allocation concealment process. Blinding was not performed in Duke et al. [20], while the blinding process was not mentioned in Singhi et al. [21]. Figure 2a and b present the results of the risk-of-bias assessment.

Primary outcomes

Mortality

“Very low certainty evidence from two RCTs involving 407 children suggests that the effect of fluid restriction on mortality at admission compared with normal fluid maintenance is uncertain (RR 1.19, 95% CI 0.77–1.85). [Figure 4]

Fig. 4.

Effect of fluid restriction vs. fluid maintenance on neurological sequelae

Sequelae

Very low certainty evidence from two RCTs involving 407 children suggests that the effect of fluid restriction on sequalae at admission compared with normal fluid maintenance is uncertain (RR 1.31, 95% CI 0.74–2.30).

Subgroup analysis

Mortality

The study by Singhi et al. [21] reported on mortality according to hyponatraemia status. Evidence showed no difference in mortality among those with or without hyponatraemia between the fluid restricted group and standard maintenance groups. [Mortality: 4 out of 15 (26.67%) vs. 0 out of 11 (0%) in the hyponatremia group and 3 out of 13 (23%) vs. 2 out of 11 (18%) in the no hyponatremia group, P = 0.48].

Neurological sequelae

The study by Singhi et al. [21] reported on sequelae according to hyponatraemia status. Evidence showed no difference in sequelae among those with or without hyponatremia between the fluid restricted group and standard maintenance groups. [Sequelae: 6 out of 15 (40%) vs. 4 out of 11 (36.36%) in the hyponatremia group and 4 out of 13 (23%) vs. 2 out of 11 (18%) in the no hyponatremia group, p = 0.48].

Discussion

Fluid restriction has traditionally been considered in the management of acute bacterial meningitis due to its possible role in reducing cerebral oedema, a common complication that can lead to increased intracranial pressure and poor neurological outcomes. Fluid restriction was originally recommended due to cases of hyponatremia seen in children with bacterial meningitis, which often results from elevated antidiuretic hormone (ADH) levels. The assumption was that restricting fluids could mitigate cerebral oedema by preventing further fluid accumulation in the brain.

Clinical dehydration (hypovolemia) has been reported in children with acute bacterial meningitis, without a significant impact on mortality rates [12, 13]. In children given full maintenance fluids and volume replacement, ADH levels normalized within 24 h. Children with meningitis given only two-thirds of maintenance fluids maintained high ADH levels, suggesting that normalising ADH requires sufficient sodium and fluid intake to address hypovolemia.

This systematic review assessed the effects of fluid restriction on mortality, neurological, and audiological outcomes in acute bacterial meningitis. Despite the rigorous search strategy across databases, only two randomized controlled trials (RCTs) were identified, indicating a limited evidence base on this topic. Both studies indicated that fluid restriction did not reduce mortality or neurological sequelae and suggested a preference for adequate fluid maintenance. No significant benefit was observed with fluid restriction compared to adequate fluids, even among patients with or without hyponatremia at admission. The appropriate administration route for the maintenance fluids is that it should be given orally or by enteric tube when possible; breastfeeding is best for infants and young children. If oral or enteric routes are not feasible, use isotonic IV fluids like Ringer’s lactate or normal saline. For patients with suspected SIADH but no shock or hypovolemia, moderate fluid restriction may be considered based on clinical judgment.

Adequate fluid maintenance is of utmost importance in decreasing the mortality and morbidity. This has to be emphasized especially in low-middle income countries, resource-poor settings, and in children with malnutrition or dehydration. In such settings, children have to be treated with intravenous fluids to maintain osmolality and prevent hyponatremia. Dehydration, fluid restriction and hyponatremia may be indirect evidence of poor prognostic factors in the treatment of acute bacterial meningitis. Intravenous maintenance fluids are favoured compared to fluid restriction in the first 48 h of clinical presentation especially in regions where patients present late or which have high mortality. But there is a lack of evidence for regions with low mortality and/or high-income countries, children presenting early or those with normal body surface area [12, 13]. There are also no studies in adults especially elderly populations to assess the effectiveness of fluid restriction or administration of fluid [12, 13].

“The main limitation of this review is the paucity of eligible evidence. Only two studies met the inclusion criteria, both conducted more than a decade ago and exclusively involving children. There is a critical need for contemporary data, particularly studies including adults and older adults, to strengthen the evidence base.” Both studies used cefotaxime or chloramphenicol—and chloramphenicol is no longer recommended for meningitis—and one study focused mainly on H. influenzae. These limitations should be considered when generalizing these results to current practice. The two studies reviewed did not adequately fulfil the objective of evaluating the impact of different fluid volumes on outcomes in acute bacterial meningitis. Importantly, all investigations were conducted within hospital settings, and none examined primary care or pre-hospital environments, where initial fluid therapy is commonly administered prior to referral for specialised care at tertiary facilities. In the post‑vaccination era there has been a substantial decline in the incidence of acute bacterial meningitis (ABM), particularly in cases with microbiological confirmation, due to widespread introduction of conjugate vaccines forms another major limitation of this review. Research focused on out-of-hospital and primary care contexts may hold greater relevance for low- and middle-income countries, where healthcare systems are generally less structured and closely monitored [13].

Conclusions

This review highlights the current ambiguity surrounding fluid restriction as a therapeutic approach for acute bacterial meningitis in children. While fluid restriction does not appear to reduce mortality or sequelae rates compared to maintenance therapy, these findings indicate the need for further studies. Expanding the evidence base through larger, more diverse trials could aid in developing tailored fluid management strategies that address the specific needs of patients with acute bacterial meningitis across different clinical and geographic settings.

Abbreviations

ABM

Acute bacterial meningitis

GBS

Group B Streptococcus

WHO

World Health Organization

ADH

Antidiuretic hormone

RCT

Randomized controlled trials

CSF

cerebrospinal fluid

HIC

High-Income Country

LMIC

Low-Middle Income Country

GRADE

Grading of Recommendations Assessment, Development and Evaluation

GDG

Guideline Development Group

ROB

Risk of bias’

RR

risk ratios

CI

confidence intervals

Author contributions

NM: performed the search strategies, analysed the results, did the systematic review and finally was instrumental in drafting the manuscript DD: performed the search strategies, analysed the results, did the systematic review TAS: performed the search strategies, analysed the results, did the systematic review NS: analysed the results and the systematic review FV: analysed the results and the systematic review NB: analysed the results and the systematic review KK: Statistician performed the search strategies and the systematic review TD: analysed the results and the systematic review All Authors were involved in reading and revising the manuscript. All authors read and approved the final manuscript.

Funding

None.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not Applicable.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.

Disclaimer

The evidence review conducted for this work formed an integral component of the development of the WHO guidelines on the diagnosis, treatment, and care of meningitis.