A systematic review and meta-analysis of Plasmodium infections and erythrocyte sedimentation rate in relation to disease severity

Abstract

Background

Given the global burden of malaria and its impact on public health, there is a pressing need to understand the role of routine hematological parameters, particularly erythrocyte sedimentation rate (ESR), in malaria diagnosis and prognosis. This systematic review and meta-analysis aimed to comprehensively examine the differences in ESR between malaria patients and non-malarial controls, as well as between severe and non-severe malaria cases.

Methods

This systematic review and meta-analysis followed the PRISMA guidelines and was registered with PROSPERO (CRD42023482332). Comprehensive searches were conducted in ProQuest, Embase, Ovid, MEDLINE, PubMed, Scopus, and Google Scholar from inception to July 1, 2025, using terms related to “erythrocyte sedimentation rate” and “malaria.” Studies were included if they reported ESR levels in malaria patients compared to non-malarial controls, or in severe versus non-severe malaria cases. Non-original studies, those without extractable ESR data, or those lacking appropriate comparison groups were excluded. Data extraction and quality assessment were independently performed by two reviewers using the Joanna Briggs Institute critical appraisal checklists. A meta-analysis using a random-effects model was conducted to estimate standardized mean differences (SMD) in ESR levels between groups. Heterogeneity was assessed with the I² statistic, and further explored through meta-regression, subgroup analyses, and sensitivity analyses. Publication bias was evaluated using funnel plots and Egger’s test.

Results

The review identified 19 studies that met the inclusion criteria. The qualitative synthesis revealed that a majority of these studies, which employed various research designs, consistently observed elevated ESR in symptomatic malaria cases across different populations and age groups. The meta-analysis confirmed significantly higher ESR levels in malaria patients than non-malarial controls (P = 0.0171, SMD: 3.0037, 95% confidence interval [CI]: 0.5347–5.4726, random effects model, I² = 97.3%, 11 studies). However, a significant publication bias was detected (Egger’s test, P = 0.005), suggesting that the pooled estimates may be influenced by selective reporting. The meta-analysis also demonstrated that severe malaria cases have higher ESR levels compared to non-severe cases (P = 0.0014, SMD: 2.2556, 95% CI: 0.8759–3.6354, random effects model, I²: 96.1%, 4 studies).

Conclusion

The findings of this study indicate that elevated ESR is frequently reported in malaria patients and may be more pronounced in severe cases. However, the very high between-study heterogeneity and the presence of publication bias limit confidence in the robustness of these associations. While ESR may hold potential as a supportive diagnostic or prognostic marker when combined with other clinical and laboratory parameters, its independent utility in malaria remains uncertain and requires further investigation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-025-11622-x.

Introduction

Malaria is a life-threatening disease caused by parasites of the genus Plasmodium. Six species are recognized as primary human pathogens: Plasmodium falciparum (P. falciparum), Plasmodium vivax (P. vivax), Plasmodium malariae (P. malariae), Plasmodium knowlesi (P. knowlesi), Plasmodium ovale curtisi (P. o. curtisi), and Plasmodium ovale wallikeri (P. o. wallikeri). In addition, several simian malaria species, such as P. cynomolgi, P. brasilianum, and P. inui, have been reported to infect humans [1–4]. According to the World Malaria Report 2024, an estimated 263 million malaria cases occurred globally in 2023, an increase of 11 million compared with 2022, underscoring the continued global burden of this vector-borne disease [5].

Routine laboratory parameters composed of complete blood count (CBC) [6, 7], liver function tests (LFTs) [8, 9], renal function tests [10, 11], C-reactive protein (CRP) [12], procalcitonin [13], and the erythrocyte sedimentation rate (ESR) [14, 15] are widely used in malaria diagnosis and management. ESR is a simple hematological test that measures the rate at which red blood cells sediment in one hour [16]. Although non-specific, ESR is elevated in various inflammatory conditions, including autoimmune diseases, infections, and malignancies [16]. In infectious diseases, it has also been proposed as a marker of childhood bone and joint infections [17, 18].

In malaria, altered ESR values have been frequently reported. For example, one study found elevated ESR in 53.5% of P. vivax and 60% of P. falciparum cases [14]. Other reports have suggested a correlation between ESR levels and malaria severity [19]. However, findings across studies remain inconsistent: some report significantly elevated ESR in malaria patients, whereas others find no difference compared with non-malarial controls [20, 21]. This inconsistency raises questions about the clinical utility of ESR in malaria diagnosis and prognosis.

Although individual studies have explored ESR changes in malaria, no prior systematic review and meta-analysis have synthesized this evidence across populations, parasite species, and disease severities. Importantly, the present systematic review and meta-analysis incorporates the most recent studies, including data up to 2025, thereby updating the evidence base, and is the first to conduct a separate quantitative analysis of severe versus non-severe malaria cases, a distinction critical for early clinical triage. Therefore, the objectives are to compare ESR levels between malaria patients and non-malarial controls, and to evaluate differences between severe and non-severe cases. The findings will help determine if ESR can serve as a useful adjunctive marker for malaria diagnosis and severity assessment, while also identifying sources of heterogeneity and providing recommendations for future research.

Methods

Guidelines for reporting and protocol registration

This systematic review and meta-analysis was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [22]. The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42023482332 [23].

Review questions

The review questions were structured using the PECO framework [24]. The target population (P) included individuals in malaria-endemic areas. The exposure of interest (E) was Plasmodium infection or severe malaria. The comparison group (C) consisted of individuals without malaria or those with non-severe cases. The outcome (O) was ESR levels. The primary review question addressed differences in ESR between malaria patients and non-malarial controls. The secondary question examined ESR differences between severe and non-severe malaria cases.

Definitions

Severe malaria was defined as an infection accompanied by clinical or laboratory evidence of vital organ dysfunction, according to the World Health Organization (WHO) [25]. Non-severe malaria referred to cases without such evidence and included both uncomplicated and asymptomatic malaria. Non-malarial controls were defined as healthy individuals or febrile patients with other illnesses in whom malaria tests were negative.

Eligibility criteria

Original studies reporting ESR levels in malaria patients compared with non-malarial controls, or in severe versus non-severe malaria cases, were included. Eligible research designs were observational studies (cross-sectional, case-control, and cohort). Case reports, small case series without comparative groups, in vitro or animal studies, conference abstracts, editorials, letters, reviews, and systematic reviews were excluded due to the lack of extractable primary data suitable for meta-analysis. Studies were further excluded if ESR outcomes were not reported or if relevant data could not be extracted. To reduce confounding, studies involving participants with co-infections or comorbidities known to influence ESR (e.g., bacterial infections, autoimmune diseases) were excluded unless separate ESR data for malaria cases were provided.

Search strategy

Main databases, including ProQuest, Embase, Ovid, MEDLINE, PubMed, and Scopus, were systematically searched to identify relevant studies. Additional records were identified through searches conducted in Google Scholar and by reviewing the reference lists of included studies. The initial search was performed between 7 and 10 November 2023. To ensure the review reflects the most current evidence, an updated search was conducted on 1 July 2025. This updated search was performed using the same search terms and strategy, which focused on ‘erythrocyte sedimentation rate’ and ‘malaria’. The general search strategy was ‘(“Blood Sedimentation” OR “Erythrocyte Sedimentation” OR “Erythrocyte Sedimentation Rate” OR “Erythrocyte Sedimentation Rates” OR “Sedimentation Rate” OR “Sedimentation Rates” OR “Sed rate” OR ESR) AND (malaria OR Plasmodium OR “Plasmodium Infection” OR “Remittent Fever“ OR “Marsh Fever” OR Paludism)’ (Table S1). There was no restriction on publication date. Only articles published in the English language were included in the review.

Study selection and data extraction

After removing duplicates, the titles and abstracts of all retrieved studies were screened for relevance based on the study objectives. Studies were included if they reported ESR levels in (i) malaria patients compared to non-malarial controls, or (ii) severe malaria patients compared to non-severe cases. Studies were excluded if they were non-original articles (e.g., conference abstracts, case reports, reviews), lacked relevant ESR data, or did not meet the predefined inclusion criteria. The full texts of potentially eligible studies were then assessed for inclusion.

A standardized data extraction form was used to collect key information from the included studies. Extracted data included the first author’s name, study location, study design, participant characteristics, Plasmodium species, and methods of malaria diagnosis and ESR measurement (Table 1). When data were missing, attempts were made to contact the study authors; if the data could not be retrieved, this was recorded as missing. Study selection and data extraction were performed independently by two review authors (MK and AM). Discrepancies were resolved through discussion or, when required, by consulting a third review author (ATS). When relevant articles were identified but the full texts could not be accessed, attempts were made to obtain them by contacting the corresponding authors directly via email.

Table 1.

Summary of data extraction items

Data Item	Description
First Author	Name of the first author of the study
Year of study conduction	Year the study was conducted
Study location	Country or region where the study was conducted
Study design	Cross-sectional, case-control, or cohort design
Participant characteristics	Age, sex, and clinical details of participants
Plasmodium species	Species of malaria parasite identified
Malaria detection method	Method used to diagnose malaria (e.g., microscopy, rapid diagnostic test)
ESR measurement method	Method used to measure ESR
ESR values	Reported ESR values in each comparison group
Severity classification	Symptoms of patients (e.g., symptomatic vs. asymptomatic; severe vs. non-severe malaria)

Quality assessment

The methodological quality of the included studies was assessed using the Joanna Briggs Institute (JBI) critical appraisal checklists appropriate for analytical cross-sectional, case-control, and cohort studies [26]. The assessment focused on key domains such as clarity of inclusion criteria, description of study subjects and settings, reliability of measurement methods, management of potential confounding factors, and appropriateness of statistical analyses (Table 2). Two authors (MK, ATS) independently assessed quality, with disagreements resolved by consensus or third-party consultation (AM).

Table 2.

Summary of quality assessment criteria

Study type	JBI assessment criteria
Cross-Sectional	8 criteria, including: clear inclusion criteria, detailed setting description, valid measurement of exposure/outcome, identification and management of confounding factors, and appropriate statistical analysis
Case-Control	10 criteria, including: comparability of groups, matching procedures, consistent identification criteria, valid exposure/outcome measurement, management of confounding factors, sufficient exposure period, and appropriate statistical analysis
Cohort	11 criteria, including: similar group recruitment, valid exposure/outcome measurement, management of confounding factors, sufficient follow-up time, completeness of follow-up, and appropriate statistical analysis

Data synthesis and statistical analysis

The summary of data synthesis and statistical analysis methods is listed in Table 3. A qualitative synthesis summarized ESR differences between malaria patients and controls, and between severe and non-severe cases. For quantitative synthesis, meta-analyses were performed using a random-effects model. Meta-analysis was performed using a random-effects model [27] to calculate pooled standardized mean differences (SMDs) in ESR levels between malaria patients and non-malarial controls, and also between severe and non-severe malaria cases. Heterogeneity was assessed with the I² statistic, with values > 50% indicating substantial heterogeneity [28]. Meta-regression was performed to explore heterogeneity by publication year, study design, geographic location, age group, Plasmodium species, clinical status, and diagnostic/measurement methods. Subgroup analyses were conducted for these covariates [29]. Outliers were identified and removed where necessary. Power analysis was conducted to assess the adequacy of study numbers. Publication bias was assessed using funnel plots and Egger’s test [30, 31]. The trim-and-fill method was applied if bias was suspected [32]. All analyses were conducted in R Studio (Version 2023.09.1 + 494) [33]. The R codes for the meta-analysis, meta-regression, and sensitivity analysis are provided in Table S2.

Table 3.

Summary of data synthesis and statistical analysis methods

Analysis Step	Description
Qualitative synthesis	Narrative summary of ESR differences between (i) malaria patients vs. non-malarial controls, and (ii) severe vs. non-severe malaria cases.
Meta-analysis	Pooled standardized SMDs of ESR levels calculated using a random-effects model.
Heterogeneity assessment	Measured using the I² statistic; I² >50% considered substantial heterogeneity.
Meta-regression	Explored sources of heterogeneity by analyzing covariates such as publication year, study design, location, age group, Plasmodium species, clinical status, malaria, and ESR measurement methods.
Subgroup analyses	Stratified analyses based on covariates to explore effect variations.
Sensitivity analysis	Tested the robustness of the meta-analysis by sequentially removing individual studies.
Outlier detection	Identified and excluded studies that significantly deviated from the pooled effect estimate.
Power analysis	Assessed whether the number of included studies was sufficient to detect a significant difference in ESR levels.
Publication bias Assessment	Evaluated using funnel plot visualization and Egger’s test for asymmetry. The trim and fill method was used to adjust for suspected small-study bias or publication bias.
Software used	All analyses were performed using R Studio (Version: 2023.09.1 + 494).

Results

Search results

A total of 1,903 records were initially identified through database searches (ProQuest, Embase, Ovid, MEDLINE, PubMed, and Scopus; n = 1,703) and Google Scholar (n = 200). From the main databases, 1,703 records were identified and 244 duplicates were removed. This left 1,459 records to be screened, of which 1,366 were excluded because they involved non-relevant populations, lacked ESR data, or were conference abstracts. Ninety-three records were sought for retrieval, but four could not be obtained. Finally, 89 reports were assessed for eligibility, and 83 were excluded for reasons such as being case reports/series, in vitro studies, animal studies, lacking ESR data in malaria patients, or other specified criteria.

From the 200 Google Scholar records, 158 were excluded because they involved non-relevant populations or outcomes of interest. Forty-two records were sought for retrieval, but eight could not be obtained. Finally, 34 reports were assessed for eligibility, and 23 were excluded for reasons such as duplication with records in the main databases, inability to extract ESR data, or other specified criteria. Finally, nineteen studies [15, 20, 21, 34–49] were included in the final review: 6 [34, 40, 43, 44, 47, 50] from the main databases, 11 [20, 21, 35–37, 39, 41, 42, 46, 48, 49] from Google Scholar, and 2 [38, 45] from reference lists (Fig. 1).

Fig. 1.

Study flow diagram showing the study selection processes

Characteristics of included studies

Of the 19 included studies, over half were conducted in Africa (10/19, 52.6%), followed by Asia (8/19, 42.1%). Cross-sectional designs were the most frequent (10/19, 52.6%). Both adults and children were represented: nine studies (47.4%) included all age groups, five (26.3%) focused on adults, and three (15.8%) on children. Symptomatic malaria was the primary focus in most studies (14/19, 73.7%). P. falciparum was the most commonly identified species (7/19, 36.8%), while the remaining studies (12/19, 63.2%) examined P. vivax, other species, or did not specify the species. Microscopy, alone or in combination with other diagnostic methods, was used in 78.9% (15/19) of the studies. For ESR measurement, the Westergren or modified Westergren method was employed in 63.2% (12/19) of the studies (Table 4, Table S3).

Table 4.

ESR in patients with malaria compared to non-malarial controls. Clinical classification (severe, uncomplicated, or non-specified) was reported in a subset of the included studies

No.	Authors	Year of study conduction	Study location	Plasmodium spp.	Age group	Clinical malaria (severe, uncomplicated, mild)*	ESR in patients with malaria
1	Albiti et al. (2014) [34]	2008–2009	Yemen	P. falciparum	< 14 years	Uncomplicated malaria	ESR was significantly increased in malaria as compared with non-malarial controls.
2	Auta et al. (2016) [20]	Not specified	Nigeria	Not specified	0 to more than 36 years	Not reported	No significant difference in ESR between malaria and non-malarial controls.
3	Boampong et al. (2010) [21]	2008–2009	Ghana	P. falciparum	Male: 5 to 64 years; female: 5 to 68 years	Not reported	No significant difference in ESR between malaria and non-malarial controls.
4	Chidoka et al. (2013) [35]	2011	Nigeria	P. falciparum	11–31	Uncomplicated malaria	ESR was significantly increased in malaria as compared with non-malarial controls.
5	Chukwuocha et al. (2010) [36]	Not specified	Nigeria	Not specified	Malaria (29), controls (163): 11–60 years	Not reported	ESR was significantly increased in malaria as compared with non-malarial controls.
6	Ebuehi et al. (2009) [37]	Not specified	Nigeria	P. falciparum	Not specified	Severe malaria	ESR was significantly increased in malaria as compared with non-malarial controls.
7	Ekanem et al. (2018) [38]	Not specified	Nigeria	Not specified	6–60 months	Not reported	(1) ESR was significantly increased in malaria as compared with non-malarial controls. (2) No significant difference in ESR between malaria alone and with septicemia.
8	Eledo et al. (2018) [39]	Not specified	Nigeria	Not specified	18–45 years	Not reported	ESR was significantly increased in malaria as compared with non-malarial controls.
9	Eriksson et al. (1989) [40]	1975–1986	Not specified	P. falciparum (86), P. vivax (111). P. ovale (50), P. malariae (1), mixed infections (10), controls (120)	1–78 years	Not reported	No significant difference in ESR between malaria and non-malarial controls.
10	Farooq et al., (2018) [41]	2017–2018	Pakistan	Not specified	Not specified	Not reported	ESR was significantly increased in malaria as compared with non-malarial controls.
11	Francis et al. (2014) [42]	Not specified	Nigeria	Not specified	5–70 years	Not reported	ESR was significantly increased in malaria as compared with non-malarial controls.
12	Hussain et al. (2013) [43]	2008–2009	India	P. falciparum (52), P. vivax (42), mixed infection (12)	2–58 years	Not reported	ESR was significantly increased in malaria as compared with non-malarial controls.
13	Karunaweera et al. (1998) [44]	Not specified	Sri Lanka	P. vivax	13–69 years	Uncomplicated malaria	ESR was significantly increased in malaria as compared with non-malarial controls.
14	Kumar et al. (2017) [45]	2012–2014	India	P. vivax	≥ 18 years	Severe malaria	ESR was significantly increased in severe malaria as compared with non-severe malaria.
15	Mathews et al. (2019) [46]	2013–2015	India	P. vivax	> 12 years	Severe and non-severe malaria	ESR was significantly increased in severe malaria as compared with non-severe malaria.
16	Roy et al. (2024) [47]	2018–2019	India	P. falciparum	14–78 years	Severe and non-severe malaria	ESR was significantly increased in severe malaria as compared with non-severe malaria.
17	Sambo et al. (2020) [48]	Not specified	Nigeria	P. falciparum	Not specified	Not reported	ESR was significantly increased in malaria as compared with non-malarial controls.
18	Teddy Charles et al. (2013) [49]	2013	Nigeria	P. falciparum	18–59 years	Not reported	ESR was significantly increased in malaria as compared with non-malarial controls.
19	Vemula et al., 2016 [15]	Not specified	India	P. falciparum (15), P. vivax (86), mixed infection (21)	Not specified	Severe and non-severe malaria	ESR was significantly increased in severe malaria as compared with non-severe malaria.

ESR Erythrocyte sedimentation rate

*Not reported in the “Clinical malaria” column indicates that the original study did not report malaria severity classification

Risk of bias

For cross-sectional studies [20, 21, 35, 40, 41, 43, 46–48, 50], most clearly defined inclusion criteria, described subjects and settings, and measured exposures and outcomes using standard criteria. However, handling of confounding factors was inconsistent. Although most studies applied appropriate statistical analyses, many did not report methods for handling confounding factors or failed to identify relevant confounding factors (Table S4).

For case-control studies [36–39, 42, 44], cases and controls were generally comparable apart from disease status. Exposure measurement was consistent, although in some studies [36, 37] the reliability of the methods was unclear, and one [39] did not use a standard method for Plasmodium identification. Confounding factors management was rarely reported, except in one study [44] that addressed it comprehensively. Outcome assessment was standardized, but statistical methods were sometimes insufficiently described [37, 42].

For cohort studies [15, 45, 49], all studies recruited comparable groups and assigned exposures appropriately. Two studies [15, 45] reported a valid and reliable exposure measurement, while the one study [49] did not clearly specify methods. Two studies did not adequately describe strategies for controlling confounding factors [15, 49]. All studies ensured participants were initially outcome-free and used valid outcome measures, but the completeness of follow-up and management of attrition were unclear.

Systematic review findings

For the difference in ESR levels between malaria patients and non-malarial controls, all included studies consistently demonstrated elevated ESR in symptomatic malaria cases across different populations and age groups [15, 35–39, 41–46, 48–50]. The rise in ESR was particularly noted in infections caused by P. falciparum and P. vivax. However, a few studies (e.g., Auta et al. [20] Boampong et al. [21] and Eriksson et al. [40]) reported no significant ESR elevation, underscoring some inconsistency.

When comparing severe versus non-severe malaria, ESR was consistently higher in severe cases across both pediatric and adult populations (e.g., Kumar et al. [45] Mathews et al. [46] Roy et al. [47] and Vemula et al. [15]). Although the number of studies with relevant clinical stratification was limited, this pattern was broadly consistent across settings.

Meta-analysis findings

The difference in ESR levels between malaria patients and non-malarial controls

The pooled analysis of 11 studies (n = 1,623 participants) indicated significantly elevated ESR in malaria patients (SMD = 3.00, 95% CI 0.53–5.47, P = 0.0171). Heterogeneity was high (I² = 97.3%, Q = 371.88, P < 0.0001), suggesting substantial variability between studies (Fig. 2). Full study-level values are provided in Table S3.

Fig. 2.

Forest plot comparing erythrocyte sedimentation rate (ESR) between malaria patients and non-malarial controls. The pooled standardized mean difference (SMD = 3.00, 95% CI: 0.53–5.47) indicates a significant elevation of ESR in malaria patients. Substantial heterogeneity was observed (I² = 97.3%, P < 0.0001) using a random-effects model, and the prediction interval highlighted wide dispersion across studies

Cumulative meta-analysis confirmed the robustness of this finding, with significance maintained across sequential study additions (all P < 0.0001, Fig. 3).

Fig. 3.

Cumulative meta-analysis of erythrocyte sedimentation rate (ESR) comparing malaria patients with non-malarial controls. The cumulative trend consistently indicates elevated ESR in malaria patients, with the pooled standardized mean difference (SMD) remaining significantly positive under the random-effects model. High between-study heterogeneity persisted throughout the sequential inclusion of studies

Meta-regression identified diagnostic method as the only significant moderator (P < 0.0001), explaining 74.4% of between-study variance (R² = 74.36%, Table 5). No other covariates (publication year, study design, age, region, Plasmodium species, or clinical status) significantly accounted for heterogeneity (all P > 0.05).

Table 5.

Meta-regression analysis of the pooled comparisons of ESR levels between malaria patients and non-malarial controls

Covariates	tau2	I2 (%)	R-squared (%)	Test for Residual Heterogeneity, P value	Test of Moderators, P value
Publication years	16.5194	99.68	3.72	< 0.0001	0.2304
Study design	17.5352	99.73	0.00	< 0.0001	0.3683
Continent	19.2963	99.70	0.00	< 0.0001	0.5740
Age group	17.4236	99.73	0.00	< 0.0001	0.3748
Plasmodium species	25.4743	99.75	0.00	< 0.0001	0.9041
Clinical status (symptomatic vs. asymptomatic malaria)	16.4579	99.74	4.08	< 0.0001	0.2270
Clinical status (severe vs. uncomplicated malaria)	15.9443	99.75	7.08	< 0.0001	0.2485
Diagnostic method for malaria	4.3989	99.15	74.36	< 0.0001	< 0.0001
Method for ESR measurement	21.8449	99.76	0.00	< 0.0001	0.9762

ESR Erythrocyte sedimentation rate

Effect sizes varied significantly by diagnostic method, publication year, and geographic region (Table 6). Studies using rapid diagnostic tests (RDTs) showed markedly higher ESR effects (SMD = 13.67) than those using microscopy (SMD = 1.20), possibly reflecting inclusion of recently resolved or co-morbid cases. African studies showed larger effects (SMD = 4.09) than Asian ones (SMD = 1.60), and older studies (2000–2009) had higher pooled SMDs than more recent ones. No meaningful subgroup differences emerged for age group, study design, symptom status, or ESR measurement method (all P >0.05)

Table 6.

Subgroup analyses of the pooled comparisons of ESR levels between malaria patients and non-malarial controls

Subgroup analyses	Test for subgroup differences (P value)	Hedges’ g [95% CI]	I2 (%)	Number of studies
Publication years	< 0.0001
2020–2023
2010–2019		2.9509 [0.0548–5.8470]	96.9	8
2000–2009		9.4852 [6.8219–12.1486]	N/A	1
Before 2000		0.2396 [−0.0024–0.4816]	30.1	2
Study design	0.1806
Case-control study		4.7298 [0.4442–9.0154]	97.7	6
Cross-sectional study		1.1286 [–0.0836–2.3408]	97.5	4
Cohort study		0.7721 [0.2464–1.2979]	N/A	1
Continent	0.0082
Africa		4.0857 [0.2753–7.8962]	97.5	7
Asia		1.6022 [0.4239–2.7805]	96.0	3
Africa, Asia, America		0.1535 [–0.0634–0.3703]	97.5	1
Age group	0.2648
Children		4.0604 [–0.9849–9.1058]	96.3	3
Adults		7.1767 [–5.4635–19.8168]	99.2	2
Children and adults		1.0478 [–0.0050–2.1006]	97.1	5
Not specified		1.9780 [1.5353–2.4206]	N/A	1
Plasmodium species	< 0.0001
P. falciparum		2.9328 [–1.1309–6.9965]	96.5	4
P. vivax		0.4166 [0.0439–0.7894]	N/A	1
P. falciparum, P. vivax		0.1535 [–0.0634–0.3703]	N/A	1
P. falciparum, P. vivax, mixed infections		2.2861 [1.8098–2.7624]	N/A	1
Not specified		4.7057 [–1.0312 to 10.4425]	98.1	4
Symptoms	0.3853
Symptomatic malaria		1.9831 [0.1615–3.8047]	96.4	8
Not specified		5.5636 [–2.3126–13.4398]	98.5	3
Methods for Plasmodium identification	< 0.0001
Microscopic method		1.1973 [0.4125–1.9821]	96.6	7
RDT		13.6708 [11.4994–15.8421]	N/A	1
Not specified		3.9304 [–1.2640–9.1248]	95.7	3
Method for ESR measurement	0.8042
Westergren method		3.0333 [–0.3532–6.4198]	97.4	7
Modified Westergren’s method		2.1248 [1.6873–2.5623]	N/A	1
Not specified		3.3341 [–2.4343–9.1025]	96.0	3

ESR Erythrocyte sedimentation rate, N/A Not assessed, RDT Rapid diagnostic test

Publication bias

The funnel plot showed an asymmetrical distribution of the pooled estimate (Fig. 4). Egger’s test indicated significant asymmetry (P = 0.005, Fig. 4), suggesting publication bias. A trim-and-fill adjustment (adding 4 imputed studies) reduced the effect to non-significance (SMD = 0.82, 95% CI − 2.13–3.77, P = 0.5844), highlighting possible inflation of the original effect.

Fig. 4.

The funnel plot showed an asymmetrical distribution of the pooled estimate

Sensitivity analysis

Sensitivity analysis showed no single study unduly influenced the pooled effect (SMD remained significant at ~ 3.00). High heterogeneity persisted (I² >93%) across all leave-one-out models. Four studies [21, 37, 39, 40] were flagged as statistical outliers due to extreme SMDs or wide CIs. Their exclusion slightly reduced heterogeneity (I² = 93.5%) and improved precision, but did not alter the overall conclusion of elevated ESR in malaria.

Power analysis

With 11 studies and an assumed SMD of − 0.41, the meta-analysis had 100% statistical power to detect a true effect (α = 0.05), confirming adequate sample size and effect magnitude (Fig. 5).

Fig. 5.

The figure visually represents the power across different effect sizes. The plateau at the top of the curve, where the power equals 1 (or 100%), shows that for the given sample size and number of studies, the power to detect an effect of size similar to or larger than the one specified (SMD of −0.41) is very high. The red dot at the very beginning of the x-axis may represent the actual effect size estimated by the meta-analysis, which falls well within the region where the power to detect the effect is maximal

The difference in ESR levels between severe malaria patients and non-severe cases

A separate meta-analysis of four studies showed significantly higher ESR in severe malaria cases (SMD = 2.26, 95% CI 0.88–3.64, P = 0.0014, I² = 96.1%, Fig. 6). This pattern held across sequential inclusion of studies (Fig. 7), though heterogeneity remained substantial, reflecting diverse clinical definitions and patient populations.

Fig. 6.

Forest plot comparing erythrocyte sedimentation rate (ESR) between severe and non-severe malaria cases. The pooled standardized mean difference (SMD = 2.26, 95% CI: 0.88–3.64) indicates significantly higher ESR in severe malaria. Substantial heterogeneity was observed (I² = 96.1%, P < 0.0001) under the random-effects model, and the prediction interval illustrates wide variability across studies

Fig. 7.

Cumulative meta-analysis comparing erythrocyte sedimentation rate (ESR) between severe and non-severe malaria. The cumulative trend consistently indicates elevated ESR in severe malaria patients, with the pooled standardized mean difference (SMD) remaining significantly positive under the random-effects model. High between-study heterogeneity persisted throughout the sequential inclusion of studies

Discussion

The primary goal of this systematic review and meta-analysis was to evaluate the differences in ESR between malaria patients and non-malarial controls. The qualitative synthesis of studies [15, 35–39, 41–46, 48–50] predominantly showed elevated ESR levels in malaria patients. The meta-analysis confirms a significant elevation of ESR in malaria patients compared to non-malarial controls. The observed elevation of ESR in symptomatic malaria cases, especially those caused by P. falciparum and P. vivax, aligns with the known inflammatory response elicited by malaria infection [51, 52]. Clinically, ESR measurement at the bedside may assist in distinguishing malaria from other causes of fever, particularly in resource-limited settings. This implication is supported by previous research, which suggested that a combination of high ESR levels, low hemoglobin, and low blood sugar can reliably predict P. vivax infections in India [43].

The elevated ESR observed in malaria may be attributable to increased immunoglobulin production, which promotes rouleaux formation and faster erythrocyte sedimentation [40]. This correlation between ESR and immunoglobulin G appears consistent irrespective of age, sex, or the density of parasitemia [21]. However, this trend was not consistent across all studies. For instance, Auta et al. [20] Boampong et al. [21] and Eriksson et al. [40] reported no significant ESR elevation in certain cases. While several factors might contribute to this variability, the specific reasons for these inconsistencies remain unclear. Possible explanations derived from the included studies may account for the observed inconsistencies in ESR levels between patients with malaria and those without the disease. First, different Plasmodium species may lead to variations in ESR elevation. For example, P. falciparum infections are often associated with more severe symptoms, which might result in higher ESR levels. However, current evidence suggests that ESR is elevated in malaria patients regardless of the infecting Plasmodium species, and in some studies, P. falciparum cases showed slightly lower ESR levels than non-P. falciparum infections [40, 43]. Nevertheless, the studies by Auta et al. [20] and Boampong et al. [21] did not specify the Plasmodium species, which could mean they included less severe Plasmodium species, possibly contributing to the lack of significant ESR elevation. Second, the different symptomatic statuses of malaria (e.g., severe, uncomplicated, or mild) could influence ESR levels, as demonstrated by several studies [15, 45–47]. Eriksson et al. [40] included a range of Plasmodium species and did not specify the severity of the cases, which could suggest a mix of mild and severe cases, potentially averaging out ESR levels and resulting in no significant elevation. Third, methodological variations in ESR measurement and malaria diagnosis may have contributed to inconsistent findings. While both Auta et al. [20] and Boampong et al. [21] used microscopy and Westergren methods for ESR estimation, the process of estimation could affect the results. Eriksson et al. [40] did not specify the method for ESR, which adds another layer of variability. Fourth, demographic and genetic differences among study populations may influence ESR responses to malaria [53]. For instance, certain populations may have innate resistance or different immune responses to malaria that could result in lower ESR elevations [54]. Fifth, the study design and the sample size could also influence the outcomes. Eriksson et al. [40] conducted a study before the year 2000 across multiple continents, possibly leading to a more heterogeneous sample that could dilute the ESR signal. Sixth, the timing of ESR measurement during the course of illness could influence the results. It is important to specify the timing of ESR measurements during the course of the illness. For example, if taken early, ESR might not have elevated yet. Additionally, similar ESR values in some malaria and non-malaria patients may reflect overlapping inflammatory responses in other febrile illnesses. A previous study proposed that ESR values at least twice the upper reference limit, alongside appropriate signs and symptoms of malaria, might indicate Plasmodium infections, especially in cases where gold standard microscopy fails to detect the parasite in the blood [21]. While findings indicate that ESR is significantly elevated in malaria patients compared to non-malarial controls, it is important to emphasize that ESR is a nonspecific marker of inflammation, which can also be elevated in other infections, such as bacterial infections [32] lung infections [55] and bone and joint infections [17, 18]. Therefore, in endemic areas where these conditions frequently coexist, a combination of markedly elevated ESR with other laboratory markers may help distinguish malaria from other febrile illnesses.

The findings from the qualitative synthesis and the subsequent meta-analysis also provide compelling evidence of higher ESR levels in severe malaria cases compared to non-severe cases. This pattern, observed consistently across different geographic locations and among diverse demographic groups including both children and adults, aligns with the hypothesis that the severity of malaria correlates with greater inflammatory response, as indicated by ESR. The cumulative meta-analysis further reinforces these findings, showing a trend of higher ESR in severe malaria cases compared to non-severe cases as more studies are added to the analysis. This indicates a consistent pattern over time and across different research contexts, strengthening the argument that ESR can be a relevant biomarker in distinguishing between severe and non-severe malaria cases. Interestingly, the previous study suggested that pretreatment ESR, done at the time of diagnosis, could predict the complications of malaria patients with a good predictive value [15]. Due to the limited number of eligible studies, meta-regression and subgroup analyses exploring ESR differences between severe and non-severe malaria were not feasible and warrant future investigation.

While the meta-analysis highlights the association between elevated ESR and malaria severity, other inflammatory markers such as C-reactive protein (CRP) and procalcitonin have also been investigated as potential biomarkers for severe malaria. A previous meta-analysis demonstrated that CRP levels are significantly higher in patients with severe malaria compared to those with uncomplicated malaria [56]. In contrast, another meta-analysis reported no significant difference in procalcitonin levels between severe and uncomplicated malaria cases [13]. In clinical practice, a combined approach using ESR alongside CRP or procalcitonin may improve diagnostic and prognostic accuracy, particularly for distinguishing malaria from other febrile illnesses and for identifying severe cases early. Another important aspect is the potential correlation between ESR elevation and parasite density. Higher parasite loads are often associated with a more robust inflammatory response in malaria, mediated by heightened levels of key pro-inflammatory cytokines such as TNF-α, IFN-γ, and IL-6 [57]. However, the studies included in this meta-analysis did not consistently report parasitemia levels alongside ESR measurements, limiting the ability to directly analyze this relationship.

The cumulative meta-analysis has revealed a dynamic and evolving pattern of ESR levels in malaria patients, with each additional study contributing to a significant SMD and maintaining statistical significance. This sequential accumulation of studies highlights the robustness and consistency of the findings over time. The subgroup analysis further reinforces these findings, presenting significant temporal, geographical, and methodological variations in ESR responses. Notably, geographic differences and Plasmodium species appeared to influence effect sizes, indicating possible regional and biological contributors to ESR variation. Additionally, the significant impact of the diagnostic method on the observed effect sizes points to the sensitivity of ESR responses to the techniques used for malaria detection. The subgroup analysis showed that studies utilizing microscopic methods showed a smaller effect size, while those employing RDTs displayed much larger effects. This large discrepancy suggests that RDT-based studies may overestimate ESR elevation in malaria cases, possibly due to the inclusion of individuals with recent but resolved infections or co-existing inflammatory conditions. These findings highlight a critical methodological concern: the heterogeneity in diagnostic techniques across studies can substantially impact the observed relationship between ESR and malaria. Therefore, interpretation of ESR as a biomarker in malaria should be made cautiously, particularly in studies relying solely on RDTs.

Sensitivity analyses further confirm the robustness of the meta-analysis, with high heterogeneity remaining a consistent feature. The identification of outliers and their subsequent removal in the random-effects model also supports the overall significance of the findings, despite the persisting heterogeneity. Lastly, a power analysis demonstrating a 100% probability of detecting a true effect confirms the adequacy of the sample size and the robustness of the observed effect. The results of qualitative and quantitative analyses provide confidence in the validity of conclusions and the likelihood that this analysis has effectively captured the true impact of malaria on ESR levels.

Despite insights from the systematic review and meta-analysis, some limitations must be acknowledged. An important limitation is the lack of standardization in ESR measurement methods across the included studies. While the Westergren method was most commonly used, some studies did not specify the technique, and a few employed automated systems, potentially introducing inter-study variability. Additionally, comorbidities such as anemia, HIV infection, and other concurrent infections may influence ESR levels independently of malaria status. A significant limitation of the meta-analysis is the substantial heterogeneity observed across the included studies (I² >95%). The heterogeneity likely reflects differences in study design, geographic settings, Plasmodium species, measurement techniques, and patient populations, including the significant influence of the malaria diagnostic method (microscopy vs. RDT). This variability limits the external validity of findings, making it difficult to generalize the results to diverse clinical settings. This underscores the need for a context-specific interpretation of ESR in malaria and highlights the importance of considering local epidemiology, diagnostic methods, and patient characteristics when applying these findings to clinical practice. The exclusion of some studies for which full-texts could not be obtained may have introduced selection bias or led to an incomplete synthesis of the available evidence. A significant publication bias was identified in the meta-analysis, as indicated by Egger’s test (P = 0.005). The trim-and-fill method also suggested that the initially observed significant association between elevated ESR and malaria may have been influenced by publication bias. Therefore, the results should be interpreted with caution.

Although ESR has been widely studied in various inflammatory and chronic conditions, including bacterial infections, viral diseases, and hematologic malignancies such as leukemia, it is generally regarded as a nonspecific marker of inflammation rather than a primary diagnostic tool in these conditions. Although ESR has potential as a biomarker for malaria severity, its nonspecific nature limits its standalone use. In malaria-endemic regions where co-infections and other febrile illnesses are common, combining ESR with other diagnostic tools such as microscopy, rapid diagnostic tests, CRP, and procalcitonin may enhance diagnostic accuracy and aid in early detection of severe cases. Future research should focus on larger, diverse populations with standardized data collection to further clarify ESR’s role in malaria diagnosis and prognosis, especially in resource-limited settings.

Conclusion

This systematic review and meta-analysis provide insights into the relationship between ESR levels and malaria. Elevated ESR levels are frequently reported in malaria patients, particularly in infections with P. falciparum and P. vivax, as well as in symptomatic cases. Severe malaria cases also tend to exhibit higher ESR compared to non-severe cases. However, the substantial between-study heterogeneity and potential publication bias highlight uncertainty regarding the strength and consistency of these associations. While the feasibility of measuring ESR at the point of care suggests possible utility for early detection of complications, its role as a reliable diagnostic or prognostic marker for malaria remains uncertain and warrants further investigation.

Abbreviations

CBC

Complete blood count

CI

Confidence interval

CRP

C-reactive protein

ESR

Erythrocyte sedimentation rate

JBI

Joanna Briggs Institute

LFTs

Liver function tests

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO

International Prospective Register of Systematic Reviews

RDTs

Rapid diagnostic tests

SMD

Standardized mean differences

WHO

World Health Organization

Authors’ contributions

MK, ATS, KUK, and AM carried out the study design, study selection, data extraction, and statistical analysis; and drafted the manuscript. FRM, KT, KW, and PW participated in critical editing the manuscript. All authors read and approved the final manuscript.

Funding

Open access funding provided by Mahidol University. KW is funded by the Australian National Health and Medical Research Council (NHMRC) Investigator Grant (2008697).

Data availability

All data relating to the present study are available in this manuscript, Table S1, Table S2, Table S3, and Table S4 files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.