FECAL INCONTINENCE RESOLVED USING METFORMIN WITHDRAWAL

Abstract

To the Editor: Glycemic control is often surprisingly good in in people with diabetes mellitus in geriatric care, although this may paradoxically be achieved by weight loss or malnutrition or other poorly understood mechanisms related to comorbidities.1,2 In this context, previous prescriptions for antidiabetic drugs may become unnecessary. It has previously been shown that supervised withdrawal of oral hypoglycemic agents (OHAs) is often possible without worsening of glycemic control in older hospitalized adults, even when nutritional support is offered.1 This letter reports a case in which fecal incontinence was resolved using metformin withdrawal. A general practitioner referred an 82-year-old woman to the geriatric service of the Geneva University Hospitals for a global assessment in the context of increasing dependency on nursing care at home. One of the most prominent complaints was worsening fecal incontinence, the management of which put a heavy burden on her husband and caregivers. She had arterial hypertension, type 2 diabetes mellitus, vascular dementia, and depression. She had a stroke 8 years earlier, without persistent motor deficit but with neurogenic pain as a sequel. Urinary and fecal incontinence problems had been present for at least 15 years. Two surgical procedures for a rectocele (in 1995) and a cystocele (in 1998), had failed to provide lasting relief. Her medications included olmesartan, metoprolol, clopidogrel, gabapentin, sitagliptin (50 mg/d), and metformin (500 mg/d). Her physical examination was unremarkable. Her body mass index was 33.7 kg/m2. She had no motor deficit and her Mini-Mental State Examination score was 18 out of 30. Her Mini Nutritional Assessment score was 10 out of 14, indicating that she was at risk of malnutrition. Laboratory values on admission were unremarkable except for hypoalbuminemia (serum albumin 3.1 g/dL) and vitamin B12 deficiency (cyanocobalamin 126 pmol/L). Her glycosylated hemoglobin (HbA1c) was 6.6%. Somewhat to our puzzlement, the patient had no fecal incontinence during her first days in the hospital. She therefore just received supportive care, and planning for her admission to a nursing home was begun. On admission, the low doses of OHA and the low HbA1c had led us to suspect that these agents were not necessary for good glycemic control, and sitagliptin and metformin had been withdrawn. A slow, progressive worsening in glycemic control was observed. On day 21, her fasting and predinner blood glucose values were 152 and 245 mg/dL, respectively, so it was decided to reintroduce metformin at an initial dose of 500 mg/d. That same day, the nursing team was confronted with the recurrence of fecal incontinence, severe enough to preclude any activities outside her room. A consultation with the neuropsychologist had to be interrupted because of the patient's discomfort. On Day 23, metformin was again interrupted. Fecal incontinence rapidly disappeared, and the patient remained symptom-free until her discharge to her home on Day 48. She was started on gliclazide, with acceptable glycemic control (fasting blood glucose <140 mg/dL) at discharge. Although the use of a number of different drugs in older individuals is usually based on rational grounds, polypharmacy is a major clinical problem because of adverse drug reactions, drug–drug interactions, and poor adherence to drug therapy.3,4 Drug side effects are often difficult to predict directly from their biological properties. If unrecognized, they can lead to severe impairments in quality of life or to the prescription of additional drugs to treat side effects. In the field of diabetes mellitus care, the most predictable drug side effect is hypoglycemia, but other, more-unusual side effects may severely affect the quality of life of older adults with diabetes mellitus. Because it has no effect on insulin secretion, metformin carries a low risk of hypoglycemia. Nonetheless, it may reduce food intake and impair vitamin B12 absorption.5,6 These effects are potentially deleterious in a patient at risk of malnutrition. Metformin is known for causing digestive symptoms, but in the present case, fecal incontinence had not been recognized as such. In practice, it is surprising how often patients or caregivers fail to recognize the association between metformin and digestive symptoms. This patient's cognitive impairment probably exacerbated this problem. A reintroduction test finally—albeit unwittingly—demonstrated the role of metformin in fecal incontinence. The present observation illustrates that careful reevaluation of antidiabetic drugs, aiming for the simplest possible regimen, may have significant benefits beyond reducing the risk of hypoglycemia. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Author Contributions: CG and UNV: acquisition of subjects and data, preparation of manuscript. MZL: acquisition of subjects and data. Sponsor's Role: The was no sponsor for this study.