Reference
Hayward G, Mort S, Hay AD, et al. D-Mannose for prevention of recurrent urinary tract infection among women: a randomized clinical trial. JAMA Intern Med. 2024;184(6):619-628.
Study Objective
To determine whether D-mannose taken daily for 6 months reduces incidence of recurrent urinary tract infections (rUTIs) in women
Key Takeaway
D-mannose did not significantly reduce recurrent UTIs in women compared to a placebo. While previous smaller studies suggested benefits, the latest evidence calls into question its effectiveness, highlighting the need for further research and consideration of alternative preventive strategies, such as cranberry extracts.
Design
Double-blinded, randomized, placebo-controlled trial conducted in 99 primary care centers in the United Kingdom
Participants
From 598 total participants (mean age 58 years), investigators randomized 303 women to D-mannose (50.7%) and 295 to placebo (49.3%).
Intervention
Two grams daily of D-mannose powder or a matched volume of fructose powder, taken as a scoop once daily. Both are white powders and taste sweet.
Primary Outcome
The primary outcome was the proportion of women experiencing at least 1 further episode of clinically suspected UTI for which they contacted ambulatory care within 6 months of randomization, determined through review of primary care records. In primary care, practitioners don’t generally rely on microbiological confirmation for guiding initial antibiotic prescribing, and in up to 1/3 of cases, it will be inconclusive.
Secondary outcomes included number of days of moderately bad (or worse) symptoms of UTI recorded in symptom diaries, time to next consultation with a clinically suspected UTI, number of clinically suspected UTIs, number of microbiologically proven UTIs, number and consumption of antibiotic courses for UTI, defined daily dose and total milligram by antibiotic type, proportion of women with resistant uropathogens cultured during an episode of acute infection, and hospital admissions related to UTI.
Key Findings
Outcome data were available for 583 participants (97.5% of the group). Of the 294 women taking D-mannose, 150 contacted their care providers about a suspected UTI (51.0%). In the placebo group, 161 of 289 women reported a clinically suspected UTI (55.7%). The difference in UTI occurrence risk between the D-mannose and placebo groups was −5% (95% CI, −13% to 3%; P=0.26).
There was no statistical difference in risk of UTI between the 2 groups. There were also no differences in the number of clinically suspected UTIs or microbiologically proven UTIs or the other secondary outcomes tracked.
Transparency
The study authors did not report any conflict of interest.
Discussion and Clinical Implications
Urinary tract infections are the most common bacterial infections that bring women in to see primary care providers. About half of all women will have a UTI at some point in their lives. Some women suffer from recurrent UTIs, defined as 2 UTIs in 5 months or 3 in a year. It is estimated that more than 400 million UTIs occur each year worldwide, and more than 200,000 of these are fatal.1
Many women with recurrent UTIs are treated with daily antibiotics, which are effective if the drugs are used regularly, but extended use leads to antibiotic-resistant UTIs.2,3
It behooves us to be able to offer our patients nonantibiotic treatments that do not lead to resistance or cause adverse side effects. For many years, D-mannose has been considered an alternative to antibiotics.
This study by Gail Hayward et al, which concludes that D-mannose is not effective and should no longer be used to treat chronic UTIs, deserves our close attention.
Naturopathic doctors have prescribed D-mannose for decades, believing that it is an effective prophylactic against UTIs. It is hard to equate the findings reported in this new study with our longstanding experience. The contrast between our assumptions and the results reported suggests the need for us to critically examine both the initial evidence and ongoing research that supports our assumption.
The earliest report that suggests D-mannose might be of use in treating urinary tract infections appears to be a 1978 paper written by 2 Swedish scientists, C Svanborg Edén and Hans-Arne Hansson, in their report, “Presence of pili of fimbriae on Escherichia coli bacteria isolated from the urine of patients with urinary tract infection was related to the ability of the bacteria to attach to human uroepithelial cell.” D-mannose was 1 of 2 compounds they identified that blocked this agglutination.4 This may well have been the paper that led us to the idea that D-mannose prevented E coli bacteria from being able to anchor onto the bladder wall. Their work was in vitro, using bacterial cultures.
Over the following years, studies examined the effect of either D-mannose in combination with cranberry extracts or cranberry alone. D-mannose is a monosaccharide found in some plants and fruits, and while numerous websites suggest that D-mannose is especially high in cranberries, no citation supporting this has yet surfaced. In fact, cranberries contain negligible amounts of D-mannose (0.04%–0.14% of dry weight).5 With fresh cranberries being about 88% water, the quantity of fresh berries that could provide an equivalent dose of D-mannose as used in these various studies would be impossible to consume.6
The assumption has been that D-mannose prevents UTIs because of its potential “to saturate the E. coli FimH structures and subsequently block E. coli adhesion to the urinary tract epithelial cells so that the bacteria are washed out by the urine.” This competitive inhibition has been considered the main mechanism for UTI prevention.7
Although small amounts of D-mannose are present in some foods, the D-mannose used in supplements is synthesized, usually starting with sugar or mannitol; it is not isolated from natural substances.8
Even if D-mannose is a simple sugar, it isn’t metabolized by humans. While rapidly absorbed, it is excreted just as rapidly, unchanged. At least 90% is absorbed in the small intestine and then quickly excreted from the blood via the kidneys. Plasma half-times may be as short as 30 minutes, and the majority is excreted by 1 hour. There has been interesting debate as to whether prevention of E coli infection is because of pharmacologic action or mechanistic action—in other words, whether D-mannose should be classified as a medical agent or medical device. The current opinion is that “healthcare products containing D-mannose indicated for UTI prophylaxis should be classified as a medical device composed of substances.”9 As interesting a debate as that might be, our concern is now simpler: namely, whether D-mannose has clinical use—that is, whether it works.
In looking back at the clinical evidence on D-mannose, we can see that it has never been definitively proven to work; in fact, the evidence in favor turns out to be rather thin. Let’s consider the relevant clinical trials.
In 2013, KranjĨec et al reported on a randomized, controlled trial of D-mannose for recurrent UTI prevention. Women who had received antibiotics to treat an acute UTI (N=308) were divided into 3 groups that received either 2 grams per day of D-mannose, 50 mg nitrofurantoin daily, or nothing. Recurrence rates were 14.6% in the D-mannose group, 20.4% in the nitrofurantoin group, and 60.8% in the group without any treatment. D-mannose use was significantly associated with a lower risk of UTI recurrence, comparable to that in the nitrofurantoin group.
In 2014 Porru et al reported results from a randomized crossover trial that included 60 women with acute, symptomatic UTI and 3 or more recurrent UTIs during the preceding 12 months. Patients were randomly assigned to antibiotic treatment with trimethoprim/sulfamethoxazole (TMP-SMZ: 2×160 mg/800 mg daily for 5 days) or therapy with D-mannose (3×1 g daily for 2 weeks), followed by prophylactic therapy over 5 months with TMP-SMZ (1×160 mg/800 mg daily for 1 week each month) or D-mannose (1×1 g daily). In the group treated with D-mannose, a significantly increased mean time to UTI recurrence (200 days) was observed in comparison to the antibiotic group (52.7 days; P<0.0001).10 These results certainly would suggest D-mannose is superior to antibiotics.
In a prospective, noncomparative study (no control group) published in 2016, Domenici et al looked at D-mannose in the treatment of acute UTI and in recurrences in 43 women.11 Significant improvement of most UTI symptoms was seen following treatment with 1.5 grams D-mannose, twice daily for 3 days and then once a day for 10 days. Patients were then randomized into 2 groups, 1 receiving D-mannose once a day for a week every other month for 6 months and the other without D-mannose. The recurrence rate for women receiving treatment (4.5%) was significantly lower than in the untreated group (33.3%; P=0.05). The mean time to onset of UTI was 43 days (±4.1 standard deviations [SD]) in the prophylaxis group and 28 days (±5.4 SD) in the other group (P=0.0001).
A 2016 open-label prospective feasibility study by Phe et al evaluated D-mannose for the prevention of UTIs in 22 patients with multiple sclerosis. D-mannose was associated with a significant decrease in the number of monthly UTIs.12
In 2020, a systematic review and meta-analysis by Lenger et al was published on the efficacy of D-mannose to reduce UTI recurrence in adult women with a history of recurrent UTIs. It makes D-mannose look pretty good. Comparison of D-mannose with placebo resulted in a pooled relative risk for rUTIs of 0.23 (95% CI, 0.14–0.37; heterogeneity=0%; D-mannose: n=125, placebo: n=123). In addition, the pooled relative risk of rUTI comparing D-mannose with preventive antibiotics was 0.39 (95% CI 0.12–1.25; heterogeneity=88%; D-mannose: n=163, antibiotics: n=163). These data suggest that D-mannose is about as effective as antibiotics.13
I recall hearing at one point that D-mannose was the active agent in cranberries that prevented E coli adherence and that this was why cranberries were an effective intervention against UTIs. Further investigation has made me realize that this claim is not accurate.14 Cranberries also appear to block E coli adhesion, but it can’t be because they contain a lot of D-mannose.15 Both cranberries and D-mannose independently appear to block E coli adhesion, though cranberries do so in a more “robust” manner.
There are a handful of small, positive studies, but should we believe them? In reading the abstracts, one can’t judge how well the studies were done We’ve seen so many proven ideas during our careers lose their footing and fall from favor.
In 2022, in a Cochrane systematic review, Cooper et al concluded that there was still insufficient evidence to support D-mannose for UTI prophylaxis. Their analysis summarized 7 RCTs (N=719). They found that “Overall, the quality of the evidence is poor. Most studies were judged to have unclear or high risk of bias across most domains. Data was sparse and addressed very few outcomes. The GRADE evaluation was rated as very low certainty evidence due to very serious limitations in the study design or execution (high risk of bias across all studies) and sparse data (single study data and small sample sizes).”16
We are left trying to guess at the relative trustworthiness of small, early studies as we attempt to weigh them against this recent large, multicenter Hayward study and a Cochrane review. Let’s back up and relook at what we know about D-mannose.
Websites often suggest that cranberries are the main dietary source of D-mannose. Well, as we’ve established, that’s not true. But still, cranberries have long been considered useful for treating UTIs, and some of the enthusiasm for D-mannose stemmed from the assumption that these sugars might be the active ingredient in cranberries. A compelling backstory like that makes a good selling point. Cranberries, like other natural products, are complex foods and are useful for UTIs for multiple reasons. Over the last 27 years, numerous trials have used cranberries alone to treat UTIs.17-28(That’s only a partial list of citations.) Cranberries have also been trialed in combination with other nutritional supplements such as probiotics,29 D-mannose,30,31 and other supplements.32
A systematic review and meta-analysis published in July 2024 looked at various cranberry products for UTI prevention. Twenty studies were included in the analysis (N=3,091). Of these, 18 studies reported a 54% lower rate of UTIs with cranberry juice compared to no treatment and a 27% reduction in comparison to placebo. In other words, there is ample and decent evidence in support of cranberries being effective.33
The evidence in support of D-mannose has been relatively weak in comparison. With the publication of this new trial by Hayward et al, the support has grown even weaker for D-mannose. We can no longer justify using D-mannose as the sole agent for prophylaxis of UTIs. We may only be able to justify its use in combination with cranberry extracts and possibly with probiotics.
Some of the discrepancies seen between the various D-mannose studies may be in how it was dosed or how patients took it. Recall that D-mannose is rapidly absorbed and rapidly eliminated. It may need to be present in the bladder throughout the day to be effective. A once-a-day dose, no matter the quantity, may not be adequate, and it may be that divided doses spread across the day could still prove to be effective. If D-mannose passes through the body in an hour, perhaps frequent doses would be more reliable.
Might there be a synergistic effect when we use cranberry in combination with D-mannose? They both interfere with E coli binding to the bladder. Possibly alternating doses of D-mannose and cranberry might provide greater antiadhesion coverage?
A once-a-day dose, no matter the quantity, may not be adequate, and it may be that divided doses spread across the day could still prove to be effective.
The initial theory from way back as to why cranberry offered protection against UTIs was that it acidified the urine and, for this reason, had a bacteriostatic effect. By the mid-1980s, this acidification theory was superseded by the current model of bacterial anti-adhesion activity. In 1984 Sobota disproved the acid urine theory and developed a bioassay that quantified “anti-adhesion activity (AAA)” of the different fractions of cranberries.34 Since then, Sobota’s AAA test has become the standard for evaluating cranberry extracts and comparing them to other agents that target E coli adhesion. More recent publications identify cranberry proanthocyanidins and flavan-3-ol oligomers as responsible for cranberry’s antiadhesion properties.35,36
A 2024 paper by Amy Howell et al reports on the relative impacts of D-mannose and cranberry extracts using Sobota’s AAA test. Her study used a crossover design to determine ex vivo AAA against P-type and type-1 fimbriated uropathogenic Escherichia coli in patients taking either 3 500-mg capsules (1,500 mg) of D-mannose (NOW) or a cranberry-fruit-juice dry extract containing 36 mg of soluble proanthocyanidins (1 265-mg cranberry juice extract [Ellura®]), using bioassays that measure urinary activity following consumption. The cranberry supplement exhibited both P type and type 1 in vitro and ex vivo AAA, while D-mannose only prevented type 1 adhesion. Cranberry also “demonstrated more robust and consistent ex vivo urinary AAA than D-mannose.”37
Perhaps it would be helpful to consider our knowledge of D-mannose in light of the hierarchies of evidence-based medicine (EBM). It is generally accepted that there is a hierarchy of evidence that runs from weakest to strongest evidence. Most EBM systems place randomized, controlled trials at a higher level than case series or expert opinions, which are relegated to the lowest level of a pyramid of trustworthiness. RCTs are at a higher level because they are designed to be unbiased and have less risk of systematic errors. Most also agree that current, well-designed systematic reviews and meta-analyses should be at the top of the evidence pyramid.
The Evidence Hierarchy
Figure adapted from: Greenhalgh T. How to read a paper: the basics of evidence-based medicine. Hoboken: John Wiley & Sons, Incorporated; 2010. Figure 2.1 A simple hierarchy of evidence for assessing the quality of trial design in therapy studies. https://libguides.mssm.edu/ebm/hierarchy
Cooper et al’s 2022 systematic review of D-mannose concluded, “There is currently little to no evidence to support or refute the use of D-mannose to prevent or treat UTIs in all populations. This review highlights the severe lack of high-quality RCTs testing the efficacy of D-mannose for UTIs in any population.” Now, in the study reviewed here, Hayward et al conclude their report on their RCT: “daily D-mannose did not reduce the proportion of women with rUTI in primary care who experienced a subsequent clinically suspected UTI. Daily D-mannose should not be recommended to prevent future episodes of clinically suspected UTI in women with rUTI in primary care” (bold font added).
Our profession’s members are unlikely to immediately change prescribing habits. Nor will our patients, who are certain that D-mannose is an effective treatment, stop using it. Many of us will continue to prescribe D-mannose against recurrent UTIs, and our patients will continue to self-prescribe it. Until the knowledge that D-mannose is less effective than we thought becomes widespread, we should at least strongly encourage our patients to take some form of cranberry extract along with D-mannose, as the current evidence says cranberry does work. To quote the 2021 systematic review and meta-analysis by Liu et al, “evidence from our updated meta-analysis indicated that cranberry supplementation significantly reduced the incidence of occurring UTIs in susceptible populations. Furthermore, cranberry may be considered as a promising adjuvant therapy for preventing UTIs in susceptible individuals.”38