2023 Volume 70 Issue 11 Pages 1103-1107
Sodium-glucose cotransporter-2 (SGLT2) inhibitors are a novel class of antidiabetic drugs. Guidelines for the proper use of SGLT2 inhibitors recommend caution regarding urinary tract infections (UTIs). However, little evidence has been reported on the relationship between SGLT2 inhibitors and UTIs in large epidemiological studies. We investigated (1) the relationship between diabetes mellitus (DM) and UTIs and (2) the relationship between SGLT2 inhibitor prescriptions and the likelihood of developing UTIs in patients with DM, using a nationwide Japanese health insurance claims database by MDV analyzer®. We found that the incidence of UTIs was significantly higher among patients with DM than among those without DM (odds ratio (OR), 1.71; 95% confidence interval (CI), 1.69–1.72, for male; OR, 1.90; 95% CI, 1.89–1.92 for female). In contrast, in male patients with DM, the prescription of SGLT2 inhibitors was negatively associated with the likelihood of developing UTIs (OR, 0.74; 95% CI, 0.72–0.75). Among female patients with DM, there was no significant difference in the incidence of UTIs with or without an SGLT2 inhibitor prescription (OR, 0.99; 95% CI, 0.96–1.01). Subgroup analyses by age confirmed similar relationships between SGLT2 inhibitor prescriptions and UTIs, except for female patients aged ≤39 years, in whom SGLT2 inhibitor prescription was negatively associated with the likelihood of developing UTIs. In conclusion, our analysis of a nationwide claims database found no evidence that SGLT2 inhibitors increase UTIs in Japanese patients with DM, regardless of sex or age.
SODIUM-GLUCOSE COTRANSPORTER-2 (SGLT2) inhibitors are a novel class of antidiabetic drugs that have recently been widely used to treat chronic heart failure and chronic kidney disease. We have previously reported that SGLT2 inhibitors prevent renal stone formation via their anti-inflammatory and anti-fibrotic effects in epidemiological, in vivo, and in vitro studies [1, 2]. In 2015, the United States (US) Food and Drug Administration announced a safety warning for physicians and patients based on 19 cases of sepsis due to urinary tract infections (UTIs) receiving SGLT2 inhibitors and updated the drug label for SGLT2 inhibitors [3]. Guidelines for the proper use of SGLT2 inhibitors also recommend caution notes stating UTIs as adverse events for patients [4]. Diabetes mellitus (DM) is a known risk factor for UTIs [5]. SGLT2 inhibitors suppress glucose reabsorption in the renal proximal tubules, resulting in glucosuria, which could theoretically cause bacterial infections and UTIs [6]. A meta-analysis from 2017 comprising 52 randomized controlled trials involving 36,689 patients reported that dapagliflozin was associated with a higher risk of UTIs than placebo [7]. Conversely, a previous population-based cohort study in the US involving 123,752 patients, including a subgroup analysis focused on dapagliflozin, found that SGLT2 inhibitor use was not associated with an increased risk of severe or non-severe UTIs [8]. The results from recent studies are consistent with these findings [9, 10].
Etiology and risk of UTIs differ by country and region [11, 12]. Furthermore, the prescribing patterns of SGLT2 inhibitors vary by region and time [13]. Therefore, evidence from other countries may not be applicable to Japan, and the latest real-world evidence in Japan is of great importance to Japanese physicians, patients, and policymakers. However, there is little evidence reported on the relationship between SGLT2 inhibitors and UTIs from large epidemiologic studies in Japan. Thus, we investigated the relationship between SGLT2 inhibitors and UTIs in patients with DM using data from a nationwide Japanese health insurance claims database. We explored the relationship between DM and the likelihood of developing UTIs and examined the relationship between SGLT2 inhibitors and UTIs in patients with DM.
In this cross-sectional study, we used the Japanese administrative database maintained by the MDV analyzer® (Medical Data Vision Co., Ltd., Tokyo, Japan). The MDV analyzer® maintains a de-identified hospital administrative database containing patients’ medical information from Japanese acute diagnosis procedure combination (DPC) hospitals [14]. The database consists of inpatient and outpatient medical care data from 480 acute DPC hospitals in different regions of Japan, covering approximately 27% of all DPC hospitals in Japan. As of May 31, 2023, the cumulative number of patients using the MDV analyzer® was 43.8 million (including data on patients who had already died). Furthermore, the MDV analyzer® contains patient-level information on demographic characteristics, diagnoses coded according to the International Classification of Diseases, 10th revision (ICD-10), and prescription information. Sex and age distributions in the source population were similar to those in the Japanese census, and several researchers have used this database [15-17].
Data on the study population (aged ≥20 years) were collected between January 1, 2022, and December 31, 2022. Patients with ICD-10 codes E10-14 and N10, 30, 34, 41, 45, 76, 151, and 390 were defined as having DM and UTI, respectively (Table 1). The target SGLT2 inhibitors, including canagliflozin, dapagliflozin, empagliflozin, ipragliflozin, luseogliflozin, and tofogliflozin, have been oral SGLT2 inhibitors approved for use in Japan since December 2020. SGLT2 inhibitors formulated in combination tablets were also considered target oral SGLT2 inhibitors for this analysis. The targeted oral SGLT2 inhibitors are listed in Table 2.
International Classification of Diseases, 10th revision codes for diabetes mellitus and urinary tract infection
| International Classification of Diseases, 10th revision codes | |
|---|---|
| Diabetes mellitus | E10 (Type 1 diabetes mellitus) |
| E11 (Type 2 diabetes mellitus) | |
| E12 (Malnutrition-related diabetes mellitus) | |
| E13 (Other specified diabetes mellitus) | |
| E14 (Unspecified diabetes mellitus) | |
| Urinary tract infection | N10 (Acute pyelonephritis) |
| N30 (Cystitis) | |
| N34 (Urethritis and urethral syndrome) | |
| N41 (Inflammatory diseases of prostate) | |
| N45 (Orchitis and epididymitis) | |
| N73 (Other female pelvic inflammatory diseases) | |
| N151 (Renal and perirenal abscess) | |
| N390 (Urinary tract infection) |
The list of oral SGLT2 inhibitors
| SGLT2 inhibitors | Canagliflozin |
| Canagliflozin/Teneligliptin | |
| Dapagliflozin | |
| Empagliflozin | |
| Empagliflozin/Linagliptin | |
| Ipragliflozin | |
| Ipragliflozin/Sitagliptin | |
| Luseogliflozin | |
| Tofogliflozin |
Odds ratio (OR) and 95% confidence interval (CI) for the likelihood of developing UTIs were estimated using logistic regression models. Analyses were stratified according to sex. Among patients with DM, we performed a subgroup analysis by age to determine whether there were differences in the observed relationships between age groups. Statistical significance was set at p < 0.05. The JMP statistical software version 12.2 (SAS Institute, Cary, NC, USA) was used for statistical analysis.
The study was approved by the Institutional Review Boards of Tohoku Medical and Pharmaceutical University Hospital (2020-2-253). The requirement for informed consent was waived because of the retrospective nature of the epidemiological study, and all personal and identifying data were anonymized.
A total of 4,675,035 males and 5,404,110 females were included in this study. Logistic regression analysis revealed that patients with DM had significantly more UTIs than those without DM (OR, 1.71; 95% CI, 1.69–1.72, for male; OR, 1.90; 95% CI, 1.89–1.92 for female) (Table 3).
The relationship between diabetes and urinary tract infections
| Male (n = 4,675,035) | N | Urinary tract infection (+), n (%) | Urinary tract infection (–), n (%) | OR | 95% CI | p value |
| Diabetes (+) | 978,003 | 68,255 (6.98) | 909,748 (93.02) | 1.71 | 1.69–1.72 | <0.0001 |
| Diabetes (–) | 3,697,032 | 155,760 (4.21) | 3,541,272 (95.79) | Ref | ||
| Female (n = 5,404,110) | N | Urinary tract infection (+), n (%) | Urinary tract infection (–), n (%) | OR | 95% CI | p value |
| Diabetes (+) | 685,353 | 61,392 (8.96) | 623,961 (91.04) | 1.90 | 1.89–1.92 | <0.0001 |
| Diabetes (–) | 4,718,757 | 231,915 (4.91) | 4,486,842 (95.09) | Ref |
CI, confidence interval; OR, odds ratio
Of the 978,003 males and 685,353 females with DM, 181,366 and 87,520, respectively, were prescribed SGLT2 inhibitors. Among males with DM, SGLT2 inhibitor prescription was negatively associated with the likelihood of developing UTIs (OR, 0.74; 95% CI, 0.72–0.75) (Table 4). In contrast, in females with DM, no significant difference was observed between the incidence of UTIs with or without SGLT2 inhibitor prescriptions (OR, 0.99; 95% CI, 0.96–1.01) (Table 4).
The relationship between SGLT2 inhibitor and urinary tract infections in patients with diabetes mellitus
| Male (n = 978,003) | N | Urinary tract infection (+), n (%) | Urinary tract infection (–), n (%) | OR | 95% CI | p value |
| SGLT2 inhibitor (+) | 181,366 | 9,955 (5.49) | 171,411 (94.51) | 0.74 | 0.72–0.75 | <0.0001 |
| SGLT2 inhibitor (–) | 796,637 | 58,300 (7.32) | 738,337 (92.68) | Ref | ||
| Female (n = 685,353) | N | Urinary tract infection (+), n (%) | Urinary tract infection (–), n (%) | OR | 95% CI | p value |
| SGLT2 inhibitor (+) | 87,520 | 7,766 (8.87) | 79,754 (91.13) | 0.99 | 0.96–1.01 | 0.3497 |
| SGLT2 inhibitor (–) | 597,833 | 53,626 (8.97) | 544,207 (91.04) | Ref |
CI, confidence interval; OR, odds ratio
Subgroup analyses by age confirmed similar relationships between SGLT2 inhibitor prescriptions and UTIs, except for female patients aged ≤39 years, in whom the prescription of SGLT2 inhibitors was negatively associated with the likelihood of developing UTIs (OR, 0.73; 95% CI, 0.62–0.85), as was observed in male patients (Table 5).
Subgroup analysis by age on the relationship between SGLT2 inhibitors and urinary tract infections in patients with diabetes mellitus
| Male (n = 978,003) | N | Urinary tract infection (+), n (%) | OR | 95% CI | p value | |
| Male (age: 20–39) | SGLT2 inhibitor (+) | 5,480 | 100 (1.82) | 0.44 | 0.36–0.55 | <0.0001 |
| SGLT2 inhibitor (–) | 16,454 | 663 (4.03) | Ref | |||
| Male (age: 40–59) | SGLT2 inhibitor (+) | 47,841 | 1,192 (2.49) | 0.65 | 0.61–0.70 | <0.0001 |
| SGLT2 inhibitor (–) | 123,472 | 4,657 (3.77) | Ref | |||
| Male (age: 60–) | SGLT2 inhibitor (+) | 128,018 | 8,663 (6.77) | 0.83 | 0.81–0.85 | <0.0001 |
| SGLT2 inhibitor (–) | 656,738 | 53,010 (8.07) | Ref | |||
| Female (n = 685,353) | N | Urinary tract infection (+), n (%) | OR | 95% CI | p value | |
| Female (age: 20–39) | SGLT2 inhibitor (+) | 2,955 | 181 (6.13) | 0.73 | 0.62–0.85 | <0.0001 |
| SGLT2 inhibitor (–) | 17,754 | 1,464 (8.25) | Ref | |||
| Female (age: 40–59) | SGLT2 inhibitor (+) | 19,220 | 1,472 (7.66) | 0.99 | 0.93–1.05 | 0.7521 |
| SGLT2 inhibitor (–) | 82,978 | 6,411 (7.73) | Ref | |||
| Female (age: 60–) | SGLT2 inhibitor (+) | 65,345 | 6,113 (9.35) | 1.02 | 0.99–1.05 | 0.2086 |
| SGLT2 inhibitor (–) | 497,101 | 45,751 (9.20) | Ref |
CI, confidence interval; OR, odds ratio
We examined the association between SGLT2 inhibitors and UTIs in patients with DM using nationwide Japanese health insurance claims data, investigating both the relationship between DM and UTI development and the impact of SGLT2 inhibitors on UTI occurrence in this population. Our study revealed that patients with DM had a significantly higher incidence of UTIs than those without DM. A US database study found that patients with type 2 DM had a significantly higher incidence of UTIs than those without DM (9.4 vs. 5.7%; OR, 1.54) [5]. The same trend was observed in males and females and in all age groups. The frequency of UTIs was higher in females than in males in all age groups, indicating that the risk of UTIs was highest in females with DM [5]. These results are consistent with those of the present study.
Furthermore, our study found no evidence that SGLT2 inhibitors increased UTIs in Japanese patients with DM, regardless of sex or age. In particular, males and young females with DM who were prescribed SGLT2 inhibitors had a significantly lower incidence of UTIs than those not prescribed SGLT2 inhibitors. Interestingly, recent population-based cohort studies in the US, United Kingdom, and Canada have also reported a decreased risk of UTIs with SGLT2 inhibitors vs. glucagon-like peptide 1 receptor agonists or insulin [8, 10]. These results are consistent with those of the present study. One possible reason is that SGLT2 inhibitors may have been prescribed to appropriate patients and avoided in patients prone to recurrent UTIs or in immunologically vulnerable patients. Another possible reason is that although the glycosuric effects of SGLT2 inhibitors create an environment that supports bacterial growth, these effects are offset by the improvement in urinary flow through osmotic diuresis [18]. Further studies are required to elucidate this mechanism.
This study has several limitations. First, we could not examine laboratory data such as HbA1c levels, urine culture results, and comorbidities in detail using our database. Second, as this was a cross-sectional study, a causal relationship between SGLT2 inhibitors and UTIs could not be established. Finally, although indications for SGLT2 inhibitors have been extended to chronic heart failure and chronic kidney disease, our findings may not be generalizable to patients without DM. Therefore, further large prospective longitudinal studies that include patients with chronic heart failure or chronic kidney disease but not DM, and those with DM, are required.
In conclusion, this cross-sectional study using a nationwide Japanese health insurance claims database found no evidence that SGLT2 inhibitors increased UTIs in patients with diabetes, regardless of sex or age.
CI, confidence interval; DM, diabetes mellitus; DPC, diagnosis procedure combination; OR, odds ratio; SGLT2, Sodium-glucose cotransporter-2; UTI, urinary tract infection; US, United States
This work was supported by the Japan Society for the Promotion of Science (No. 22K16822). We would like to thank Editage (www.editage.com) for English language editing.
None of the authors have any potential conflicts of interest associated with this research.
