Association of uric acid levels with severity of Plasmodium infections: a systematic review and meta-analysis

Elevated uric acid (UA) levels have been reported in malaria patients and are particularly prominent in severe malaria cases. This study aims to synthesize the difference in UA levels between malaria patients and uninfected controls, and between patients with severe and non-severe malaria. A comprehensive literature search was carried out across databases such as Embase, MEDLINE, Ovid, PubMed, Scopus, ProQuest, and Google Scholar to identify relevant studies for inclusion. The methodological quality of the included studies was evaluated independently by two reviewers using the JBI critical appraisal tool for observational studies. A meta-analysis was performed to calculate the pooled effect sizes, expressed as Hedges' g, with 95% confidence intervals (CIs). The Hedges' g was pooled using the random-effects model. An initial search yielding a total of 1122 articles, and a final total of 19 studies being included in the review. Elevated UA levels were observed more prominently in malaria patients, especially those with severe manifestations, when compared to uninfected controls. The conducted meta-analysis demonstrated a significant elevation in UA levels in patients suffering from malaria as compared to uninfected controls (P < 0.01, Hedges’s g = 1.40, 95% CI 0.84–1.95, I2 = 95.81, 16 studies). The conducted meta-analysis demonstrated a significant elevation in UA levels in patients suffering from severe malaria as compared to non-severe malaria (P < 0.01, Hedges’s g = 3.45, 95% CI 1.06–5.83, I2 = 98.73, 6 studies). In summary, these findings provide valuable insights into the potential use of UA as a biomarker for malaria infection and determination of its severity. Further research is needed to validate these findings and to explore the underlying mechanisms that contribute to the elevation of UA levels during malaria infection.

Quality assessment.The methodological quality of the included studies was evaluated independently by two reviewers (SK, MK) using the JBI critical appraisal tool for observational studies 25 .Studies were assessed based on selection, comparability, and outcome for cohort and case-control studies, and selection, comparability, and exposure for cross-sectional studies.Discrepancies were resolved through discussion, or a third reviewer (KUK) was consulted when necessary.
Data synthesis and analysis.Thematic synthesis was employed to capture these overarching themes and narratives, synthesizing the collective qualitative insights 26 .The meta-analysis was performed to calculate the pooled effect sizes, expressed as Hedges' g, with 95% confidence intervals (CIs).The Hedges' g was pooled using the random-effects model 27 .Heterogeneity among studies was assessed using the I 2 statistic, with values > 50% indicating substantial heterogeneity 28 .Subgroup analyses were performed based on publication years, study design, geographical continent, age group of participants, species of Plasmodium, diagnostic methods for malaria, and the type of blood samples collected.Meta-regression analyses were conducted to explore potential sources of heterogeneity.Sensitivity analysis, or leave-one-out analysis, was conducted to assess the robustness of the results.Publication bias was examined visually using funnel plots and statistically with Egger's test, and Characteristics of studies.This analysis comprised 19 studies, primarily published from 2010 to 2023 (73.7%), and employed cross-sectional (63.2%), cohort (21.1%), and case-control designs (15.8%).The geographical distribution was nearly balanced between Africa (42.1%) and Asia (57.9%), with a preponderance of studies from Nigeria and India.The studies predominantly focused on P. falciparum (52.6%) and P. vivax (10.5%).A subset of studies investigated multiple Plasmodium species, with 15.8% examining both P. falciparum and P. vivax, and another 10.5% looking into mixed infections involving P. falciparum, P. vivax, and other species.Participant cohorts ranged across all age groups, with a slightly higher proportion of studies focusing on adults (42.1%).Microscopy was the predominant method for malaria detection (63.2%), with some studies utilizing combined methodologies.The blood samples for uric acid measurement were equally divided between serum and plasma (42.1% each), with some studies unspecified (Table 1, Table S2).

Quality of studies.
The quality of the studies was assessed using the JBI critical appraisal checklist for observational studies.Among the cross-sectional studies, consistent adherence to the checklist outcomes was observed in the studies conducted by Aqeel et al. 29,30 , Ayede et al. 31 , Bertrand et al. 20 , Bhardwaj et al. 32 , Karua and Kishan 41 , Okon et al. 34 , Prabhu et al. 42 , and Iwalokun et al. 19 .However, some studies such as Erel et al. 38 , Idonije et al. 4 , and Thurnham et al. 35 had an unclear status regarding the inclusion criteria and addressing confounding www.nature.com/scientificreports/factors.Among the case-control studies, Ebrahim et al. 37 , Hatim et al. 40 , and Olisekodiaka et al. 39 lacked identification of confounding factors and strategies to address them.In the cohort studies, Chandrashekhar et al. 33 and Lopera-Mesa et al. 17 lacked identification of confounding factors, strategies to address them, and details on follow-up time.Das et al. 36 study also lacked identification of confounding factors, strategies to address them, and details on follow-up time.Moreover, Das BS and Nanda NK 6 study was unclear regarding the utilization of strategies to address incomplete follow-up (Table S3).
Qualitative synthesis of UA levels.From thematic synthesis, the predominant theme that emerged from studies was the general consensus that malaria patients exhibited higher UA levels compared to uninfected controls 6,19,20,30,[32][33][34][37][38][39][40]42 . This obsevation was further supported by narratives and discussions emphasizing the clinical implications of these elevated levels.Furthermore, several studies emphasized an intensified theme revolving around even higher UA levels in severe malaria cases compared to non-severe ones 17,19,30,33,34,41 .An interesting narrative thread found in the four studies posits a direct relationship between the intensity of malaria parasitemia and rising UA levels 17,31,34,39 .However, contrasting themes were evident in studies by Aqeel et al. 29 and Idonije et al. 4 , where narratives did not underscore a significant difference in UA levels between the two groups (Table 2).www.nature.com/scientificreports/ UA levels in patints suffering from malaria as compared to uninfected controls.In the quantitative synthesis, sixteen studies were deemed eligible for inclusion, owing to the presence of quantitative data 4,6,19,20,[29][30][31][32][33][34][35][37][38][39][40]42 .The conducted meta-analysis demonstrated a significant elevation in UA levels in patients suffering from malaria as compared to uninfected controls (P < 0.01, Hedges's g = 1.40, 95% CI 0.84-1.95,I 2 = 95.81, 16 studies,Fig. 2). Fuhermore, meta-regression analyses were performed, factoring in publication years, study design, geographical continent, age group of participants, species of Plasmodium, diagnostic methods for malaria, and the type of blood samples collected.The analyses demonstrated that both the year of publication and the employed study design significantly influenced the pooled effect estimate (P < 0.05, refer to Table S4).This evidence suggests that these two variables should be given due consideration while interpreting the results.The subgroup analyses of UA levels between malaria cases and uninfected controls have been conducted.For studies published between 2010 and 2023, a statistically significant higher UA level in malaria patients (P < 0.01, Hedges' g = 1.93, 95% CI 1.11-2.54,I 2 = 95.88) was identified across 11 studies.However, studies published before 2010 did not present a statistically significant difference.In terms of study design, both cross-sectional and case-control studies showed significantly higher UA levels in malaria cases (P < 0.01), while cohort studies did not.Geographically, studies from both Africa and Asia consistently revealed higher UA levels in malaria cases (P < 0.01).When assessing age groups, children and adults had significantly higher UA levels (P < 0.01), while groups with unspecified age or encompassing all age groups did not demonstrate statistical significance.UA levels varied across different Plasmodium species, with significantly higher UA levels identified in P. falciparum and mixed infection cases (P < 0.01).Cases with P. vivax also had higher UA levels, but the significance level was less robust (P = 0.04).The diagnostic method for malaria (microscopy or microscopy/RDT) and types of blood samples (serum or plasma) also showed significant differences in UA levels.However, UA levels in cases where diagnostic methods and types of blood samples were not specified did not reach statistical significance (Table 3).
UA levels in patients suffering from severe malaria as compared to non-severe malaria.In the quantitative synthesis, six studies were deemed eligible for inclusion, owing to the presence of quantitative data 17,19,30,[32][33][34] .The conducted meta-analysis demonstrated a significant elevation in UA levels in patients suffering from severe malaria as compared to non-severe malaria (P < 0.01, Hedges's g = 3.45, 95% CI = 1.06-5.83,I 2 = 98.73, 6 studies, Fig. 3).Furthermore, meta-regression analyses were performed, factoring in publication years, study design, geographical continent, age group of participants, species of Plasmodium, diagnostic methods for malaria, and the type of blood samples collected.The analyses divulged that none of these factors significantly influenced the pooled effect estimate (P > 0.05, Table S4).

Sensitivity analysis.
The robustness of the meta-analysis was evaluated using a leave-one-out analysis, which tests the influence of individual studies on the overall meta-analysis result by systematically removing each study and re-running the analysis.When applied to the comparison of UA levels between malaria patients     www.nature.com/scientificreports/and uninfected controls, this analysis revealed consistently significant elevations in UA levels in malaria patients across all re-run analyses (P < 0.05 in all cases, as depicted in Fig. 4).These results demonstrate the robustness of the meta-analysis findings, as the overall significance was maintained irrespective of the exclusion of any single study.
On the other hand, the results were different when this leave-one-out analysis was applied to the comparison of UA levels between severe and non-severe malaria cases.In this case, the significance of the results varied depending on the study that was excluded (as shown in Fig. 5, with P < 0.05 in some reruns and P > 0.05 in others).This variation in results indicates that the meta-analysis outcomes in this context were not robust, and the overall conclusions were sensitive to the removal of individual studies.

Publication bias.
Visual inspection of the funnel plot unveiled an asymmetrical distribution (as depicted in Fig. 6), suggesting potential bias in the meta-analysis comparing UA levels between malaria cases and uninfected controls.Further substantiating this, Egger's test revealed a significant small-study effect (P < 0.01), indicating the presence of publication bias likely attributable to the disproportionate influence of smaller studies.
To correct for this observed bias, the trim-and-fill method was implemented.This statistical technique adjusts the overall effect size by estimating and correcting for the number and outcomes of hypothetically missing studies due to publication bias.After applying this method, the revised results continued to indicate a significant increase in UA levels among malaria patients compared to uninfected controls (Table S5).

Discussion
The study results shed light on the intricate link between UA levels and malaria, and underscore the significance of certain study-related variables, such as the year of publication and study design, in determining the strength and direction of this association.A meta-analysis of 16 studies revealed a significant elevation in UA levels in patients afflicted with malaria as compared to uninfected controls.Notably, the year of publication and the Figure 5.The leave-one-out method showed the re-run meta-analysis of the difference in the uric levels between patients with severe and non-severe malaria.CI confidence interval.employed study design emerged as significant factors influencing the pooled effect estimate.This raises important questions about potential changes in research methodology, laboratory techniques, or other unmeasured confounding variables that might have changed over the years and could affect the reliability of the association between UA levels and malaria.Furthermore, the results of the subgroup analyses show nuances in UA levels between malaria cases and uninfected controls.Studies conducted more recently (2010-2023) found a significant increase in UA levels among malaria patients, whereas older studies did not.It is possible that recent improvements in diagnostic and analytical techniques might have contributed to these results.As for the study design, both cross-sectional and case-control studies revealed a significant increase in UA levels in malaria cases, suggesting that these designs might be more sensitive at detecting this difference.The meta-analysis also reveals that the geographic location and age group of the participants play a crucial role in determining the UA levels.Studies conducted in Africa and Asia consistently showed elevated UA levels in malaria patients, which could reflect the high malaria burden in these regions 1 .
Interestingly, UA levels varied between P. falciparum and P. vivax infections.This variability could be attributed to the unique pathogenesis and disease manifestations between P. falciparum and P. vivax infections 43,44 .The fact that cases of P. falciparum and mixed infections exhibited significantly higher UA levels suggests a potentially more severe inflammatory response associated with these types of infections.For different age groups, both adult and pediatric patients had higher UA levels, underscoring the potential of UA as a biomarker for malaria across all age groups.Nevertheless, UA may be used as a biomarker for malaria severity only in children because there was no difference in UA between severe and non-severe malaria among studies that enrolled adults.
The current meta-analysis included six studies due to the availability of quantitative data 17,19,30,[32][33][34] , underscored a significant increase in UA levels in patients with severe malaria compared to those with non-severe malaria.Elevated UA levels appear to be associated with severe malaria; however, this association seems to be influenced by factors such as geographical location and the age group of participants, as demonstrated in the subgroup meta-analysis.Given the observed heterogeneity among the studies reviewed, the findings from the meta-analysis should be approached with caution.The diversity in methodologies and objectives of the primary studies inherently adds complexity to a comprehensive interpretation.While this review identifies a potential link between UA levels and malaria severity, it is essential for readers and researchers to contextualize each primary study's aims and methods when drawing conclusions from these insights.Moreover, delving deeper into the mechanisms underlying this association is vital, not only for a clearer understanding of malaria's pathophysiology but also for uncovering potential therapeutic targets, enhancing predictions of disease severity, and tailoring treatments more effectively.
Previous in vitro investigations have demonstrated that the induction of the innate immune response to Plasmodium infection, specifically in human dendritic cells, is facilitated by the presence of UA precipitates derived from P. falciparum.This process involves the maturation of dendritic cells, leading to elevated expression levels of cell surface co-stimulatory molecules such as CD40, CD80, and CD86 45,46 .In a similar vein, a reduction in UA levels in vivo has the potential to mitigate the host's inflammatory response and alleviate associated pathology 46 .A previous study revealed that the formation of UA through the degradation of Plasmodium-derived hypoxanthine/xanthine triggers the release of TNF, which is a cytokine associated with the response to Plasmodium infections 47 .Moreover, inhibiting the formation of UA significantly diminishes the secretion of cytokines such as TNF, IL-6, IL-1β, and IL-10, which are known to be involved in the response to P. falciparum infections and are associated with the pathogenesis of malaria 48 .Another study indicated that patients with P. vivax malaria exhibit lower levels of uric acid during the acute phase compared to the convalescent phase, suggesting clinical improvement resulting from antimalarial treatment or as a negative finding in P. vivax malaria 49 .Conversely, it has been reported that UA levels increase during the acute phase of P. falciparum malaria and correlate with disease severity 17 .The elevated levels of UA contribute to the pathogenesis of P. falciparum malaria by inducing endothelial pathology, as evidenced by the basal plasma levels of UA and endothelial biomarkers, which rise during episodes of uncomplicated malaria and escalate further with increasing disease severity 50 .
The present study is subject to certain limitations.Firstly, there was a publication bias detected in the metaanalysis results, which may lead to an overestimation of the effect size.Secondly, our meta-analysis demonstrated a high I 2 value, indicating substantial heterogeneity among the included studies.This heterogeneity, resulting from differences in study design, population demographics, geographical location, diagnostic methods, and other factors, complicates the task of drawing definitive conclusions.Thirdly, although our meta-regression analyses factored in several variables, there may still be unconsidered confounding factors that could potentially influence the observed relationship between malaria severity and UA levels.Fourthly, the limited number of studies included in our analysis constrains the breadth of our conclusions, making it challenging to generalize across all malaria patients or geographical locations.Lastly, the limited data available for both severe and non-severe malaria patients may impinge on the reliability of the comparison made between these two groups.Consequently, future research should aim to address these limitations for a more comprehensive understanding of the topic.

Conclusion
In light of the objective to synthesize the difference in UA levels between malaria patients and uninfected controls, as well as between patients with varying malaria severity, the review found evidence supporting the potential role of UA as a distinguishing biomarker.Elevated UA levels were observed more prominently in malaria patients, especially those with severe manifestations, when compared to uninfected controls.However, it is crucial to approach these findings with caution due to certain study-related variables that could influence results.While the findings underscore UA's potential as an indicative marker for malaria infection and its severity, further research is imperative to validate these observations and delve deeper into the mechanisms prompting the surge in UA levels during a malaria infection.

Figure 2 .
Figure 2. The forest plot shows the difference in the uric levels between malaria patients and uninfected controls.CI confidence interval, N number of participants, SD standard deviation.

Figure 3 .
Figure 3.The forest plot shows the difference in the uric levels between patients with severe malaria and those with non-severe malaria.CI confidence interval, N number of participants, SD standard deviation.

Figure 4 .
Figure 4.The leave-one-out method showed the re-run meta-analysis of the difference in the uric levels between malaria patients and uninfected controls.CI confidence interval.

Figure 6 .
Figure 6.The funnel plot showing an asymmetrical distribution of the effect estimates between the middle line of the plot.CI confidence interval.

Table 1 .
Characteristics of studies.UA uric acid, RDT rapid diagnostic test, QBT quantitative buffy coat assay.

Table 2 .
Outcomes of included studies.UA uric acid.

Table 3 .
Subgroup analyses of UA levels between malaria cases and uninfected controls.CI confidence interval, N/A not assessed, RDT rapid diagnostic test.

Table 4 .
Subgroup analyses of UA levels between severe and non-severe malaria.CI confidence interval, N/A not assessed, RDT rapid diagnostic test.