Handgrip strength in patients with type 2 diabetes correlates with diabetic polyneuropathy. A single-center, retrospective observational study in Japanese patients

Abstract

Nerve conduction studies (NCS) are the standard method for diagnosing diabetic polyneuropathy. Because a clear association between handgrip strength and diabetic neuropathy can serve as a screening tool, the present study evaluated the association between handgrip strength and NCS and diabetes-related complications. A total of 436 patients with type 2 diabetes (T2D) who were admitted to our hospital between April 1, 2018 and March 31, 2023, and evaluated using Baba’s diabetic neuropathy classification (BDC) were included. The participants were grouped by sex using the grip strength tertile method to assess correlations with the prevalence of diabetic microvascular complications in the high-handgrip group (HG), middle-handgrip group (MG), and low-handgrip group (LG). The percentage of BDC-0 was 65% in the HG, 54% in the MG, and 36% in the LG. Furthermore, none of the participants in the HG had BDC-3/4, whereas 4% in the MG and 15% in the LG had BDC-3/4. The morbidity progression of diabetic neuropathy was seen in the order of LG, MG, and HG (p < 0.001). Patients with T2D and advanced diabetic neuropathy had decreased handgrip strength. Early evaluation of BDC and other NCS should be considered if decreased handgrip strength is evident.

Introduction

Diabetic polyneuropathy (DPN), a microvascular complication of diabetes, occurs relatively early compared with other microvascular complications [1]. The gold standard for diagnosing DPN is nerve conduction studies (NCS) [2]. Baba’s diabetic neuropathy classification (BDC) has been proposed to assess nerve conduction velocity in patients with DPN and is useful for quantitatively assessing neuropathy [3, 4]. We used BDC to accurately assess neuropathy in patients with diabetes at the time of admission. In our previous studies [5], BDC was correlated with the prevalence of diabetic retinopathy, diabetic nephropathy, macrovascular disease, and carotid artery thickening in patients with type 2 diabetes during educational hospitalization.

DPN is a complication that can reduce the quality of life of patients with diabetes [6, 7] and is a risk factor for diabetic foot gangrene [8-10] and unconscious hypoglycemia [11]. Large clinical studies have shown that good long-term glycemic control can prevent the onset and progression of diabetic neuropathy [12, 13]. Early intervention is desirable in patients who have already developed DPN. In contrast, patients with advanced DPN and only negative symptoms tend to be diagnosed late. Although NCS are the standard method for diagnosing DPN, they are difficult to perform regularly in an outpatient setting without the proper medical facilities. Therefore, a simpler and more convenient index-based screening method for diabetic neuropathy is needed.

Handgrip strength is an indicator of muscle mass and can be easily assessed in outpatient settings. A positive correlation has been reported between the treatment response and handgrip strength in patients with type 2 diabetes on an outpatient basis [14]. Although the relationship between the pathological progression of DPN and handgrip strength [15, 16] has been elucidated, few studies have directly evaluated the relationship between handgrip strength and BDC. This study aimed to clarify the correlation between handgrip strength and BDC, a standard method for diagnosing DPN, and its effectiveness as a screening tool for diabetes-related complications.

Materials and Methods

Study design

This was a single-center, retrospective, cross-sectional study of patients with type 2 diabetes admitted to the Department of Diabetes, Endocrinology and Metabolism at the Kawasaki Medical School Hospital. A total of 561 patients between April 1, 2018 and March 31, 2023 were included. The study protocol, including the opt-out informed consent, was approved by the Institutional Review Board of the Kawasaki Medical School (No. 6362-00). This study was conducted in accordance with the principles of the Declaration of Helsinki. The flowchart of the participant selection process is shown in Fig. 1. Of the 561 patients with type 2 diabetes, those younger than 20 years at the time of admission (n = 5), those with diabetes-complicated pregnancies (n = 2), those with malignant complications (n = 59), and those taking corticosteroids or immunosuppressive drugs (n = 12), or both, were excluded. Patients whose nerve conduction velocity (n = 19) or handgrip strength (n = 28) were not assessed were subsequently excluded. Finally, 436 patients with type 2 diabetes, who were evaluated for grip strength and BDC with nerve conduction studies were included. Study participants were grouped by sex using the handgrip strength tertile method. Men with a handgrip strength of 37–54 kg and women with a handgrip strength of 21–40 kg were categorized in the high-handgrip strength group (HG, n = 142). Men with a handgrip strength of 30–36 kg and women with a handgrip strength of 15–20 kg were in the middle-handgrip strength group (MG, n = 154). Men with a handgrip strength of 11–29 kg and women with a handgrip strength of 5–14 kg were defined as the low-handgrip strength group (LG, n = 138).

Fig. 1  The flowchart of the participants and exclusions in this study

Baba’s diabetic neuropathy classification assessment methods

NCS were conducted in all patients with type 2 diabetes admitted to our hospital by a skilled technician using an MEB-2306 or MEB-2312 system (Nihon Kohden Corporation, Tokyo, Japan). NCS were performed on the upper and lower extremities on one side and on the more symptomatic side if DPN symptoms were present. A neurologist analyzed the obtained data. The BDC evaluation method was the same as previously reported. Sensory nerve conduction velocity (SCV), sensory nerve action potential (SNAP) of the sural nerve, motor nerve conduction velocity (MCV), and compound muscle action potential (CMAP) of the tibial nerve were evaluated [3, 4] (Fig. 2 [5]). When no abnormalities were observed in the nerve conduction study, the designation was BDC 0 (normal). When the MCV of the tibial nerve was <42 m/s, the F-wave latency of the tibial nerve was prolonged, or the A-wave of the tibial nerve or the SCV of the sural nerve was <42 m/s, the patient was classified as BDC 1 (mild). When the SNAP of the sural nerve was <5 μV, the patient was classified as BDC 2 (moderate). BDC 3 (moderate to severe) was classified when the CMAP of the tibial nerve was 2–5 mV, and BDC 4 (severe) was classified when the CMAP was <2 mV. BDC 3 and BDC 4 are clinically severe neurological dysfunction conditions and were analyzed as the same group in this study. As a result, 226 patients were classified as BDC 0, 113 as BDC 1, 70 as BDC 2, and 27 as BDC 3–4.

Fig. 2  How to evaluate Baba’s diabetic classification by nerve conduction studies [5]

Diagnostic criteria for DPN

As diagnostic criteria for diabetic neuropathy, our facility uses the simple diagnostic criteria for diabetic polyneuropathy proposed by the Diabetic Neuropathy Study Group of Japan [17, 18]. If a patient has a history of diabetes and has been ruled out for other peripheral neuropathies, they are diagnosed with diabetic neuropathy if they meet two of the following three criteria: (1) subjective symptoms based on DPN (symptoms of bilateral foot numbness or pain, or abnormal sensations in the toes or soles), (2) decreased or absent bilateral Achilles tendon reflexes, or (3) decreased vibration sensation in the bilateral medial malleoli. In addition to the above three conditions, if abnormalities in conduction velocity, amplitude or latency are observed in two or more nerves in a nerve conduction test, or if there is clinically apparent diabetic autonomic neuropathy (DAN), the condition is diagnosed as diabetic neuropathy. For vibration sensation, a C128 tuning fork was placed on the endocondyle of the tibia, the left and right vibration sensing times were measured, and the average value was calculated. If the vibration time was less than 10 seconds, it was diagnosed as a decrease in the sense of vibration. The coefficient of variation of the R-R interval (CVR-R) was assessed using a 12-lead electrocardiographic device. The CVR-R was calculated from the interval variation of the R wave in 100 consecutive heartbeats during deep breathing. Seven patients with atrial fibrillation were excluded from the evaluation because it was not possible to evaluate the CVR-R. A CVR-R of less than 2.0 was used as the criterion for diagnosing DAN [19, 20].

Collection of patient information

The method for collecting information to assess patients’ backgrounds was similar to that reported previously [5]. Age, duration of diabetes, smoking history, alcohol consumption history, and current medications were obtained upon admission for all participants. Smokers were divided into current, past, and never smokers. Participants who consumed more than 20 g/day of ethanol three or more times per week were considered regular drinkers. All patients were asked by their physicians upon admission whether they had a previous stroke or myocardial infarction. Weight, body mass index (BMI), abdominal circumference, handgrip strength, blood pressure, and pulse were measured upon admission. Handgrip strength was measured on the right and left sides, and the value on the stronger side was used for the analysis. The day after admission, body composition was analyzed using an InBody770 (Muranaka Medical Instruments Co., Ltd., Osaka, Japan). Blood glucose, hemoglobin A1c (HbA1c), lipids, and liver and kidney function were assessed using fasting blood tests the day after admission. The modified Davis staging system was used to evaluate diabetic retinopathy [21]. Ophthalmoscopy was used to classify the disease into non-diabetic retinopathy, simple diabetic retinopathy, preproliferative retinopathy (PPDR), and proliferative diabetic retinopathy (PDR). PPDR was diagnosed when soft leukocoria was observed in addition to the above findings. PDR was diagnosed when neovascularization, vitreous hemorrhage, and tractional retinal detachment were observed in addition to the above findings. Diabetic nephropathy was evaluated based on estimated glomerular filtration rate (eGFR) and albuminuria according to the staging system of the Japan Diabetes Society in the following stages: Stage 1: eGFR >30 mL/min/1.73 m² and urinary albumin <30 mg/gCr; Stage 2: eGFR >30 mL/min/1.73 m² and urinary albumin; Stage 3: eGFR ≥30 mL/min/1.73 m² and urinary albumin ≥300 mg/gCr; Stage 4: eGFR <30 mL/min/1.73 m². Patients in Stage 5 (renal failure) were excluded. The carotid intima-media thickness (IMT) was evaluated by an ultrasound technician using an Aplio series ultrasound system (CANON MEDICAL SYSTEMS, Tochigi, Japan). The carotid IMT was evaluated in the common carotid artery wall 10 mm centered from the carotid sinus. Two IMT points were measured within a 10-mm range, and the mean value was defined as the average IMT. The maximum diameter within the same measurement range was defined as the maximum IMT. The ankle-brachial index (ABI) and cardio-ankle vascular index (CAVI) were calculated by measuring the bilateral upper arm and ankle blood pressure in the supine position and the ratio of the blood pressure in the higher arm to that in the ankle on each side. The lower of the left and right ABIs were included in the analysis.

Statistical analysis

Data were expressed as median (interquartile range). The primary endpoint was the correlation between handgrip strength and BDC. The secondary endpoint was the diabetes-related parameters at admission that could correlate with handgrip strength. To evaluate the differences in parameters among the three groups (HG, MG, and LG), we used the Kruskal-Wallis test, the Dunn test as a post hoc test, and the chi-square test. To evaluate whether the DPN indicators BDC and CVR-R are independent factors that can be used to estimate grip strength, we conducted multiple regression analyses using age, sex, duration of diabetes, BMI, HbA1c, and BDC or CVR-R as explanatory variables. We used dummy variables when using sex, BDC, and CVR-R as the explanatory variables. JMP Pro (17.0.0) was used for analysis, and Microsoft EXCEL for Mac (16.84) was used for tabulation. Images for the graphical abstract were generated using the AI image generation model, DALL-E, developed by OpenAI.

Results

Clinical characteristics of study participants

The clinical characteristics of the study participants are summarized in Table 1. The mean age of the participants was 65 (51–73) years, with an average diabetes duration of 11 (3–19) years. This study included 260 men and 176 women. No differences were observed in the HbA1c levels at admission between the groups. Older age, longer disease duration, and significantly lower BMI and abdominal circumference were seen in the LG than in the HG (p < 0.001 for each). Higher skeletal muscle mass was observed in the HG than in the other groups, although the difference was not statistically significant (p = 0.072). Diastolic blood pressure was significantly higher in the HG than in the MG and LG (p < 0.001). Albumin and alanine transaminase (ALT) levels were significantly higher in the HG than in the LG (p < 0.001). Urea nitrogen and creatinine levels were higher in the LG (p < 0.001) than in the other groups. During hospitalization, sulfonylureas, glinides, thiazolidine, and alpha-glucosidase inhibitors were used more frequently for diabetes treatment in the LG than in the HG or MG. In contrast, sodium-glucose co-transporter 2 (SGLT2) inhibitors were used more frequently in the HG than in the MG or LG (p = 0.008). Hypertension medications were used significantly less frequently in the HG than in the MG and LG (p = 0.007). In addition, we evaluated the trends in arterial stiffness indices in each group. The carotid mean/max IMT was significantly greater in the LG than in the HG (p < 0.001). The ABI was significantly lower in the LG than in the HG and MG (p = 0.005), and the CAVI was higher in the HG, MG, and LG (p < 0.001).

Table 1 Comparison of various parameters in patients classified by handgrip strength

Parameter	HG (n = 142)	MG (n = 154)	LG (n = 138)	p value
Male/female	87/57	87/67	86/52	0.583
Age (years)	53 (44–63)	66 (56–72)	73 (65–78)	<0.001
Duration of diabetes (years)	7 (2–13)	10 (3–19)	18 (9–27)	<0.001
Body weight (kg)	76 (69–88)	67 (59–79)	61 (54–70)	<0.001
BMI (kg/m2)	28.1 (24.8–31.7)	26.0 (22.9–29.4)	24.6 (21.8–26.9)	<0.001
Abdominal circumference (cm)	96 (89–105)	93 (84–101)	88 (80–96)	<0.001
Grip strength (kg)	38 (26–41)	30 (18–34)	19 (12–25)	<0.001
Body fat mass in InBody770 (kg)	22 (16–31)	21 (15–27)	20 (13–27)	0.418
Fat-free body mass in InBody770 (kg)	42 (34–49)	39 (33–44)	38 (34–45)	0.085
Skeletal muscle mass in InBody770 (kg)	27 (22–30)	21 (17–24)	20 (18–25)	0.072
Smoking history (current/past/never, %)	35/20/45	29/23/48	25/30/45	0.217
Drinking history (none or occasional/regular, %)	67/33	67/33	73/27	0.425
History of old myocardial infarction (%)	6	10	15	0.047
History of old cerebral infarction (%)	4	5	14	0.006
Systolic blood pressure (mmHg)	132 (116–143)	129 (118–146)	132 (116–143)	0.730
Diastolic blood pressure (mmHg)	80 (71–92)	76 (67–84)	72 (66–84)	<0.001
Pulse rate (beats/min)	80 (72–91)	78 (70–89)	81 (70–94)	0.432
HbA1c (mmol/mol)	84 (69–104)	82 (65–101)	84 (74–102)	0.102
Triglyceride (mmol/L)	1.4 (1.0–2.0)	1.4 (1.1–2.0)	1.2 (1.0–1.7)	0.036
Total cholesterol (mmol/L)	4.4 (3.6–5.5)	4.5 (3.9–5.2)	4.3 (3.7–4.9)	0.133
LDL-cholesterol (mmol/L)	2.5 (2.1–3.4)	2.6 (2.2–3.2)	2.4 (1.9–3.0)	0.011
HDL-cholesterol (mmol/L)	1.1 (0.9–1.2)	1.1 (0.9–1.4)	1.1 (1.0–1.4)	0.038
Total protein (g/L)	71 (67–73)	70 (67–74)	70 (66–74)	0.333
Albumin (g/L)	42 (40–45)	42 (39–44)	40 (37–43)	<0.001
AST (U/L)	20 (16–27)	21 (16–30)	19 (15–25)	0.168
ALT (U/L)	24 (17–39)	21 (15–32)	17 (12–26)	<0.001
Urea nitrogen (mmol/L)	5.0 (4.3–6.1)	5.2 (3.9–6.5)	6.2 (5.0–7.9)	<0.001
Creatinine (μmol/L)	61 (50–74)	65 (53–81)	71 (57–99)	<0.001
Diagnosis of DPN (%)	29	46	65	<0.001
Subjective symptoms related to DPN in the plantar surfaces of both feet (%)	16	27	40	0.002
Loss of Achilles tendon reflex (%)	26	30	45	0.019
Decrease in vibration sensation (%)	21	25	49	<0.001
BDC-0/BDC-1/BDC-2/BDC-3/4 (n)	93/34/17/0	83/42/23/6	50/37/20/21	<0.001
CVR-R (2.0% or more/less than 2.0%, n)	128/16	120/32	98/30	0.014
Mean IMT (mm)	0.75 (0.67–0.88)	0.81 (0.69–0.95)	0.91 (0.89–1.10)	<0.001
Max IMT (mm)	0.96 (0.84–1.16)	1.04 (0.86–1.23)	1.23 (1.00–1.50)	<0.001
ABI	1.11 (1.06–1.17)	1.12 (1.02–1.18)	1.08 (0.97–1.15)	0.005
CAVI	7.9 (6.8–8.9)	8.8 (7.7–9.8)	9.6 (8.5–10.5)	<0.001
Percentage of drug at time of admission
Insulin injections	23	31	32	0.167
GLP-1 receptor agonist	17	19	22	0.641
Sulfonylurea or Glinide	13	19	29	0.002
Dipeptidyl peptidase-4 inhibitor	38	38	46	0.247
Biguanide	52	45	36	0.397
Thiazolidine	8	5	14	0.037
Alpha glucosidase inhibitor	5	6	13	0.034
Sodium glucose co-transporter 2 inhibitor	42	28	27	0.008
Hypercholesterolemia medications	43	47	55	0.118
Hypertension medications	38	49	55	0.007

Data are presented as median (interquartile range). HG, high grip strength; MG, middle grip strength; LG, low grip strength; BDC, Baba’s diabetic neuropathy classification; BMI, body mass index; DPN, diabetic polyneuropathy; CVR-R, coefficient of variation of R-R interval; IMT, intima-media complex of the carotid artery; GLP-1, glucagon like peptide-1. The results were analyzed by Kruskal-Wallis test or chi-square test.

Relationship between handgrip strength and DPN

The percentage of participants diagnosed with DPN in this study was 29% for HG, 46% for MG, and 65% for LG, and the percentage increased with decreased grip strength (p < 0.001, Table 1). The percentage of participants with subjective symptoms and absent Achilles tendon reflexes associated with DPN also showed a similar trend (p = 0.002, p = 0.019, respectively). The duration of vibration of the lower limb using a tuning fork, an index used to evaluate vibration sensation, was 12 (11–15) seconds for the HG, 12 (10–14) seconds for the MG, and 10 (7–12) seconds for the LG, which were significantly shorter than those for the HG (p = 0.026, p < 0.001, respectively, Fig. 3B). The percentage of people in the LG group who had a reduced sense of vibration was 49%, compared to 21% in the HG group and 25% in the MG group, and was significantly higher in the LG group (p < 0.001, Table 1). The percentages of BDC among participants in the HG, MG, and LG are shown in Fig. 3A. The percentage of BDC-0 was 65%, 54%, and 36% in the HG, MG, and LG, respectively. Additionally, none of the participants in the HG had BDC-3/4, whereas 4% of the MG and 15% of the LG had BDC-3/4. The pathological progression of diabetic neuropathy was observed in the following order: LG, MG, and HG (p < 0.001). The CVR-R at deep breaths was 3.9 (2.7–5.5)% for the HG, 3.6 (2.2–5.0)% for the MG, and 3.0 (2.0–4.4)% for the LG, with significantly lower CVR-R for the LG than in the HG (p = 0.002, Fig. 3C).

Fig. 3  Percentage of Baba’s diabetic neuropathy classification (BDC) in participants in the high grip strength group (HG), middle grip strength group (MG) and low grip strength group (LG) (A). The results were analyzed by the chi-square test. Vibration time at endocondyle of lower limb using C128 tuning fork in each group (B). Coefficient of variation of R-R interval (CVR-R) during deep breathing in each group (C). Data are expressed as IQR (error bars are 90% CI). * p < 0.05, ** p < 0.005, *** p < 0.001, Kruskal-Wallis test was performed, and Dunn test was performed as a post hoc test in comparison with HG.

Next, multiple regression analysis was performed to determine whether BDC and handgrip strength were independently associated with other factors (Table 2). Handgrip strength was independently associated with BDC after adjusting for age, sex, diabetes duration, BMI, and HbA1c. When handgrip strength was evaluated according to BDC, there was no significant difference between BDC-0 and BDC-1. However, in cases with BDC-2 and BDC-3/4, significantly lower handgrip strength was observed than in BDC-0. Even when analyzed separately by sex, the handgrip strength of the participants in the BDC-3/4 category was significantly lower than that of the participants in BDC-0 (Supplementary Tables 1, 2). These results suggest a significant correlation between handgrip strength and BDC and that BDC is an independent factor defining handgrip strength.

Table 2 Grip strength at each BDC when adjusted for multiple factors

	Handgrip strength	p value
BDC-0	25.9 (25.1–26.8)	—
BDC-1	26.0 (24.8–27.3)	0.889
BDC-2	23.9 (22.3–25.4)	0.027
BDC-3/4	20.9 (18.4–23.4)	<0.001

Data are presented as estimated value (95%CI). Adjusted for age, sex, duration of diabetes, body mass index, and HbA1c at admission. BDC, Baba’s diabetic neuropathy classification. The p-values for significance tests with the criterion being BDC-0 are listed.

Using the same method, we also evaluated the relationship between the CVR-R and handgrip strength (Table 3). Participants with a CVR-R of less than 2.0% had lower handgrip strength than participants with a normal CVR-R. In the analysis based on sex, only women with decreased CVR-R had significantly lower handgrip strength (Supplementary Tables 3, 4). Participants with abnormalities in BDC and the CVR-R, which are indicators of diabetic neuropathy, were found to have decreased grip strength.

Table 3 Grip strength at each CVR-R when adjusted for multiple factors

	Handgrip strength	p value
CVR-R ≥2.0%	25.9 (25.2–27.3)	—
CVR-R <2.0%	23.8 (22.3–25.3)	0.014

Data are presented as estimated value (95%CI). Adjusted for age, sex, duration of diabetes, body mass index, and HbA1c at admission. CVR-R, coefficient of variation of R-R interval. The p-value for the significance test based on a CVR-R >2.0% is shown.

Handgrip strength and other diabetes microvascular complications

We found an association between handgrip strength and BDC. Next, we evaluated the correlation between handgrip strength and other diabetes-related complications. The percentage of diabetic retinopathy progression in each group is shown in Fig. 4A. The percentages of patients with some form of diabetic retinopathy on admission were 22% in the HG, 25% in the MG, and 45% in the LG. PPDR or PDR was noted in 3%, 6%, and 14% of the HG, MG, and LG, respectively, with a significantly higher prevalence of diabetic retinopathy in the LG (p < 0.001). No difference in the stage of diabetic nephropathy was observed between the groups (Fig. 4B).

Fig. 4  Percentage (A) of participants in the high grip strength (HG), medium grip strength (MG), and low grip strength (LG) groups with the revised Davis classification of diabetic retinopathy; NDR, non-diabetic retinopathy; SDR, simple diabetic retinopathy; PPDR, preproliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy. Percentage of Stage classification of diabetic nephropathy in each group (B). The results were analyzed by the chi-square test.

Discussion

In this study, handgrip strength was correlated with BDC, one of the methods to assess diabetic neuropathy. BDC was an independent factor affecting handgrip strength after adjusting for several parameters. Handgrip strength is a parameter that can be easily measured in outpatient settings. Our data indicated that handgrip strength may have clinical significance as a screening tool for diabetic neuropathy (Graphical Abstract).

Graphical Abstract

When the relationship between BDC and handgrip strength was evaluated, a significant correlation was found between handgrip strength and BDC, even after adjusting for age, sex, disease duration, BMI, and HbA1c, indicating that BDC is an independent factor that determines handgrip strength. The same result was obtained with the CVR-R. Therefore, we hypothesized that DPN was affecting grip strength. No correlation was observed with skeletal muscle mass in the body composition analysis at each stage of BDC or with the presence or absence of a decreased CVR-R (Supplementary Table 5). In this study, reduced handgrip strength in the participants with DPN may not be due to a decrease in skeletal muscle mass. Long-term DPN can also lead to decreased skeletal muscle mass [22]. In contrast, previous cross-sectional studies have reported that DPN causes muscle weakness in the distal parts of the limbs [23]. It has also been reported that in populations with DPN, muscle strength per unit volume is lower than that in populations without DPN [24]. Functionally related neuromuscular deficits may appear in severe or long-term DPN [25]; in some cases, exercise training-based interventions may be useful for preventing the onset or progression of DPN [26, 27]. A balanced training program of aerobic and anaerobic exercises effectively manages DPN [27]. In this study, there were many cases of severe DPN in participants with decreased handgrip strength, and exercise training may be effective for such patients.

Chronic diseases, including type 2 diabetes, are associated with risk factors for sarcopenia [28]. Sarcopenia is characterized by reduced skeletal muscle mass, physical function, glucose metabolism, and myokine secretion [29], leading to reduced insulin sensitivity and an increased incidence of diabetes [30]. Previous meta-analyses have found a significant association between DPN and sarcopenia in patients with diabetes [31, 32]. The association between reduced skeletal muscle mass and DPN has been reported. However, most studies have been conducted in older populations, and the reciprocal effects are unclear, as both DPN and the risk of sarcopenia increase with age [28]. In the present study, a trend was observed between handgrip strength and skeletal muscle mass, and a significant correlation was found between BDC and handgrip strength. An independent association between handgrip strength and BDC was observed after adjusting for factors strongly correlated with DPN, such as age, sex, duration of diabetes, and BMI. When examining patients with type 2 diabetes who have reduced handgrip strength, it must also be considered that DPN progression may be present.

In this study, there was a significant correlation between handgrip strength and each DPN index, and several parameters showed a significant correlation among the HG, MG, and LG. Compared with the HG, the MG and LG were characterized by older age and a longer duration of diabetes. The LG had a high proportion of patients with a history of myocardial and cerebral infarction and impaired renal function, with significant carotid IMT and a decrease in the ABI. In patients with type 2 diabetes, the frequency of macrovascular disorders and renal dysfunction increases as the duration of the disease increases [33, 34]. In addition, BMI, abdominal circumference, ALT, and albumin levels were high in the HG. This study included participants of various ages; however, a significant correlation between BMI and handgrip strength in older individuals has been reported [35]. Diastolic blood pressure has been found to be an independent factor affecting handgrip strength [36]. In this study, sulfonylureas, glinides, and thiazolidines were used more frequently in the LG, whereas SGLT2 inhibitors were used more frequently in the HG. As a prescription pattern for older patients with diabetes, the use of sulfonylureas and glinides increases with age, whereas the use of SGLT2 inhibitors decreases with age [37]. These factors may affect grip strength. The parameters that showed differences between the groups listed above may influence handgrip strength. Based on the results of this observational study, future studies should be conducted with a uniform patient background.

This study has several limitations. This was a retrospective cohort study of patients with type 2 diabetes in an educational inpatient setting, which may differ from the typical profile of patients with type 2 diabetes in an outpatient setting. In addition, all participants in this study were Japanese, and the interpretation of the results may differ depending on race and region. Second, when we used BDC to evaluate patients at our hospital, we only tested one side of the body, except in cases where there was a clear difference in symptoms between the left and right sides. In DPN, the symptoms generally appear symmetrically on both sides and in many cases, it is possible to evaluate them correctly by testing one side. However, incorrect results are possible if there is a concomitant unilateral neuropathy that is not related to diabetes. And there are no international standards for the CVR-R as a diagnostic method for DAN, and in this study, the cutoff value was set at less than 2.0% CVR-R, with reference to empirical standards. However, since CVR-R decreases with age, there is a possibility of overdiagnosis of DAN, especially in a group with many elderly participants, as in this study. Finally, this study used a three-quartile grouping of handgrip strength based on sex. This grouping differs from the commonly used sarcopenia criterion of handgrip strength. While the correlation between handgrip strength loss and BDC within a specific population can be evaluated, it is not possible to calculate a cutoff value for handgrip strength for the general population.

Taken together, patients with type 2 diabetes and decreased handgrip strength may develop diabetic neuropathy. Because handgrip strength is a simple indicator that can be easily measured in an outpatient setting with limited examination time, early testing similar to BDC or other NCS should be considered when decreased handgrip strength is evident.

Acknowledgements

We are very grateful to Dr. Shoji Hemmi, Department of Neurology, Kawasaki Medical School, for his help in evaluating BDC in hospitalized patients with type 2 diabetes on a routine basis. The abstract of this report was presented at the 62nd General Meeting of the Chugoku-Shikoku Regional Meeting of the Japan Diabetes Society (Okayama). This manuscript was edited by Elsevier Language Editing Services.

Competing Interests

H.K. has received honoraria for lectures, received scholarship grants, and received research grant from Novo Nordisk Pharma, Sanofi, Eli Lilly, Boehringer Ingelheim, Taisho Pharma, Sumitomo Pharma, Takeda Pharma, Ono Pharma, Daiichi Sankyo, Mitsubishi Tanabe Pharma, Kissei Pharma, MSD, AstraZeneca, Astellas, Novartis, Kowa, Abbott. K.K. has been an advisor to, received honoraria for lectures from, and received scholarship grants from Novo Nordisk Pharma, Sanwa Kagaku, Taisho Pharma, Kowa, Sumitomo Pharma, Mitsubishi Tanabe Pharma, Astellas, Boehringer Ingelheim. S.N. has received honoraria for lectures from Novo Nordisk Pharma and Daiichi Sankyo. T.M. is a member of Endocrine Journal’s Editorial Board. All other authors have no conflict of interests.

Author Contributions

Y.I. designed the study. Y.I., S.N., M.K., Y.Ki., Y.W., T.S., E.N., M.K., T.S., K.D., Y.O., H.I., J.S., Y.F., Y.Ka., T.K., M.S., T.M., and H.K. treated patients and collected data. Y.I. analyzed the data. Y.I., S.N., T.K., M.S., T.M. and H.K. contributed to discussion. K.K. supervised the project. Y.I. wrote the manuscript. H.K. reviewed and edited the manuscript.

Funding Statement

The authors declare no funding statement.

Data Availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Supplementary Table 1 Grip strength at each BDC when adjusted for multiple factors in male participants

	Handgrip strength	p value
BDC-0	33.4 (32.1–34.7)	—
BDC-1	33.1 (31.6–34.6)	0.762
BDC-2	29.9 (27.7–32.1)	0.009
BDC-3/4	27.3 (23.7–31.0)	0.003

Data are presented as estimated value (95%CI). Adjusted for age, duration of diabetes, body mass index, and HbA1c at admission. BDC, Baba’s diabetic neuropathy classification. The p-values for significance tests with the criterion being BDC-0 are listed.

Supplementary Table 2 Grip strength at each BDC when adjusted for multiple factors in female participants

	Handgrip strength	p value
BDC-0	18.2 (17.1–19.3)	—
BDC-1	18.9 (16.8–20.9)	0.581
BDC-2	18.2 (16.1–20.4)	0.990
BDC-3/4	14.2 (10.9–17.6)	0.029

Data are presented as estimated value (95%CI). Adjusted for age, duration of diabetes, body mass index, and HbA1c at admission. BDC, Baba’s diabetic neuropathy classification. The p-values for significance tests with the criterion being BDC-0 are listed.

Supplementary Table 3 Grip strength at each CVR-R when adjusted for multiple factors in male participants

	Handgrip strength	p value
CVR-R ≥2.0%	33.1 (32.1–34.1)	—
CVR-R <2.0%	31.4 (29.3–33.4)	0.151

Data are presented as estimated value (95%CI). Adjusted for age, duration of diabetes, body mass index, and HbA1c at admission. CVR-R, coefficient of variation of R-R interval. The p-value for the significance test based on a CVR-R >2.0% is shown.

Supplementary Table 4 Grip strength at each CVR-R when adjusted for multiple factors in female participants

	Handgrip strength	p value
CVR-R ≥2.0%	18.5 (17.6–19.4)	—
CVR-R <2.0%	15.9 (13.8–17.9)	0.022

Data are presented as estimated value (95%CI). Adjusted for age, duration of diabetes, body mass index, and HbA1c at admission. CVR-R, coefficient of variation of R-R interval. The p-value for the significance test based on a CVR-R >2.0% is shown.

Supplementary Table 5 Skeletal muscle mass in InBody770 when participants are divided into groups using BDC or CVR-R

	BDC-0 (n = 174)	BDC-1 (n = 92)	BDC-2 (n = 54)	BDC-3/4 (n = 22)	p value
Skeletal muscle mass (kg)	25 (21–30)	26 (22–30)	25 (20–30)	23 (19–29)	0.419
	CVR-R ≥2.0% (n = 275)		CVR-R <2.0% (n = 57)		p value
Skeletal muscle mass (kg)	26 (21–30)		23 (21–29)		0.295

Data are presented as median (interquartile range). BDC, Baba’s diabetic neuropathy classification; CVR-R, coefficient of variation of R-R interval. The results were analyzed by Kruskal-Wallis test.

References

• 1   Dyck  PJ,  Herrmann  DN,  Staff  NP,  Dyck  PJ (2013) Assessing decreased sensation and increased sensory phenomena in diabetic polyneuropathies. Diabetes 62: 3677–3686.
• 2   Dyck  PJ,  Karnes  JL,  Daube  J,  O’Brien  P,  Service  FJ (1985) Clinical and neuropathological criteria for the diagnosis and staging of diabetic polyneuropathy. Brain 108(Pt 4): 861–880.
• 3   Himeno  T,  Kamiya  H,  Nakamura  J (2020) Lumos for the long trail: strategies for clinical diagnosis and severity staging for diabetic polyneuropathy and future directions. J Diabetes Investig 11: 5–16.
• 4   Baba  M,  Suzuki  C,  Ogawa  Y (2018) Severity grading system of diabetic neuropathy in type-2 diabetes by nerve conduction study: five-year prospective study on occurrence of diabetic foot, macroangiopathic events, and eventual death. Jpn J Clin Neurophysiol 46: 71–77 (In Japanese).
• 5   Iwamoto  Y,  Nakanishi  S,  Itoh  T,  Nakao  E,  Sugisaki  T, et al. (2022) Correlation of Baba’s diabetic neuropathy classification with various diabetes-related complications. Front Endocrinol (Lausanne) 13: 1054934.
• 6   Ziegler  D (2020) Diabetic polyneuropathy. Internist (Berl) 61: 243–253 (In German).
• 7   Pop-Busui  R,  Boulton  AJ,  Feldman  EL,  Bril  V,  Freeman  R, et al. (2017) Diabetic neuropathy: a position statement by the American Diabetes Association. Diabetes Care 40: 136–154.
• 8   Volmer-Thole  M,  Lobmann  R (2016) Neuropathy and diabetic foot syndrome. Int J Mol Sci 17: 917.
• 9   Lechleitner  M,  Abrahamian  H,  Francesconi  C,  Kofler  M,  Sturm  W, et al. (2019) Diabetic neuropathy and diabetic foot syndrome (Update 2019). Wien Klin Wochenschr 131: 141–150 (In German).
• 10   Boulton  AJ (2014) Diabetic neuropathy and foot complications. Handb Clin Neurol 126: 97–107.
• 11   Mishriky  BM,  Cummings  DM,  Powell  JR (2022) Diabetes-related microvascular complications—a practical approach. Prim Care 49: 239–254.
• 12   Pop-Busui  R,  Herman  WH,  Feldman  EL,  Low  PA,  Martin  CL, et al. (2010) DCCT and EDIC studies in type 1 diabetes: lessons for diabetic neuropathy regarding metabolic memory and natural history. Curr Diab Rep 10: 276–282.
• 13   Braffett  BH,  Gubitosi-Klug  RA,  Albers  JW,  Feldman  EL,  Martin  CL, et al. (2020) Risk factors for diabetic peripheral neuropathy and cardiovascular autonomic neuropathy in the diabetes control and complications trial/epidemiology of diabetes interventions and complications (DCCT/EDIC) study. Diabetes 69: 1000–1010.
• 14   Nakanishi  S,  Shimoda  M,  Kimura  T,  Sanada  J,  Fushimi  Y, et al. (2024) The impact of handgrip strength and waist circumference on glycemic control: prospective, observational study using outpatient clinical data in Japanese patients with type 2 diabetes mellitus. J Diabetes Investig 15: 892–898.
• 15   Oh  TJ,  Kang  S,  Lee  JE,  Moon  JH,  Choi  SH, et al. (2019) Association between deterioration in muscle strength and peripheral neuropathy in people with diabetes. J Diabetes Complications 33: 598–601.
• 16   Yasemin  Ö,  Seydahmet  A,  Özcan  K (2019) Relationship between diabetic neuropathy and sarcopenia. Prim Care Diabetes 13: 521–528.
• 17  Japan Diabetic Neuropathy Study Group (2001) Simplified diagnostic criteria for diabetic distal symmetric polyneuropathy (revised edition). Peripheral Nerve 12: 225–227.
• 18  Japan Diabetic Neuropathy Study Group (2006) Revision of simple diagnostic criteria for diabetic polyneuropathy. Peripheral Nerve 17: 101–103.
• 19   Miyamoto  M,  Kotani  K,  Yagyu  H,  Koibuchi  H,  Fujii  Y, et al. (2010) The correlation between CVR-R and carotid atherosclerosis in type 2 diabetes mellitus patients with diabetic neuropathy. J Physiol Anthropol 29: 149–152.
• 20   Sugimoto  K,  Miyaoka  H,  Sozu  T,  Sekikawa  N,  Wada  R, et al. (2024) Associations of age-adjusted coefficient of variation of R-R intervals with autonomic and peripheral nerve function in non-elderly persons with diabetes. J Diabetes Investig 15: 186–196.
• 21   Takahashi  H,  Tampo  H,  Arai  Y,  Inoue  Y,  Kawashima  H (2017) Applying artificial intelligence to disease staging: Deep learning for improved staging of diabetic retinopathy. PLoS One 12: e0179790.
• 22   Allen  MD,  Major  B,  Kimpinski  K,  Doherty  TJ,  Rice  CL (2014) Skeletal muscle morphology and contractile function in relation to muscle denervation in diabetic neuropathy. J Appl Physiol (1985) 116: 545–552.
• 23   Van Eetvelde  BLM,  Lapauw  B,  Proot  P,  Vanden Wyngaert  K,  Helleputte  S, et al. (2021) The impact of diabetic neuropathy on the distal versus proximal comparison of weakness in lower and upper limb muscles of patients with type 2 diabetes mellitus: a cross-sectional study. J Musculoskelet Neuronal Interact 21: 464–474.
• 24   Hilton  TN,  Tuttle  LJ,  Bohnert  KL,  Mueller  MJ,  Sinacore  DR (2008) Excessive adipose tissue infiltration in skeletal muscle in individuals with obesity, diabetes mellitus, and peripheral neuropathy: association with performance and function. Phys Ther 88: 1336–1344.
• 25   Allen  MD,  Doherty  TJ,  Rice  CL,  Kimpinski  K (2016) Physiology in medicine: neuromuscular consequences of diabetic neuropathy. J Appl Physiol (1985) 121: 1–6.
• 26   Colberg  SR,  Sigal  RJ,  Fernhall  B,  Regensteiner  JG,  Blissmer  BJ, et al. (2010) Exercise and type 2 diabetes: the American College of Sports Medicine and the American Diabetes Association: joint position statement executive summary. Diabetes Care 33: 2692–2696.
• 27   Streckmann  F,  Zopf  EM,  Lehmann  HC,  May  K,  Rizza  J, et al. (2014) Exercise intervention studies in patients with peripheral neuropathy: a systematic review. Sports Med 44: 1289–1304.
• 28   Purnamasari  D,  Tetrasiwi  EN,  Kartiko  GJ,  Astrella  C,  Husam  K, et al. (2022) Sarcopenia and chronic complications of type 2 diabetes mellitus. Rev Diabet Stud 18: 157–165.
• 29   Mesinovic  J,  Zengin  A,  De Courten  B,  Ebeling  PR,  Scott  D (2019) Sarcopenia and type 2 diabetes mellitus: a bidirectional relationship. Diabetes Metab Syndr Obes 12: 1057–1072.
• 30   Srikanthan  P,  Karlamangla  AS (2011) Relative muscle mass is inversely associated with insulin resistance and prediabetes. Findings from the third National Health and Nutrition Examination Survey. J Clin Endocrinol Metab 96: 2898–2903.
• 31   Wannarong  T,  Sukpornchairak  P,  Naweera  W,  Geiger  CD,  Ungprasert  P (2022) Association between diabetic peripheral neuropathy and sarcopenia: a systematic review and meta-analysis. Geriatr Gerontol Int 22: 785–789.
• 32   Andreassen  CS,  Jakobsen  J,  Ringgaard  S,  Ejskjaer  N,  Andersen  H (2009) Accelerated atrophy of lower leg and foot muscles—a follow-up study of long-term diabetic polyneuropathy using magnetic resonance imaging (MRI). Diabetologia 52: 1182–1191.
• 33   Russo  MP,  Grande-Ratti  MF,  Burgos  MA,  Molaro  AA,  Bonella  MB (2023) Prevalence of diabetes, epidemiological characteristics and vascular complications. Arch Cardiol Mex 93: 30–36.
• 34   LeRoith  D,  Biessels  GJ,  Braithwaite  SS,  Casanueva  FF,  Draznin  B, et al. (2019) Treatment of diabetes in older adults: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 104: 1520–1574.
• 35   Pua  YH,  Tay  L,  Clark  RA,  Thumboo  J,  Tay  EL, et al. (2023) Associations of height, weight, and body mass index with handgrip strength: a Bayesian comparison in older adults. Clin Nutr ESPEN 54: 206–210.
• 36   Körei  AE,  Kempler  M,  Istenes  I,  Vági  OE,  Putz  Z, et al. (2017) Why not to use the handgrip test in the assessment of cardiovascular autonomic neuropathy among patients with diabetes mellitus? Curr Vasc Pharmacol 15: 66–73.
• 37   Foresta  A,  Succurro  E,  Baviera  M,  Macaluso  G,  Ojeda-Fernández  L, et al. (2023) Prescribing trends of glucose-lowering drugs in older adults from 2010 to 2021: a population-based study of Northern Italy. Diabetes Res Clin Pract 202: 110742.