Do geriatrics require dose titration for antidiabetic agents?R Shastry1, P Adhikari2, A Kamath3, M Chowta1, S Ullal1, MRSM Pai1
1 Department of Pharmacology, Kasturba Medical College, Manipal University, Mangalore, Karnataka, India
2 Department of Medicine, Kasturba Medical College, Manipal University, Mangalore, Karnataka, India
3 Department of Community Medicine, Kasturba Medical College, Manipal University, Manipal, Karnataka, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.123153
Source of Support: None, Conflict of Interest: None
Objective: To evaluate the antidiabetic drug dosage differences between geriatric and nongeriatric diabetics with reference to duration of disease and creatinine clearance (Crcl). Materials and Methods: Prospective study conducted for 6 months in a tertiary care hospital. Patients with type 2 diabetes mellitus were grouped into geriatric (age ≥60 years) and nongeriatric (age <60 years). Patients' demographic data, duration of diabetes, medication, and serum creatinine were recorded. Crcl was calculated using Cockcroft-Gault formula. Doses of sulfonylureas (SU) were converted into equivalent doses, taking glibenclamide as standard. Univariate analysis was done for comparison of drug doses between groups. Result: A total of 320 geriatric and 157 nongeriatric diabetics completed the study. The duration of diabetes and Crcl adjusted dose reduction of glibenclamide (mean dose: Geriatrics 7.2±0.4 mg, nongeriatrics 9.6±0.7 mg; P=0.01) and gliclazide (mean dose: Geriatrics 85.5±11.5 mg, nongeriatrics 115.3±32.7 mg; P=0.42) was 25%, glimepiride (mean dose: Geriatrics 1.62±0.13 mg, nongeriatrics 2.1±0.18 mg; P=0.06) was 22%. Glipizide did not require dose reduction. Mean converted equivalent dose of sulfonylurea monotherapy was significantly lower in geriatrics than nongeriatrics (3.2±0.5 vs 6.4±1.02 mg; P=0.01) and showed 50% dose reduction. Mean dose of metformin was lower in geriatrics (901±32.2 mg vs 946.7±45.8 mg; P=0.45) and showed 5% reduction in dosage. There was no difference in the mean drug doses of thiazolidinediones and insulin between the groups. Conclusion: A substantial dose reduction of glibenclamide (25%), gliclazide (25%), glimepiride (22%), and metformin (5%) in geriatrics compared to nongeriatrics was observed. Smaller dosage formulations like 0.75 mg glibenclamide, 0.5 mg glimepiride, 20 mg gliclazide, and 250 mg metformin may be of value in geriatric diabetic practice.
Keywords: Creatinine clearance, diabetes mellitus, equivalent dose, nongeriatrics
The Indian aged population is currently the second largest in the world.  Problems faced by this segment are numerous and current health policies do not adequately address improvement of their health status. , Diabetes mellitus (DM) is a classic example of a disease that increases with age.  Almost 50% of type 2 diabetic patients are over 60 years of age. 
Age related changes in kidney and liver function are the most important physiological changes of the human body which require adjustment of drug selection and dosage. The impact of diabetes on renal impairment changes with increasing age.  The progressive decline in renal function that occurs with age may result in slower elimination of drugs that are partially or completely cleared by kidneys, leading to toxicity and additive side effects when dosages are not adjusted. 
The evidence to specifically guide diabetes treatment in older adults is limited and largely based on extrapolation from studies of younger adults with diabetes. To date, there has been no landmark diabetes trial conducted in a geriatric population.  There are no clear guidelines regarding dosage of drugs for elderly diabetic patients. Hence, this study was planned to evaluate the antidiabetic drug dosage differences between geriatric and nongeriatric patients with reference to duration of disease and creatinine clearance (Crcl).
Prospective study conducted for a period of 6 months in a tertiary care teaching hospital. Patients with type 2 diabetes mellitus were identified and grouped into geriatric (age ≥ 60 years) and nongeriatric (age <60 years). They were included in the study after obtaining written informed consent. The study was approved by the Institutional Ethics committee (IEC). Patient's demographic data, duration of diabetes, and medication details were recorded in a structured proforma. Laboratory investigations included fasting blood sugar (FBS), glycosylated hemoglobin (HbA1c), and serum creatinine. Crcl was calculated using Cockcroft-Gault formula.
Patients are followed-up for 6 months without significant alteration in the drug dosage for stabilization of glycemic control. Patients who needed frequent drug dosage changes due to repeated acute illness were not enrolled in the study. Drug treatment was not altered during the 6 months follow-up period. The prescriptions of these patients were analyzed and the investigators did not interfere with the dose and their titration. Mean FBS over 6 months was calculated. HbA1c was done after a minimum of 3 months during the 6 month follow-up period.
Both individual and total antidiabetic drug dosages were taken into account for comparison between groups. Doses of sulfonylureas (SU) were converted into equivalent doses, and glibenclamide was taken as standard for easy comparison. 
Comparison of demographics, duration of disease, drug doses and laboratory parameters was done by using statistical package for Social Sciences (SPSS) version 11.5. Students' unpaired t test was used to compare the different parameters between the two groups. Univariate analysis was done for comparison of drug doses of two groups adjusted for duration of disease and Crcl. All the analyses were done at 5% significance.
A total of 477 diabetic patients were included for the study (geriatrics n=320, nongeriatrics n=157). Demographic data and laboratory values of both groups are given in [Table 1]. [Table 2] shows the comparison of duration and Crcl adjusted doses of antidiabetic drugs in geriatrics and nongeriatrics. Duration and Crcl adjusted mean doses of glibenclamide and SU monotherapy were significantly lower in geriatrics compared to nongeriatrics. It is clear from the table that even after adjustment for duration of DM and Crcl, age independently affected geriatric drug dosage for SU (glibenclamide, glimepiride) while metformin, thiazolidinediones (TZD) and insulin dose reduction did not appear clinically or statistically significant. No deterioration of glycemic control or hypoglycemia was noted during the study period.
In our study of comparison of 320 elderly diabetic patients with 157 younger diabetics, we found significant differences in duration of diabetes, Crcl and drug dosages of most of the antidiabetics. Even when drug dosages were adjusted for duration of disease, body mass index (BMI), and Crcl; age appeared to independently influence drug dosage.
The mean glibenclamide dose was significantly lower in geriatric population compared to nongeriatric population (geriatrics 7.2±0.4 mg, nongeriatrics 9.6±0.7 mg; P=0.01). Since all the patients were taken for the study after stabilization of glycemic status for 6 months, this difference is clinically significant. Although the apparent difference was 2 mg, the duration (10 years) and Crcl (69.7 ml/min) adjusted dose reduction was 25%. This indicates that reduction in dosage of glibenclamide starts with 25% and may go up to 25-70% in patients with longer duration of diabetes and very low Crcl. Glibenclamide is most likely to accumulate during renal dysfunction as it yields two active metabolites and the clearance of these metabolites appears delayed.  This leads to prolonged hypoglycemia in the elderly and hence its avoidance is recommended in elderly diabetics. However, our study shows that glibenclamide can be safely used in elderly provided the dose is titrated to the lowest available dose of 1.25 mg. Formulations of smaller doses such as 0.75 mg may be of value in the treatment of elderly diabetics with very long duration and low Crcl and studies to substantiate this are needed. This is the first study where elderly showed significant difference in the drug dose even when adjusted for duration of disease and Crcl indicating that this dosage reduction is age related. In geriatric diabetics with lower Crcl further dosage reductions are necessary. Kidney Disease Outcomes Quality Initiative (KDOQI)  recommends that glibenclamide should be avoided and glipizide or gliclazide can be used without any dose adjustment in patients with stage 3-5 chronic kidney disease. There are no clinical trials in elderly people on glibenclamide, hence exact dosage recommendations cannot be made. However, formulations such as 0.5, 0.75, and 1 mg may be of great value in titrating geriatric drug dosage and tailoring it to individual needs. With smaller initial dosages and gradual upward titration its safety in the elderly is equivalent to that in the younger people. However, head-to-head comparison of elderly diabetics with younger diabetics on glibenclamide monotherapy or add on therapy with metformin with slow titration is necessary before we recommend it for elderly people.
Gliclazide too needed dosage adjustment. In our study population, dosage reduction was almost 25% because of longer duration of diabetes and lower Crcl and hence starting dose can be as low as 20 mg or modified release preparations with 15 mg.
Glimepiride also required lower dosage when adjusted with duration of disease and Crcl. Mean glimepiride dose for the elderly diabetics was 1.6 versus 2.1 mg for younger diabetics. The reduction was 22% when adjusted with duration of disease and Crcl. Although a total reduction of only about 0.5 mg/day was seen in our study, it is clinically significant because the staring dose of glimepiride is 1 mg/day. Hence, staring dose of glimepiride for elderly could be 0.5 mg.
No dosage adjustment was required for glipizide (mean doses in geriatrics and nongeriatrics was approximately 9.2 mg) which is in accordance with earlier reviews which state that among the second generation SU, glipizide is suited for elderly or patients with mild renal insufficiency as it has a shorter half-life and no active metabolite. 
The duration (7.5 years) and Crcl (59.4 ml/min) adjusted doses of SU (converted equivalent doses) when used as monotherapy in our population showed almost 50% dosage reduction in geriatrics compared to nongeriatrics which is clinically and statistically significant. As a whole when we adjust the doses of SU with duration of diabetes and Crcl, it showed 25-50% dose reduction of SU irrespective of its use as monotherapy or in combination therapy.
In case of metformin, the duration (8.9 years) and Crcl (71.3 ml/min) adjusted dose reduction was only 5% because the mean glomerular filtration rate in these patients was >60 ml/min which is one of the reasons for minimum dose reduction. Hence, for patients with low Crcl, (between 30 and 60 ml) according to previous literature dosage adjustment is necessary. 
The difference in dosage of TZDs between the groups was not statistically or clinically significant; this could be because the sample size for subgroup analysis of TZD usage was small. However, literature also mentions that TZDs are almost entirely metabolized by liver and the major metabolites do not accumulate in patients with stage 3-4 chronic kidney disease, hence no dosage adjustment is necessary. , The efficacy is similar in old and young type 2 diabetics.
There was no significant difference in doses of insulin therapy between the two groups.
All oral antidiabetic doses were converted into single unit to give total oral doses of drugs. Even the total oral antidiabetics received by the geriatric population was lower compared to nongeriatrics. When adjusted for duration of disease (7.72 years) and Crcl (69.78 ml/min) there was no significant reduction in dosage indicating that individuals with normal Crcl do not need dosage adjustment. However, patients with lower Crcl and longer duration of disease need dosage adjustments.
Crcl usually declines with age. Hence, clearance of most of the drugs reduces with age and the dosage requires adjustment depending upon Crcl. Therefore, in geriatric population it is necessary to adjust doses of oral hypoglycemic agents (OHA) with respect to Crcl to achieve a similar degree of glycemic control and avoid hypoglycemia.
A substantial dose reduction of glibenclamide (25%), gliclazide (25%), glimepiride (22%), and metformin (5%) in geriatrics compared to nongeriatrics when adjusted for duration of disease and Crcl was observed. Hence, geriatric diabetics need drug dosage adjustment. A smaller starting dosage formulation like 0.75 mg glibenclamide, 0.5 mg glimepiride, 20 mg gliclazide, and 250 mg metformin may be of value in geriatric diabetic practice.
[Table 1], [Table 2]