Annals of African Medicine

: 2022  |  Volume : 21  |  Issue : 1  |  Page : 26--33

Treatment outcomes of graded dose of empagliflozin in type-2 diabetes: A real world study

Deep Dutta1, Meha Sharma2, Sameer Aggarwal3, Anshita Agarwal4, Anil Dhall5,  
1 Department of Endocrinology, Center for Endocrinology, Diabetes, Arthritis and Rheumatism Super Speciality Clinics, New Delhi, India
2 Department of Rheumatology, Center for Endocrinology, Diabetes, Arthritis and Rheumatism Super Speciality Clinics, New Delhi, India
3 Department of Endocrinology, Apex Hospital, Rohtak, Haryana, India
4 Department of Endocrinology, Dr. Ram Manohar Lohia Hospital, New Delhi, India
5 Department of Cardiology, Janakpuri Super Speciality Hospital, New Delhi, India

Correspondence Address:
Deep Dutta
Department of Endocrinology, Center for Endocrinology, Diabetes, Arthritis and Rheumatism Super-speciality Clinics, Plot 107 & 108, Sector 12A Dwarka, New Delhi - 110 075


Background: Costs are important cause of therapeutic noncompliance in type-2 diabetes mellitus (T2DM). Half-tablet empagliflozin (EMPA)-25 mg has lowest monthly cost among all EMPA preparations; data is unavailable on efficacy of half EMPA-25. This study compared real world weight loss and glycaemic outcomes of 10 mg versus 12.5 mg versus 25 mg of EMPA. Methods: Data, retrospectively captured from records of 2 different centresfor patients > 35 years-age having T2DM on EMPA as part of standard pharmacotherapy for T2DM, having > 6 months follow-up data available was analysed. Patients were in 3-groups depending on EMPA dosage: Group 1 on EMPA 10 mg/day (1-tablet EMPA-10), Group-2 on EMPA 12.5 mg/day (half-tablet EMPA-25), and Group 3 on EMPA 25 mg/day (1-tablet EMPA-25). Primary endpoints were glycaemic efficacy and weight-loss. Results: Of 3601 records screened, data from 599 patients (184, 239 and 176 in Group-1, 2 and 3 respectively) was analysed. All 3 groups were comparable with regards to sex, blood pressure, haemoglobin, renal function, medications use. Group-3 were significantly older, had longest diabetes duration, highest HbA1c and lowest body mass index. Glycaemic efficacy was comparable among groups (ΔHbA1c Groups 1-3: −0.9 (−1.9 – 0.0), −1.0 (−1.8 – 0.5) and − 1.0 (−1.5 – 0.22], respectively; P = 0.363). Patients on EMPA 12.5 or 25 mg/d had significantly higher total (−1.4 [−3.0 –0.2] vs. −0.3 [−2.4 – 1.32] kg; P = 0.028) and percent weight-loss (−1.75% [−4.15 – 0.26] vs. −0.44% [−3.11 – 1.39]; P = 0.039), and significantly higherfraction achieving HbA1c < 5.7% (12% vs. 0; P = 0.021), compared to EMPA-10. Conclusion: Half EMPA-25 is the most cost effective way of using EMPA in clinical practice.

How to cite this article:
Dutta D, Sharma M, Aggarwal S, Agarwal A, Dhall A. Treatment outcomes of graded dose of empagliflozin in type-2 diabetes: A real world study.Ann Afr Med 2022;21:26-33

How to cite this URL:
Dutta D, Sharma M, Aggarwal S, Agarwal A, Dhall A. Treatment outcomes of graded dose of empagliflozin in type-2 diabetes: A real world study. Ann Afr Med [serial online] 2022 [cited 2022 Sep 28 ];21:26-33
Available from:

Full Text


India currently has an overall 9% and 14%–18% prevalence of diabetes and prediabetes, respectively,[1],[2] with one of the highest global rates of prediabetes to diabetes progression.[3] The CARRS study from New Delhi showed a very high prevalence of prediabetes/diabetes of 72.7%.[4] Two recent studies from Delhi reported very high obesity prevalence (71.5% and 69.29% in a cohort of 1473 and 5336 patients, respectively).[5],[6] Weight loss has a critical role in improving glycemic control in people living with type-2 diabetes mellitus (T2DM), and may even lead to diabetes reversal.[7] SGLT2i have become popular for managing diabesity because of good glycaemic efficacy and weight-loss property.

EMPA is currently available at doses of 10 mg and 25 mg tablets for clinical use in India, costing INR 47 (INR 4.7/mg) and INR 57 (INR 2.28/mg) per tablet, respectively.[8],[9] From costing point of view, half-tablet of EMPA 25 mg would provide 12.5 mg of EMPA at INR 28.5, which would be cost-effective than taking 1.5 tablets of EMPA 10 mg (EMPA-10) at INR 70.5 (247% higher), or even 1-tablet of EMPA 10 mg (165% higher). The monthly cost of therapy for one tablet EMPA-10, half tablet EMPA-25 and full tablet EMPA-25 taken once daily are INR 1410, INR 855, and INR 1710, respectively. The corresponding monthly treatments cost for metformin 1000 mg, pioglitazone 15 mg, and sulfonylurea glipizide 5 mg is INR 3/tablet (monthly cost INR 180 for 2 g therapy per day), INR 2/tablet (monthly cost INR 120 for 30 mg therapy per day) and INR 0.53/tablet (monthly cost INR 63.6 for 20 mg therapy per day), respectively.[10]

Taking half-tablet of EMPA-25 significantly brings down monthly costs. However, no data are available on glycaemic efficacy and weight loss properties of half-tablet of EMPA-25. This study evaluated the real-world outcomes on weight loss and glycemia of different graded doses of EMPA (10 mg vs 12.5 mg vs 25 mg) along with all the other different anti-diabetes agents inT2DM.


The electronic medical record database of 2 different endocrinology centers in New Delhi was screened for the duration of the study was from January 2018 to April 2019. Patients with T2DM of more than 35 years-age were considered. Patients with associated severe chronic co-morbid states like chronic liver disease (Child's B or C), renal disease (glomerular filtration rate < 60 mL/min as calculated by CKD-EPI formula), cardiac disease (including coronary artery disease and heart failure), malignancies, active infection (tuberculosis, HIV, viral hepatitis), post organ transplant, patients on psychiatry medications, and those with chronic autoimmune disorders (lupus, scleroderma), were excluded. Furthermore, patients with a history of hospital admission in the past 6 months were also excluded. Patients with prior use of SGLT2 inhibitors were excluded. Incomplete records were excluded from the analysis. Details of other medications being used as per standard care were noted. Details of other anti-diabetes medications being used were noted. Patients with at least 6-months follow-up available were included. The study was approved by the institutional ethics committee. Data on anthropometry, glycemia, lipid parameters and spot urine albumin creatinine ratio (ACR) as a measure of microvascular complication at baseline and after 6 months follow up. Information was noted with regards to occurrence of different adverse drug reactions (ADRs), specifically, hypoglycaemia, genital infections, complicated upper urinary tract infections, fractures, euglycemic ketosis, and any other event reported by the patients. The entire flow of patient recruitment is elaborated in [Figure 1]. All patients had dedicated sessions with a dietician for ½ h during their initial visit and were counseled regarding low carbohydrate hypo-caloric diet.{Figure 1}

Patients were put into one of three groups depending on their EMPA dose: Group 1 were on EMPA 10 mg/day (1 tablet of EMPA-10), Group 2 were on EMPA 12.5 mg/day (half tablet of EMPA-25), and Group 3 were on EMPA 25 mg/day (1 tablet of EMPA-25).

Patients in Group-2 (EMPA 12.5 mg/d) were given a demonstration how to split the EMPA-25 tablet. EMP-25 tablet is relatively a flat elongated tablet making splitting easier. Using preferably a pill-cutter, the patients were shown how to cut the EMPA-25 tablet into two halves. In case a pill-cutter was not available, the patients were explained that a small kitchen knife can also be used to cut the tablet into 2 equal halves. The patients were asked to keep one of the halves in the tablet package carefully to be used the next day. They were reassured that sometimes the halves may not be exactly from the middle, but since the patient himself/herself will only be taking the other half of the tablet the subsequent day, it would average out and result in an overall intake of 15 EMPA-25 tablets over a period of 30 days.

Statistical analysis

Data analysis was performed using the Statistical Package for Social Sciences (SPSS) version 20.0 (IBM, Chicago, Illinois, USA). Kolmogorov–Smirnov test was used to check the normality of the variable distribution. Normally distributed variables were expressed as mean ± standard deviation. Skewed variables were expressed as median (25th–75th percentile). ANOVA was used for comparing three or more study groups. Chi-square test was used for categorical variables. A priori alpha of P < 0.05 considered statistically significant.


A total of 3601 medical records were screened, of which 1239 patients were on EMPA. Data from 599 patients, who fulfilled all inclusion and exclusion criteria, and for which at least 6-month follow-up available was analyzed. Of 599 patients, 184, 239 and 176 patients were on EMPA 10 mg/d (Group-1), EMPA 12.5 mg/d (Group-2) and EMPA 25 mg/d (Group-3) respectively. The details demographic, anthropometric, glycemic, metabolic, and medication profile of patients are elaborated in [Table 1].{Table 1}

The patients in all 3-groups (Group 1-3) were comparable with regards to sex distribution, systolic blood pressure, hemoglobin, renal function (creatinine), lipid parameters, and microvasculature damage (ACR) [Table 1]. The groups were comparable with regard to use of different anti-diabetes medications [Table 1]. However, the patients in Group-3 were significantly older, had the longest duration of diabetes, had the lowest body weight and body mass index (BMI) as compared to patients in Group-1 and 2 [Table 1]. Baseline HbA1c was highest in Group-3, followed by Group-2 and 1, which was statistically different [Table 1].

Following 6-months treatment, both absolute and percent weight loss was highest in Group-3, followed by Group-2 and Group-1 (P = 0.083 and 0.07, respectively) [Table 1]. Group-1 patients virtually had no weight loss. It must be highlighted that Group-3 patients lost the maximum weight, in spite of having the lowest absolute weight and BMI of all the groups, highlighting the additional weight loss potential of higher doses of EMPA. In terms of glycemic efficacy, the fall in HbA1c after 6-months therapy was not statistically different among groups [Table 1]. Similar HbA1c in all groups after 6-months of treatment, in the background of higher baseline HbA1cin Group-3, points toward slightly higher glycaemic efficacy of EMPA-25. A greater percent of patients in Group-2 achieved Hba1c < 5.7% as compared to other groups [Table 1].

Post-hoc analysis between each of the groups re-confirmed this observation. Both absolute and percent weight loss was significantly higher inGroup-3 as compared to Group-1 [Table 2]. The absolute and percent weight loss was similar in Group-2 and 3 [Table 3]. Patients in Group-2 continued to have a significantly higher absolute weight loss and a higher percent weight-loss, which approached statistical significance when compared to Group-1 [Table 4]. This highlights the superiority of EMPA 12.5 mg/day and 25 mg/day over 10 mg/day with regards to weight loss.{Table 2}{Table 3}{Table 4}

The medication profile, glycemic efficacy, weight loss properties of Group-2 and Group-3 was very similar and higher than that of Group-1 [Table 3]. Patients in Group-3 were older, had a higher baseline HbA1c, and lower body weight. Hence, a separate analysis of the study outcomes was done for the patient receiving EMPA 10 mg/d vs. those receiving EMPA 12.5 mg/d or 25 mg/d [Table 5]. Patients receiving EMPA 12.5 or 25 mg/d were significantly older and had a significantly higher total and percent weight loss (in the background of comparable baseline BMI), and significantly higher number of people achieving HbA1c < 5.7% [Table 5]. There were 3 reports of severe hypoglycemia, necessitating a visit to the hospital emergency department, 41 reports (6.84%) of mild self-limiting hypoglycemia, 27 reports (4.51%) of mild lower genital infection, and 2 reports of upper urinary tract infection involving the kidneys. ADR occurrence was comparable across groups [Table 1].{Table 5}


A meta-analysis published in 2014, which included ten studies with 6203 participants on EMPA 10 mg or 25 mg per day, demonstrated the mean change in HbA1C was 0.62% for EMPA 10 mg and 0.66% for EMPA 25 mg,[11] highlighting glycaemic efficacy was comparable for low and high doses. Pharmacodynamics studies have demonstrated that the quantum of increased urine glucose excretion with EMPA-10 and EMPA-25 is on an average 64 g/d and 78 g/d, respectively.[12] Further, short-term 8-day use of graded dose of EMPA from 10 to 100 mg/d in people with T2DM lead to only a marginal increase in urine glucose excretion from 77.9 gm/d to 89.8 g/d.[12]

This study highlighted for the first time the significantly better weight loss potential of EMPA 12.5 mg/d over 10 mg/d, and comparable to that of 25 mg/d. This makes half tablet of EMPA-25 a very attractive and cost-effective treatment option in diabesity. In different clinical trials also, EMPA-25 consistently had better weight loss outcomes as compared to EMPA-10, when used alone, or in combination with other medications.[13] In terms of glycaemic efficacy, HbA1c reduction was comparable over 3 months for all the 3 doses of EMPA evaluated in this study. However, a higher baseline HbA1c in patients on EMPA 25 mg/d points towards a trend of marginally better glycaemic efficacy of EMPA-25. Also a greater number of patients on EMPA 12.5 mg/d or 25 mg/d achieved HbA1c < 5.7% (38.8 mmol/ml) as compared to EMPA 10 mg/d. Hence even from glycemic point of view EMPA 12.5 mg/d scores over 10 mg/d, and compares favorably with 25 mg/d (cost perspective). In different clinical trials, the glycaemic efficacy of EMPA-10 and EMPA-25 though not statistically significantly different when used along with other diabetes medications, it tended to be slightly better in some scenarios for EMPA-25.[13]

In terms of improving cardiovascular outcomes in T2DM, both EMPA-10 and EMPA-25 had similar outcomes.[1],[13] In a more recent meta-analysis involving data from 15 RCTs (5374 in EMPA group and 2517 in control group), both EMPA-10 and EMPA-25 were demonstrated to have similar, weight-loss property, glycaemic efficacy and side-effect profile.[14]

T2DM medications are taken probably lifelong, which turns out to be for 30–40 years in India, where the peak incidence is in 30s–40s.[1] Lack of affordability is the most common cause for poor compliance to diabetes medications, resulting in greater glycaemic variability, worse glycaemic control, and over the years increased burden of end-organ damage. “Tablet splitting” not something new and has been in practice for a long time across the world. Freeman et al. in a review of PubMed (1966-June 2011) and International Pharmaceutical Abstract (1975-June 2011) found 17 studies dealing with different clinical outcomes, patient acceptance, or economic benefits of “tablet splitting.”[15] Patients with chronic disorders, which often needed life-long therapy, were most commonly doing “tablet-splitting” viz those on statins, anti-hypertensive medications, and anti-psychotics. Their main conclusion was “tablet splitting” did not seem to effect clinical outcomes.[15] The authors' personal observation are that tablet splitting is commonly practiced in India with regards to diabetes medications, as especially with relatively costlier medications like SGLT2 inhibitors.

The limitations of this study include the lack of matching of study groups at baseline, especially with regards to age and body weight. These are limitations intrinsic to real-world studies, where matching and randomization is not possible. Hence, we have focussed on percent weight loss and not absolute weight loss, which would not be affected by the baseline weight/BMI. Other limitations include the short study period of 6 months, making it difficult to assess long-term weight loss, and the lack of robust data on adherence. However, as a department policy, we always encourage our patients to carry medicines with them whenever they come for visits to the outpatient department (for checking and verification) and collect back empty packs of medicines from patients during these follow-up visits to ensure good compliance of medication intake.

To summarize, this study, for the first time, documents the glycemic efficacy, durability, and weight loss potential of half tablet of EMPA-25 taken once a day over a period of 6 months. Half tablet of EMPA-25 has comparable glycaemic efficacy with full tablet of EMPA-10 and full tablet of EMPA-25. Half tablet of EMPA-25 has a better weight loss profile than full tablet of EMPA-10 and similar to that of full tablet of EMPA-25. Half EMPA-25 is the most cost-effective way of using EMPA in clinical practice.


First report to highlight the glycaemic efficacy and weight loss with half tablet of empagliflozin-25 mg (EMPA25) (12.5 mg/d) in diabetesHalf-tablet of EMPA25 (cost INR 28.5) is cheaper than 1-tablet EMPA 10 mg (EMPA10) (165% higher cost) and costs half of full EMPA25Glycaemic efficacy of EMPA 12.5 mg/d was similar to 10 mg/d and 25 mg/dPatients on half EMPA25 had significantly higher total and percent weight loss,Patients on half EMPA25 hadsignificantly higher people achieving HbA1c < 5.7% (38.8 mmol/ml) compared to EMPA10.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Dutta D, Mukhopadhyay S. Intervening at prediabetes stage is critical to controlling the diabetes epidemic among Asian Indians. Indian J Med Res 2016;143:401-4.
2Dutta D, Choudhuri S, Mondal SA, Mukherjee S, Chowdhury S. Urinary albumin : creatinine ratio predicts prediabetes progression to diabetes and reversal to normoglycemia: Role of associated insulin resistance, inflammatory cytokines and low vitamin D. J Diabetes 2014;6:316-22.
3Mondal SA, Dutta D, Kumar M, Singh P, Basu M, Selvan C, et al. Neck circumference to height ratio is a reliable predictor of liver stiffness and nonalcoholic fatty liver disease in prediabetes. Indian J Endocrinol Metab 2018;22:347-54.
4Deepa M, Grace M, Binukumar B, Pradeepa R, Roopa S, Khan HM, et al. CARRS Surveillance Research Group. High burden of prediabetes and diabetes in three large cities in South Asia: The Center for cArdio-metabolic Risk Reduction in South Asia (CARRS) Study. Diabetes Res Clin Pract 2015;110:172-82.
5Singla R, Garg A, Singla S, Gupta Y. Temporal change in profile of association between diabetes, obesity, and age of onset in urban India: A brief report and review of literature. Indian J Endocrinol Metab 2018;22:429-32.
6Dutta D, Jaisani R, Khandelwal D, Ghosh S, Malhotra R, Kalra S. Role of metformin, sodium-glucose cotransporter-2 (SGLT2) inhibitors, glucagon-like peptide-1 (GLP-1) receptor agonists, and orlistat based multidrug therapy in glycemic control, weight loss, and euglycemia in diabesity: A real-world experience. Indian J Endocrinol Metab 2019;23:460-7.
7Lean ME, Leslie WS, Barnes AC, Brosnahan N, Thom G, McCombie L, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT): An open-label, cluster-randomised trial. Lancet 2018;391:541-51.
8New Delhi, India. Empagliflozin 10 mg Tablet Online Purchase. Available from: https://www. mg-tablet-322717. [Last accessed on 2019 Oct 01].
9New Delhi, India. Empagliflozin 25 mg Tablet Online Purchase. Available from: https://www. mg-tablet-322816. [Last accessed on 2019 Oct 01].
10Sharma M, Kumar M, Dutta D.Hydroxychloroquine in diabetes and dyslipidaemia: Primum nonnocere. Diabet Med 2020;37:1404-5.
11Liakos A, Karagiannis T, Athanasiadou E, Sarigianni M, Mainou M, Papatheodorou K, et al. Efficacy and safety of empagliflozin for type 2 diabetes: A systematic review and meta-analysis. Diabetes Obes Metab 2014;16:984-93.
12Heise T, Seman L, Macha S, Jones P, Marquart A, Pinnetti S, et al. Safety, tolerability, pharmacokinetics, and pharmacodynamics of multiple rising doses of empagliflozin in patients with type 2 diabetes mellitus. Diabetes Ther 2013;4:331-45.
13Frampton JE. Empagliflozin: A review in type 2 diabetes. Drugs 2018;78:1037-48.
14Zhang YJ, Han SL, Sun XF, Wang SX, Wang HY, Liu X, et al. Efficacy and safety of empagliflozin for type 2 diabetes mellitus: Meta-analysis of randomized controlled trials. Medicine (Baltimore) 2018;97:e12843.
15Freeman MK, White W, Iranikhah M. Tablet splitting: A review of the clinical and economic outcomes and patient acceptance. Second of a 2-part series. Part 1 was published in May 2012 (Consult Pharm 2012;27:239-53). Consult Pharm 2012;27:421-30.