Abstract
Background
Numerous studies demonstrated the risk factors for urological complications in patients with diabetes before sodium–glucose co-transporter 2 inhibitor (SGLT2i) became commercially available. This study aimed to comprehensively investigate urological characteristics in patients with type 2 diabetes (T2DM) after SGLT2i became commercially available.
Methods
We examined 63 outpatients with T2DM suspected of bacteriuria based on urinary sediment examinations. Urine cultures were performed, and lower urinary tract symptoms (LUTS) were assessed via questionnaires. Patients with bacteriuria were assessed using ultrasonography to measure post-void residual volume (PVR). Utilizing demographic and laboratory data, a random forest algorithm predicted LUTS, bacteriuria, and symptomatic bacteriuria (SB).
Results
Thirty-two patients had LUTS and 31 had bacteriuria. High-density lipoprotein cholesterol level was crucial in predicting LUTS, while age was crucial in predicting bacteriuria. In predicting SB among patients with bacteriuria, creatinine level and estimated glomerular filtration rate were crucial. Our models had high predictive accuracy for LUTS (area under the curve [AUC] = 0.846), followed by bacteriuria (AUC = 0.770) and SB (AUC = 0.938) in receiver operating characteristic curve analysis. These predictors were previously reported as risk factors for urological complications. Although SGLT2i use was not an important predictor in our study, all SGLT2i users with bacteriuria had SB and exhibited higher PVR compared to non-SGLT2i users with bacteriuria.
Conclusion
This study’s random forest model highlighted distinct essential predictors for each urological condition. The predictors were consistent before and after SGLT2i became commercially available.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13340-023-00687-1.
Keywords: Diabetes, Bacteriuria, LUTS, SGLT2i
Introduction
Diabetes is associated with various urological complications. The prevalence of lower urinary tract complications in patients with diabetes is higher than that of other major complications, such as neuropathies and nephropathies [1]. Several studies have shown that urological complications have a negative effect on the quality of life of patients with diabetes [2–4]. Lower urinary tract symptoms (LUTS) are a common urological complication in patients with diabetes [1, 5], and is associated with decreased quality of life [4], cardiovascular disease [6], and mortality [7, 8]. Residual urine was more frequently observed in women with diabetes than in women without diabetes [9]. Another observational study showed that high post-void residual volume (PVR) increased the risk of LUTS [10]. Similarly, bacteriuria is known to be associated with LUTS [11], and to affect bladder contractility and voiding efficiency [12].
In patients with diabetes, diabetes treatment-associated urological complications should be considered. A high prevalence of LUTS, including pollakiuria and nocturia, was reported in patients being treated with sodium-glucose cotransporter 2 inhibitors (SGLT2i) [13]. Nevertheless, few studies have focused on the urological characteristics of patients with diabetes after SGLT2i became commercially available. We hence investigated the urological characteristics of patients with type 2 diabetes (T2DM) using the LUTS score and PVR quantified by abdominal ultrasonography, and also analyzed the relationship between SGLT2i treatment and urological characteristics.
Materials and methods
Study design
This cross-sectional study was conducted at Toda Chuo General Hospital in Japan from September 2017 to January 2018. The study design is shown in Fig. 1. Data of urine cultures and a questionnaire on LUTS were obtained from outpatients with type 2 diabetes suspected of having bacteriuria (urinary sediment examination of ≥ 104 colony-forming units [cfu]/mL). In addition, patients with positive urine cultures were assessed using abdominal ultrasonography to measure PVR. Urinary sediment examination was repeated on the day of the ultrasonography to confirm the persistence of bacteriuria. Prostate volumes were evaluated in men. Information regarding demographic data was obtained using questionnaires. Information on bidet toilet use was also collected as it is widely used in Japan [14].
Fig. 1.
Study design. Patients suspected of having bacteriuria on urinary sediment examination were assessed using urine cultures and questionnaires about their lower urinary tract symptoms (LUTS), including the International Prostate Symptom Score (IPSS), Overactive Bladder Symptom Score (OABSS), and International Consultation on Incontinence Questionnaire Short Form (ICIQ-SF). Patients with positive urine cultures were considered to have bacteriuria and were classified as either symptomatic or asymptomatic based on the LUTS score. Additionally, abdominal ultrasonography was performed to measure post-void residual urine volumes.
The exclusion criteria included patients who were pregnant or who had a previous diagnosis of urinary tract abnormalities. This study was approved by the Ethics Committee of Toda Chuo General Hospital (study approval no.: 0496). All the participants provided informed consent before their inclusion in the study.
Measurement and definitions
Bacteriuria was defined as the presence of at least 105 cfu/mL of 1 or 2 bacterial species in clean-voided midstream urine cultures [15, 16]. LUTS were assessed using the International Prostate Symptom Score (IPSS), Overactive Bladder Symptom Score (OABSS) [17], and International Consultation on Incontinence Questionnaire Short Form (ICIQ-SF) [18]. We used the IPSS to evaluate LUTS in both men and women. Although the IPSS was developed to assess benign prostate hyperplasia, it is also known to be applicable for assessing LUTS in both men and women [19]. IPSS consists of 7 questions regarding feeling of incomplete emptying, pollakisuria, intermittency, urgency, slow stream, straining to void, and nocturia. IPSS-QoL measures the quality life components. OABSS consists of the following 4 components: daytime frequency score, nocturia score, urgency score, and urge incontinence score. An overactive bladder (OAB) was defined as 2 or more points on the urgency score, and a total score of 3 or more points. ICIQ-SF consists of 4 components evaluating the frequency, severity, impact on quality of life of urinary incontinence and a self-diagnostic item concerning urine leakage. Patients with LUTS were defined as those diagnosed as having an OAB or moderate-to-severe score on the IPSS (IPSS ≥ 8), and patients with symptomatic bacteriuria (SB) were defined as patients with both bacteriuria and LUTS. Asymptomatic bacteriuria (ASB) was defined as having bacteriuria without LUTS. Based on the clinical guidelines for female Lower Urinary Tract Symptoms [20], the PVR was considered significant when ultrasonography findings demonstrated a volume larger than 100 mL.
Machine learning using random forest
Random forest analysis was performed using R package “randomForest” in R version 4.2.2 software. The number of trees was set at 500. Optimal mtry was determined using the tuneRF function in the package. A total of 63 patients who tested positive in their urine culture and questionnaire were included in the analysis of predicting LUTS and bacteriuria. For the prediction of SB among patients with bacteriuria, 31 patients with bacteriuria were included. Two datasets were randomly split into a training data set (80%) and a test data set (20%) to assess the accuracy of these models. The median decrease in the Gini (MDG) index was used to assess the importance of a feature in the models.
Statistical analysis
Statistical analyses were performed using GraphPad Prism version 7. The distribution of the data was validated by the D’Agostino-Pearson normality test. For the continuous variables that passed the normality test, differences between groups were assessed by the 2-tailed Student t-test. Correlations of the laboratory variables were calculated using the Pearson’s correlation test. For the continuous variables that did not pass the normality test, the Mann–Whitney’s test was performed. Correlations of the laboratory variables were calculated using the Spearman’s correlation test. The Fisher’s exact test was used for the categorical variables. A P-value of less than 0.05 was considered to indicate a statistically significant difference between groups.
Results
Sixty-three patients suspected of having bacteriuria with urinary sediments were included in the analysis. The patients’ demographic and urological characteristics are shown in Table 1. When the patients were classified according to their LUTS score, patients with LUTS were older (P = 0.002), had longer durations of diabetes (P = 0.027), lower high-density lipoprotein cholesterol (HDL-C) levels (P = 0.002), higher blood urea nitrogen levels (P < 0.001), and lower estimated glomerular filtration rates (eGFR) (P = 0.029) than patients without LUTS (Supplementary Table 1). Among the 63 patients suspected of having bacteriuria, 31 had positive urine cultures. Compared with patients with negative urine culture for bacterial growth, patients with bacteriuria were significantly older (P = 0.001), had a lower body mass index (BMI) (P = 0.017), and had higher rates of angiotensin converting enzyme inhibitor or angiotensin receptor blocker use (P = 0.021) (Supplementary Table 2). There was no difference in LUTS presentation, except for the urge incontinence score in the OABSS (P = 0.019).
Table 1.
Demographic and urological characteristics of the total cohort
| Total cohort (n = 63) | |
|---|---|
| Characteristic | |
| Age (years) | 66 (53–74) |
| Female | 44 (69.8) |
| Body mass index (kg/m2) | 25.2 (23.2–27.8) |
| Duration of diabetes (years) | 7.0 (4.0–13.0) |
| Patients with retinopathy | 19 (30.2) |
| Patients with albuminuria† | 33 (52.4) |
| Medication | |
| Metformin | 33 (52.4) |
| DPP-4i | 34 (54.0) |
| SGLT2i | 14 (22.2) |
| GLP-1RA | 13 (20.7) |
| Sulfonylurea | 14 (22.2) |
| Insulin | 17 (27.0) |
| ACE-I/ARB | 25 (39.7) |
| CCB | 13 (20.6) |
| Statin | 38 (61.9) |
| Diuretic | 7 (11.1) |
| Biomarker | |
| White blood cell count (/μL) | 6,810 (5,515–7,820) |
| Hemoglobin (g/dL) | 13.0 (12.0–14.2) |
| HbA1c (%) | 6.9 (6.5–7.3) |
| Total cholesterol (mg/dL) | 177 (162–198) |
| Triglycerides (mg/dL) | 113 (85–166) |
| LDL cholesterol (mg/dL) | 106 (87–118) |
| HDL cholesterol (mg/dL) | 51 (45–64) |
| Uric acid (mg/dL) | 5.5 (4.7–6.1) |
| BUN (mg/dL) | 16 (13–21) |
| Creatinine (mg/dL) | 0.7 (0.6–0.9) |
| BUN/ Creatinine ratio | 21.2 (17.2–25.7) |
| eGFR (mL/min/1.73 m2) | 67 (52–79) |
| Urological assessment | |
| Bidet user | 31 (49.2) |
| IPSS | 5 (2–10.0) |
| IPSS-QoL score | 2 (1–3) |
| Total OABSS | 3 (1–6) |
| Daytime frequency score | 1 (0–1) |
| Nocturia score | 1 (1–2) |
| Urgency score | 0 (0–3) |
| Urge incontinence score | 0 (0–1) |
| OAB | 21 (33.3) |
| ICIQ-SF score | 0 (0–8) |
| Infecting organism | |
| Escherichia coli | 23 (36.5) |
| Klebsiella pneumoniae | 3 (4.7) |
| Streptococcus agalactiae | 1 (1.6) |
| Enterococcus faecalis | 1 (1.6) |
| Morganella morganii | 1 (1.6) |
| Staphylococcus hominis | 1 (1.6) |
Values are shown as the median (25th and 75th percentile) or number (%).
† Patients with albuminuria are those with a urinary albumin/creatinine ratio of > 30.
DPP-4i: dipeptidyl peptidase-4 inhibitor; SGLT2i: sodium glucose cotransporter 2 inhibitor; GLP-1RA: glucagon-like peptide-1 receptor agonist; ACE-I: angiotensin coenzyme inhibitor; ARB: angiotensin receptor blocker; CCB: calcium channel blocker; LDL: low-density lipoprotein; HDL: high-density lipoprotein; BUN: blood urea nitrogen; eGFR: estimated glomerular filtration rate; IPSS: International Prostate Symptom Score; IPSS-QoL score: IPSS quality of life score; OABSS: Overactive Bladder Symptom Score; OAB: overactive bladder; ICIQ-SF: International Consultation on Incontinence Questionnaire Short Form; PVR: post void residual urine volume. *P < 0.05, **P < 0.01, and ***P < 0.001 by the 2-tailed Student t-test or Mann–Whitney’s test for 2-group comparisons of the continuous variables, and the Fisher’s exact test for comparisons of the categorical variables.
Among the 31 patients with bacteriuria, 29 patients underwent ultrasonography. Among them, 19 (65.5%) patients had SB, and 10 (34.5%) patients had ASB. The demographic and clinical characteristics of these patients are summarized in Supplementary Table 3. There were no significant differences in the demographic characteristics or medication use rates between the patients with SB and those with ASB. In contrast, patients with SB had significantly higher white blood cell counts (P = 0.046), lower HbA1c levels (P = 0.032) and HDL-C levels (P = 0.006) than those with ASB. Although we could not detect significant differences, patients with SB had a higher median PVR (P = 0.070) and a higher rate of patients with a significant PVR (P = 0.134) than those with ASB. We did not observe a significant correlation between PVR and LUTS (Fig. 2). The bacteria detected in the urine cultures were similar between the 2 groups. Escherichia coli was the most common organism in both groups. Of the 29 patients with bacteriuria, 5 (17.2%) patients were SGLT2i users, and all of them were symptomatic. In contrast, of the 24 patients with bacteriuria who were non-SGLT2i users, 14 (58.3%) patients were symptomatic. Median PVR and the frequency of patients with a significant PVR were significantly higher in SGLT2i users with bacteriuria than in non-SGLT2i users with bacteriuria (Supplementary Table 4).
Fig. 2.
Correlation analysis between post-void residual (PVR) volume and lower urinary tract symptoms. (A-C) Correlation analysis demonstrated that PVR did not correlate with LUTS assessed using International Prostate Symptom Score(A), Overactive Bladder Symptom Score (OABSS) (B), or International Consultation on Incontinence Questionnaire Short Form (ICIQ-SF) (C) among patients with bacteriuria (n = 29); Pearson’s correlation of determination r = 0.219, P = 0.327 (A); r = − 0.131, P = 0.523 (B). Spearman’s correlation of determination r = − 0.266, P = 0.190 (C).
To predict the urological condition of a patient (LUTS, bacteriuria, and SB) from their demographic and metabolic parameters, we next used the random forest method. The accuracy, sensitivity, and specificity of the random forest model were 76.9%, 75.0%, and 80.0%, respectively, in predicting LUTS, 76.9%, 75.0%, and 80.0%, respectively, in predicting bacteriuria, and 85.7%, 50.0%, and 100%, respectively, in predicting SB. Figure 3 (A–C) shows the MDG index in predicting each condition. A higher MDG suggests a higher importance of the variable. HDL-C was identified as a crucial variable in predicting LUTS, age as a crucial variable in predicting bacteriuria, and creatinine (Cr) and eGFR levels as crucial variables in predicting SB among patients with bacteriuria. The receiver operating characteristic (ROC) curves of each model are shown in Fig. 4 (A–C). Our models had high predictive accuracy for LUTS (area under the curve [AUC] = 0.75, 95%CI = 0.408–1.000), followed by bacteriuria (AUC = 0.83, 95%CI = 0.584–1.000), and SB (AUC = 0.90, 95%CI = 0.623–1.000).
Fig. 3.
Importance of variables in predicting LUTS (A), bacteriuria (B), and SB (C) in the random forest algorithm. (A–C) The X-axis shows the mean decrease in the Gini index, and the Y-axis represents the importance of each variable. HDL-C: high-density lipoprotein cholesterol; BUN: blood urea nitrogen; BUN/Cr: BUN/ creatinine ratio; eGFR: estimated glomerular filtration rate; Cr: creatinine; Hb: hemoglobin: TC; total cholesterol; WBC: white blood cell count; TG: triglycerides; BMI: body mass index; LDL-C: low-density lipoprotein cholesterol; UA: uric acid; SGLT2i: sodium glucose cotransporter 2 inhibitor; ACEi/ARB: angiotensin coenzyme inhibitor or angiotensin receptor blocker; SU: sulfonylurea, CCB: calcium channel blocker; GLP-1RA: glucagon-like peptide-1 receptor agonist; DPP-4i: dipeptidyl peptidase-4 inhibitor.
Fig. 4.
Receiver operating characteristic (ROC) curve analysis of our random forest models. (A–C) Area under the curve (AUC) of the ROC curve is 0.75 (95%CI: 0.408–1.000) for predicting lower urinary tract symptoms (LUTS) (A), 0.83 (95%CI: 0.584–1.000) for predicting bacteriuria (B), and 0.90 (95%CI: 0.623–1.000) for predicting symptomatic bacteriuria (SB).
Discussion
In this study, we clarified the urological characteristics of patients with T2DM suspected of having bacteriuria. Numerous studies demonstrate the association between diabetes and LUTS or bacteriuria. On the other hand, few studies have focused on bacteriuria and LUTS in patients with T2DM especially after SGLT2i became commercially available. Our prediction models included covariates known to be associated with urological complications in patients with T2DM, such as age [15, 21, 22], duration of type 2 diabetes [22], renal function [23], BMI [21], dyslipidemia [22, 24, 25], and glycemic control [22, 26]. Because these studies used regression analysis, the number of variables was limited in the analysis. In this study, we used a random forest method to perform exhaustive analysis utilizing many variables commonly measured in daily practice and identified predictors of urological complications. Therefore, our study is of great significance as an observational study.
Our study suggests that LUTS and bacteriuria have different predictors. HDL-C level was a stronger predictor of LUTS than the other variables. Although the mechanism remained to be elucidated, atherosclerosis causes LUTS via ischemia and oxidative stress in the bladder [27]. HDL-C is known to be the marker of atherosclerosis. Previous studies showed that reduced HDL-C is independently associated with LUTS after covariate adjustment [24, 25]. Patients with atherosclerosis showed higher LUTS score compared with those without atherosclerosis, suggesting the association between atherosclerosis and LUTS [28]. As observed in our study, HDL-C was lower in patients with atherosclerosis compared with those without atherosclerosis and LDL-C was comparable between the two groups in the study. Thus, HDL-C may be a better marker of atherosclerosis to predict LUTS than LDL-C. To predict bacteriuria, age is the most important variable, with a high MDG value. This is consistent with previous studies showing age as a risk factor for bacteriuria [15, 22]. Aging induces various pathologies associated with bacteriuria including age-related bladder dysfunction [29], age-related atrophy of the vaginal mucosa and urethra due to estrogen deficiency in women [30], and functional changes in the immune system [31]. In predicting SB among patients with bacteriuria, Cr and eGFR levels showed high MDG values, but blood urea nitrogen (BUN) and BUN/Creatinine ratio showed low MDG values, suggesting that dehydration rather than renal function is a crucial variable in predicting LUTS among patients with bacteriuria. There has been little focus on the association between renal function and LUTS. A previous study focused on LUTS in older patients without diabetes and showed the association between renal function and LUTS [32]. Our cohort for predicting SB was older than that of predicting LUTS and bacteriuria (Supplementary Table 3). Our results might reflect the observed trend in older patients. In addition, using the smaller cohort in predicting SB might result in different predictors from LUTS and bacteriuria, and lower MDG values. However, all models had an accuracy of more than 70%, and a high AUC in the ROC curve, suggesting the effectiveness of our prediction models. Since very few studies focused on both bacteriuria and LUTS in patients with diabetes in a single study, further studies are required to elucidate the inconsistency. In summary, the predictors in our models were consistent with previous studies. Those studies were investigated before SGLT2i became commercially available, and our study demonstrated a similar trend even after SGLT2i became commercially available. While both dyslipidemia and age have been reported to be associated with LUTS and bacteriuria, the variations in predictors suggest that the contribution of each condition may differ for each complication.
The association between glycemic control and urological complications is controversial. Whereas some cross-sectional studies did not identify an association between glycemic control and LUTS [33] or bacteriuria [15], other studies showed that an increase in HbA1c level was associated with LUTS [22, 26]. In our study, HbA1c level was lower in patients with SB than in those with ASB, but HbA1c level was not identified as a crucial variable in predicting SB. Similarly, HbA1c level was not a crucial variable in predicting LUTS or bacteriuria. These results suggested that current glycemic control may not be an accurate predictor of urological complications. Our overall study population had good glycemic control (median HbA1c level: 6.9%) and may not be representative enough of the population of diabetes patient to draw statistically robust conclusions regarding the association between glycemic control and urological complications.
Ultrasonography revealed no significant difference in PVR between patients with SB and those with ASB (Supplementary Table 3) and no correlation between PVR and LUTS score, while it detected a small number of patients with low LUTS score but high PVR (Fig. 2). Examination by a specialist should be considered in patients with a high residual urine volume [20]. Our finding indicates the LUTS questionnaire is not sufficient to identify such patients, and PVR may be useful as an adjunctive test. We could not identify the predictors due to a limited sample size (data not shown). Further studies are required to predict those patients.
In our study, no significant difference was observed in the number of SGLT2i users between patients with or without LUTS (Supplementary Table 1) as opposed to a previous study [13]. This discrepancy can be attributed to differences in the inclusion criteria and patient characteristics. One major difference is that the previous report focused only on males with T2DM. Similar to the result for LUTS, there was no significant difference in the number of SGLT2 users between patients with or without bacteriuria (Supplementary Table 2). Thus, our prediction model did not detect SGLT2i use as a crucial predictor of urological complications. This suggested the minimal effect of SGLT2i on the onset of LUTS and bacteriuria. On the contrary, all SGLT2i users with bacteriuria were symptomatic, although we could not detect a significant difference in the number of patients with symptomatic bacteriuria between SGLT2i users and non-SGLT2i users, owing to the insufficient power for statistical comparison (P = 0.134). This might suggest that SGLT2i use can exacerbate the risk of LUTS in patients who already have risk factors of developing LUTS, such as bacteriuria. In SGLT2i users with bacteriuria, significantly higher PVR were observed than non-SGLT2i users with bacteriuria. Osmotic diuresis caused by SGLT2i use may contribute to detrusor muscle hypertrophy and hyperactivity, leading to an increase in PVR. Similar to our study, Hall et al. [34] reported a case of urosepsis in a man with a significant PVR who was prescribed SGLT2i. The prevalence of patients with T2DM and a PVR of 100 mL or larger is higher in our study than in previous studies conducted prior to the commercial use of SGLT2i [10]. An increased PVR is known to be associated with urinary tract infections (UTIs) [35, 36]. It remains unclear whether SGLT2i use is associated with an increased risk of developing UTIs [37–39]. The differences in the conclusions among the studies may be owing to differences in baseline characteristics of the patients, particularly regarding their urological conditions. However, it is hard to draw conclusions about the effect of SGLT2i on urological issues because the sample size was small in our study.
We also analyzed the association between urological problems and bidet toilet use, because bidet toilets are widely used in Japan [14], and several studies have suggested an association between bidet toilet use and urogenital complications [40, 41]. We found that bidet toilet use was not a crucial variable in our model for predicting LUTS or bacteriuria.
Our study had several limitations. First, this was a cross-sectional study, and therefore we could not determine causation. Second, the sample size was relatively small. Third, the urine samples for urinary sediment examinations and urine cultures were not obtained simultaneously, and spot urine cultures were obtained. This might have caused an overestimation of the prevalence of bacteriuria. However, all of the 29 patients with bacteriuria showed significant bacteriuria (≥ 104 cfu/mL) on urinary sediment examination during 2 consecutive outpatient visits, suggesting the consistent presence of bacteriuria. Fourth, we did not measure the variable associated with neuropathy, which was previously reported to be associated with LUTS [42].
In conclusion, our prediction models identified the predictors utilizing demographic and laboratory data commonly measured in daily practice. The predictors were consistent before and after SGLT2i became commercially available. Although SGLT2i use was not an important predictor in our study, SGLT2i users with bacteriuria showed higher PVR than non-SGLT2i users with bacteriuria, which might suggest the possible effect of SGLT2i on the urological condition in high-risk patients with urological complications.
Supplementary Information
Below is the link to the electronic supplementary material.
Funding
This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.
Data Availability
The data that support the findings of this study are available on request from the corresponding author.
Declarations
Conflict of interest
Financial interests: Ryo Suzuki received honoraria, research funding, subsidies, and donations from Novo Nordisk Inc., Sanofi K.K., Sumitomo Pharma Co., Ltd., Astellas Pharma Inc., Kowa co., Ltd., MSD K.K., Eli Lilly Japan K.K., Mitsubishi Tanabe Pharma Co., Teijin Pharma, Ltd., Ono Pharmaceutical Co., Ltd., Daiichi Sankyo co., Ltd, Sanwa Kagaku Kenkyusho, Ltd., Takeda Pharmaceutical Co., Ltd. Masato Odawara received honoraria, subsidies, or donations from Sumitomo Pharma Co., Ltd., Taisho Pharmaceutical Co., Ltd., Novartis Pharma K.K., Eli Lilly Japan K.K., Astellas Pharma Inc., Ono Pharmaceutical Co., Ltd., Kyowa Hakko Kirin Co., Ltd., Daiichi Sankyo Co., Ltd., Takeda Pharmaceutical Co., Ltd., Mitsubishi Tanabe Pharma Corporation, Nippon Boehringer Ingelheim Co., Ltd., Merck Sharp & Dohme Corporation, AstraZeneca K.K., Novo Nordisk Pharma Ltd., and Sanofi K.K. The rest of the authors declare that they have no conflict of interest.
Ethical approval
All procedures followed were approved by the Ethics Committee of Toda Chuo General Hospital (study approval no.: 0496, approval date of revised version: 6-May-2021) and were in accordance with the Helsinki Declaration of 1964 and later versions. All the participants provided written informed consent before their inclusion in the study.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Daneshgari F, Liu G, Birder L, Hanna-Mitchell AT, Chacko S. "Diabetic bladder dysfunction: current translational knowledge," (in eng) J Urol. 2009;182(6 Suppl):S18–26. doi: 10.1016/j.juro.2009.08.070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Xu D, Gao J, Wang X, Huang L, Wang K. Prevalence of overactive bladder and its impact on quality of life in 1025 patients with type 2 diabetes in mainland China. J Diabetes Complications. 2017;31(8):1254–1258. doi: 10.1016/j.jdiacomp.2017.05.001. [DOI] [PubMed] [Google Scholar]
- 3.Kupelian V, et al. Prevalence of lower urinary tract symptoms and effect on quality of life in a racially and ethnically diverse random sample: the Boston Area Community Health (BACH) Survey. Arch Intern Med. 2006;166(21):2381–2387. doi: 10.1001/archinte.166.21.2381. [DOI] [PubMed] [Google Scholar]
- 4.Jacobson AM, et al. Relationship of urologic complications with health-related quality of life and perceived value of health in men and women with type 1 diabetes: the Diabetes Control and Complications Trial/Epidemiology of Interventions and Complications (DCCT/EDIC) cohort. Diabetes Care. 2015;38(10):1904–1912. doi: 10.2337/dc15-0286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hill SR, Fayyad AM, Jones GR. Diabetes mellitus and female lower urinary tract symptoms: a review. Neurourol Urodyn. 2008;27(5):362–367. doi: 10.1002/nau.20533. [DOI] [PubMed] [Google Scholar]
- 6.Tanaka Y, et al. Association of lower urinary tract symptoms based on the international prostate symptom score and cardiovascular disease. Circ J. 2021;85(11):2092–2099. doi: 10.1253/circj.CJ-21-0278. [DOI] [PubMed] [Google Scholar]
- 7.Akerla J, et al. Lower urinary tract symptoms and mortality among finnish men: the roles of symptom severity and bother. J Urol. 2022;207(6):1285–1294. doi: 10.1097/JU.0000000000002450. [DOI] [PubMed] [Google Scholar]
- 8.Chung MS, Chuang YC, Lee JJ, Lee WC, Chancellor MB, Liu RT. Prevalence and associated risk factors of nocturia and subsequent mortality in 1,301 patients with type 2 diabetes. Int Urol Nephrol. 2014;46(7):1269–1275. doi: 10.1007/s11255-014-0669-2. [DOI] [PubMed] [Google Scholar]
- 9.Lee WC, Wu HP, Tai TY, Liu SP, Chen J, Yu HJ. Effects of diabetes on female voiding behavior. J Urol. 2004;172(3):989–992. doi: 10.1097/01.ju.0000136255.83054.0c. [DOI] [PubMed] [Google Scholar]
- 10.Appa AA, et al. Clinical predictors and significance of postvoid residual volume in women with diabetes. Diabetes Res Clin Pract. 2013;101(2):164–169. doi: 10.1016/j.diabres.2013.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sorrentino F, et al. "Associations between individual lower urinary tract symptoms and bacteriuria in random urine samples in women," (in eng) Neurourol Urodyn. 2015;34(5):429–433. doi: 10.1002/nau.22607. [DOI] [PubMed] [Google Scholar]
- 12.Ishida S, et al. Urodynamic risk factors of asymptomatic bacteriuria in men with non-neurogenic lower urinary tract symptoms. World J Urol. 2022;40(12):3035–3041. doi: 10.1007/s00345-022-04194-x. [DOI] [PubMed] [Google Scholar]
- 13.N. C. Chilelli et al., "Lower urinary tract symptoms (LUTS) in males with type 2 diabetes recently treated with SGLT2 inhibitors-overlooked and overwhelming? A retrospective case series," (in eng), Endocrine, vol. 59, no. 3, pp. 690–693, 03 2018, 10.1007/s12020-017-1301-x. [DOI] [PubMed]
- 14.Tsunoda A. Bidet Toilet Use May Cause Anal Symptoms and Nosocomial Infection. J Anus Rectum Colon. 2021;5(4):335–339. doi: 10.23922/jarc.2021-027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.S. E. Geerlings et al., "Asymptomatic bacteriuria may be considered a complication in women with diabetes. Diabetes Mellitus Women Asymptomatic Bacteriuria Utrecht Study Group," (in eng), Diabetes Care, vol. 23, no. 6, pp. 744–9, Jun 2000, 10.2337/diacare.23.6.744. [DOI] [PubMed]
- 16.Renko M, Tapanainen P, Tossavainen P, Pokka T, Uhari M. "Meta-analysis of the significance of asymptomatic bacteriuria in diabetes," (in eng) Diabetes Care. 2011;34(1):230–235. doi: 10.2337/dc10-0421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Homma Y, et al. Symptom assessment tool for overactive bladder syndrome–overactive bladder symptom score. Urology. 2006;68(2):318–323. doi: 10.1016/j.urology.2006.02.042. [DOI] [PubMed] [Google Scholar]
- 18.E. Karantanis, M. Fynes, K. H. Moore, and S. L. Stanton, "Comparison of the ICIQ-SF and 24-hour pad test with other measures for evaluating the severity of urodynamic stress incontinence," Int Urogynecol J Pelvic Floor Dysfunct, vol. 15, no. 2, pp. 111–6; discussion 116, Mar-Apr 2004, 10.1007/s00192-004-1123-2. [DOI] [PubMed]
- 19.Okamura K, Usami T, Nagahama K, Maruyama S, Mizuta E. "The relationships among filling, voiding subscores from International Prostate Symptom Score and quality of life in Japanese elderly men and women," (in eng) Eur Urol. 2002;42(5):498–505. doi: 10.1016/s0302-2838(02)00438-4. [DOI] [PubMed] [Google Scholar]
- 20.S. Takahashi et al., "Clinical Guidelines for Female Lower Urinary Tract Symptoms (second edition)," Int J Urol, vol. 28, no. 5, pp. 474–492, May 2021, doi: 10.1111/iju.14492. [DOI] [PubMed]
- 21.James R, Hijaz A. Lower urinary tract symptoms in women with diabetes mellitus: a current review. Curr Urol Rep. 2014;15(10):440. doi: 10.1007/s11934-014-0440-3. [DOI] [PubMed] [Google Scholar]
- 22.Dai M, et al. Incidence and risk factors of asymptomatic bacteriuria in patients with type 2 diabetes mellitus: a meta-analysis. Endocrine. 2023;82(2):263–281. doi: 10.1007/s12020-023-03469-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ishay A, Lavi I, Luboshitzky R. Prevalence and risk factors for asymptomatic bacteriuria in women with Type 2 diabetes mellitus. Diabet Med. 2006;23(2):185–188. doi: 10.1111/j.1464-5491.2005.01758.x. [DOI] [PubMed] [Google Scholar]
- 24.Tai HC, et al. "Metabolic syndrome components worsen lower urinary tract symptoms in women with type 2 diabetes," (in eng) J Clin Endocrinol Metab. 2010;95(3):1143–1150. doi: 10.1210/jc.2009-1492. [DOI] [PubMed] [Google Scholar]
- 25.Feng T, et al. Serum cholesterol and risk of lower urinary tract symptoms progression: results from the Reduction by Dutasteride of Prostate Cancer Events study. Int J Urol. 2017;24(2):151–156. doi: 10.1111/iju.13265. [DOI] [PubMed] [Google Scholar]
- 26.E. Papaefstathiou, K. Moysidis, P. Sarafis, E. Ioannidis, and K. Hatzimouratidis, "The impact of Diabetes Mellitus on Lower urinary tract symptoms (LUTS) in both male and female patients," Diabetes Metab Syndr, vol. 13, no. 1, pp. 454–457, Jan - Feb 2019, doi: 10.1016/j.dsx.2018.11.009. [DOI] [PubMed]
- 27.Nomiya M, Andersson KE, Yamaguchi O. Chronic bladder ischemia and oxidative stress: new pharmacotherapeutic targets for lower urinary tract symptoms. Int J Urol. 2015;22(1):40–46. doi: 10.1111/iju.12652. [DOI] [PubMed] [Google Scholar]
- 28.Azab S. The impact of atherosclerosis on lower urinary tract function. Aging Male. 2013;16(3):108–111. doi: 10.3109/13685538.2013.795532. [DOI] [PubMed] [Google Scholar]
- 29.Andersson KE, Boedtkjer DB, Forman A. The link between vascular dysfunction, bladder ischemia, and aging bladder dysfunction. Ther Adv Urol. 2017;9(1):11–27. doi: 10.1177/1756287216675778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kasyan G, Berketova TY, Rogozin AK, Pushkar DY. Asymptomatic bacteriuria in postmenopausal women with diabetes mellitus. Cent European J Urol. 2013;66(3):320–326. doi: 10.5173/ceju.2013.03.art17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ligon MM, Joshi CS, Fashemi BE, Salazar AM, Mysorekar IU. Effects of aging on urinary tract epithelial homeostasis and immunity. Dev Biol. 2023;493:29–39. doi: 10.1016/j.ydbio.2022.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Soma O, et al. Relationship between frailty and lower urinary tract symptoms among community-dwelling adults. Low Urin Tract Symptoms. 2020;12(2):128–136. doi: 10.1111/luts.12292. [DOI] [PubMed] [Google Scholar]
- 33.Tai HC, Tai TY, Yang WS, Wang SW, Yu HJ. "Associations between lower urinary tract dysfunction and glycemic control in women with type 2 diabetes: a cross-sectional study," (in eng) J Diabetes Complications. 2016;30(3):415–419. doi: 10.1016/j.jdiacomp.2016.01.002. [DOI] [PubMed] [Google Scholar]
- 34.V. Hall, J. Kwong, D. Johnson, and E. I. Ekinci, "Caution advised with dapagliflozin in the setting of male urinary tract outlet obstruction," (in eng), BMJ Case Rep, vol. 2017, May 2017, doi: 10.1136/bcr-2017-219335. [DOI] [PMC free article] [PubMed]
- 35.Huang Z, Xiao H, Li H, Yan W, Ji Z. Analysis of the incidence and risk factors of male urinary tract infection following urodynamic study. Eur J Clin Microbiol Infect Dis. 2017;36(10):1873–1878. doi: 10.1007/s10096-017-3007-7. [DOI] [PubMed] [Google Scholar]
- 36.Stern JA, Hsieh YC, Schaeffer AJ. Residual urine in an elderly female population: novel implications for oral estrogen replacement and impact on recurrent urinary tract infection. J Urol. 2004;171(2 Pt 1):768–770. doi: 10.1097/01.ju.0000107261.64927.b3. [DOI] [PubMed] [Google Scholar]
- 37.A. D. Kaze, M. Zhuo, S. C. Kim, E. Patorno, and J. M. Paik, "Association of SGLT2 inhibitors with cardiovascular, kidney, and safety outcomes among patients with diabetic kidney disease: a meta-analysis," Cardiovasc Diabetol, vol. 21, no. 1, p. 47, Mar 23 2022, doi: 10.1186/s12933-022-01476-x. [DOI] [PMC free article] [PubMed]
- 38.Li D, Wang T, Shen S, Fang Z, Dong Y, Tang H. Urinary tract and genital infections in patients with type 2 diabetes treated with sodium-glucose co-transporter 2 inhibitors: A meta-analysis of randomized controlled trials. Diabetes Obes Metab. 2017;19(3):348–355. doi: 10.1111/dom.12825. [DOI] [PubMed] [Google Scholar]
- 39.Vasilakou D, et al. "Sodium-glucose cotransporter 2 inhibitors for type 2 diabetes: a systematic review and meta-analysis," (in eng) Ann Intern Med. 2013;159(4):262–274. doi: 10.7326/0003-4819-159-4-201308200-00007. [DOI] [PubMed] [Google Scholar]
- 40.Kiuchi T, Asakura K, Nakano M, Omae K. "Bidet toilet use and incidence of hemorrhoids or urogenital infections: A one-year follow-up web survey," (in eng) Prev Med Rep. 2017;6:121–125. doi: 10.1016/j.pmedr.2017.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Tsunoda A, Takahashi T, Arika K, Kubo S, Tokita T, Kameda S. "Survey of electric bidet toilet use among community dwelling Japanese people and correlates for an itch on the anus," (in eng) Environ Health Prev Med. 2016;21(6):547–553. doi: 10.1007/s12199-016-0578-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Bansal R, Agarwal MM, Modi M, Mandal AK, Singh SK. Urodynamic profile of diabetic patients with lower urinary tract symptoms: association of diabetic cystopathy with autonomic and peripheral neuropathy. Urology. 2011;77(3):699–705. doi: 10.1016/j.urology.2010.04.062. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author.