Hypothesis / aims of study
Metformin is the most commonly used glucose lowering agent for type 2 diabetes. Recent evidence suggests metformin facilitates the proliferation and migration of stem cells, which in turn have shown to improve bladder overactivity in animal models. The objective of this study was to determine if metformin use was associated with a lower rate of overactive bladder (OAB) medication use.
Study design, materials and methods
We conducted a retrospective nested case-control cohort study using population-based health-care administrative databases in Ontario, Canada, between 2002-2019. Our cohort was defined as diabetics aged 66 years or older. Cases were defined as patients in our cohort who had received a prescription for an OAB medication, and then matched with up to 4 controls based on age, sex, cohort entry date, and diabetes diagnosis date. Exposure was defined as a new prescription for metformin prior to receiving a prescription for an OAB medication. Adjusted odds ratios (aOR) were estimated using conditional logistic regression. Sensitivity analysis was done to assess for the relationship between cumulative days’ supply of metformin and use of OAB medications.
Results
Study population:
We identified 2,233,084 patients with DM2 in Ontario, Canada, age 69 and older. Of these patients we identified 16,549 cases who received a prescription for an OAB medication and matched them with 64,171 controls. Of the case patients, 31% had a new prescription for metformin during the study period. The mean age was 78.6 (SD 6.43) and 64% were female. Most patients were from urban areas (88%). Of patients with a CCI (35.2%), most had a score of 0. The most common comorbidity was coronary artery disease (8.6%). Mean and median use of metformin was 4.4 and 3.8 years, respectively. Cases had a mean and median distinct drug prescription of 24.8 (SD 12.2) and 23 (IQR 16-31), respectively. Male patients had a mean of 0.5 (SD 1.1) distinct BPH medication prescriptions. In terms of healthcare utilization in the three years prior to index date, cases had a mean of 1.3 (SD 2.7) emergency department visits, 68.1 (SD 55) office visits to any physician, 1.4 (SD 2.8) urology visits, and 0.27 (SD 1.2) endocrinology visits. After age and sex matching, most baseline characteristics did not differ significantly between the case and control groups except cases had a lower CCI score, more drug prescriptions, and more visits to a physician.
Metformin use and OAB:
We found a small positive association between OAB medication use and metformin use (aOR = 1.07, 95% CI = 1.03-1.12). In adjusted analysis, as comorbidity increases from CCI 1 to CCI 3 the odds of OAB exposure among patients receiving metformin decreases, as compared to CCI 0. The odds of OAB exposure were increased for patients with increased number of unique prescriptions, BPH prescriptions, ED visits, or urology visits.
Secondary Analysis:
On multivariable analysis, summed days’ supply of metformin was also associated with OAB medication use except when summed metformin days was >2220, at which point no association was found. Of note, the association found is minimal, but statistically significant.
Interpretation of results
Prior human studies have not examined the direct effect of metformin on bladder overactivity. We postulated that as metformin is known to facilitate proliferation of stem cells, and stem cell injections into bladder tissue have improved bladder overactivity in animal models, that patients on metformin may have lower rates of bladder overactivity. Metformin has also been shown to reduce oxidative stress and inflammation, both of which are risk factors for organ dysfunction. One recent study examined the effect of metformin on bladder tissue, where rats with partial bladder outlet obstruction (BOO) were treated with metformin. After short term use (two weeks) the rats had lower baseline bladder pressures and a reduced inflammatory reaction. After long term use (nine weeks), the rats had resolution of bladder fibrosis and improved bladder compliance [1]. This study suggests that metformin may inhibit bladder remodeling and alleviate bladder dysfunction in BOO.
In contrast with our hypothesis, our population-based study found that metformin was not associated with a reduced rate of OAB medications among older diabetics. In fact, we found that metformin use in this population slightly increased the risk of use of OAB medication, when controlling for other factors. It is possible that this increase in risk of OAB medication use may be due to early urologic complications of diabetes, where those patients on metformin may represent those with worse disease, compared to diabetics not yet treated with medications. Urologic complications of diabetes are common and associated with lower urinary tract dysfunction in 50-80% of patients, ranging from OAB storage symptoms to the voiding symptoms of underactive bladder. Urodynamic studies in diabetic patients have shown a 55% rate of detrusor overactivity, especially in the early stages of diabetes [2]. The pathophysiology behind overactivity in diabetic bladders is multifactorial and due to autonomic neuropathy from microvascular damage, deficits in the regulation of detrusor contractility, and changes to the bladder urothelium leading to alteration in the excitability of sensory fibers. The duration of hyperglycemia and level of blood glucose in diabetics have been shown to be risk factors for worsening bladder dysfunction, with improved glucose control preventing bladder dysfunction progression [3].
It is well known that older patients are at risk of polypharmacy and multimorbidity. We found that the odds of OAB exposure were increased for patients with increased number of unique prescriptions. Our study cohort had a mean age of 78, with a mean number of 19 distinct drug prescriptions over the study period. It is possible that urinary side effects of multiple medications, or urinary symptoms due to multiple medical comorbidities, may contribute to the increased use of OAB medications in this population.