The Effects of Sodium–Glucose Cotransporter 2 Inhibitors on Body Composition in Type 2 Diabetes Mellitus: A Narrative Review

Abstract

Body composition is related to cardiometabolic disorders and is a major driver of the growing incidence of type 2 diabetes mellitus (T2DM). Altered fat distribution and decreased muscle mass are related to dysglycemia and impose adverse health-related outcomes in people with T2DM. Hence, improving body composition and maintaining muscle mass is crucial in T2DM. Sodium-glucose cotransporter 2 (SGLT2) inhibitors are novel glucose-lowering medications gaining popularity because of their cardiorenal-protective effects and weight-lowering characteristics. However, reports on myopathy secondary to SGLT2 inhibitor treatment raised a safety concern. The importance of maintaining muscle mass in people with T2DM necessitates further investigation to explore the impact of novel medications on body composition. In this review, we discussed current evidence on the impact of SGLT2 inhibitors on body composition in people with T2DM.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13300-023-01481-7.

Key Summary Points

Why carry out this study?

Altered body composition plays a significant role in the pathogenesis and prognosis of type 2 diabetes mellitus.

Sodium–glucose cotransporter 2 (SGLT2) inhibitors are new glucose-lowering drugs that reduce body weight, but their effect on body composition remains uncertain.

This review aimed to investigate the association of SGLT2 inhibitors and body composition in type 2 diabetes mellitus.

What was learned from the review?

SGLT2 inhibitors reduce total body weight in people with type 2 diabetes mellitus. Reduction of body weight is primarily due to the loss of fat mass.

In long-term studies, loss of fat-free mass approximately contributes 35% of body weight reduction. This change is comparable to the body composition changes reported after lifestyle interventions and bariatric surgery.

The impact of SGLT2 inhibitors on body composition significantly differs from sulfonylureas and dipeptidyl peptidase 4 inhibitors, but it is similar to glucagon-like peptide 1 receptor agonists.

Introduction

Type 2 diabetes mellitus (T2DM) is a growing public health burden, affecting more than 400 million individuals worldwide [1]. The condition is associated with long-term adverse outcomes, namely cardiovascular disease (CVD), chronic kidney disease (CKD), as well as increased risk of mortality [2]. Modifiable risk factors including dietary habits and alterations in body composition influence the burden of T2DM [3]. In this context, the significance of body composition extends beyond measures such as body weight (BW) and body mass index (BMI). Other factors such as abnormal fat distribution, particularly visceral adiposity, and reduction of lean mass are more crucial [4, 5]. Thus, understanding the effects of therapeutic interventions on body composition in T2DM is clinically significant.

Sodium–glucose cotransporter 2 (SGLT2) inhibitors are a novel class of oral glucose-lowering medications that increase renal glucose excretion while posing little risk of hypoglycemia [6]. They also reduce blood pressure and have been associated with cardiorenal protection [7–9]. SGLT2 inhibitors have been associated with weight reduction through fat mass loss, but concerns have been raised regarding their potential impact on muscle mass [10]. Rare cases of sarcopenia and myopathy have been reported after using SGLT2 inhibitors, particularly in older individuals taking statins [11, 12]. However, recent animal studies revealed opposite trends, i.e., increased skeletal muscle mass and hand grip strength after treatment with SGLT2 inhibitors [13].

Considering the mentioned controversial findings, it is important to evaluate body composition changes associated with SGLT2 inhibitors. In this review we summarized current evidence and the key underlying mechanisms related to the effects of SGLT2 inhibitors on body composition, focusing on fat mass and fat-free mass.

Type 2 Diabetes Mellitus and Body Composition

Changes in body composition influence the risk of developing T2DM through a variety of mechanisms such as insulin resistance, metabolic dysfunction, and de novo lipogenesis in ectopic tissues [14, 15]. There is also a bidirectional interplay between T2DM and loss of muscle mass giving rise to a vicious cycle [16]. Sarcopenia, characterized as progressive age-related muscle mass and function decline, is more common in people with T2DM and is influenced mainly by insulin resistance [17, 18]. Sarcopenia contributes to poor glycemic control and increased risk of adverse outcomes such as frailty, particularly among the aging population [17, 19].

Impaired insulin action in skeletal muscle disrupts multiple pathways, thereby promoting protein degradation and muscle catabolism, while also hampering protein synthesis [20]. A chronic inflammatory state, oxidative stress, increased reactive oxygen species, and accumulation of advanced glycation end-products (AEGs) also exacerbate muscle loss in people with diabetes [21–23].

Effective weight management and improving body composition are crucial in the management of T2DM and achieving better glycemic control while minimizing diabetes-related chronic complications [24, 25]. Current guidelines emphasize on the importance of sustained weight loss and muscle mass restoration through individualized therapeutic approaches, dietary modifications, and regular exercise [14, 26–29].

Mechanism of Action of Sodium–Glucose Cotransporter 2 Inhibitors on Body Composition

SGLT2 inhibitors reduce renal glucose reabsorption, leading to decreased blood glucose levels independent of insulin release [30]. In a healthy adult, almost all of the filtered glucose is reabsorbed, primarily by SGLT2 proteins [31]. These proteins are expressed in the proximal convoluted tubules of the kidneys and actively transport glucose into the interstitium [32]. In individuals with diabetes, SGLT2 proteins are paradoxically upregulated, resulting in excessive glucose reabsorption despite high blood glucose levels [33]. Inhibition of SGLT2 transporters promotes renal glucose excretion and improves glucose homeostasis and insulin sensitivity. Additional metabolic benefits include an early diuretic effect and calorie loss in the urine [34], causing a significant weight reduction [6, 35]. Despite sustained urinary glucose excretion, compensatory adaptation mechanisms attenuate excess water and calorie loss associated with long-term SGLT2 inhibition. These mechanisms include upregulation of the renin–angiotensin–aldosterone system, increased appetite, and changes in energy expenditure/substrate utilization [36, 37].

SGLT2 inhibitors also induce gluconeogenesis while enhancing the uptake, utilization, and catabolism of fatty acids, leading to various effects on lipid metabolism, including glucagon release, ketogenesis, and free fatty acid oxidation and mobilization [38, 39]. Consequently, SGLT2 inhibitors are linked to alterations in body composition even during prolonged use.

Methods for Evaluating Body Composition

Body composition refers to the proportion of different tissues in the human body, such as fat, muscle, bone, and water [40]. Methods for analyzing body composition are categorized as direct (e.g., cadaveric) or indirect methods, with the latter estimating body composition using mathematical equations based on known parameters [41]. Traditionally, body composition is divided into fat mass and fat-free mass, but more detailed models further divide fat-free mass into water, proteins, and minerals [42] (Table 1).

Table 1.

Different compartments of body composition

Compartment	Definition
Fat mass	Mass of adipose tissue
Fat-free mass	Total body mass except fat mass
Lean body mass	Fat-free mass except for mineral content (bones)
Skeletal muscle mass	Lean body mass minus connective tissue, skin, and other organs

Simple anthropometric measurements such as BMI, waist circumference, and waist-to-height ratio offer quick assessments of body composition but lack detailed information [43, 44]. Computed tomography (CT) scan and magnetic resonance imaging (MRI) are the gold standard methods for evaluating body composition; however, high cost and low accessibility limit their widespread use [45].

Common methods for measuring body composition include bioelectrical impedance analysis (BIA) and dual-energy X-ray absorptiometry (DXA). BIA rapidly assesses body composition by measuring impedance after passing a small electrical current through the body. It estimates total body water, fat-free mass, and fat mass using single or multiple frequencies [46]. DXA has higher precision in the assessment of body composition. It uses X-ray imaging to measure bone density, fat mass, and lean mass [42]. Both DXA and BIA offer segmental assessments of body composition in regions like arms, legs, and trunk.

When interpreting the findings for clinical decision-making, it is crucial to consider the limitations of each technique. BIA results rely on a constant, population-based level of hydration and body fluid distribution, making them susceptible to other factors such as target population, stage of obesity, fluid overload, dehydration, and recent exercise prior to testing [47, 48]. DXA offers greater precision but requires trained technicians and uses estimation algorithms that may not be applicable to certain populations [49]. Additionally, caution should be exercised when using terms like fat-free mass and lean body mass interchangeably with skeletal muscle mass, as there are distinctions between these compartments (Table 1).

Methods

We searched PubMed/Medline, Scopus, and Web of Science from inception to July 2023 for studies that assessed changes in body composition (including fat mass and fat-free mass) among patients with T2DM treated with SGLT2 inhibitors. The search included Medical Subject Heading (MeSH) terms for “body composition” OR “muscle mass” OR “lean body mass” OR “anthropometric indices” AND “type 2 diabetes mellitus” AND “sodium-glucose cotransporter 2 inhibitors”. The following drug names were also included: “empagliflozin”, “dapagliflozin”, “Farxiga”, “Xigduo”, “canagliflozin”, “Jardiance”, “Glyxambi”, “ertugliflozin”, “Stegaltro”, “Steglujan”, “Qtren”, “ipragliflozin”, “sotagliflozin”, “Zynquista”, “tofogliflozin”, “luseogliflozin”, “licogliflozin”, “Novartis”, and “Phlorizin”. The search was limited to randomized clinical trials (RCT), and additional manual searching was performed.

The search yielded 1100 results, of which 1080 were excluded. Exclusion criteria were animal studies, duplicated studies, studies with unrelated topics, inappropriate study designs, inaccessible full text, and incomplete data. Finally, 20 articles were included for review in this study. Quality assessment of the included articles were conducted using National Institute of Health (NIH) Tool for Quality Assessment of Controlled Intervention Studies (available from www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools). The quality of studies were rated as “good”, “fair”, and “poor”.

To facilitate report and comparing the findings, we refer to changes in the lean body mass and skeletal muscle mass as the fat-free mass in this review, acknowledging the differences between these definitions.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Results and Discussion

We identified 20 RCTs investigating body composition change after treatment with SGLT2 inhibitors and details of the reviewed studies are presented in Tables 1 and 2. The method of body composition assessment is unspecified in one study [50]. Eleven studies used DXA [51–61], six studies used BIA [62–67], and two used a combination of DXA, BIA, and MRI [68, 69]. Also, studies had mostly fair to good qualities (supplementary material, Table S1).

Table 2.

Characteristics of clinical trials reporting changes in body composition with SGLT2is treatment

Study	Year, country	Population	Intervention	Sample size	Age (years)	Baseline BMI (kg/m2)	Concurrent medications
Bolinder et al. [53]	2012, multicenter	T2DM	Dapagliflozin 10 mg/day	91	60.6 ± 8.2	32.1 ± 3.9	Exclusive treatment with metformin
			Placebo	89	60.8 ± 6.9	31.7 ± 3.9
Bolinder et al. [52]	2014, multicenter	T2DM	Dapagliflozin 10 mg/day	69	60.6 ± 8.2	32.1 ± 3.9	Exclusive treatment with metformin
			Placebo	71	60.8 ± 6.9	31.7 ± 3.9
Blonde et al. [51]	2016, multicenter	T2DM	Canagliflozin 100 mg/day	63	64.3 ± 6.6	30.9 ± 4.8	Antihyperglycemic agents
			Canagliflozin 300 mg/day	73	63.0 ± 6.0	31.6 ± 4.3
			Placebo	75	64.2 ± 6.4	32.0 ± 5.5
Fadini et al. [62]	2017, Italy	T2DM	Dapagliflozin 10 mg/day	15	66.3 ± 1.8	28.4 ± 1.4	Oral glucose-lowering drugs or insulin
			Placebo	16	61.0 ± 1.8	32.8 ± 1.4
Inoue et al. [69]	2019, Japan	T2DM	Ipragliflozin 50 mg/day	24	60.5 ± 9.8	27.9 ± 4.0	Insulin alone or plus oral hypoglycemic agents
			Placebo	24	60.8 ± 12.1	27.7 ± 4.5
Chehregosha et al. [56]	2021, Iran	T2DM/NAFLD	Empagliflozin 10 mg	35	50.5 ± 8.4	30.9 ± 3.3	NR
			Pioglitazone	34	52.5 ± 7.9	29.4 ± 3.7
			Placebo	37	51.8 ± 7.8	30.2 ± 4.4
Lauritsen et al. [58]	2021, Denmark	T2DM	Empagliflozin 25 mg/day vs. placebo (crossover design)	13	62 ± 6	31.5 ± 5.0	Metformin
Horibe et al. [68]	2022, Japan	T2DM	Dapagliflozin 5 mg/day	26	59.7 ± 12.0	28.0 ± 4.0	Oral hypoglycemic agents other than SGLT2i
			Placebo	24	62.3 ± 6.5	27.6 ± 3.8
Brandt-Jacobsen et al. [54]	2023, Denmark	T2DM	Empagliflozin 25 mg/day	38	65.7 ± 9.1	32.8 ± 5.6	Unspecified glucose-lowering treatment
			Placebo	40	66.4 ± 8.7	30.3 ± 5.9
Nakaguchi et al. [60]	2020, Japan	T2DM	Empagliflozin 10 mg/day	31	66.3 ± 9.5	25.8 ± 4.1	Insulin
			Liraglutide	30	67.2 ± 9.0	26.4 ± 4.6
McCrimmon et al. [59]	2020, multicenter	T2DM	Canagliflozin 300 mg/day	90	58.6 ± 10.1	32.3 ± 5.5	Metformin
			Semaglutide	88	57.8 ± 9.9	32.6 ± 6.4
Cefalu et al. [55]	2013, multicenter	T2DM	Canagliflozin 100 mg/day	111	NR	NR	Metformin
			Canagliflozin 300 mg/day	102
			Glimepiride	96
Kitazawa et al. [64]	2020, Japan	T2DM	Tofogliflozin 20 mg/day	33	57.3 ± 11.4	25.3 ± 3.9	Metformin and DPP4 inhibitors
			Glimepiride	31	57.6 ± 9.3	25.4 ± 3.8
Wolf et al. [61]	2021, Brazil	T2DM	Dapagliflozin 10 mg/day	44	58 ± 7	30 (7)	Up to two oral hypoglycemic agents
			Glibenclamide	45	58 ± 7	30 (7)
Tsurutani et al. [50]	2018, multicenter	T2DM	Ipragliflozin 50 mg/day	60	53.5 ± 11.72	28.8 (6.3)	Patients with prior use of SGLT2is or incretin-related agents were excluded
			Sitagliptin	59	54.0 ± 10.7	28.5 (5.2)
Zeng et al. [67]	2022, Taiwan	T2DM	Empagliflozin 25 mg/day	46	58.9 ± 9.9	27.7 ± 5.0	Premixed insulin with or without OAD
			Linagliptin	51	58.7 ± 10.2	28.0 ± 3.5
Kato et al. [63]	2017, Japan	T2DM	Dapagliflozin 5 mg/day vs. control (Cross-over design)	Preceding group = 27	48.7 ± 11.5	30.3 ± 5.3	Insulin or antidiabetic drugs
				Following group = 29	49.4 ± 11.8	29.6 ± 4.9
Shimizu et al. [65]	2018, Japan	T2DM/NAFLD	Dapagliflozin 5 mg/day	33	56.2 ± 11.5	27.6 ± 4.7	Three OAD with or without insulin
			Control	24	57.1 ± 13.8	28.3 ± 3.5
Yamakage et al. [66]	2020, Japan	T2DM	Dapagliflozin 5 mg/day	27	58.4 ± 13.0	31.3 ± 7.6	Sulfonylureas, biguanides, alpha-glucosidase, DPP4 inhibitors, or their combination
			Control	27	60.7 ± 11.9	30.7 ± 6.2
Han et al. [57]	2020, Korea	T2DM	Ipragliflozin 50 mg/day	30	52.5 ± 10.3	30.4 ± 5.4	Metformin and pioglitazone combination for at least 8 weeks
			Control	15	56.7 ± 11.8	30.2 ± 2.5

T2DM type 2 diabetes, BMI body mass index, NR not reported, SGLT2i sodium–glucose cotransporter 2 inhibitors, DPP4 dipeptidyl peptidase 4, OAD oral antidiabetic drugs

Data are reported as mean ± standard deviation or median (interquartile range)

In the following sections, we present and discuss the findings based on the study duration and the comparators.

Effect of Treatment Duration on Body Composition

There were four studies with long-term duration (follow-up > 24 weeks). They showed that SGLT2 inhibitors are associated with a reduction of 2.5–4.5 kg in BW, 2–2.6 kg in fat mass, and 0.9–1.5 kg in fat-free mass, compared to the baseline. Notably, loss of fat-free mass comprised 25–36% of the total weight reduction [51, 52, 55, 59] (Table 3). In the remaining studies with shorter duration of follow-up, loss of fat-free mass ranged widely from 0 to 20% of total weight loss in three studies [61, 63, 66], 20–50% in seven studies [53, 54, 60, 65, 67–69], and more than 50% in three studies [57, 62, 64] (Fig. 1). In one study, the reduction in fat-free mass was greater than the overall weight loss, and the fat-free mass loss-to-BW loss ratio exceeded 1 [58].

Table 3.

Body composition changes following SGLT2 inhibitors treatment

Author, year, country		Assessment method	Duration (weeks)	Body composition indices in SGLT2is groupa (kg)
				Baseline BW	BW change	Baseline fat mass	Fat mass change	Baseline fat-free massb	Fat-free mass change
Compared to placebo
1	Bolinder et al. (2012), [53]	DXA	24	92.1	− 2.96*	33.6	− 2.22*	56.2	− 1.1*
2	Bolinder et al. (2014), [52]	DXA	102	92.1	− 4.54*	33.7	− 2.80	55.3	− 1.30
3	Blonde et al. [51]	DXA	26	100: 88.9 300: 93.2	− 2.5* − 3.2*	32.2 33.8	− 1.9* − 2.4*	51.2 53.2	− 0.9* − 1.2*
4	Fadini et al. [62]	BIA	12	NR	− 3.1*	NR	− 0.1	NR	− 2.9*
5	Inoue et al. [69]	DXA/BIA	24	72.34	− 2.78*	DXA: 23.29 BIA: 22.14	DXA: − 2.07* BIA: − 2.21*	DXA: 41.63 BIA: 47.14	DXA: − 0.6 BIA: − 0.56
6	Chehregosha et al. [56]	DXA	24	82.2	− 2.7*	NR	NR	NR	NR
7	Lauritsen et al. [58]	DXA	4	95.2	− 0.6	31.4	− 0.2	60.4	− 1.0*
8	Horibe et al. [68]	DXA/ BIA	24	73.29	− 2.40*	DXA: 25.43 BIA: 23.27	DXA: − 2.32* BIA: − 1.73*	DXA: 45.86 BIA: 47.92	DXA: − 0.17 BIA: − 0.64
9	Brandt–Jacobsen et al. [54]	DXA	13	97.1	− 1.40*	30.2	− 0.9*	67.4	− 0.49
Compared to GLP-1 receptor agonists
10	Nakaguchi et al. [60]	DXA	24	69.0	− 1.5	19.2	− 0.7	46.1	− 0.6
11	McCrimmon et al. [59]	DXA	52	87.6	− 4.1	32.5	− 2.62	51.3	− 1.48
Compared to sulfonylureas
12	Cefalu et al. [55]	DXA	52	100 mg: 84.4 300 mg: 85.9	− 4.4* − 4.2*	28.2 29.3	NR	47.7 44.6	− 0.9* − 1.1*
13	Kitazawa et al. [64]	BIA	24	67.0	− 2.0*	19.4	− 0.7*	47.6	− 1.3*
14	Wolf et al. [61]	DXA	12	81.6	− 2.7*	29.9	− 2.0*	51.3	− 0.34*
Compared to DPP4 inhibitors
15	Tsurutani et al. [50]	NR	12	NR	− 2.2*	NR	NR	NR	NR
16	Zeng et al. [67]	BIA	24	71.4	− 1.55*	20.8	− 1.02	46.5	− 0.44*
Compared to conventional treatment
17	Kato et al. [63]	BIA	12	Grp 1: 80.0 Grp 2: 81.7	Grp 1: − 1.2 Grp 2: − 3.2	NR	Grp 1: − 1.39 Grp 2: − 1.97	NR	Grp 1: − 0.12 Grp 2: − 0.22
18	Shimizu et al. [65]	BIA	24	73.6	− 2.9*	NR	NR	27.8	− 0.9*
19	Yamakage et al. [66]	BIA	24	80.5	− 3.2*	NR	NR	25.9	0.1
20	Han et al. [57]	DXA	24	84.2	− 1.6*	24.7	− 1.0	56.6	− 0.8

SGLT2 sodium–glucose cotransporter 2 inhibitors, BW body weight, T2DM type 2 diabetes mellitus, DXA dual-energy X-ray absorptiometry, BIA bioelectrical impedance analysis, NR not reported, NAFLD nonalcoholic fatty liver disease, Grp group

*Statistically significant between-group P value

^aData are presented as mean or median values

^bThe label fat-free mass includes muscle mass, lean mass, and fat-free mass

Fig. 1.

Changes in total body weight and fat-free mass (kg) following SGLT2 inhibitor treatment

The study of Fadini et al. also demonstrated that dapagliflozin treatment for 12 weeks causes a high proportional loss of fat-free mass [62]. They reported that BIA-derived fat-free mass was reduced by 2.9 kg (accounting for over 90% of BW reduction), with a total body water loss of 2.4 kg. Bioelectrical impedance vector analysis (BIVA) in this study showed that dapagliflozin mainly reduced the fluid content of the body.

Changes in visceral and subcutaneous adipose tissue were investigated in 10 studies [53, 56, 57, 59, 63, 65–69]. SGLT2 inhibitors reduced visceral and subcutaneous adipose tissue greater than placebo; however, this difference was only significant in the study of Bolinder et al. [53]. The difference in changes were not significant when SGLT2 inhibitors were compared to compared to semaglutide and linagliptin [59, 67]. Additional details regarding these changes can be found in supplementary material, Table S2.

Unless contraindicated, glucose-lowering medications are prescribed for a lifetime. As mentioned earlier, the effects of SGLT2 inhibitors on body composition vary on the basis of the duration of treatment. Initial weight reduction is attributed to the loss of fluid and calories. In the long term, BW reduction becomes attenuated as a result of counter-regulatory mechanisms, and studies indicate that it plateaus after about 26 weeks [70]. Consequently, in the long term, SGLT2 inhibitors result in a modest decrease in BW, around 2–3 kg [71].

Body compartments with high water content, namely fat-free mass, are particularly affected by acute fluid loss [62]. In contrast, longer assessments of patients with T2DM who were treated with empagliflozin showed insignificant changes in body water and fat-free mass [67]. These findings support the notion that fat-free mass is potentially more affected early following SGLT2 inhibitor treatment [58].

Two previous meta-analyses have explored the impact of SGLT2 inhibitors on body composition and their overall findings are consistent with those observed in long-term studies [72, 73]. The meta-analysis by Pan and colleagues reported that the mean difference [95% confidence interval] when comparing SGLT2 inhibitors with the control group was − 2.73 kg [− 3.32 to − 2.13] for BW, − 1.16 kg [− 2.01 to − 0.31] for fat mass, − 0.76 kg [− 1.53 to 0.01] for lean mass, and − 1.01 kg [− 1.91 to − 0.11] for skeletal muscle mass [73]. A network meta-analysis by Ida et al. revealed that canagliflozin and dapagliflozin significantly reduced fat-free mass compared to placebo, accounting for 20–30% of the total weight reduction [72].

Weight reduction interventions generally result in both fat mass and fat-free mass losses, albeit to different extents, depending on their mechanism of action, concurrent medications, baseline BMI, and gender [74, 75]. Current evidence suggests that loss of fat-free mass contributes to about one-third of the total reduction in BW following lifestyle interventions and bariatric surgery [76–79] (supplementary material, Table S3). Changes in body composition associated with SGLT2 inhibitor treatment are comparable to other weight reduction interventions. Nonetheless, it is crucial to investigate the impact of acute loss of fat-free mass on patient outcomes, especially among vulnerable individuals.

Sodium–Glucose Cotransporter 2 Inhibitors Versus Other Comparators

In this review, nine placebo-controlled studies were included. The between-group mean difference of changes in total BW and fat-free mass were about − 1.5 to − 3 kg and + 0.27 to − 0.9 kg, respectively [51–54, 56, 58, 62, 68, 69] (supplementary material, Fig. S1). MRI analyses on iliopsoas muscle surface area after 24 weeks of treatment also showed no significant alterations compared to placebo [68, 69]. In a sub-study of the CANTATA-SU trial, the effects of add-on canagliflozin treatment on body composition over 26 weeks were assessed, with both 100 mg and 300 mg per day doses [51]. Both doses caused significant changes in BW, fat mass, and fat-free mass compared to placebo, while the proportional loss of fat-free mass was consistent at approximately 37%.

In three studies, sulfonylureas were the comparator drugs [55, 61, 64]. Our review found that sulfonylureas increased BW, fat mass, and fat-free mass across all three studies, resulting in significant differences when compared to SGLT2 inhibitors. The mean difference was − 3 to − 5 kg for total BW and − 1.2 to − 2.2 kg for fat-free mass. Sulfonylureas are insulin secretagogues agents and are associated with weight gain [80]. Taking note of these differences between sulfonylureas and SGLT2 inhibitors helps us tailor individualized treatment plans. For patients with T2DM who are overweight, SGLT2 inhibitors might be considered as a preferred treatment option as they address glucose control and body composition simultaneously.

Two studies compared the effects of SGLT2 inhibitors to dipeptidyl peptidase 4 (DPP4) inhibitors on body composition [50, 67]. The first involved Asian patients with uncontrolled T2DM despite receiving premixed insulin, and was conducted over 24 weeks [67]. In this study, baseline BMIs were 28.0 ± 3.5 kg and 27.7 ± 5.0 kg/m² in linagliptin and empagliflozin groups, respectively. As expected, empagliflozin reduced BW by 1.8 kg compared to linagliptin. The authors also report a difference in change of − 1.39 kg in fat-free mass [67]. The other study, which lasted for a shorter duration of 12 weeks, was conducted among Japanese people with T2DM and BMI over 28 kg/m² [50]. It showed that the changes in BW and skeletal muscle index (SMI) with ipragliflozin compared to sitagliptin were − 1.61 kg and − 0.035 kg/m², respectively. Further studies are needed to explore the effects of SGLT2 inhibitors vs DPP4 inhibitors on fat-free mass.

Two studies compared the effects of SGLT2 inhibitors and glucagon-like peptide 1 (GLP-1) receptor agonists on body composition in patients with T2DM, utilizing DXA as the assessment method [59, 60]. Nakaguchi et al. reported results from a 24-week treatment period, indicating a nonsignificant between-group difference of − 0.1 kg for fat-free mass [60]. Another study, conducted over 52 weeks, also reported a nonsignificant mean difference of − 0.78 kg (95% confidence interval − 1.61, 0.04) for fat-free mass. Two reviews exploring current evidence on this issue reported comparable effects of GLP-1 receptor agonists and SGLT2 inhibitors on fat-free mass [81, 82]. Moreover, a network meta-analysis found that semaglutide was associated with a greater reduction in BW and fat-free mass, followed by canagliflozin and dapagliflozin [72]. Although current evidence supports the comparable effects of GLP-1 receptor agonists and SGLT2 inhibitors on body composition, limited head-to-head comparison data warrant caution in making treatment decisions. On the other hand, certain patients with T2DM benefit from combination therapy of GLP-1 receptor agonists and SGLT2 inhibitors, including those with multiple risk factors, atherosclerotic cardiovascular disease, and patients not reaching specific treatment goals (e.g., obesity) [83, 84]. Further research is needed to explore the outcomes of such combination therapies on body composition.

Other studies included in this review, investigated the effects of add-on therapy with SGLT2 inhibitors among patients with T2DM who were taking oral antidiabetic medications and/or insulin [57, 63, 65, 66]. Significant reduction in total BW was reported with SGLT2 inhibitor therapy, without a clinically meaningful effect on fat-free mass [57, 63, 65, 66].

Strengths, Limitations, and Future Direction

In this study, we performed a systematic search, and included published RCTs. Additionally, the overall quality of the included studies was evaluated. These strengths enabled us to comprehensively examine how SGLT2 inhibitors affect body composition in individuals with T2DM. Nevertheless, there were some limitations when interpreting the findings. The major limitation was that data were largely heterogeneous in relation to the study duration, method of body composition assessment, and patients’ characteristics. In addition, investigating the changes in body composition was not the primary outcome in most of the studies. Also, a few studies considered background medications or the effect of multiple treatment regimens. For instance, Nakaguchi et al. compared the effects of treatment with empagliflozin versus liraglutide among patients who were taking insulin as the baseline regimen [60]. This study reported smaller changes in all compartments of body composition compared to McCrimmon et al.’s study, which also reported the effects of SGLT2 inhibitors versus GLP-1 receptor agonists on body composition [59]. These findings further highlight the importance of fractional changes in fat-free mass when assessing the safety of SGLT2 inhibitors.

Current meta-analyses have not fully addressed the heterogeneities in patient characteristics including comparator drugs, method of body composition assessment, and study duration [72, 73]. Pan and colleagues performed subgroup analysis based on different comparators, but the duration of treatment was not considered [73]. The method of body composition assessment was also not considered in the study of Ida and colleagues [72]. A major limitation is the lack of a universally recognized standard method to measure body composition. As mentioned earlier, BIA is affected by fixed assumptions of hydration status. This is highly significant when coexisting comorbidities, mainly CKD, are present [85]. Future trials should prioritize using more precise methods like DXA and diligently account for potential confounding factors.

Conclusion

Current evidence suggests that weight loss associated with SGLT2 inhibitor treatment is mainly due to the reduction of fat mass. Loss of fat-free mass contributes to one-third of total BW loss. This change is comparable to the body composition changes reported after lifestyle interventions, bariatric surgery, and GLP-1 receptor agonist treatment.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Mohammad E. Khamseh contributed to the study conception and design. Soodeh Jahangiri, Mohammad E. Khamseh, and Mojtaba Malek were involved in the drafting of the manuscript. Mohammad E. Khamseh, Mojtaba Malek, and Sanjay Kalra contributed to reviewing and editing the manuscript. All authors read and approved the final manuscript.

Funding

No funding was received for this study or publication of this article.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Declarations

Conflict of Interest

Soodeh Jahangiri, Mojtaba Malek, Sanjay Kalra, and Mohammad E. Khamseh have nothing to disclose.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.