Safety of Empagliflozin in Patients with Type 2 Diabetes Mellitus in Saudi Arabia: A Post-Authorisation Safety Study ()
1. Introduction
Type 2 diabetes mellitus (T2DM) is one of the most common chronic metabolic disorders affecting people within the age group of 20 to 79 years, with a global prevalence rate of 9.3% [1] [2] . According to the International Diabetes Federation report, 537 million people are currently living with diabetes worldwide, while 73 million people are affected in the Middle East and North Africa region, with adult prevalence rate in Saudi Arabia being 17.7% [3] . Saudi Arabia has the second-highest prevalence of T2DM in the Middle East and stands seventh in terms of global prevalence. It represents a major public health concern due to associated high levels of morbidity and mortality [4] [5] [6] [7] [8] .
Diabetes mellitus is characterised by elevated levels of blood glucose, attributable to the malfunctioning of the feedback loops between insulin action and insulin secretion, thereby culminating into a metabolic imbalance [2] . Pharmacological interventions for T2DM have witnessed a paradigm shift from conventional biguanides (metformin), thiazolidinediones and insulin secretagogues (sulphonylureas and meglitinide analogues) to novel sodium-glucose co-transporter type 2 (SGLT2) inhibitors, incretin mimetics (dipeptidyl peptidase-4 [DPP-4] inhibitors, and glucagon-like peptide 1 receptor agonists) [9] [10] [11] . However, the cornerstone of T2DM management is lifestyle modification, exercise, cessation of smoking, healthy diet and weight management [12] .
An important drug category for T2DM used in the recent past is SGLT2 inhibitors that act by diminishing the blood glucose level via partial blockage of glucose reabsorption from the kidney back to the bloodstream, thereby improving the diabetes control [1] . It is also evident that SGLT2 inhibitors improve the renal and cardiovascular outcomes [13] . Several studies have reported that SGLT2 inhibitors are well tolerated and exhibit dose-dependent effectiveness, particularly empagliflozin [1] [14] [15] [16] [17] [18] . Empagliflozin is being widely used as a monotherapy or as an add-on combination therapy; however, it is associated with certain adverse events (AEs) such as urinary tract infection (UTI), diabetic ketoacidosis (DKA) and hypoglycaemia [17] [19] [20] [21] . Based on its safety and effectiveness profile, the use of empagliflozin could be paramount during the Ramadan period as this period showcases the aggravation of several risk factors in patients with T2DM, including hypoglycaemia, hyperglycaemia, DKA, dehydration and thrombosis [10] [22] .
Therefore, parallel to the launch of SGLT2 inhibitors in Saudi Arabia, post-authorisation safety study (PASS) was conducted upon the request of Saudi Food and Drug Authority (SFDA) to monitor the safety aspects of empagliflozin in patients with T2DM.
2. Patients and Methods
2.1. Study Design and Participant Selection
This local, comparator, non-interventional, regulatory post-marketing study was conducted to monitor the safety of empagliflozin in patients with T2DM. “New users” design was used wherein comparison was made between empagliflozin (10 or 25 mg) versus DPP-4 inhibitors. A total of 25 sites participated in the study, of which only 18 sites enrolled a total of 1502 patients. The details of the study sites have been incorporated in supplementary Table S1. The study was conducted from 2018 to 2020; each patient was followed up for 12 months after the index period, inclusive of Ramadan period. The inclusion criteria for patients in both groups, namely empagliflozin and DPP-4 inhibitors, were primary diagnosis of T2DM performed in Saudi Arabia, having signed informed consent form, at least 18 years of age at the index period, not used other SGLT2 or DPP-4 inhibitors during the previous 12 months of the index period. The exclusion criteria included known hypersensitivity or contraindication to empagliflozin or DPP-4 inhibitor, or any other excipient and the patient being prescribed fixed dose combination (FDC) of SGLT2 inhibitors with DPP-4 inhibitors.
Index period was defined as the date on which each identified new user received the index prescription for empagliflozin or DPP-4 inhibitor.
Ramadan period was defined as the first day of Ramadan to 29th day of Ramadan based on the Islamic Hijri calendar; for 2019: 05 May to 04 Jun. 2019 (±1 to 2 days); for 2020: 23 Apr to 23 May 2020 (±1 to 2 days).
2.2. Ethical Considerations
The ethics committee approval was obtained from the Medical Services General Directorate, Research and Ethics Committee, Western Region, Saudi Arabia for 23 sites out of 25 (IRB Registration No. H-02-T-078). One site was approved by IRB, King Fahad Medical Centre (IRB Registration No. H-01-R-012) while another site was approved by Research Center, International Medical Center (IRB Registration No. 2018-12-102). The study was conducted in accordance with the Declaration of Helsinki. The patients signed the informed consent form before enrolling in the study. The study was conducted in accordance with the laws, regulations, and relevant guidelines of Saudi Arabia. (NCT03764631)
2.3. Methodology
All patients were followed up for 12 months after the index period (inclusive of Ramadan period) or until any of the specified conditions were met (death, specific exclusion criteria, the last continuous treatment of the index drug [empagliflozin or DPP-4 inhibitor], or new treatment episode started with any of the other index drugs). All the patients attended 4 follow-up visits at baseline, Week 16 ± 2, Week 32 ± 2 and Week 52. The investigators assessed and recorded the preliminary study parameters such as medical history, glycated haemoglobin (HbA1c), blood glucose level, renal functions, physical examinations, concomitant mediations, comorbidities, prior medication exposure, AEs, hospitalisation, etc. at successive visits, wherever applicable, into the electronic case report forms.
The incidence and time of first occurrence of DKA, severe UTI, volume depletion and dehydration were the primary outcomes and were evaluated in the index period (Diabetes ketoacidosis is defined as a serious complication of diabetes characterised by high level of ketones in the body due to lack of insulin and low food intake; severe UTI is defined as pyelonephritis or urosepsis; volume depletion is defined as the reduction in the extracellular fluids; and dehydration is defined as the loss of total body water that leads to hypertonicity). The secondary outcomes were diabetes-associated complications evaluated during the Ramadan period. Metformin, insulin, and treatment complexity status, along with analysis of AEs were conducted across both the treatment groups at the index and follow-up period along with Ramadan period.
2.4. Analysis
The sample size of 1500 patients was planned as per the SFDA requirements; as per regulations to detect DKA, 750 patients need to be enrolled in each group. Propensity scores were used for the quantitative analysis of probability of receiving empagliflozin at the index date for new users of both empagliflozin and DPP-4 inhibitors. Sensitivity analysis was performed to evaluate potential bias and confounding. The data collected were represented as arithmetic means, standard deviations (SDs), medians, minimum and maximum values for continuous data and confidence intervals (CI), wherever applicable. All statistical tests were conducted with a 2-sided significance level α of 0.05 and analysis was performed using statistical analysis software (SAS®) version 9.4.
3. Results
3.1. Study Population
Overall, 1502 patients were enrolled in the study, of which 93.7% (n = 1408) completed 1 year of treatment and their treatment status was still ongoing at the end of study. Of the remaining patients, 5.7% (n = 86) terminated the study early due to loss to follow-up (4.9%, n = 74); switching of treatment (0.4%, n = 6); AE-related termination (0.1%, n = 1); and reasons not specified (0.3%, n = 5). The data were unavailable for 0.5% of patients (n = 8) after completion of 1 year of treatment.
Overall, 64.2% (n = 965) of the enrolled patients were male and the mean age of the enrolled patients at index date was 52.5 ± 10.5 years (Table 1). The overall median duration of T2DM at index period was 53 months and more than 60% (n = 982) of the patients had a family history of T2DM. The most common medical condition across both the groups was metabolism and nutritional disorders (36.6%, n = 550), followed by vascular disorders (20.3%, n = 305) (with respect to System Organ Class [SOC]), while hypertension (20%, n = 301) followed by hyperlipidaemia (14.6%, n = 220) and dyslipidaemia (11.9%, n = 178) were the primary conditions as per the preferred term [PT]). More than 90% of the patients reported no prior insulin use. The mean HbA1c, fasting blood glucose (FBG), complete blood count (CBC), liver function tests, renal function parameters and lipid profile were comparable across the 2 treatment groups. Majority of the patients (81.3%, n = 1221) were on prior medications, with 77.8% (n = 1168) being on anti-hyperglycaemic agents and 19.5% on lipid-modifying agents (19.5%, n = 293).
3.2. Primary Outcomes
Only 0.1% (n = 1) of patient in both the treatment groups reported the occurrence of DKA; the first occurrence in the index period in the empagliflozin and the DPP-4 inhibitor groups was 365 days and 344 days, respectively. However, 0.8% (n = 6) patients in the empagliflozin group and 0.9% (n = 7) patients in the DPP-4 inhibitor group (total: <1%, n = 13) exhibited the occurrence of volume depletion; the time to first occurrence in the index period in empagliflozin and DPP-4 inhibitor groups was 169 ± 131 days and 256.1 ± 110.5 days, respectively. The treatment-adjusted incidence rate ratio of DKA and volume depletion in empagliflozin/DPP-4 inhibitors groups was 0.981 and 0.841, respectively (Table 2). None of the patients reported the occurrence of UTI and dehydration across both treatment groups. Similar result was observed in the follow-up period.
3.3. Secondary Outcomes
None of the patients in the 2 treatment groups had the incidence of DKA, severe UTI, volume depletion or dehydration in the index and follow-up periods during Ramadan.
3.4. Metformin Status at Baseline
Out of 751 patients in each treatment group, 271 patients in the empagliflozin group and 331 patients in the DPP-4 inhibitors group were on metformin FDC therapy at baseline. The occurrence of DKA in those on metformin was reported by 0.4% and 0.3% (n = 1 in each case) of patients in the empagliflozin and DPP-4 inhibitor groups, respectively. Out of the 13 patients (<1%) reporting volume depletion, 1.1% (n = 3) were on empagliflozin and 1.2% (n = 4) were on DPP-4 inhibitor; metformin was used as FDC in both cases at baseline (Table 3). None of the patients on metformin FDC at baseline reported the incidence of DKA in the index period. No patients across both the treatment groups reported the incidence of severe UTI or dehydration in the index period. Similar results were obtained during the follow-up period.
Abbreviation: DPP-4: Dipeptidyl peptidase-4.
![]()
Table 2. Incidence of diabetic ketoacidosis and volume depletion in the index period.
Abbreviation: DKA: Diabetic ketoacidosis; DPP-4: Dipeptidyl peptidase-4. Note: The mean ± SD duration of the index period and follow-up period was 350.6 ± 66.7 days and 351 ± 65.7 days, respectively. Crude incidence rate is the proportion of patients with events in the respective period.
![]()
Table 3. Incidence of diabetic ketoacidosis and volume depletion in index period by metformin status.
Abbreviation: DKA: Diabetic ketoacidosis; DPP-4: Dipeptidyl peptidase-4.
None of the patients across both treatment groups had the incidence of DKA, UTI, volume depletion and dehydration in the index and follow-up periods during Ramadan.
3.5. Treatment Complexity Status at Baseline
The patients in each treatment group were divided based on the treatment complexity status: on mono glucose-lowering drug (GLD) treatment (empagliflozin group: 116 patients; DPP-4 inhibitor group: 113 patients), dual GLD treatment (empagliflozin group: 343 patients; DPP-4 inhibitor group: 393 patients), triple GLD treatment (empagliflozin group: 231 patients; DPP-4 inhibitor group: 209 patients), quadruple GLD treatment (empagliflozin group: 43 patients; DPP-4 inhibitor group: 29 patients), quintuple GLD treatment (empagliflozin group: 12 patients; DPP-4 inhibitor group: 6 patients), sextuple GLD treatment (empagliflozin group: 4 patients; DPP-4 inhibitor group: 1 patients), septuple GLD treatment (empagliflozin group: 1 patient; DPP-4 inhibitor group: 0 patients) and octuple GLD treatment (empagliflozin group: 1 patient; DPP-4 inhibitor group: 0 patients).
It was reported that 2.3% (n = 1) and 0.3% patients (n = 1) reported DKA in the empagliflozin group and DPP-4 inhibitor groups while being on quadruple GLDs and dual GLDs therapy, respectively, at the baseline. All 13 patients (<1%) reporting volume depletion was on GLD combinations at baseline (dual, triple, quadruple, and quintuple (Table 4). None of the patients in either of the 2 treatment groups had the incidence of severe UTI or dehydration in the index period. Similar results were reported in the follow-up period.
None of the patients in 2 treatment groups had the incidence of DKA, severe UTI, volume depletion or dehydration in the index and follow-up periods during Ramadan.
![]()
Table 4. Incidence of diabetic ketoacidosis and volume depletion in index period by glucose-lowering drug status.
Abbreviation: DKA: Diabetic ketoacidosis; DPP-4: Dipeptidyl peptidase-4; GLD: Glucose-lowering drugs.
3.6. Insulin Use at Baseline
The primary outcomes were evaluated based on the use of insulin at baseline, wherein 50 patients in the empagliflozin group and 26 patients in the DPP-4 inhibitors group confirmed the usage.
The occurrence of DKA was reported in 2% of the patients (n = 1) in the empagliflozin group using insulin and 0.1% (n = 1) of the patient on DPP-4 inhibitor without insulin at baseline. Out of the 13 patients (<1%) reporting volume depletion, only 2% of the patients (n = 1) in the empagliflozin group confirmed the use of insulin (Supplementary Table S2). None of the patients in the 2 treatment groups had the incidence of severe UTI or dehydration in the index period. Similar results were reported in the follow-up period.
None of the patients in the 2 treatment groups had the incidence of DKA, severe UTI, volume depletion or dehydration in the index and follow-up periods during Ramadan.
3.7. Adverse Event Summary
Overall, 8.1% (n = 121) of patients reported AEs, wherein 7.7% of patients (n = 58) were treated with empagliflozin and 8.4% of the patients (n = 63) with DPP-4 inhibitor in the index period. A total of 0.3% of the patients (n = 4) reported AEs in the index period during Ramadan, where 0.4% (n = 3) were from the empagliflozin group and 0.1% (n = 1) was treated with DPP-4 inhibitors. A detailed representation of different types of AEs is provided in Table 5.
![]()
Table 5. Classification of different adverse events observed.
Abbreviation: AE: Adverse event; AESI: Adverse event of special interest; DPP-4: Dipeptidyl peptidase-4; SAE: Serious adverse event. Note: The mean ± SD duration of the index period and follow-up period during Ramadan was 30.8 ± 3.0 days.
3.7.1. In Relation to SOC and PT
On further analysis based on the SOC and PT in the index period, maximum events reported were gastrointestinal disorders (26 events), followed by nervous system disorders (23 events) and general disorders and administration (20 events). With respect to PT, none of the AEs were present in >1% of the patients in the 2 treatment groups. In the index period during Ramadan, a total of 0.3% (n = 4) patients reported 4 AEs, where 0.4% (n = 3) of patients in the empagliflozin group reported 3 AEs (infections and infestations [corona virus infection], metabolism and nutrition disorder [hypoglycaemia] and renal and urinary disorder [chromaturia] [1 patient each]) and 0.1% (n = 1) patient in the DPP-4 inhibitors group reported 1 AE (gastrointestinal disorder [nausea]).
3.7.2. Incidence of AEs in Relation to Study Treatment
A total of 8 empagliflozin-related AEs were observed in 1.1% of patients (n = 8) (hypoglycaemia: 3; thirst: 2; fungal infection: 2; genital fungal infection:1). However, most of these events were mild, except for 1 event each of hypoglycaemia and fungal infection. During the Ramadan period, 0.4% (n = 1) patient reported 1 empagliflozin-related hypoglycaemia.
3.7.3. Incidence of Adverse Events of Special Interest
A total of 0.3% of patients (n = 4) reported 5 adverse events of special interest (AESIs) in the index period (empagliflozin group: 0.1% [n = 1] patient each of DKA and UTI; DPP-4 inhibitor group: 0.1% [n = 1] patient reported DKA, 0.1% [n = 1] patient reported 1 event each of acute kidney injury and end-stage renal disease).
3.7.4. Serious Adverse Events
A total of 0.5% (n = 8) of patients reported 10 serious AEs (SAEs) in the index period, where 6 patients were from the empagliflozin group, and 2 patients were from the DPP-4 inhibitor group. Of 6 patients reporting 7 SAEs in the empagliflozin group, 1 patient each reported 1 SAE of myocardial ischaemia, anal fissure, hernia, DKA and obesity, and 1 patient reported 2 SAEs (pneumonia and lung adenocarcinoma). Of the 2 patients treated with DPP-4 inhibitor, 3 SAEs were reported (1 patient reported 1 event of DKA and another patient reported 2 SAEs [acute kidney injury and end-stage renal disease]).
3.8. Other Variables
Comparable decline was reported in mean HbA1c levels from baseline to visits 2, 3 and 4 in both treatment groups. No major changes were observed in mean weight, body mass index, pulse rate, systolic blood pressure and diastolic blood pressure across different visits. A total of 8.3% patients (n = 1221) affirmed using concomitant medication, primarily anti-hyperglycaemic medications (empagliflozin group: 82.8% [n = 622]; DPP-4 inhibitor group: 79.8% [n = 599]), followed by lipid-modifying agents (empagliflozin group: 20.1% [n = 151]; DPP-4 inhibitor group: 20% [n = 150]); and agents acting on renin-angiotensin system (empagliflozin group: 14.2% [n = 105]; DPP-4 inhibitor group: 12.3% [n = 92]). Similar results were reported in the index and follow-up periods during Ramadan.
4. Discussion
Previous clinical studies have established the safety and effectiveness of SGLT2 inhibitors in significantly improving the glycaemic control along with exhibiting a well-tolerated safety profile [1] [14] [23] [24] [25] [26] [27] . The pharmacokinetic and exposure-response studies of empagliflozin substantiated no significant change in the risk of UTI, hypoglycaemia or volume depletion [15] and a well-documented favourable benefit-risk profile was established [28] . Empagliflozin has been incorporated as an add-on to metformin therapy and as monotherapy, it is well tolerated in both cases [19] [21] [29] . However, certain studies have shed light on hypoglycaemic AEs, UTIs and volume depletion across treatment groups using SGLT2 during different studies [21] [30] .
The association of T2DM with DKA was established in 41.4% of the patients with T2DM in a hospital-based retrospective analysis, thereby highlighting the risk factor in patients with T2DM [31] . A separate study presented DKA as a major concern in using SGLT2, where hospitalisation was observed in 4 patients, thereby reinforcing the need to establish the safety profile of these medications [32] . It was studied that DKA is prevalent in almost one-third of the patients presenting with T2DM, causing an increase in the hospitalisation and economic burden [33] [34] . In a post-authorisation safety and effectiveness study of dapagliflozin in Saudi Arabia, 12.3% of the patients with T2DM reported the AEs, with UTI and DKA at 1% and 0.2% of all the events, respectively [14] . Another paramount concern in patients with T2DM is UTI that warrants a routine urine analysis. It was established that patients with T2DM more than 40 years of age and female patients in particular are at a higher risk of asymptomatic UTI, although the overall prevalence of UTI in T2DM is nearly 11.5% [35] [36] [37] [38] . A retrospective analysis of tofogliflozin, an SGLT2 inhibitor, exhibited significant glucose-lowering competencies with low associated risk of electrolyte abnormalities and dehydration [39] . Ertugliflozin, an SGLT2 inhibitor, did not lead to significant volume depletion (1.2% - 1.9%; dose-dependent) as reported in the randomised double-blind trial [40] . The key trends observed in the current study using empagliflozin were in corroboration with the previously published clinical evidence, where less than 1% of the patients reported T2DM-associated complications (0.1% patients [n = 1] reported DKA each from empagliflozin and DPP-4 inhibitor group; <1% patients reported volume depletion across both the groups). None of the patients reported UTI or dehydration. Thus, the obtained results reinforced the safety profile of empagliflozin.
The medical ramifications of fasting during Ramadan are still unexplored. However, because of the fasting in patients with T2DM, insulin secretion is perturbed, resulting in T2DM-associated complications such as hypoglycaemia, hyperglycaemia, DKA, dehydration and thrombosis along with a significant lowering in the blood lipid profile, blood pressure, HbA1c level [10] [22] [41] . A meta-analysis reported that there was an improvement in the HbA1c and hypoglycaemic control when patients with T2DM were treated with SGLT2 during Ramadan, with no major AEs [42] . The DIA-RAMADAN, a real-world study and VIRTUE study evaluated the safety profile of anti-diabetic treatment during the Ramadan period which reported reductions in HbA1c, weight control and fasting plasma glucose [22] [43] . The outcome of the present study with empagliflozin confirmed that none of the patients reported any incidence of DKA, UTI, volume depletion or dehydration during the Ramadan period, which was in line with the established literature.
Thus, the present study provides insights on the safety aspect of empagliflozin, an SGLT2 inhibitor, which was exhibited to be safe during the 1-year period in Saudi Arabia, parallel to its launch in the region. However, we do acknowledge certain limitations of the study. The incidences of DKA, severe UTI, volume depletion and dehydration were not analysed across different age groups, male versus female, duration of disease and varying doses of study treatment. The sample size was determined as per the SFDA requirement, which could be received as limited considering the anticipated incidences of outcomes. Laboratory investigations (CBC, lipid profile, and liver and renal function tests) were not performed for subsequent visits, which further limited the assessment of the impact of study treatment on laboratory parameters.
The strength of the study lies in the fact that it brings out the safety data specific to Saudi Arabia for empagliflozin in patients with T2DM as a regulatory mandate. Furthermore, a PASS for T2DM encompassing and comparing the diabetes-associated complications in the index period and Ramadan period in a single study is practically unavailable. Thus, this study might form a foundation for future follow-up studies on an elaborate scale to yield more profound outcomes.
5. Conclusion
The study evinced empagliflozin in reducing blood glucose levels in patients with T2DM in Saudi Arabia when administered as per the local prescribing information. The outcome was also evaluated during the Ramadan period when all the safety and tolerability parameters were adequately met. Overall, no safety concerns were recognised during the study, thereby underpinning the favourable benefit/risk ratio of empagliflozin. Currently, no update on the empagliflozin prescribing information is deemed necessary. Nonetheless, extensive follow-up studies are warranted to address the study limitations and further complement the study outcomes.
Acknowledgements
The authors would like to thank Bushra Nabi and Yukti Singh from IQVIA, India for their medical writing and editing support, and Joyson Kurian from Clintec, an IQVIA company for drafting of the report.
Funding
The study was funded by Boehringer Ingelheim.
Disclosure
Dr. Sameh Rakha is the employee of Boehringer Ingelheim.
Data Availability Statement
Data are available on request due to privacy/ethical restrictions. The data that support the findings of this study are available on request from the corresponding author, Dr. Saud Alsifri. The data are not publicly available due to restrictions (containing information that could compromise the privacy of research participants).