The Association Between Second-Line Oral Antihyperglycemic Medication on Types of Dementia in Type 2 Diabetes: A Nationwide Real-World Longitudinal Study

Abstract

Background:

There are few reports that evaluated the association between various types of dementia and dual oral therapy with antihyperglycemic medication.

Objective:

The goal of this study was to investigate the association between treatment of dual antihyperglycemic medication and dementia subclass in type 2 diabetes mellitus using the Korean National Health Insurance System.

Methods:

This study included 701,193 individuals with diabetes prescribed dual oral therapy between 2009 and 2012 from the Korean National Health Insurance Service Database, which were tracked until 2017. All-cause, Alzheimer’s (AD) and vascular dementia (VaD) were investigated by dual oral therapy. Adjustments were made for age, sex, income, diabetes duration, hypertension, dyslipidemia, smoking, drinking, exercise, body mass index, glucose level, and estimated glomerular filtration rate.

Results:

Dual therapy with metformin (Met) + dipeptidyl peptidase-4 inhibitor (DPP-4i), Met + thiazolidinedione (TZD), and sulfonylurea (SU) + thiazolidinediones (TZD) were significantly associated with all-cause dementia (HR = 0.904, 0.804, and 0.962, respectively) and VaD (HR = 0.865, 0.725, and 0.911, respectively), compared with Met + SU. Met + DPP-4i and Met + TZD were associated with significantly lower risk of AD (HR = 0.922 and 0.812), compared with Met + SU. Dual therapy with TZD was associated with a significantly lower risk of all-cause dementia, AD, and VaD than nonusers of TZD (HR = 0.918, 0.925 and 0.859, respectively).

Conclusion:

Adding TZD or DPP-4i instead of SU as second-line anti-diabetic treatment may be considered for delaying or preventing dementia. Also, TZD users relative to TZD non-users on dual oral therapy were significantly associated with lower risk of various types of dementia.

Keywords

Alzheimer’s disease dementia dipeptidyl-peptidase IV inhibitors population type 2 diabetes mellitus 2 4-thiazolidinedione

INTRODUCTION

The incidence of type 2 diabetes mellitus (T2D) is increasing rapidly and is a growing global health problem. The International Diabetes Foundation (IDF) reports there were 366 million adults with diabetes worldwide in 2011 and there will be 552 million by 2030 [1]. To control the glucose status in T2D and decrease diabetes complications, the emergence of new anti-diabetic medications has occurred over the last two decades. Along with the discovery of heterogeneity in diabetes pathogenesis [2], the number of patients on dual and triple combination therapy has steadily increased [3, 4]. According to the guidelines for diabetes management, metformin (Met) is recommended as first-line therapy, and various other oral and injectable pharmacological agents and can be combined with it, considering their efficacy, safety, side effects, costs, and co-morbidities such as atherosclerotic cardiovascular disease, heart failure, or chronic kidney disease [5].

T2D is one of the risk factors to increase the risk of dementia. Based on the epidemiological Rotterdam study by Ott et al., T2D is supposed to nearly double the risk of dementia [6 –8]. However, currently, the management of dementia is mainly symptomatic therapy with little effect on delaying its progression or preventing it. A 2013 meta-analysis of 28 prospective observational studies suggested that patients with diabetes are at higher risk of all-cause dementia, Alzheimer’s disease (AD), and vascular dementia (VaD) [9]. T2D shares common pathophysiological components with dementia, such as cardiovascular risk factors, glucose toxicity, insulin resistance, and inflammation [10]. Therefore, anti-diabetic therapy has become an interesting approach to preventing dementia.

Although there was a report stating that there is no evidence supporting a glucose-lowering strategy to delay or prevent dementia [11], diabetes medications may influence the development and progression of cognitive decline in humans with diabetes [12]. Recently, eight human epidemiologic retrospective observational studies with more than 3 years of follow-up have shown associations between T2D medications and dementia risk [13 –20] (see Supplementary Table 1). By using active comparative medication, the risk of dementia was decreased in Met versus sulfonylurea (SU), DPP4 inhibitor (DPP-4i) versus SU, and Met + thiazolidinediones (TZD) versus Met + SU [13 –15]. By using ever-users versus never-users, ever-users of Met, TZD, and DPP-4i diminished the risk of dementia more than never-users [17 –19]. However, in real-world clinical settings, various drug combinations exist. Only Bohlken et al. revealed the benefit of Met, TZD, and Met + SU and the disadvantage of insulin for reducing the risk of dementia in various mono- and dual therapy combinations [16]. Also, there are few reports that evaluated dementia subclasses, and DPP-4i showed inconsistent results for AD and VaD [14, 19].

Therefore, the aim of this study was to analyze the association between dual oral therapy with glucose-lowering drugs and various types of dementia.

METHODS

Study design, populations, and setting

We used a population-based retrospective observational cohort study established by the National Health Insurance Service (NHIS) of Korea between 2009 and 2012, traced to 2017. NHIS is a compulsory single-payer national healthcare coverage system and has covered approximately 50 million population of South Korea since 1989. Detailed information about the NHIS claim database has been made available [21]. Briefly, the NHIS database has access to the medical records of the entire Korean population, and they undergo general health screenings every 2 years. The NHIS claims database contains information on diagnoses (International Classification of Diseases, 10th revision (ICD-10) code), prescription records, and demographics. The NHIS claim database includes major risk factor variables but lacks many medical records and laboratory tests such as body size, brachial blood pressure, and fasting blood chemistry data. Therefore, we combined the NHIS claims database with the health screening examination dataset.

From the NHIS claims database, we extracted 2,703,713 people among those who had undergone health check-ups, including those who had a blood sugar level of 126 or higher at their health screening examination or those who were prescribed anti-diabetes medication and were given a diagnosis of T2D (ICD-10 code, E11 to E14) between January 2009 and December 2012. The exclusion criteria are as follows: age < 40, no anti-diabetic medication within 1 year of the examination, those who had developed dementia before the examination or developed dementia within 1 year of the examination, and those with missing data (See Fig. 1). The study cohort was followed up to 2017. All subjects provided written informed consent, and this study protocol was approved by the Institutional Review Board of the Sungkyunkwn University School of Medicine/Kangbuk Samsung Hospital (IRB No. 2018-01-036) and the Institutional Review Board of the NHIS (number: NHIS-2019-1-218). Informed consent was waived by the IRB.

Fig. 1

Flow chart of the sample selection process.

The study therapies were defined as follows: Met, SU, TZD, DPP-4i, Meglitinide (Megl), α-glucosidase inhibitor (AGI), and insulin because the use of these medication classes was common at the time of the baseline of the study (in contrast to other drug classes such as SGLT-2 inhibitors and GLP-1 agonists). Among the 1,578,322 people who used anti-diabetic medications, there were many differences in baseline characteristics for each group with only an insulin regimen, oral monotherapy, dual oral therapy, and triple oral therapy (Supplementary Table 2). The association between anti-diabetic medications and dementia was investigated by stratifying with/without insulin treatment due to different baseline characteristics. Based on Korean guideline of diabetes management, Met has been usually used as first line pharmacotherapy to treat people with T2D. After preliminary analysis, we focused on the association between dual oral therapy and various types of dementia by comparing Met + SU group as reference group.

Study outcomes

The primary outcome was the first diagnosis of all-cause dementia (ICD-10 codes, F00∼03, F107, G30, G310, G311, G318, and a prescription of anti-dementia medication), AD (ICD-10 codes, F00, G30, and a prescription of anti-dementia medication), or VaD (ICD-10 codes, F01, and a prescription of anti-dementia medication).

Covariates

Baseline comorbidities of patients including hypertension and dyslipidemia were analyzed based on existing corresponding ICD-10 codes in the claims data and medication history. The study covariates included age at index date, gender, duration of diabetes, income, presence of hypertension, presence of dyslipidemia, smoking, drinking, presence of regular exercise, body mass index, systolic and diastolic blood pressure, total cholesterol, fasting glucose, and estimated glomerular filtration rate.

Statistical analyses

Data are presented as the mean±SD for normally distributed continuous variables and as proportions for categorical variables. ANOVA was used to analyze the continuous variables, and the differences between nominal variables were compared with the chi-square test. Annual event rates are described as the number of events per 1000 person-years (PY). Hazard ratios (HR) and their corresponding 95% confidence intervals (CI) were calculated using Cox proportional hazard models for investigating the association between dementia and diabetes medication, after adjusting for age, sex, income, hypertension, dyslipidemia, DM duration, smoking, drinking, exercise, body mass index (BMI), glucose, and the estimated GFR (glomerular filtration rate). Subgroup analyses differentiated by multiple cardiovascular risk factors were subsequently performed. All p-values were two-sided, and a value < 0.05 was considered statistically significant. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).

RESULTS

Characteristics of study population

Among the 1,778,322 patients receiving diabetes medication, baseline characteristics according to the number of insulin and oral medications were presented in Supplementary Table 2. A total of 701,193 patients with dual oral therapy were selected in the cohort, averaging 60.6 years old. Patients in the SU + TZD group demonstrated a longer duration of diabetes and those in the Met + SU group showed lower income. The baseline characteristics of each group are presented in Table 1.

Table 1

Baseline characteristics of combination therapy group without insulin. Data are presented as mean±S.D. or percent

	Met + SU	Met + DPP-4i	Met + TZD	Met + AGI	SU + TZD	SU + AGI	Other dual therapy	p
Number	419,442	134,195	37,401	12,320	53,382	30,261	14,192
Age (y)	60.7±10.2	57.9±10.1^*	58.7±9.7^*	61.8±10.2^*	60.1±9.7^*	63.7±10.0^*	61.4±10.2^*	<0.0001
Gender, male (%)	55.9	56.1^*	57.1^*	51.3^*	57.6^*	54.6^*	56.4^*	<0.0001
DM duration, ≥5 y (%)	91.4	85.7^*	90.6^*	92.1^*	96.8^*	96.6	93.4^*	<0.0001
Income, low 20% (%)	23.3	21.00^*	20.3^*	22.3^*	21.3	23.0	21.6^*	<0.0001
Hypertension (%)	65.1	60.0^*	61.3^*	65.4^*	64.3^*	67.7	65.6	<0.0001
Dyslipidemia (%)	47.0	56.0^*	60.0^*	51.7^*	57.3^*	48.0^*	53.7^*	<0.0001
Smoking (%)								<0.0001
Non	59.2	57.2^*	58.4^*	62.7^*	58.1^*	62.3^*	59.3^*
Ex	18.1	20.3^*	21.3^*	18.5^*	19.2^*	18.1^*	20.8^*
Current	22.7	22.5^*	20.3^*	18.9^*	22.7^*	19.7^*	19.9^*
Drinking (%)								<0.0001
Non	63.4	62.6^*	61.5^*	68.6^*	62.4^*	68.2^*	67.4^*
Mild	27.9	29.1^*	30.3^*	24.7^*	29.2^*	24.5^*	25.7^*
Heavy	8.8	8.4^*	8.2^*	6.7^*	8.4^*	7.3^*	7.0^*
Regular exercise (%)	21.9	22.9^*	24.3^*	22.2^*	23.0	22.1	23.4^*	<0.0001
Body Mass Index (kg/m²)	25.2±3.2	25.2±3.3^*	25.6±3.3^*	24.6±3.3^*	25.8±3.5^*	24.7±3.2^*	24.7±3.3^*	<0.0001
SBP (mmHg)	129.6±15.7	126.9±15.2^*	126.9±14.9^*	127.7±15.4^*	128.5±15.2^*	129.3±15.9^*	128.2±15.9^*	<0.0001
DBP (mmHg)	78.9±9.9	77.7±9.9^*	77.3±9.7^*	77.7±9.7^*	77.5±9.8^*	77.9±10.0^*	77.4±10.1^*	<0.0001
Total cholesterol (mg/dL)	191.4±41.0	183.4±41.9^*	187.3±40.8	186.1±40.2^*	191.6±42.1^*	190.2±42.5^*	183.9±41.3^*	<0.0001
Fasting glucose (mg/dL)	147.3±50.4	142.6±46.3^*	130.9±40.0^*	133.4±41.1^*	143.0±48.7^*	143.8±49.6^*	143.8±48.1^*	<0.0001
eGFR (mL/min/m²)	84.4±35.0	86.3±37.4^*	85.1±39.1^*	82.2±31.1^*	82.5±33.9^*	79.0±38.6^*	80.0±40.6^*	<0.0001

DM, diabetes mellitus; SBP, systolic blood pressure; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate. ^*p < 0.05, for statistical difference compared with Met + SU group.

The association between oral anti-diabetic medication and dementia type in dual combination therapy

Table 1 shows that people treated with Met + DPP-4i or Met + TZD versus those treated with Met + SU or other SU combinations were younger and had relatively shorter diabetes duration and relatively higher income. A total of 63,546 cases of any dementia were observed during the 6,050,444 person-years comprising the follow-up period. Compared with dual therapy with Met + SU, those treated with Met+ DPP-4i, Met + TZD, and SU + TZD therapy were associated with a lower risk of all-cause dementia, AD, and VaD in the non-adjusted model. After full adjustments, compared with dual therapy with Met + SU, Met + DPP-4i, Met + TZD, and SU + TZD therapy were associated with a lower risk of all-cause dementia (HR (95% CI) = 0.904 (0.879–0.929), 0.804 (0.767–0.844), and 0.962 (0.929–0.996), respectively) and VaD (HR (95% CI) = 0.865 (0.806–0.928), 0.725 (0.64–0.822), and 0.91 1(0.833–0.995), respectively). Also, compared with the group treated with SU + Met, those treated with Met + DPP-4i and Met + TZD were associated with a significantly lower risk of AD (HR (95% CI) = 0.922 (0.894–0.951) and 0.812 (0.77–0.857)) except for SU + TZD (HR (95% CI) = 0.971 (0.934–1.01)) (Table 2). However, combination dual oral therapy with insulin did not show significant results in lowering dementia (Supplementary Tables 3 and 4). Also, there were no significant results in triple oral therapy (data not shown).

Table 2

Incidence rates and hazard ratios of dementia, Alzheimer’s dementia, and vascular dementia by dual oral anti-diabetic medication exposure

	Without insulin
	N	n	PY	Incident rate	MODEL 1 HR	MODEL 2 HR	MODEL 3 HR
				(Per 1,000 PY)	(95% CI)	(95% CI)	(95% CI)
All-cause dementia
Met + SU	419,442	31,770	2936018.5	10.82	1 (ref.)	1 (ref.)	1 (ref.)
Met + DPP-4i	134,195	6,046	887640.41	6.81	0.656 (0.638, 0.674)	0.905 (0.88, 0.931)	0.904 (0.879, 0.929)
Met + TZD	37,401	1,760	272680.1	6.45	0.587 (0.559, 0.615)	0.773 (0.737, 0.811)	0.804 (0.767, 0.844)
Met + AGI	12,320	1,070	86643.5	12.35	1.137 (1.070, 1.208)	1.029 (0.968, 1.093)	1.014 (0.954, 1.077)
SU + TZD	53,382	3,522	388984.35	9.05	0.820 (0.792, 0.849)	0.934 (0.902, 0.967)	0.962 (0.929, 0.996)
SU + AGI	30,261	3,267	211409.38	15.45	1.419 (1.369, 1.471)	1.044 (1.007, 1.082)	1.021 (0.985, 1.059)
Other dual therapy	14,192	1,164	96847.2	12.02	1.122 (1.059, 1.19)	1.059 (0.998, 1.122)	1.045 (0.985, 1.108)
Alzheimer’s disease
Met + SU	419,442	25,382	2,936,018.5	8.65	1 (ref.)	1 (ref.)	1 (ref.)
Met + DPP-4i	134,195	4,855	887,640.41	5.47	0.665 (0.645, 0.686)	0.925 (0.897, 0.954)	0.922 (0.894, 0.951)
Met + TZD	37,401	1,417	272,680.1	5.20	0.589 (0.559, 0.622)	0.784 (0.743, 0.827)	0.812 (0.77, 0.857)
Met + AGI	12,320	848	86,643.5	9.79	1.127 (1.052, 1.207)	1.017 (0.950, 1.089)	1.001 (0.935, 1.072)
SU + TZD	53,382	2,833	388,984.35	7.28	0.823 (0.791, 0.855)	0.943 (0.907, 0.98)	0.971 (0.934, 1.010)
SU + AGI	30,261	2,588	211,409.38	12.24	1.404 (1.348, 1.462)	1.027 (0.986, 1.069)	1.006 (0.966, 1.047)
Other dual therapy	14,192	947	96,847.2	9.78	1.145 (1.073, 1.222)	1.08 (1.012, 1.153)	1.067 (1.000, 1.138)
Vascular dementia
Met + SU	419,442	4,985	2936018.5	1.70	1 (ref.)	1 (ref.)	1 (ref.)
Met + DPP-4i	134,195	942	887640.41	1.06	0.647 (0.603, 0.694)	0.862 (0.804, 0.925)	0.865 (0.806, 0.928)
Met + TZD	37,401	259	272680.1	0.95	0.551 (0.486, 0.624)	0.695 (0.613, 0.788)	0.725 (0.640, 0.822)
Met + AGI	12,320	178	86643.5	2.05	1.206 (1.039, 1.401)	1.107 (0.953, 1.285)	1.098 (0.946, 1.276)
SU + TZD	53,382	543	388984.35	1.40	0.808 (0.739, 0.883)	0.890 (0.814, 0.973)	0.911 (0.833, 0.995)
SU + AGI	30,261	521	211409.38	2.46	1.444 (1.320, 1.581)	1.100 (1.005, 1.204)	1.076 (0.983, 1.178)
Other dual therapy	14,192	184	96847.2	1.90	1.130 (0.975, 1.309)	1.065 (0.919, 1.234)	1.050 (0.906, 1.217)

N, cases followed; n, incident cases of dementia; PY, Person-Years; HR, hazard ratio; CI, confidence interval; SU, Sulfonylurea; Met, Metformin; TZD, thiazolidinediones; DPP-4i, dipeptidyl peptidase 4 inhibitor; Megl, Meglitinide; AGI, alpha-glucosidase inhibitor; MODEL 1, unadjusted; MODEL 2, adjusted for MODEL 1 + gender, income, hypertension, dyslipidemia, duration of diabetes; MODEL 3, adjusted for MODEL 2 + smoking, drinking, exercise, body mass index, fasting glucose, estimated glomerular filtration rate.

The effect of TZD on dual oral therapy and the types of dementia

TZD users showed a lower risk of dementia compared with nonusers of TZD on dual oral anti-diabetic medications (Table 3). After adjustment, compared with the nonusers of TZD, dual therapy with TZD was associated with a significantly lower risk of all-cause dementia, AD, and VaD (HR (95% CI) = 0.918 (0.892–0.944), 0.925 (0.896–0.955), and 0.859 (0.798–0.924)).

Table 3

Incidence rates and hazard ratios of dementia, Alzheimer’s dementia, and vascular dementia by TZD and non-TZD based dual oral therapy exposure

	N	n	PY	Incident rate	MODEL 1	MODEL 2	MODEL 3
					HR (95% CI)	HR (95% CI)	HR (95% CI)
All-cause dementia
without TZD	609892	43266	3604967.31	12.00177319	1 (ref.)	1 (ref.)	1 (ref.)
with TZD	91301	5333	574063.13	9.289919037	0.76 (0.739, 0.782)	0.886 (0.861, 0.912)	0.918 (0.892, 0.944)
Alzheimer’s disease
without TZD	609892	34579	3604967.31	9.592042597	1 (ref.)	1 (ref.)	1 (ref.)
with TZD	91301	4291	574063.13	7.47478766	0.761 (0.737, 0.785)	0.894 (0.866, 0.923)	0.925 (0.896, 0.955)
Vascular dementia
without TZD	609892	6799	3604967.31	1.88600878	1 (ref.)	1 (ref.)	1 (ref.)
with TZD	91301	813	574063.13	1.416220547	0.74 (0.688, 0.795)	0.832 (0.774, 0.895)	0.859 (0.798, 0.924)

N, cases followed; n, incident cases of dementia; PY, Person-Years; HR, hazard ratio; CI, confidence interval; TZD, thiazolidinediones; MODEL 1: unadjusted; MODEL 2, adjusted for MODEL 1 + gender, income, hypertension, dyslipidemia, duration of diabetes; MODEL 3: adjusted for MODEL 2 + smoking, drinking, exercise, body mass index, fasting glucose, estimated glomerular filtration rate.

DISCUSSION

This retrospective study, including more than 1,578,000 patients with T2D with long term follow-up (median follow-up 6.94 years), provided clinical evidence of the comparative effects of TZD and DPP-4i versus other second-line oral anti-diabetic medications on the risk all-cause dementia, AD, and VaD. Also, compared with that of Met + SU, the dual combination of TZD without Met showed a significantly lower risk of all-cause dementia and VaD. However, there was no benefit of combination with insulin or the use of more than dual therapy to prevent dementia in this observational study.

In this study, any combination of TZD in dual oral therapy showed benefit in all-cause dementia and VaD. TZD activates the proliferator-activated receptor-γ (PPAR-γ) that has insulin sensitivity [22], and anti-inflammatory, anti-oxidative, anti-acetylcholinesterase, and neuroprotective activity in vascular disease animal models [23]. It is known that TZDs may decrease recurrent stroke and related vascular events both in patients with T2D and even in those with insulin resistance without diabetes [24, 25]. Vascular dementia can also be caused by a series of small strokes. The authors have searched for an index showing insulin resistance well [26, 27]. Our findings support that selecting TZDs as second-line therapy may be good option to prevent VaD. In AD, TZD dual combination except SU + TZD revealed the benefit in AD. The effect of TZDs on insulin sensitivity induces anti-inflammatory and anti-amyloidogenic effects in AD [28]. Further study may be helpful to show why combination of SU + TZD than that of Met + SU did not lower AD events.

The real-world benefit of TZD in dementia is supported by large population-based retrospective observational research with two Taiwan National Health Insurance datasets and one German general practice dataset [15 –17]. Tseung et al. showed that TZD ever-users versus never-users was associated with a lower risk of dementia (HR = 0.716 95% CI 0.545–0.940) [17]. In a matched case-control German study, Bohlken et al. showed that the risk of dementia in general practice, regardless of the prescription duration, was reduced in TZD ever-users versus never-users (OR = 0.80, 95% CI 0.68–0.95), but not for Met (OR = 0.96, 95% CI 0.88–1.04), DPP-4i (OR = 0.99, 95% CI 0.91–1.07), or SU (OR = 1.03, 95% CI 0.96–1.10), and it was increased in insulin users (OR = 1.34, 95% CI 1.24–1.44) [16]. In combination with Met, Lu et al. reported that 204,323 Taiwan patients with T2D and≥65 years that MET + TZD was significantly superior to Met + SU in the risk of developing dementia (HR = 0.56) [15]. On the other hand, a nationwide nested case-control study, enrolled from 1995 to 2012, reported that TZD ever-users, including rosiglitazone, versus never-users did not have significant odds of developing dementia (OR = 0.89, 95% CI 0.76–1.03), but further analysis of dual therapy with TZD was not conducted [20]. In Korea, there are two TZDs, lobeglitazone and pioglitazone, approved by the Ministry of Food and Drug Safety. Rosiglitazone was excluded by enrolment starting in 2009 in our study due to controversy about its role in cardiovascular events [29]. The advantage of TZD for preventing any type of dementia becomes apparent in dual therapy.

DPP-4i increases the glucose-dependent insulin response and reduces glucagon secretion by inhibiting GLP-1 degradation [30]. Wium-Andersen et al. showed DPP-4i ever-users had lower odds of dementia than never-users (OR = 0.80, 95% CI 0.74–0.88) and the Met + DPP-4i dual combination was associated with lower odds of dementia than no anti-diabetes medication (OR = 0.70, 95% CI 0.58–0.86) [20]. In subclasses of dementia, Kim, et al. demonstrated that DPP-4i new users versus SU new users over 60 years old had a lower risk of all-cause dementia and AD (HR = 0.66, 95% CI 0.56–0.78 and HR = 0.64, 95% CI = 0.52–0.79), but not VaD (HR = 0.56, 95% CI 0.38–1.14) over 4 years [14]. Although a meta-analysis of three large randomized controlled trials of DPP-4i (the SAVOR-TIMI 53, EXAMINE, and TECOS trial) versus placebo showed no significant reduction in the risk of stroke [31], a predisposing factor for VaD, the median follow-up period (1.5∼3 years) might be too short to see the difference in these events. On the other hand, Taiwan national data of those aged≥50 years by Chen et al. showed DPP-4i new users versus non DPP-4i users had a lower risk of all-cause dementia (HR = 0.80, 95% CI 0.68–0.88) and VaD (HR = 0.61, 95% CI 0.40–0.68), but not AD (HR = 0.89, 95% CI 0.71–1.27) with a mean duration of 7 years of follow-up. The difference between these studies may be attributed to their study designs, their demographics and their population sample sizes, as well as the duration of follow up. Compared to SU, a beneficial effect of DPP-4i on AD was more evident in dual therapy with Met in our study.

Also, the combination with insulin lost the benefits of protective dual therapy on dementia in our study (Supplementary Table 4). The beneficial effect of DPP-4i and TZD as second-line therapy on dementia went away in cases of insulin combination. This is consistent with the study by Bohlken et al., where they showed that insulin was associated with an increased risk of developing dementia (OR = 1.34, 95% CI 1.24–1.44) and a combination with insulin was a risk factor of dementia [16]. A pooled analysis of 5 other cohorts also showed insulin users versus non-users were associated with an increased risk of new-onset dementia (pooled HR = 1.58, 95% CI 1.18–2.12) [32]. Although patients treated with insulin are likely to have a longer duration of diabetes, poor glucose control, and more comorbidities than those on oral hypoglycemic agents, the association between dementia and insulin still remains high even after adjustment of these confounding factors. However, caution is needed in interpreting these observational studies. There are some reports that intranasal insulin might be associated with improvement in brain function and cognitive changes [33, 34]. Insulin itself should not be mistaken as a cause of dementia. A meta-analysis of five observational studies reported a significantly increased risk of dementia in patients who had hypoglycemic events (pooled OR = 1.68, 95% CI 1.45–1.95) [35]. Several possible mechanisms have been suggested from animal and human studies [36 –38], where hypoglycemia might induce neuronal cell death in brain memory areas, increase adrenaline levels and cause endothelial dysfunction. People who take insulin injections among their diabetes drugs would have frequent events of hypoglycemia and that could be one reason why TZD, DPP-4i, and Met have better effects on dementia than insulin medication.

There were some limitations to this analysis. First, this was a retrospective observational study. Therefore, the actual mechanism of the association between anti-diabetic medications and dementia cannot be clarified. Second, discrepancies between the actual diagnosis and claim data might be possible. The accuracy of diagnoses in claim databases is debatable. T2D and dementia prevalence may be underestimated because only subjects seeking a medical evaluation can be identified. Third, sodium glucose co-transporter 2 (SGLT2) inhibitors and glucagon like peptide 1 (GLP1) receptor agonists were not included in the analysis of this study, because of not commonly used during the observation period. Fourth, the monotherapy and more than dual oral therapy in real clinical environment is beyond the scope of this study. It is recommended that Met usually be prescribed first in monotherapy according to Korean guideline of diabetes management [39]. Also, it tends to be difficult to maintain more than dual oral therapy without changing the prescription.

In conclusion, we investigated whether different types of anti-diabetic medications and treatment regimens (combinations of anti-diabetic drugs) were associated with the risk of developing dementia. Adding a TZD or DPP-4i instead of SU as second-line treatment to first-line metformin might be considered for delaying dementia. TZD combined with other than Met in dual therapy also had a significantly lower risk of dementia.

DISCLOSURE STATEMENT

Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/20-1535r1).

Footnotes

The supplementary material is available in the electronic version of this article: .

References

Saeedi

, Petersohn

, Salpea

, Malanda

, Karuranga

, Unwin

, Colagiuri

, Guariguata

, Motala

, Ogurtsova

, Shaw

, Bright

, Williams

; IDF Diabetes Atlas Committee (2019) Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res Clin Pract 157, 107843.

Defronzo

(2009) Banting Lecture. From the triumvirate to the ominous octet: A new paradigm for the treatment of type 2 diabetes mellitus. Diabetes 58, 773–795.

, Kim

, Park

, Jung

, Kwon

, Park

, Song

, Han

, Lee

, Ko

; Task Force Team for Diabetes Fact Sheet of the Korean Diabetes Association (2016) Trends of antidiabetic drug use in adult type 2 diabetes in Korea in 2002-2013: Nationwide population-based cohort study. Medicine 95, e4018.

Wilkinson

, Douglas

, Stirnadel-Farrant

, Fogarty

, Pokrajac

, Smeeth

, Tomlinson

(2018) Changing use of antidiabetic drugs in the UK: Trends in prescribing 2000–2017. BMJ Open 8, e022768.

Davies

, D’Alessio

, Fradkin

, Kernan

, Mathieu

, Mingrone

, Rossing

, Tsapas

, Wexler

, Buse

(2018) Management of hyperglycemia in type 2 diabetes, 2018. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 41, 2669–2701.

Ott

, Stolk

R P

, Hofman

, van Harskamp

, Grobbee

D E

, Breteler

(1996) Association of diabetes mellitus and dementia: The Rotterdam Study. Diabetologia 39, 1392–1397.

Huang

, Chung

, Leu

, Lin

, Chiu

, Hsu

, Chiang

, Huang

, Chen

, Lin

, Chen

, Chan

(2014) Diabetes mellitus and the risk of Alzheimer’s disease: A nationwide population-based study. PLoS One 9, e87095.

Ohara

, Doi

, Ninomiya

, Hirakawa

, Hata

, Iwaki

, Kanba

, Kiyohara

(2011) Glucose tolerance status and risk of dementia in the community: The Hisayama Study. Neurology 77, 1126–1134.

Gudala

, Bansal

, Schifano

, Bhansali

(2013) Diabetes mellitus and risk of dementia: A meta-analysis of prospective observational studies. J Diabetes Investig 4, 640–650.

10.

Ninomiya

(2014) Diabetes mellitus and dementia. Curr Diab Rep 14, 487.

11.

Areosa Sastre

, Vernooij

, González-Colaço Harmand

, Martínez

(2017) Effect of the treatment of type 2 diabetes mellitus on the development of cognitive impairment and dementia. Cochrane Database Syst Rev 6, CD003804.

12.

Verma

, Despa

(2019) Contributing factors to diabetic brain injury and cognitive decline. Diabetes Metab 43, 560–567.

13.

Orkaby

, Cho

, Cormack

, Gagnon

, Driver

(2017) Metformin vs sulfonylurea use and risk of dementia in US veterans aged≥65 years with diabetes. Neurology 89, 1877–1885.

14.

Kim

, Jeon

, Kim

, Lee

, Moon

, Han

(2018) Risk of dementia in older patients with type 2 diabetes on dipeptidyl-peptidase IV inhibitors versus sulfonylureas: A real-world population-based cohort study. J Clin Med 8, 28.

15.

, Yang

, Li

, Hsieh

, Ou

(2018) Lower risk of dementia with pioglitazone, compared with other second-line treatments, in metformin-based dual therapy: A population-based longitudinal study. Diabetologia 61, 562–573.

16.

Bohlken

, Jacob

, Kostev

(2018) Association between the use of antihyperglycemic drugs and dementia risk: A case-control study. J Alzheimers Dis 66, 725–732.

17.

Tseng

(2018) Pioglitazone reduces dementia risk in patients with type 2 diabetes mellitus: A retrospective cohort analysis. J Clin Med 7, 306.

18.

Tseng

(2019) Metformin and the risk of dementia in type 2 diabetes patients. Aging Dis 10, 37–48.

19.

Chen

, Chung

, Lu

, Tzeng

, Lee

, Su

, Kuo

, Liu

, Hsieh

, Chien

(2020) Association between the use of dipeptidyl peptidase 4 inhibitors and the risk of dementia among patients with type 2 diabetes in Taiwan. J Clin Med 9, 660.

20.

Wium-Andersen

, Osler

, Jorgensen

, Rungby

, Wium-Andersen

(2019) Antidiabetic medication and risk of dementia in patients with type 2 diabetes. A nested case-control study. Eur J Endocrinol 181, 499–507.

21.

Song

, Jung

, Song

, Park

, Kwon

, Cha

, Park

, Lee

, Ko

, Lee

(2014) Background and data configuration process of a nationwide population-based study using the Korean National Health Insurance System. Diabetes Metab J 38, 395–403.

22.

Inzucchi

(2002) Oral antihyperglycemic therapy for type 2 diabetes: Scientific review. JAMA 287, 360–372.

23.

Sharma

, Singh

(2011) Behavioral and biochemical investigations to explore pharmacological potential of PPAR-gamma agonists in vascular dementia of diabetic rats. Pharmacol Biochem Behav 100, 320–329.

24.

Dormandy

, Charbonnel

, Eckland

, Erdmann

, Massi-Benedetti

, Moules

, Skene

, Tan

, Lefèbvre

, Murray

, Standl

, Wilcox

, Wilhelmsen

, Betteridge

, Birkeland

, Golay

, Heine

, Korányi

, Laakso

, Mokán

, Norkus

, Pirags

, Podar

, Scheen

, Scherbaum

, Schernthaner

, Schmitz

, Skrha

, Smith

, Taton

; PROactive Investigators (2005) Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive study (PROspective pioglitAzone Clinical Trial In macroVascular Events): A randomized controlled trial. Lancet 366, 1279–1289.

25.

Kernan

, Viscoli

, Furie

, Young

, Inzucchi

, Gorman

, Guarino

, Lovejoy

, Peduzzi

, Conwit

, Brass

, Schwartz

, Adams

, Berger

, Carolei

, Clark

, Coull

, Ford

, Kleindorfer

, O’Leary

, Parsons

, Ringleb

, Sen

, David-Spence

, Tanne

, Wang

, Winder

(2016) Pioglitazone after ischemic stroke or transient ischemic attack. N Engl J Med 374, 1321–1331.

26.

Song

, Hwang

, Ahn

, Park

(2016) Comparison of the usefulness of the updated homeostasis model assessment (HOMA2) with the original HOMA1 in the prediction of type 2 diabetes mellitus in Koreans. Diabetes Metab J 40, 318–325.

27.

Hong

, Han

, Park

(2021) The insulin resistance by triglyceride glucose index and risk for dementia: Population-based study. Alzheimers Res Ther 13, 9.

28.

Heneka

, Reyes-Irisarri

, Hull

, P Kummer

(2011) Impact and therapeutic potential of PPARs in Alzheimer’s disease. Curr Neuropharmacol 9, 643–650.

29.

Nissen

, Wolski

(2007) Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 356, 2457–2471.

30.

Lovshin

, Drucker

(2009) Incretin-based therapies for type 2 diabetes mellitus. Nat Rev Endocrinol 5, 262–269.

31.

Mannucci

, Mosenzon

, Avogaro

(2016) Analyses of results from cardiovascular safety trials with DPP-4 inhibitors: Cardiovascular outcomes, predefined safety outcomes, and pooled analysis and meta-analysis. Diabetes Care 39, S196–S204.

32.

Weinstein

, Davis-Plourde

, Conner

, Himali

, Beiser

, Lee

, Rawlings

, Sedaghat

, Ding

, Moshier

, van Duijn

, Beeri

, Selvin

, Ikram

, Launer

, .Haan

, Seshadri

(2019) Association of metformin, sulfonylurea and insulin use with brain structure and function and risk of dementia and Alzheimer’s disease: Pooled analysis from 5 cohorts. PLoS One 14, e0212293.

33.

Craft

, Claxton

, Baker

, Hanson

, Cholerton

, Trittschuh

, Dahl

, Caulder

, Neth

, Montine

, Jung

, Maldjian

, Whitlow

, Friedman

(2017) Effects of regular and long-acting insulin on cognition and Alzheimer’s disease biomarkers: A pilot clinical trial. J Alzheimers Dis 57, 1325–1334.

34.

Mustapic

, Tran

, Craft

, Kapogiannis

(2019) Extracellular vesicle biomarkers track cognitive changes following intranasal insulin in Alzheimer’s disease. J Alzheimers Dis 69, 489–498.

35.

Mattishent

, Loke

(2016) Bi-directional interaction between hypoglycaemia and cognitive impairment in elderly patients treated with glucose-lowering agents: A systematic review and meta-analysis. Diabetes Obes Meta 18, 135–141.

36.

Languren

, Montiel

, Julio-Amilpas

, Massieu

(2013) Neuronal damage and cognitive impairment associated with hypoglycemia: An integrated view. Neurochem Int 63, 331–343.

37.

Fisher

, Hepburn

, Smith

, Frier

(1992) Responses of peripheral blood cells to acute insulin-induced hypoglycaemia in humans: Effect of alpha-adrenergic blockade. Horm Metab Res Suppl 26, 109–110.

38.

Wright

, Newby

, Stirling

, Ludlam

, Macdonald

, Frier

(2010) Effects of acute insulin-induced hypoglycemia on indices of inflammation: Putative mechanism for aggravating vascular disease in diabetes. Diabetes Care 33, 1591–1597.

39.

Kim

, Ko

, Kim

, Kang

, Noh

, Kim

, Park

, Hur

, Chon

, Moon

, Kim

, Rhee

, Lee

, Kim

, Rhee

, Chun

, Yu

, Kim

, Kwon

, Park

; Committee of Clinical Practice Guidelines, Korean Diabetes Association (2019) 2019 clinical practice guidelines for type 2 diabetes mellitus in Korea. Diabetes Metab J 43, 398–406.