Efficacy and safety of glucagon-like peptide-1 receptor agonists in the elderly versus non-elderly patients with type 2 diabetes mellitus: insights from a systematic review

Abstract

This systematic review aimed to compare the influence of glucagon-like peptide-1 receptor agonists (GLP-1RAs) on the efficacy and safety of elderly patients with type 2 diabetes and younger individuals. A comprehensive search of PubMed, Embase, and Web of Science databases was conducted up to September 2022. The summary standard means difference and odds ratios were calculated. Thirteen articles were included in the analysis. The incidence of adverse events (AEs) leading to discontinuation was higher in elderly patients (OR = 0.67, 95% CI 0.47 to 0.96, p = 0.028). However, no significant differences were observed in weight loss (SMD = 0.03, 95% CI –0.12 to 0.19, p = 0.686), HbA1c% (SMD = –0.02, 95% CI –0.11 to 0.08, p = 0.715), FBG levels (SMD = –0.03, 95% CI –0.11 to 0.06, p = 0.537), and the incidence of overall AEs (OR = 0.85, 95% CI 0.71 to 1.01, p = 0.072), serious AEs (OR = 0.68, 95% CI 0.45 to 1.04, p = 0.077), nausea (OR = 0.91, 95% CI 0.81 to 1.03, p = 0.140), vomiting (OR = 0.95, 95% CI 0.79 to 1.13, p = 0.532), diarrhea (OR = 0.86, 95% CI 0.72 to 1.02, p = 0.081), and hypoglycemia (OR = 1.22, 95% CI 0.90 to 1.65, p = 0.193). In conclusion, while certain AEs leading to discontinuation may be more prevalent in older patients, GLP-1RAs are effective for weight loss and lead to decreased glucose concentrations with a low rate of complications in elderly patients.

DIABETES has emerged as a widespread epidemic, with its prevalence increasing at an alarming rate. Type 2 diabetes (T2D) is the predominant form, and it is projected that over the next 15 years, more than 82 million people with T2D in developed countries will be aged 65 or older [1]. Aging is a risk factor for T2D development [2]. Elderly patients with T2D face a heightened risk of cardiovascular complications, renal failure, retinopathy, and neuropathy. Furthermore, multiple comorbidities, polypharmacy, and malnutrition in this population can further elevate the risk of adverse events (AEs) such as hypoglycemia, volume loss, impaired renal function, heart failure, neuropathy, cognitive impairment, fractures, and drug interactions [3-5]. Consequently, the selection of antidiabetic drugs in elderly patients necessitates careful consideration due to their unique safety concerns and requirements. Sulfonylureas, known to heighten the risk of hypoglycemia due to unregulated insulin release, are not recommended for the elderly population [6, 7]. While insulin therapy is effective, it carries a high risk of causing hypoglycemia, particularly challenging for elderly patients with impaired vision, arthritis, or cognitive dysfunction [8]. On the other hand, thiazolidinediones are associated with an increased risk of congestive heart failure and fractures, making them less suitable for elderly individuals [9]. In summary, there is a strong need for novel glycemic modulators to address the unique challenges of managing diabetes in the elderly, improving glycemic control without exacerbating age-related comorbidities [10, 11]. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have demonstrated significant improvements in glycemic parameters and weight reduction [12, 13]. These agents function by activating the glucagon-like peptide-1 receptor in the pancreas, enhancing insulin release, and reducing glucagon release in a glucose-dependent manner, thereby minimizing the risk of hypoglycemia [14]. Effects of GLP-1RAs on the central nervous system and gastrointestinal tract result in decreased appetite and delayed glucose absorption due to slower gastric emptying [15-18].

While acknowledging the benefits of GLP-1RAs in elderly patients, it is crucial to recognize the occurrence of gastrointestinal-related AEs and overall AEs associated with their usage. Older adults with diabetes have been reported to be at an increased risk of experiencing gastrointestinal side effects, such as anorexia, nausea, vomiting, and abdominal discomfort [19]. Age-related pharmacokinetic and pharmacodynamic changes significantly influence drug absorption, distribution, metabolism, and clearance, underscoring the need for careful consideration when prescribing antidiabetic or other medications for elderly patients [20]. Hence, the objective of this systematic review is to compare the efficacy and safety of GLP-1RAs in older and younger patients.

Materials and Methods

Search strategy

We conducted a comprehensive search of electronic databases, which included PubMed, Embase, and Web of Science, spanning from establishment to September 2022. The search employed the following keywords: GLP-1RAs, liraglutide, dulaglutide, semaglutide, lixisenatide, exenatide, diabetes, diabetes mellitus, T2D, 65 years old, and elderly. Specifically, we focused on studies published in English and manually examined the reference lists of the identified publications.

Study selection

Our study aimed to compare the efficacy and safety of GLP-1RAs in elderly and non-elderly patients diagnosed with T2D. All articles included in our review were retrospective studies. We classified patients with T2D into two age groups: those below 65 years old and those aged 65 years and above. Our systematic review scrutinized various outcomes, including weight loss, HbA1c%, fasting blood glucose (FBG) levels, the incidence of AEs, serious adverse events (SAEs), AEs leading to discontinuation, hypoglycemia, nausea, vomiting, and diarrhea. Articles were excluded from our analysis if they were abstract publications, contained duplicate data, had overlapping populations, lacked relevant data, or were non-clinical publications.

Data extraction

For all the studies included, two investigators independently extracted the following information and variables from each original literature: document name, first author’s name, publication year, country, GLP-1RAs, sample size, age, follow-up time, weight loss, HbA1c%, FBG, AEs, SAEs, AEs leading to discontinuation, occurrences of hypoglycemia, nausea, vomiting, and diarrhea. In cases of disagreement between the two investigators, any differences were resolved through discussion.

Statistical analysis and quality assessment

Following the preset protocol registered with PROSPERO 2022 (CRD42022378763), our analysis adhered to the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statements [21]. The meta-analysis was conducted using Stata 16.0 (version 16.0, Stata Corporation, College Station, TX, USA). Odds ratios (ORs) and 95% confidence intervals (CIs) were computed for the variables.

Statistical heterogeneity was assessed using I², where a value of more than 50% indicated high heterogeneity. In cases where I² > 50%, a random-effects model was employed to calculate the standard mean difference (SMD) along with a 95% confidence interval for the variable. Subgroup analyses and sensitivity analyses were conducted as needed to explore potential sources of heterogeneity.

Publication bias was evaluated through Begg’s and Egger’s tests, considering statistical significance at p < 0.05. If publication bias was identified, the trim and fill method was applied to examine whether it influenced the results. The quality of retrospective studies was appraised using the Newcastle-Ottawa Quality Assessment Scale (NOS) rating scale.

Results

In our initial search, we identified 459 studies. After a comprehensive full-text review, 17 studies were considered, and ultimately, 13 studies [22-34] involving 11,864 patients were included in our meta-analysis (Fig. 1). The basic characteristics are presented in Table 1, while the pertinent clinical results of the included studies are outlined in Table 2. The quality of the incorporated literature was found to be satisfactory, as depicted in Supplementary Table 1.

Fig. 1

Identification of eligible articles.

Table 1

Basic characteristics and clinical results of included studies.

Article	Year	Country	Article type	GLP-1RAs	Sample size	Age (y)	Follow-up time
Bode BW [22]	2011	USA	Retrospective study	liraglutide	711/185	<65/≥65	26 weeks
Boustani MA [23]	2016	USA	Retrospective study	dulaglutide	1,401/318	<65/≥65	26 weeks
Frias JP [24]	2021	USA	Retrospective study	dulaglutide	456/156	<65/≥65	52 weeks
Hamano K [25]	2017	Japan	Retrospective study	dulaglutide	386/77	<65/≥65	26 weeks
Kuang J [26]	2020	China	Retrospective study	dulaglutide	130/54	<65/≥65	26 weeks
Kuang J	2020	China	Retrospective study	dulaglutide	141/59	<65/≥65	26 weeks
Lingvay I [27]	2018	USA	Retrospective study	liraglutide	161/38	<65/≥65	26 weeks
Lingvay I	2018	USA	Retrospective study	liraglutide	216/76	<65/≥65	26 weeks
Lingvay I	2018	USA	Retrospective study	liraglutide	210/68	<65/≥65	26 weeks
Onishi Y [28]	2016	Japan	Retrospective study	dulaglutide	641/214	<65/≥65	26 weeks
Pantalone KM [29]	2018	Canada	Retrospective study	dulaglutide	205/69	<65/≥65	26 weeks
Pencek R [30]	2012	USA	Retrospective study	exenatide	1,613/454	<65/≥65	30 weeks
Raccah D [31]	2015	France	Retrospective study	lixisenatide	1,748/379	<65/≥65	26 weeks
Warren M [32]	2018	USA	Retrospective study	semaglutide	110/20	<65/≥65	56 weeks
Warren M	2018	USA	Retrospective study	semaglutide	332/77	<65/≥65	56 weeks
Warren M	2018	USA	Retrospective study	semaglutide	316/88	<65/≥65	56 weeks
Warren M	2018	USA	Retrospective study	semaglutide	281/79	<65/≥65	56 weeks
Warren M	2018	USA	Retrospective study	semaglutide	102/29	<65/≥65	56 weeks
Yabe D [33]	2022	Japan	Retrospective study	semaglutide	34/12	<65/≥65	30 weeks
Yabe D	2022	Japan	Retrospective study	semaglutide	109/30	<65/≥65	56 weeks
Yamada Y [34]	2021	Japan	Retrospective study	semaglutide	27/21	<65/≥65	26 weeks
Yamada Y	2021	Japan	Retrospective study	semaglutide	96/34	<65/≥65	57 weeks

Values are all given as young/old age group; NR, not reported; GLP-1RAs, glucagon-like peptide-1 receptor agonists; y, years; USA, the United States of America.

Table 2

Comparison of outcomes between young and old groups.

Article	HbA1c (%)	FBG (mmol/L)	AEs (%)	SAEs (%)	AEs leading to discontinuation (%)	Hypoglycemia (%)	Nausea (%)	Vomiting (%)	Diarrhea (%)
Bode BW [22]	–1.31 (1.83)/–1.32 (3.59)	–2.1 (4.58)/–1.8 (9.16)	76.1/78.4	4.8/7.6	7.2/9.7	8.0/4.3	20.4/20.5	8.4/5.4	10/13.5
Boustani MA [23]	–1.29 (0.86)/–1.24 (0.86)	NR	NR	NR	NR	NR	20.3/22.6	10.1/10.4	11.6/14.2
Frias JP [24]	–1.59 (1.32)/–1.33 (1.32)	NR	NR	7.5/10.9	12.3/14.7	NR	13.6/16.0	6.4/6.4	7.5/8.3
Hamano K [25]	–1.48 (0.91)/–1.72 (0.93)	NR	65.0/64.9	NR	NR	NR	7.0/3.9	2.3/1.3	9.3/6.5
Kuang J [26]	–1.43 (0.06)/–1.45 (0.06)	–2.17 (2.18)/–2.58 (3.49)	NR	NR	NR	7.5/1.8	NR	NR	NR
Kuang J	–1.64 (0.13)/–1.60 (0.13)	–2.28 (2.53)/–2.42 (3.86)	NR	NR	NR	14.0/20.0	NR	NR	NR
Lingvay I [27]	–2.0 (1.1)/–1.6 (1.0)	NR	NR	NR	NR	26.1/15.8	NR	NR	NR
Lingvay I	–1.3 (0.9)/–1.3 (0.8)	NR	NR	NR	NR	30.6/36.0	NR	NR	NR
Lingvay I	–1.8 (1.1)/–1.8 (1.0)	NR	NR	NR	NR	30.0/23.5	NR	NR	NR
Onishi Y [28]	NR	NR	NR	NR	NR	NR	6.9/7.9	NR	NR
Pantalone KM [29]	–1.6 (0.73)/–1.8 (0.73)	NR	NR	NR	NR	NR	NR	NR	NR
Pencek R [30]	–0.9 (1.17)/–1.0 (2.34)	–1.17 (3.25)/–1.00 (5.84)	NR	NR	NR	NR	37.8/41.4	14.1/13.7	10.5/10.4
Raccah D [31]	NR	NR	68.8/71.8	NR	NR	NR	25.5/29	9.7/14.2	8.1/9.2
Warren M [32]	–1.6 (0.83)/–1.2 (0.83)	NR	NR	NR	NR	NR	NR	NR	NR
Warren M	–1.6 (0.40)/–1.5 (0.40)	NR	NR	NR	NR	NR	NR	NR	NR
Warren M	–1.5 (0.84)/–1.7 (0.84)	NR	NR	NR	NR	NR	NR	NR	NR
Warren M	–1.8 (1.99)/–1.3 (1.99)	NR	NR	NR	NR	NR	NR	NR	NR
Warren M	–1.9 (0.24)/–1.8 (0.24)	NR	NR	NR	NR	NR	NR	NR	NR
Yabe D [33]	–2.3 (0.97)/–1.8 (0.97)	NR	61.8/83.3	2.9/0.0	0.0/8.3	NR	14.7/0.0	5.9/0.0	5.9/8.3
Yabe D	NR	NR	85.3/90.0	4.6/3.3	5.5/0.0	NR	16.5/13.3	4.6/6.7	15.6/10.0
Yamada Y [34]	–1.8 (0)/–1.8 (0)	NR	63.0/81.0	0.0/0.0	0.0/9.5	NR	11.1/4.8	NR	7.4/4.8
Yamada Y	–1.8 (1.27)/–2.3 (1.27)	NR	82.3/94.1	5.2/5.9	3.1/14.7	NR	10.4/5.9	NR	7.3/8.8

Values are all given as young/old age group; NR, not reported; HbA1c, haemoglobin A1c; FBG, fasting blood glucose; AEs, adverse events; SAEs, serious adverse events.

Efficacy

Weight loss

A total of seven articles [22, 24, 26, 30, 32-34], encompassing thirteen investigations, examined weight loss in patients with T2D who received treatment with GLP-1RAs across age groups <65 and ≥65 years (Fig. 2A). Employing a random-effects model, our findings indicated no significant difference in weight loss between these age groups (SMD = 0.03, 95% CI –0.12 to 0.19, p = 0.686). Furthermore, the analysis revealed substantial heterogeneity for this outcome measure (I² = 75.3%, p = 0.000). Sensitivity analyses performed on studies exhibiting considerable heterogeneity demonstrated that the results were not solely influenced by any single study, and subsequent subgroup analyses were conducted (Table 3).

Fig. 2

Forest plot of (A) Weight loss, (B) HbA1c%, (C) FBG.

Table 3

Subgroup analysis of weight loss.

	Studies	<65 years old	>65 years old	SMD	95%CI	I2
Countries
Asia	6	515	222	0.01	–0.36 to 0.37	77.8%
Non-Asia	7	3,861	1,066	0.03	–0.15 to 0.21	77%
GLP-1RAs
Liraglutide	2	711	185	–0.16	–0.32 to 0	—
Dulaglutide	2	271	113	0.01	–0.60 to 0.62	86.9%
Exenatide	1	1,613	454	0.11	0.01 to 0.22	—
Semaglutide	3	1,269	347	0.04	–0.23 to 0.31	75.9%
Follow-up time
26 weeks	3	1,009	319	–0.12	–0.41 to 0.16	71.2%
>26 weeks	6	1,358	381	0.09	–0.16 to 0.34	75.3%

SMD, summary standard means; GLP-1RAs, glucagon-like peptide-1 receptor agonists.

We observed that younger individuals demonstrated more pronounced weight loss when treated with liraglutide, as delineated by the country of the study, specific GLP-1RAs types, and duration of follow-up. Other subgroup analyses did not identify any sources of heterogeneity. Assessment of publication bias using Begg’s ( p = 1.000) and Egger’s ( p = 0.915) tests indicated the absence of bias across the included studies.

HbA1c%

In total, eleven articles [22-27, 29, 30, 32-34], encompassing eighteen studies, examined the impact of GLP-1RAs on changes in HbA1c levels across different age groups in patients with T2D (Fig. 2B). Utilizing a random-effects model, our analysis revealed no significant difference in HbA1c level changes between the two age groups (SMD = –0.02, 95% CI –0.11 to 0.08, p = 0.715). Notably, substantial heterogeneity was observed (I² = 63.6%, p = 0.000), prompting the performance of sensitivity and subgroup analyses (Table 4). However, these analyses failed to identify any sources of heterogeneity. Furthermore, an assessment for publication bias using Begg’s test (p = 0.202) and Egger’s test (p = 0.694) indicated the absence of publication bias.

Table 4

Subgroup analysis of HbA1c%.

	Studies	<65 years old	>65 years old	SMD	95%CI	I2
Countries
Asia	6	765	248	0.12	–0.16 to 0.41	69.6%
Non-Asia	7	6,049	1,635	–0.05	–0.15 to 0.04	59.4%
GLP-1RAs
Liraglutide	2	587	182	–0.10	–0.31 to 0.12	38.3%
Dulaglutide	6	2,263	577	0.09	–0.12 to 0.31	76.0%
Exenatide	1	1,613	454	0.07	–0. 04 to 0.17	—
Semaglutide	3	1,250	351	–0.12	–0.36 to 0.13	71.9%
Follow-up time
26 weeks	7	3,561	944	0.02	–0.11 to 0.15	62.5%
>26 weeks	6	1,638	485	–0.10	–0.30 to 0.09	67.5%

SMD, summary standard means; GLP-1RAs, glucagon-like peptide-1 receptor agonists.

FBG

In aggregate, three articles [22, 26, 30], encompassing four research studies, elucidated the impact on FBG levels in two distinct age groups of patients following the administration of GLP-1RAs (Fig. 2C). The outcomes derived from the fixed-effect model indicated no statistically significant difference in the FBG change between the two age groups subsequent to GLP-1RAs intake (SMD = –0.03, 95% CI –0.11 to 0.06, p = 0.537). The analysis further demonstrated a lack of heterogeneity (I² = 0, p = 0.642).

Safety

Adverse events

All seven studies from five articles [22-24, 28, 33] investigated the incidence of AEs following GLP-1RAs treatment in both age groups (Fig. 3A). Our analysis led to the conclusion that the incidence of AEs did not significantly differ between older and younger patients (OR = 0.85, 95% CI 0.71 to 1.01, p = 0.072), with no observed heterogeneity (I² = 0, p = 0.493). However, Begg’s test (p = 0.035) and Egger’s test (p = 0.017) indicated the presence of publication bias. Therefore, it became imperative to employ the trim and fill method to assess the stability of the pooled result. The trim and fill methods were applied, revealing that the pooled result remained robust despite the identified publication bias.

Fig. 3

Forest plot of (A) AEs, (B) SAEs (C) AEs leading to discontinuation.

Serious adverse events

Five studies from four articles [22, 24, 33, 34] investigated the incidence of SAEs following treatment with GLP-1RAs in two age groups, as illustrated in Fig. 3B. Employing a fixed-effects model, our analysis indicated that there was no significant difference in the incidence of SAEs between the two groups after GLP-1RAs treatment (OR = 0.68, 95% CI 0.45 to 1.04, p = 0.077). No significant heterogeneity was observed between the studies (I² = 0, p = 0.951). However, Egger’s test (p = 0.034) identified publication bias. To assess the stability of the pooled results, we employed the trim and fill methods. Unfortunately, the trim and fill methods indicated that the pooled result was not robust and suggested a need for additional studies to enhance reliability.

Serious adverse events leading to discontinuation

Four articles [22, 24, 33, 34], encompassing six studies, reported AEs leading to discontinuation (Fig. 3C). The results indicated a higher incidence in the old age group compared to the young age group (OR = 0.67, 95% CI 0.47 to 0.96, p = 0.028; I² = 10.2%, p = 0.350). However, Egger’s test (p = 0.034) detected publication bias. To assess the stability of the pooled results, we employed the trim and fill methods that demonstrated that the pooled result was robust.

Hypoglycemia

Six studies of the three articles [22, 26, 27] reported the incidence of hypoglycemia following GLP-1RAs treatment in both age groups (Fig. 4A). Using a fixed-effects model, the results showed that the incidence of hypoglycemia after treatment was not significantly different between the two age groups (OR = 1.22, 95% CI 0.90 to 1.65, p = 0.193, I² = 41.8%, p = 0.127). Neither Begg’s test (p = 0.260) nor Egger’s test (p = 0.224) found publication bias.

Fig. 4

Forest plot of (A) Hypoglycemia, (B) Nausea, (C) Vomiting, (D) Diarrhea.

Nausea

In total, 10 articles [22-26, 28, 30, 33, 34], which encompassed 13 studies, reported the incidence of nausea after undergoing GLP-1RAs treatment across two age groups. The fixed-effect model indicated that there was no significant difference between the two groups, and no heterogeneity was found in these studies (OR = 0.91, 95% CI 0.81 to 1.03, p = 0.140, I² = 0, p = 0.596) (Fig. 4B). Begg’s test (p = 0.002) and Egger’s test (p = 0.000) both identified publication bias. Therefore, the trim and fill methods were employed to assess the stability of the pooled result, indicating that the pooled result was robust.

Vomiting

In ten studies, encompassing eight articles [22-26, 30, 31, 33], the incidence of vomiting after the administration of GLP-1RAs was investigated in both age groups. Utilizing a fixed-effect model, our analysis led us to the conclusion that there was no significant difference in the incidence of vomiting between the two age groups following the administration of GLP-1RAs (OR = 0.95, 95% CI 0.79 to 1.13, p = 0.532) (Fig. 4C). The study results exhibited low heterogeneity (I² = 12.8%, p = 0.325). Additionally, no publication bias was identified, as evidenced by Begg’s test (p = 0.858) and Egger’s test (p = 0.217).

Diarrhea

Eight articles [22-24, 26, 30, 31, 33, 34] have data from eleven studies, indicating the incidence of diarrhea after agonist use in two age groups. The fixed-effect model study revealed no significant difference in the incidence of diarrhea between the two age groups (OR = 0.86, 95% CI 0.72 to 1.02, p = 0.081) (Fig. 4D). No heterogeneity was observed in these studies (I² = 0, p = 0.909). Neither the Begg’s test (p = 0.876) nor the Egger’s test (p = 0.955) showed any evidence of publication bias.

Discussion

To the best of our knowledge, our meta-analysis is the first one that scrutinized the efficacy and safety of GLP-1RAs in elderly diabetic patients compared to their younger counterparts. The current analysis established that GLP-1RAs can effectively treat elderly patients with T2D. Treatment engendered significant weight loss in both older adults and younger individuals; however, no notable difference in weight loss emerged between the two age groups. Pencek and Bode’s findings align with ours, although Kuang posited that the weight loss effect in the older group surpassed that in the younger group [22, 26, 30]. This discrepancy might be ascribed to the fact that older diabetics often contend with more severe metabolic dysfunction. Additionally, advancing age elevates the susceptibility to frailty, a condition characterized by heightened vulnerability, diminished physiological reserves, and impaired ability to withstand physical or psychological stress [35-37]. Frailty can be assessed using the Fried criteria, incorporating three self-reported variables (weight loss, exhaustion, and reduced physical activity) and two measurements (weakness evaluated by grip strength and slowness measured by gait speed) [38]. Frailty is, however, associated with an accelerated muscle loss that extends beyond that induced solely by aging. The principal characteristic of frailty is the loss of muscle fibers, a condition known as sarcopenia, thereby establishing a connection between sarcopenia and frailty [38, 39]. Weight loss is another characteristic of frailty, although it is not a mandatory criterion for diagnosing frailty, and obesity can also be associated with the condition [40]. Therefore, contingent on overall body weight, varying loss of muscle fibers, and the ratio of body adipose tissue to muscle tissue, frailty can be associated with a spectrum of metabolic changes and differing degrees of insulin resistance. We propose categorizing these as at least two distinct metabolic ‘phenotypes.’ The anorexic malnourished (AM) frail phenotype, characterized by substantial muscle loss and reduced insulin resistance, represents one end of the spectrum. In contrast, the sarcopenic obese (SO) frail phenotype, marked by increased visceral fat and insulin resistance, occupies the other end [41]. Presently, no hypoglycemic agents are specifically tailored for older adults with concurrent diabetes and frailty. Sinclair AJ, therefore, advocates a pragmatic approach, favoring glucose-lowering agents promoting weight loss for the SO phenotype. Caution is advised when using GLP-1RAs in older patients with the AM phenotype to mitigate the risk of hypotension, dehydration, weight loss, and falls in this vulnerable group [41].

Our meta-analysis revealed that blood glucose levels were effectively improved in both age groups. Furthermore, no significant differences were observed in the changes of FBG and HbA1c% between the two groups. Pantalone and Boustani have reached similar conclusions [23, 29]. In contrast, Onishi Y [28] suggests that GLP-1RAs may have a less favorable impact on reducing HbA1c% levels among older patients compared to younger individuals. This discrepancy could be attributed to the gradual decline in β cell quality and function, reduced insulin secretion, and the increased challenge in achieving optimal blood glucose control in older individuals with longer diabetes duration. Notably, GLP-1RAs are increasingly categorized based on their duration of action as long-acting (abiglutide, dulaglutide, once-weekly exenatide, liraglutide, semaglutide, and tamsuloside) and short-acting (twice-daily exenatide (EBID), and lixisenatide). The mechanism of action of long-acting GLP-1RAs predominantly involves stimulating β-cells to increase insulin secretion, while short-acting GLP-1RAs primarily decrease blood glucose levels by delaying gastric emptying [42]. In a study by Htike ZZ, dulaglutide treatment exhibited the most significant reduction in HbA1c and FPG compared to exenatide, followed by liraglutide and once-weekly exenatide. Albiglutide was associated with the lowest risk of nausea and diarrhea, and once-weekly exenatide presented the lowest risk of vomiting [43]. Additionally, Guo et al. demonstrated that semaglutide may have a more conspicuous antiobesity effect and lower gastrointestinal adverse events than liraglutide and exenatide [44]. Therefore, when prescribing GLP-1RAs, it is essential to consider their varying effectiveness in alignment with the patient’s dosing objectives and requirements [45, 46].

In terms of safety, a notable discrepancy in the incidence of AEs leading to discontinuation was observed between the two age groups. Specifically, the older group exhibited a higher incidence compared to the younger patient group. However, there were no significant differences in the incidence of AEs and SAEs between the two groups. Several factors contribute to the discontinuation of anti-diabetic treatment in older patients with T2D. These factors encompass frailty, inadequate or irregular nutritional intake, cognitive impairment affecting the recognition or response to low blood sugar levels, increased use of multiple medications, non-compliance with prescribed medications, and impaired renal or hepatic function [3]. Hypoglycemia emerges as a crucial consideration when selecting antidiabetic medications. Our findings suggest that there is no significant difference in the occurrence of hypoglycemia between older and younger patients. However, various studies have reported a higher incidence of hypoglycemia in the older patient group compared to younger patients [25, 31, 32]. This disparity could potentially result from impaired counterregulatory responses to hypoglycemia in older patients [47], along with age-related cognitive deficits and a lack of awareness of impending hypoglycemic events. Consequently, older individuals are more vulnerable to experiencing hypoglycemic events, leading to potential complications such as arrhythmias, dizziness, falls, and confusion [48]. Furthermore, frail elderly individuals face an increased risk of hypoglycemia, dehydration, and weight loss when using GLP-1RAs [41]. Therefore, caution should be exercised when administering GLP-1RAs to elderly patients with impaired counter-regulatory responses to blood glucose, cognitive impairment, and frailty.

Patients with suboptimal glycemic control or a protracted disease course frequently experience impaired gastric motility, leading to delayed gastric emptying and subsequent gastrointestinal symptoms. In this study, we conducted an analysis of the prevalence of gastrointestinal symptoms, including but not limited to nausea, vomiting, and diarrhea. The analysis of the studies revealed comparable findings regarding the occurrence of nausea, vomiting, and diarrhea in both patient groups. It is worth noting that gastrointestinal function tends to gradually decline in older adults, resulting in diminished drug tolerance compared to younger individuals. Moreover, older individuals with diabetes exhibit an elevated susceptibility to gastrointestinal side effects [29]. Therefore, it is imperative for elderly patients to remain vigilant towards the development of gastrointestinal issues while undergoing treatment with GLP-1RAs. Based on our findings, it is noteworthy that various articles suggest the occurrence of early gastrointestinal side effects during treatment, which are typically transient in nature and diminish within a few days or weeks [49-51]. Furthermore, studies have demonstrated a significant reduction in the initial occurrence of gastrointestinal adverse events with an increase in the number of follow-up visits [52]. Moreover, researchers advocate the administration of intervention to patients at the initiation of treatment, instead of discontinuing it in response to mild symptoms that may arise early on. It is postulated that if patients can endure the initial gastrointestinal side effects, subsequent adverse events may be mitigated over time [53].

Limitations

The median duration of this study was 26 weeks, a period that may not accurately capture the long-term impact of GLP-1RAs. However, the analysis of the reported studies reveals several limitations, even though it does provide valuable insights for clinical practice. Firstly, the included studies were retrospective analyses that employed diverse design methods and raw data. Secondly, the selection of patients was somewhat restricted, as it excluded individuals with significant medical conditions such as liver or kidney disease, as well as those with physical frailty. Thirdly, there is a possibility that other potential confounding factors, such as the use of concurrent hypoglycemic medications, might have influenced the outcomes.

Conclusion

This study discovered no significant difference in efficacy between older and younger patients treated with GLP-1RAs. Furthermore, GLP-1RAs demonstrated effectiveness in promoting weight loss and reducing glucose concentrations with minimal complications in elderly patients. Our study revealed that older patients had a higher propensity to discontinue drug therapy due to adverse events. The judgment of frailty, awareness of adverse events, and dedication of more follow-up time could enhance the care of elderly patients. Additionally, there is an ongoing need for studies with substantial sample sizes and long-term follow-up, specifically concentrating on elderly patients who are using GLP-1RAs. These studies would help establish appropriate treatment guidelines for elderly individuals with diabetes.

Acknowledgments

We would like to express our heartfelt gratitude to all authors who provided published data for our systematic review.

This study was funded by Natural Science Foundation of Sichuan Province (24NSFSC2777), Chengdu Medical Research Project (2022291), Key Projects of Chengdu University School of Clinical Medicine and Affiliated Hospital (Y202202), Chengdu University Research Initiation Programme (2081923030).

Conflict of Interest

The authors declare that they have no conflict of interest.

Reference

• 1   Wild  S,  Roglic  G,  Green  A,  Sicree  R,  King  H (2004) Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 27: 1047–1053.
• 2   Salinas  RM,  Hiriart  M,  Acosta  I,  Sosa  AL,  Prince  MJ (2016) Type 2 diabetes mellitus as a risk factor for dementia in a Mexican population. J Diabetes Complications 30: 1234–1239.
• 3   Inzucchi  SE,  Bergenstal  RM,  Buse  JB,  Diamant  M,  Ferrannini  E, et al. (2012) Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 35: 1364–1379.
• 4   Inzucchi  SE,  Bergenstal  RM,  Buse  JB,  Diamant  M,  Ferrannini  E, et al. (2015) Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care 38: 140–149.
• 5   Bramlage  P,  Gitt  AK,  Binz  C,  Krekler  M,  Deeg  E, et al. (2012) Oral antidiabetic treatment in type-2 diabetes in the elderly: balancing the need for glucose control and the risk of hypoglycemia. Cardiovasc Diabetol 11: 122.
• 6   Bryan  J,  Crane  A,  Vila-Carriles  WH,  Babenko  AP,  Aguilar-Bryan  L (2005) Insulin secretagogues, sulfonylurea receptors and K(ATP) channels. Curr Pharm Des 11: 2699–2716.
• 7   Kahn  SE,  Haffner  SM,  Heise  MA,  Herman  WH,  Holman  RR, et al. (2005) Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. N Engl J Med 355: 2427–2443.
• 8   Ober  SK,  Watts  S,  Lawrence  RH (2006) Insulin use in elderly diabetic patients. Clin Interv Aging 1: 107–113.
• 9  Canadian Diabetes Association Clinical Practice Guidelines Expert Committee,  Meneilly  GS,  Knip  A,  Tessier  D (2013) Diabetes in the elderly. Can J Diabetes 37(Suppl): S184–S190.
• 10   Chau  D,  Edelman  SV (2001) Clinical management of diabetes in the elderly. Clin Diabetes 19: 172–175.
• 11   Shiba  T,  Ishii  S,  Okamura  T,  Mitsuyoshi  R,  Pfarr  E, et al. (2017) Efficacy and safety of empagliflozin in Japanese patients with type 2 diabetes mellitus: a sub-analysis by body mass index and age of pooled data from three clinical trials. Diabetes Res Clin Pract 131: 169–178.
• 12   Fineman  MS,  Cirincione  BB,  Maggs  D,  Diamant  M (2012) GLP-1 based therapies: differential effects on fasting and postprandial glucose. Diabetes Obes Metab 14: 675–688.
• 13   Garber  AJ (2012) Novel GLP-1 receptor agonists for diabetes. Expert Opin Investig Drugs 21: 45–57.
• 14   Yabe  D,  Eto  T,  Shiramoto  M,  Irie  S,  Murotani  K, et al. (2016) Effects of DPP-4 inhibitor linagliptin and GLP-1 receptor agonist liraglutide on physiological response to hypoglycaemia in Japanese subjects with type 2 diabetes: a randomized, open-label, 2-arm parallel comparative, exploratory trial. Diabetes Obes Metab 19: 442–447.
• 15   Kirkman  MS,  Briscoe  VJ,  Clark  N,  Florez  H,  Haas  LB, et al. (2012) Diabetes in older adults: a consensus report. J Am Geriatr Soc 60: 2342–2356.
• 16   Schjoldager  BT,  Mortensen  PE,  Christiansen  J,  Orskov  C,  Holst  JJ (1989) GLP-1 (glucagon-like peptide 1) and truncated GLP-1, fragments of human proglucagon, inhibit gastric acid secretion in humans. Dig Dis Sci 34: 703–708.
• 17   Wettergren  A,  Schjoldager  B,  Mortensen  PE,  Myhre  J,  Christiansen  J, et al. (1993) Truncated GLP-1 (proglucagon 78–107-amide) inhibits gastric and pancreatic functions in man. Dig Dis Sci 38: 665– 673.
• 18   Daniele  G,  Iozzo  P,  Molina-Carrion  M,  Lancaster  J,  Ciociaro  D, et al. (2015) Exenatide regulates cerebral glucose metabolism in brain areas associated with glucose homeostasis and reward system. Diabetes 64: 3406–3412.
• 19   Liu  FP,  Dong  JJ,  Yang  Q,  Xue  XZ,  Ren  ZF, et al. (2015) Glucagon-like peptide 1 receptor agonist therapy is more efficacious than insulin glargine for poorly controlled type 2 diabetes: a systematic review and meta-analysis. J Diabetes 7: 322–328.
• 20   Mangoni  AA,  Jackson  SH (2004) Age-related changes in pharmacokinetics and pharmacodynamics: basic principles and practical applications. Br J Clin Pharmacol 57: 6–14.
• 21   Moher  D,  Liberati  A,  Tetzlaff  J,  Altman  DG,  PRISMA  Group (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6: e1000097.
• 22   Bode  BW,  Brett  J,  Falahati  A,  Pratley  RE (2011) Comparison of the efficacy and tolerability profile of liraglutide, a once-daily human GLP-1 analog, in patients with type 2 diabetes >/=65 and <65 years of age: a pooled analysis from phase III studies. Am J Geriatr Pharmacother 9: 423–433.
• 23   Boustani  MA,  Pittman  I 4th,  Yu  M,  Thieu  VT,  Varnado  OJ, et al. (2016) Similar efficacy and safety of once-weekly dulaglutide in patients with type 2 diabetes aged >/=65 and <65 years. Diabetes Obes Metab 18: 820–828.
• 24   Frias  JP,  Bonora  E,  Nevárez Ruiz  L,  Hsia  SH,  Jung  H, et al. (2021) Efficacy and safety of dulaglutide 3.0 and 4.5 mg in patients aged younger than 65 and 65 years or older: Post hoc analysis of the AWARD-11 trial. Diabetes Obes Metab 23: 2279–2288.
• 25   Hamano  K,  Nishiyama  H,  Matsui  A,  Sato  M,  Takeuchi  M (2017) Efficacy and safety analyses across 4 subgroups combining low and high age and body mass index groups in Japanese phase 3 studies of dulaglutide 0.75 mg after 26 weeks of treatment. Endocr J 64: 449–456.
• 26   Kuang  J,  Zhu  J,  Liu  S,  Li  Q (2020) Efficacy and safety of once-weekly dulaglutide in elderly chinese patients with type 2 diabetes: a post hoc analysis of AWARD-CHN studies. Diabetes Ther 11: 2329–2339.
• 27   Lingvay  I,  Handelsman  Y,  Linjawi  S,  Vilsbøll  T,  Halladin  N, et al. (2019) Efficacy and safety of ideglira in older patients with type 2 diabetes. Endocr Pract 25: 144–155.
• 28   Onishi  Y,  Oura  T,  Nishiyama  H,  Ohyama  S,  Takeuchi  M, et al. (2016) Subgroup analysis of phase 3 studies of dulaglutide in Japanese patients with type 2 diabetes. Endocr J 63: 263–273.
• 29   Pantalone  KM,  Patel  H,  Yu  M,  Fernández Landó  L (2018) Dulaglutide 1.5 mg as an add-on option for patients uncontrolled on insulin: subgroup analysis by age, duration of diabetes and baseline glycated haemoglobin concentration. Diabetes Obes Metab 20: 1461–1469.
• 30   Pencek  R,  Blickensderfer  A,  Li  Y,  Brunell  SC,  Anderson  PW (2012) Exenatide twice daily: analysis of effectiveness and safety data stratified by age, sex, race, duration of diabetes, and body mass index. Postgrad Med 124: 21–32.
• 31   Raccah  D,  Miossec  P,  Esposito  V,  Niemoeller  E,  Cho  M, et al. (2015) Efficacy and safety of lixisenatide in elderly (>/=65 years old) and very elderly (>/=75 years old) patients with type 2 diabetes: an analysis from the GetGoal phase III programme. Diabetes Metab Res Rev 31: 204–211.
• 32   Warren  M,  Chaykin  L,  Trachtenbarg  D,  Nayak  G,  Wijayasinghe  N, et al. (2018) Semaglutide as a therapeutic option for elderly patients with type 2 diabetes: pooled analysis of the SUSTAIN 1–5 trials. Diabetes Obes Metab 20: 2291–2297.
• 33   Yabe  D,  Yamada  Y,  Kaku  K,  Nishida  T,  Sato  T, et al. (2022) Efficacy and safety of once-weekly semaglutide in Japanese individuals with type 2 diabetes by baseline age and body mass index. J Diabetes Investig 13: 1161–1174.
• 34   Yamada  Y,  Yabe  D,  Hertz  CL,  Horio  H,  Nakamura  J, et al. (2022) Efficacy and safety of oral semaglutide by baseline age in Japanese patients with type 2 diabetes: a subgroup analysis of the PIONEER 9 and 10 Japan trials. Diabetes Obes Metab 24: 321–326.
• 35   Bergman  H,  Ferrucci  L,  Guralnik  J,  Hogan  DB,  Hummel  S, et al. (2007) Frailty: an emerging research and clinical paradigm-issues and controversies. J Gerontol A Biol Sci Med Sci 62: 731–737.
• 36   Abdelhafiz  AH,  Koay  L,  Sinclair  AJ (2016) The effect of frailty should be considered in the management plan of older people with type 2 diabetes. Future Sci OA 2: FSO102.
• 37   Wang  Y,  Shao  X,  Liu  Z (2022) Efficacy and safety of sodium-glucose co-transporter 2 inhibitors in the elderly versus non-elderly patients with type 2 diabetes mellitus: a meta-analysis. Endocr J 69: 669–679.
• 38   Fried  LP,  Tangen  CM,  Walston  J,  Newman  AB,  Hirsch  C, et al. (2001) Cardiovascular health study collaborative research group. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 56: M146–M156.
• 39   Rockwood  K,  Mitnitski  A (2007) Frailty in relation to the accumulation of deficits. J Gerontol A Biol Sci Med Sci 62: 722–727.
• 40   Watanabe  D,  Yoshida  T,  Watanabe  Y,  Yamada  Y,  Kimura  M, et al. (2020) A U-shaped relationship between the prevalence of frailty and body mass index in community-dwelling Japanese older adults: the Kyoto-Kameoka study. J Clin Med 9: 1367.
• 41   Sinclair  AJ,  Pennells  D,  Abdelhafiz  AH (2022) Hypoglycaemic therapy in frail older people with type 2 diabetes mellitus-a choice determined by metabolic phenotype. Aging Clin Exp Res 34: 1949–1967.
• 42   Lu  JM (2019) The role of glucagon-like peptide-1 receptor agonists in type 2 diabetes in Asia. Adv Ther 36: 798–805.
• 43   Htike  ZZ,  Zaccardi  F,  Papamargaritis  D,  Webb  DR,  Khunti  K, et al. (2017) Efficacy and safety of glucagon-like peptide-1 receptor agonists in type 2 diabetes: a systematic review and mixed-treatment comparison analysis. Diabetes Obes Metab 19: 524–536.
• 44   Guo  X,  Zhou  Z,  Lyu  X,  Xu  H,  Zhu  H, et al. (2022) The antiobesity effect and safety of GLP-1 receptor agonist in overweight/obese patients without diabetes: a systematic review and meta-analysis. Horm Metab Res 54: 458–471.
• 45   Meneilly  GS,  Cheung  E,  Tuokko  H (1994) Altered responses to hypoglycemia of healthy elderly people. J Clin Endocrinol Metab 78: 1341–1348.
• 46   Kuwata  H,  Yabe  D,  Murotani  K,  Fujiwara  Y,  Haraguchi  T, et al. (2021) Effects of glucagon-like peptide-1 receptor agonists on secretions of insulin and glucagon and gastric emptying in Japanese individuals with type 2 diabetes: a prospective, observational study. J Diabetes Investig 12: 2162–2171.
• 47   Pencek  R,  Brunell  SC,  Li  Y,  Hoogwerf  BJ,  Malone  J (2012) Exenatide once weekly for the treatment of type 2 diabetes mellitus: clinical results in subgroups of patients using different concomitant medications. Postgrad Med 124: 33–40.
• 48   Nauck  MA (2011) Incretin-based therapies for type 2 diabetes mellitus: properties, functions, and clinical implications. Am J Med 124: S3–S18.
• 49   Drucker  DJ,  Buse  JB,  Taylor  K,  Kendall  DM,  Trautmann  M, et al. (2008) Exenatide once weekly versus twice daily for the treatment of type 2 diabetes: a randomised, open-label, non-inferiority study. Lancet 372: 1240–1250.
• 50   Buse  JB,  Rosenstock  J,  Sesti  G,  Schmidt  WE,  Montanya  E, et al. (2009) Liraglutide once a day versus exenatide twice a day for type 2 diabetes: a 26-week randomised, parallel-group, multinational, open-label trial (LEAD-6). Lancet 374: 39–47.
• 51   Pratley  R,  Nauck  M,  Bailey  T,  Montanya  E,  Cuddihy  R, et al. (2011) One year of liraglutide treatment offers sustained and more effective glycaemic control and weight reduction compared with sitagliptin, both in combination with metformin, in patients with type 2 diabetes: a randomised, parallel-group, open-label trial. Int J Clin Pract 65: 397–407.
• 52   Tofé  S,  Argüelles  I,  Mena  E,  Serra  G,  Codina  M, et al. (2019) Real-world GLP-1 RA therapy in type 2 diabetes: a long-term effectiveness observational study. Endocrinol Diabetes Metab 2: e00051.
• 53   Sun  F,  Yu  K,  Yang  Z,  Wu  S,  Zhang  Y, et al. (2012) Impact of GLP-1 receptor agonists on major gastrointestinal disorders for type 2 diabetes mellitus: a mixed treatment comparison meta-analysis. Exp Diabetes Res 2012: 230624.