Efficacy of D-Mannose in the Management of Urinary Tract Infections: From Acute Treatment to Long-term Prophylaxis — A Systematic Review and Meta-analysis
Article information
Abstract
Purpose
To evaluate the efficacy of D-mannose for preventing recurrent urinary tract infections (UTIs), including in kidney transplant recipients, and to clarify discrepancies between early open-label and placebo-controlled trials.
Materials and Methods
PubMed, Embase, CENTRAL (Cochrane Central Register of Controlled Trials), and Web of Science were searched from inception to February 2026. Randomized controlled trials (RCTs) comparing D-mannose with placebo, no treatment, antibiotics, or active controls were included. Risk of bias (RoB) was assessed using RoB 2, and certainty of evidence using GRADE (Grading of Recommendations, Assessment, Development, and Evaluations).
Results
Eleven RCT reports involving 1,724 participants were included; 10 independent study populations involving 1,631 participants contributed to quantitative analyses. In placebo-controlled trials, D-mannose did not significantly reduce UTI recurrence or persistence versus placebo (risk ratio [RR], 0.38; 95% confidence interval [CI], 0.06–2.36). No-treatment comparisons showed reduced recurrence (RR, 0.23; 95% CI, 0.08–0.66), but were more vulnerable to expectation, performance, and detection biases. The exploratory combined estimate favored D-mannose (RR, 0.28; 95% CI, 0.12–0.66), but certainty was very low. Antibiotic comparisons were inconclusive (RR, 0.43; 95% CI, 0.18–1.05), whereas proanthocyanidins active-control comparisons favored D-mannose-containing regimens (RR, 0.57; 95% CI, 0.40–0.82), including data from kidney transplant recipients.
Conclusions
Current placebo-controlled evidence does not establish superiority of D-mannose over placebo. Apparent benefits were mainly driven by no-treatment comparisons with very low certainty. D-mannose remains biologically plausible but clinically uncertain; adequately powered, double-blind, placebo-controlled trials are needed before firm recommendations can be made.
HIGHLIGHTS
Placebo-controlled trials did not establish superiority of D-mannose over placebo for urinary tract infection recurrence or persistence. Apparent benefits were mainly driven by no-treatment comparisons, which had very low certainty of evidence. Further adequately powered, double-blind, placebo-controlled trials are needed before firm clinical recommendations can be made.
INTRODUCTION
Urinary tract infections (UTIs) are among the most prevalent bacterial infections globally, affecting approximately 150 million individuals annually [1,2]. Women are disproportionately affected, with more than 50% experiencing at least one episode during their lifetime [3,4]. A significant clinical challenge lies in the high rate of recurrence; approximately 25%–30% of women with an initial UTI will experience a recurrence within 6 months [5]. While antibiotics remain the cornerstone of both acute treatment and prophylaxis, their prolonged use contributes significantly to the global crisis of antimicrobial resistance and disrupts the host microbiome, often leading to adverse events (AEs) such as Clostridioides difficile infection and fungal vaginitis [6-10]. Consequently, there is an urgent need for effective nonantibiotic alternatives that can reduce reliance on antimicrobial agents without compromising clinical outcomes.
D-mannose, an inert monosaccharide naturally found in fruits, has emerged as a promising alternative strategy. Its mechanism of action is well-established: D-mannose inhibits the adhesion of Escherichia coli—the causative pathogen in over 80% of UTIs—to the urothelium [2,3]. By binding to the FimH adhesins located at the tips of type 1 fimbriae of E. coli, D-mannose prevents bacterial colonization and facilitates the clearance of pathogens through urination [11]. Given its excellent safety profile and minimal impact on systemic metabolism, D-mannose is increasingly recommended in clinical guidelines as a preventive option for recurrent UTIs [6,12].
Despite its growing popularity, the clinical evidence supporting D-mannose remains conflicting. Although several systematic reviews have been published, they were often limited by small sample sizes, significant heterogeneity, and the inclusion of nonrandomized studies [12,13]. Furthermore, a recent large-scale, high-quality randomized controlled trial (RCT) published in 2024 reported findings that challenge earlier optimistic conclusions regarding its efficacy [14]. Additionally, previous reviews have primarily focused on prophylaxis in generally healthy women, often overlooking its potential role in acute treatment or in special populations, such as kidney transplant recipients who are uniquely vulnerable to UTIs [12,15,16].
To address these gaps and controversies, we conducted a systematic review and meta-analysis of all available RCTs to date. This study aims to provide a comprehensive and up-to-date evaluation of the efficacy and safety of D-mannose compared to placebo or antibiotics, covering the full spectrum of management from acute treatment to long-term prophylaxis across diverse patient populations.
MATERIALS AND METHODS
1. Study Protocol and Eligibility Criteria
This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and the Participants, Interventions, Comparators, Outcomes, and Study (PICOS) design framework [17,18]. The study protocol was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO: CRD420261305485). Ethical review and approval were waived for this study because it is a systematic review and meta-analysis of previously published studies and does not involve direct contact with human subjects or the use of identifiable private information, in accordance with the Bioethics and Safety Act of the Republic of Korea.
The inclusion criteria were: (1) adult women with acute or recurrent UTIs, or patients undergoing urological procedures (including kidney transplant recipients); (2) studies comparing D-mannose, either as monotherapy or as part of a D-mannose-containing regimen, with placebo, no treatment, antibiotics, or active controls such as proanthocyanidins (PAC); and (3) reporting of clinical outcomes such as symptom cure rate, recurrence rate, or AEs. We included only RCTs published as full-text peer-reviewed articles to ensure the highest level of evidence. Observational studies, case reports, review articles, editorials, scientific congress abstracts, and studies without available full text were strictly excluded.
2. Search Strategy
A comprehensive search was performed in PubMed, Embase, the Cochrane Central Register of Controlled Trials (CENTRAL), and Web of Science for articles published from inception to February 2026. The search strategy was designed based on the PICOS framework, combining MeSH (medical subject headings) and free-text terms relevant to (1) population (e.g., "urinary tract infections," "cystitis," "pyelonephritis," "bacteriuria"), (2) intervention (e.g., "mannose," "D-mannose," "monosaccharides"), and (3) study design (specific filters for randomized controlled trials).
Boolean operators (AND, OR) were used to combine these concepts (e.g., [Population] AND [Intervention] AND [RCT Filter]). No language restrictions were applied. To ensure a thorough review, reference lists of eligible studies and relevant systematic reviews were manually screened. The detailed PubMed search strategy is provided in Supplementary Table 1, and similar strategies were adapted for Embase, CENTRAL, and Web of Science using appropriate database-specific syntax and subject headings.
3. Study Selection and Extraction
Two researchers (JYJ and YJM) independently screened titles and abstracts, followed by full-text review. Extracted data included: primary author, publication year, country, study design, and patient characteristics. The primary outcomes were the recurrence rate of UTIs (for prophylaxis studies) and symptom cure rate (for acute treatment studies). Secondary outcomes included time to recurrence and AEs. Any discrepancies were resolved through consensus or consultation with a third reviewer (HDJ). Reports that could not be assessed as full-text peer-reviewed RCT articles were documented separately, including the reason for nonretrieval or exclusion before full-text eligibility assessment.
4. Quality Assessment and Certainty of Evidence
The methodological quality of the included RCTs was assessed using the Cochrane Risk of Bias tool (RoB 2). We evaluated 5 domains: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result.
Furthermore, the certainty of the evidence for each outcome was appraised using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach. The evidence was classified as high, moderate, low, or very low quality based on considerations of risk of bias (RoB), inconsistency, indirectness, imprecision, and publication bias.
5. Statistical Analysis
All statistical analyses and visualizations were performed using R ver. 4.3.1 (R Foundation for Statistical Computing, Austria). The 'meta' and 'metafor' packages were utilized for meta-analysis calculations, the 'robvis' package was employed to generate risk-of-bias plots, and the 'PRISMA2020' package was used to generate the PRISMA 2020 flow diagram.
For dichotomous variables, such as recurrence rates and AEs, risk ratios (RRs) and 95% confidence intervals (CIs) were calculated using the Mantel-Haenszel method. For continuous variables, Mean differences and 95% CIs were reported. Statistical heterogeneity among the included studies was evaluated using the chi-square test (significance level of p<0.10) and quantified using the I2 statistic. To further investigate potential sources of heterogeneity and identify outliers, L'Abbé plots and radial (Galbraith) plots were visually inspected.
Given the anticipated clinical and methodological heterogeneity (e.g., varying dosages, different control groups), a random-effects model was employed for all meta-analyses to provide a conservative estimate, regardless of the I2 value. Because placebo and no-treatment controls are not clinically or methodologically equivalent, placebo-controlled trials were analyzed separately and considered the primary estimate of the specific treatment effect of D-mannose. No-treatment comparisons were analyzed separately as supportive evidence, and the combined placebo/no-treatment analysis was retained only as an exploratory pooled estimate. Potential small-study effects were explored using funnel plots when at least 5 studies were available for a comparison. Formal tests for funnel plot asymmetry, such as Egger test, were planned only when at least 10 studies were available and were therefore not performed in the present analysis. Sensitivity analyses were performed by restricting the analysis to placebo-controlled and/or double-blind trials. Prediction intervals were calculated for random-effects analyses with substantial heterogeneity when a sufficient number of studies was available, and were interpreted cautiously in analyses with few studies.
RESULTS
1. Eligible Studies
A total of 954 records were initially identified through database searching. After removing 576 duplicates, 378 records remained for screening. Following title and abstract screening, 349 records were excluded, and 29 reports were sought for retrieval. Of these, 18 reports could not be assessed as full-text RCT articles because peer-reviewed full texts were unavailable or the records were available only as non–full-text publications, such as conference abstracts, trial registry records, protocols, or abstract-only records. The remaining 11 full-text reports were assessed for eligibility. Ultimately, 11 full-text RCT reports met the inclusion criteria and were included in the review (Fig. 1) [14,19-28].
PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) 2020 flow diagram of study selection.
The baseline characteristics of the 11 included RCT reports are summarized in Table 1. Because Salinas-Casado et al. [24] was a preliminary report of the final RITUMAN study published in 2020, it was retained in the qualitative description but excluded from pooled quantitative analyses to avoid double-counting participants. All included trials were RCTs comparing the clinical efficacy of D-mannose against placebo, antibiotics, no treatment, or active controls. The study populations encompassed a broad spectrum of patients, ranging from premenopausal and postmenopausal women with recurrent UTIs to kidney transplant recipients requiring prophylaxis. The publication dates of the included studies ranged from 2014 to 2024. The contributing studies for each quantitative comparison are cited separately below when reporting the corresponding pooled analyses.
Baseline characteristics of the included randomized controlled trials
2. Quality Assessment and Publication Bias
The RoB for the 11 included RCTs was evaluated using the Cochrane Risk of Bias tool (RoB 2) (Fig. 2). The overall RoB varied according to the study design. Most studies, including Kranjčec et al. [19], Porru et al. [20], and Lenger et al. [26], adequately described random sequence generation and allocation concealment, resulting in a 'low risk' assessment for the 'randomization process' domain.
Risk of bias assessment (RoB 2) of included randomized controlled trials using RoB 2 (traffic light plot; studies labeled by reference number).
However, in the domains of 'deviations from intended interventions' and 'measurement of the outcome,' a significant number of studies were rated as raising 'some concerns' or being at 'high risk.' This was primarily due to the open-label design employed in earlier trials, where the lack of blinding could potentially influence subjective symptom reporting.
In contrast, recent double-blind, placebo-controlled trials, including Hayward et al. [14] and Salvatore et al. [27], used more rigorous blinding procedures, thereby minimizing performance and detection biases. The large trial by Hayward et al. [14] was particularly influential because of its high methodological quality and null findings. The study by Rau et al. [28], involving kidney transplant recipients, showed low risk in the randomization domain, although some concern remained regarding outcome measurement (Supplementary Fig. 1).
Because no comparison included 10 or more studies, formal tests for funnel plot asymmetry were not performed. For the largest comparison group (D-mannose vs. placebo or no treatment; k=6), an exploratory funnel plot showed visual asymmetry (Supplementary Fig. 2). Smaller studies with larger standard errors tended to favor D-mannose, whereas the largest and most precise study by Hayward et al. [14] was positioned near the null effect. This pattern suggests possible small-study effects, although the limited number of studies precludes definitive conclusions regarding publication bias. For the comparisons against antibiotics (k=3) and active controls (k=2), funnel plots and formal asymmetry tests were not performed because too few studies were available.
3. Comparison 1: D-mannose Versus Placebo or No Treatment
A total of 6 RCTs (k=6, n=1,095) compared D-mannose with either placebo or no treatment [14,19,21,23, 26,27]. Because these comparator types are not clinically or methodologically equivalent, placebo-controlled and no-treatment trials were analyzed separately, as summarized in Fig. 3.
Forest plot of Comparison 1 (D-mannose vs. placebo or no treatment) for urinary tract infection recurrence or persistence (risk ratio [RR], random-effects model). Placebo-controlled trials are presented first as the primary subgroup for estimating the specific treatment effect of D-mannose, whereas no-treatment comparisons are presented as supportive analyses. The overall combined placebo/no-treatment estimate should be interpreted as exploratory because these comparator types are not clinically or methodologically equivalent. CI, confidence interval.
In the placebo-controlled analysis, 2 double-blind RCTs (k=2, n=653) were included [14,27]. D-mannose did not significantly reduce UTI recurrence or persistence compared with placebo (RR, 0.38; 95% CI, 0.06–2.36; I2=93.2%).
In contrast, 4 no-treatment comparisons (k=4, n=442) showed a statistically significant reduction in UTI recurrence or persistence with D-mannose (RR, 0.23; 95% CI, 0.08–0.66; I2=87.4%) [19,21,23,26]. However, these trials were more vulnerable to expectation effects, performance bias, behavioral changes associated with taking an intervention, and differences in follow-up intensity, particularly because several outcomes were symptom-based and several studies used open-label designs. In addition, the 95% prediction interval was wide and crossed the line of no effect (95% prediction interval, 0.002–25.42), indicating substantial uncertainty regarding the expected effect in future settings.
The combined placebo/no-treatment analysis was retained only as an exploratory pooled estimate. This estimate favored D-mannose (RR, 0.28; 95% CI, 0.12–0.66; I2=94.1%), but the 95% prediction interval was wide and crossed the line of no effect (95% prediction interval, 0.01–5.35). The test for subgroup differences did not show a statistically significant difference between placebo-controlled and no-treatment comparisons (p=0.640), although this analysis was limited by the small number of studies and substantial within-subgroup heterogeneity. To further explore the substantial heterogeneity in this comparison, L'Abbé and radial plots were generated (Supplementary Fig. 3), suggesting that variation in treatment effects was partly related to differences in baseline event rates across comparator groups.
4. Comparison 2: D-mannose Versus Antibiotics
Three RCTs (k=3, n=401) directly compared D-mannose with prophylactic antibiotics [19,20,22]. The pooled analysis showed no statistically significant difference in the risk of UTI recurrence between the D-mannose and antibiotic groups (RR, 0.43; 95% CI, 0.18–1.05; p=0.064) (Fig. 4). Although the point estimate favored D-mannose, the CI crossed the line of no effect (RR, 1.0); therefore, this finding should be interpreted as no statistically significant difference rather than formal evidence of noninferiority. Heterogeneity was substantial (I2=78%). To investigate this heterogeneity, L'Abbé and radial plots were generated (Supplementary Fig. 4), which suggested that the variance in outcomes was driven by differences in the event rates of the control arms rather than outliers.
Forest plot of Comparison 2 (D-mannose vs antibiotics) for recurrent urinary tract infection prevention (risk ratio [RR], random-effects model), including subgroup analyses by follow-up duration/indication (long-term follow-up vs. postprocedure/short follow-up). CI, confidence interval.
In the subgroup analysis based on follow-up duration and indication, similar trends were observed. Two studies contributed to the long-term prophylaxis subgroup [19,20], for which the RR was 0.39 (95% CI, 0.12–1.25). One study contributed to the postprocedure short-term prophylaxis subgroup [22], for which the RR was 0.68 (95% CI, 0.12–3.87). Neither subgroup showed a statistically significant difference.
5. Comparison 3: D-Mannose-Containing Regimens Versus PAC Alone
Two independent RCT populations (k=2, n=238) compared D-mannose-containing regimens with PAC alone [25,28]. Although Salinas-Casado et al. [24] also evaluated a PAC-based active-control comparison, it was a preliminary report of the final RITUMAN study published in 2020 [25] and was therefore excluded from the quantitative synthesis to avoid double-counting participants. This analysis included data from kidney transplant recipients from the Manotras study [28]. D-mannose-containing regimens were associated with a significantly lower risk of UTI recurrence compared with PAC alone (RR, 0.57; 95% CI, 0.40–0.82; p=0.002) (Fig. 5). Unlike other comparisons, this analysis showed zero heterogeneity (I2=0%, p=0.71), indicating a consistent treatment effect across the 2 independent study populations.
Forest plot of Comparison 3 (D-mannose-containing regimens vs. proanthocyanidins [PAC] alone) for recurrent urinary tract infection prevention (risk ratio [RR], random-effects model). Salinas-Casado et al. 2018 was excluded from the quantitative synthesis because it was a preliminary report of the final RITUMAN study. PAC, proanthocyanidins; CI, confidence interval.
6. Certainty of Evidence (GRADE Assessment)
The certainty of the evidence was assessed using the GRADE approach (Table 2). For the placebo-controlled comparison, the certainty of evidence was rated as very low because only 2 studies were available, the CI was wide, and heterogeneity was substantial. For the no-treatment comparisons, the certainty of evidence was also rated as very low because the apparent benefit was derived mainly from trials vulnerable to expectation effects, performance bias, and symptom-based outcome assessment. The combined placebo/no-treatment estimate was therefore interpreted only as exploratory and was downgraded for serious RoB, serious inconsistency, imprecision, and suspected small-study effects.
Summary of findings (GRADE evidence profile)
For the comparison against antibiotics, the certainty of evidence was rated as low because the estimate was imprecise, with wide CIs crossing the line of no effect, and heterogeneity was substantial (I2=78%). Therefore, noninferiority cannot be formally concluded.
For the comparison against PAC alone, including one study in kidney transplant recipients, the certainty of evidence was rated as moderate. The evidence was downgraded for imprecision because only 2 RCTs with a relatively small total sample size were available, but it was not downgraded for inconsistency because heterogeneity was negligible (I2=0%).
DISCUSSION
This systematic review and meta-analysis included 11 RCT reports involving 1,724 participants, of which 10 independent study populations involving 1,631 participants contributed to the quantitative analyses after excluding one overlapping preliminary report. Overall, our findings suggest that the apparent benefit of D-mannose depends strongly on the comparator used. When placebo- controlled trials were analyzed separately, D-mannose did not show a statistically significant benefit over placebo, and the estimate was imprecise with substantial heterogeneity. In contrast, significant effects were observed mainly in no-treatment comparisons, which are more vulnerable to expectation effects, behavioral changes, differential follow-up intensity, and symptom- driven outcome assessment. Therefore, the combined placebo/no-treatment estimate should be interpreted as exploratory rather than as a definitive estimate of the specific treatment effect of D-mannose. These findings are consistent with the cautious tone of the contemporary evidence syntheses: the Cochrane review led by Cooper et al. [12] emphasized that the certainty of evidence for D-mannose remains limited, largely due to small trials and methodological concerns, while Kyriakides et al. [13] similarly highlighted heterogeneity and inconsistency across studies. Importantly, by incorporating data from kidney transplant recipients, our study extends the clinical context beyond generally healthy women and suggests that D-mannose-based strategies may also be relevant in high-risk populations with limited nonantibiotic prophylactic options [15,16,28].
A key observation is the divergence between notreatment or open-label comparisons and double-blind placebo-controlled evidence. In particular, the large, high-quality trial by Hayward et al. [14] reported no statistically significant benefit of D-mannose compared with placebo, contrasting with earlier positive studies and contributing heavily to the uncertainty of pooled estimates. This distinction is clinically important because placebo-controlled trials are better suited to estimate the specific effect of D-mannose, whereas no-treatment controls may overestimate benefit through expectation effects, intervention-associated behavioral changes, and differential follow-up. This discrepancy has been noted as practice-relevant “new evidence” in the accompanying commentary by Raphael and Huang [29], who underscored that high-quality data may challenge the optimism generated by earlier studies. Differences in internal validity likely explain part of this divergence: open-label designs may amplify perceived benefits when outcomes are symptom-driven or when care-seeking behavior is influenced by expectations. Another plausible contributor is that non-specific effects related to trial participation can reduce recurrence risk regardless of the active intervention. Supporting this, Hooton et al. [30] conducted a randomized clinical trial showing that increasing daily water intake in premenopausal women with recurrent UTIs significantly reduced UTI episodes over follow-up, demonstrating that behavioral changes such as hydration can meaningfully alter recurrence risk. Therefore, if participants in pill-taking groups (including placebo arms) increase fluid intake or adopt other preventive behaviors, the measured incremental benefit of D-mannose over placebo may be diminished in blinded trials. Consistent with this, our heterogeneity exploration (including L’Abbé plots) suggests that treatment effects vary with baseline recurrence risk in the control arm, indicating that apparent benefit may be larger in settings where “no treatment” is the comparator and baseline risk is high.
Despite the uncertainty in placebo-controlled comparisons, our antibiotic comparison offers a clinically meaningful perspective. Across trials directly comparing D-mannose with prophylactic antibiotics, there was no statistically significant difference in recurrence risk, although CIs were wide and crossed unity [19,20,22]. Although the available trials did not show a statistically significant difference between D-mannose and prophylactic antibiotics, the wide CIs and low certainty of evidence preclude conclusions regarding equivalence or noninferiority. In contemporary practice, where antimicrobial stewardship is increasingly emphasized, the potential value of D-mannose lies in its role as an “antibiotic-sparing” option that may reduce antibiotic exposure and its downstream harms—particularly microbiome disruption and antibiotic-associated complications—while not directly exerting bactericidal selective pressure [7-10]. For example, Butler et al. [31] examined outpatient treatment for uncomplicated UTI and reported that antibiotic exposure is associated with clinically meaningful AEs, reinforcing the importance of minimizing unnecessary antibiotic use when safe alternatives exist. From a mechanistic standpoint, D-mannose is understood to interfere with the binding of uropathogenic E. coli via FimH-mediated adhesion rather than bacterial killing, providing a biologically plausible rationale for a lower propensity to drive resistance compared with antibiotics [11,32-34]. Classic experimental work by Connell et al. [32] demonstrated that type 1 fimbrial expression enhances E. coli virulence in the urinary tract, while Martinez et al. [33] showed that type 1 pili facilitate bacterial interaction with bladder epithelial cells, and Chen et al. [34] further supported an in vivo role for FimH during UTI beyond simple mannose binding—together reinforcing the biological plausibility of targeting this pathway. Nevertheless, clinical decision-making should remain cautious: if D-mannose is used as an alternative to antibiotics, patients with recurrent symptoms, systemic features, or risk factors for complicated infection should still be managed according to guideline-based pathways [6,35,36]. In this context, the IDSA/ESCMID (Infectious Diseases Society of America/European Society of Clinical Microbiology and Infectious Diseases) guideline update by Gupta et al. [35] remains a widely cited benchmark for the management of acute uncomplicated cystitis and pyelonephritis, and the AUA/CUA/SUFU (American Urological Association/Canadian Urological Association/Society of Urodynamics, Female Pelvic Medicine & Urogenital Reconstruction) guideline led by Anger et al. [36] provides practical recommendations for recurrent uncomplicated UTIs, including risk stratification and approaches that align with stewardship principles.
A unique contribution of this meta-analysis is the integration of evidence in kidney transplant recipients. UTIs are common after renal transplantation and are associated with immunosuppression, anatomical and functional urinary tract factors, and frequent exposure to antimicrobials [15,16]. Chuang et al. [15], in a retrospective review across transplant centers, illustrated the frequency and clinical relevance of posttransplant UTIs, and more recent reviews, including the one by Pinchera et al. [16], have emphasized that UTIs remain an “open challenge” in transplant patients due to recurrent infection risk and complex management decisions. In the Manotras trial, Rau et al. [28] evaluated a D-mannose–containing prophylactic strategy against an active control in de novo kidney transplant recipients and reported improved outcomes for UTIs and asymptomatic bacteriuria, and our pooled analysis of active-control studies demonstrated consistent effects with negligible heterogeneity. From a practical standpoint, these findings are clinically noteworthy because kidney transplant recipients often face limitations with long-term antibiotic prophylaxis due to drug interactions, AEs, and concerns regarding nephrotoxicity and resistant organisms [15,16]. Moreover, the IDSA guideline on asymptomatic bacteriuria by Nicolle and colleagues emphasizes that screening for and treating asymptomatic bacteriuria in renal transplant recipients beyond the early posttransplant period do not improve outcomes and may even cause harm through increased antibiotic exposure [37]. In this context, D-mannose emerges as a particularly compelling stewardship-informed strategy; it may offer a nonantibiotic pathway to manage recurrent symptomatic infections in this vulnerable population while aligning with the broader goal of avoiding unnecessary and potentially harmful antimicrobial therapy for bacteriuria. However, transplant-specific conclusions should still be viewed as preliminary: the available RCT evidence in this population is currently limited, and additional trials with standardized endpoints and longer follow-up are needed before firm recommendations can be made [28].
This review also highlights important methodological and clinical challenges for future research. First, heterogeneity in comparator type (placebo vs. no treatment vs. antibiotics vs. other active supplements), dosing regimens, and outcome definitions likely contributed to inconsistent effect estimates across studies [12-14,19-28]. Second, earlier trials frequently lacked blinding and sometimes relied on symptom-based recurrence without uniform microbiological confirmation, increasing the RoB and complicating cross-study comparability [12,13,19-25]. Third, small-study effects are plausible given the observed funnel plot asymmetry, where smaller studies tended to favor D-mannose while the largest, most precise trial was near the null [14]. These considerations emphasize the need for adequately powered, double-blind, placebo-controlled RCTs with harmonized endpoints (e.g., culture-confirmed recurrence, time-to-event outcomes, and standardized adverse-event reporting), alongside prespecified subgroup analyses (e.g., premenopausal vs postmenopausal women, high baseline recurrence risk, and special populations such as transplant recipients) [6,12-16,28,36]. Finally, future trials should contextualize D-mannose within the broader landscape of nonantibiotic prophylaxis. For instance, Sihra et al. [38] provided an overview of nonantibiotic prevention strategies and the challenges of variable evidence quality across interventions, while Harding et al. [39] demonstrated in a multicenter noninferiority trial that methenamine hippurate can be an effective antibiotic-sparing alternative to antibiotic prophylaxis for recurrent UTIs, illustrating that multiple stewardship-aligned preventive pathways may coexist [38,39].
CONCLUSIONS
D-mannose remains a biologically plausible and generally well-tolerated nonantibiotic strategy, but current placebo-controlled evidence does not establish its superiority over placebo. The apparent benefit was driven mainly by no-treatment comparisons, which were judged to provide very low-certainty evidence and are more susceptible to expectation effects, performance bias, and symptom-driven outcome assessment. Evidence from antibiotic comparisons was imprecise and does not allow a formal conclusion of noninferiority, while findings from PAC active-control comparisons were based on only 2 independent RCT populations and should be considered hypothesis-generating. Therefore, D-mannose cannot currently be recommended as a proven alternative to established prophylactic strategies. Further adequately powered, double-blind, placebo-controlled trials with standardized outcome definitions are needed to clarify whether D-mannose has a clinically meaningful role in selected patients.
Supplementary Materials
Supplementary Table 1 and Supplementary Figs. 1-4 are available at https://doi.org/10.14777/uti.2652010.005.
Detailed search strategy for PubMed database
Risk of bias summary plot (RoB 2; weighted bar chart).
Funnel plot for Comparison 1 (D-mannose vs. placebo or no treatment).
L'Abbé and radial (Galbraith) plots for Comparison 1.
L'Abbé and radial (Galbraith) plots for Comparison 2.
Notes
Grant/Fund Support
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Research Ethics
The study protocol was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO: CRD420261305485). Ethical review and approval were waived for this study because it is a systematic review and meta-analysis of previously published studies and does not involve direct contact with human subjects or the use of identifiable private information, in accordance with the Bioethics and Safety Act of the Republic of Korea.
Conflict of Interest
The authors have nothing to disclose.
Author Contribution
Data curation: JYJ, YJM, JYL; Formal analysis: JYL; Funding acquisition: Not applicable-Methodology: JYJ, YJM, JYL; Project administration: HDJ, LK, JYL; Visualization: JYL; Writing - original draft: JYJ, JYL; Writing - review & editing: JYJ, YJM, DHK, HDJ, LK, JYL.