Skip to main content
NEJM Evidence homepage



Decreased blood insulin concentrations resulting from reduced pancreatic β-cell insulin secretion and elevated insulin clearance (IC) could be involved in impaired glucose metabolism in diabetes. Recently, we reported a patient with type 2 diabetes mellitus (T2DM) who had decreased blood insulin concentrations and elevated IC.


For this study, we recruited patients with newly diagnosed, treatment-naïve T2DM and measured the metabolic clearance rate of insulin (MCRI) determined by a hyperinsulinemic-euglycemic clamp examination. We defined elevated IC as an MCRI of more than 700 ml/min/m2. Using this tentative cutoff, we identified patients with T2DM with elevated IC and investigated their clinical characteristics.


We enrolled 101 patients in this study; 78.2% were men. Patients had a mean age of 54.1 years, a median body-mass index (BMI) of 25.1 kg/m2 (interquartile range [IQR], 22.9 to 28.4 kg/m2), a median hemoglobin A1c of 10.0% (IQR, 8.0 to 12.3%), and a median MCRI of 655 ml/min/m2 (IQR, 562 to 810 ml/min/m2). Our case definition for elevated IC was met by 44 patients whose median MCRI was 842 ml/min/m2 (IQR, 747 to 975 ml/min/m2) compared with those without elevated IC (570 ml/min/m2; IQR, 500 to 628 ml/min/m2). On the basis of this division, fasting blood glucose and insulin levels were 178 mg/dl (IQR, 140 to 218 mg/dl) and 4.2 mU/l (IQR, 2.7 to 5.5 mU/l), respectively, in patients with elevated IC compared with 146 mg/dl (IQR, 128 to 188 mg/dl) and 9.6 mU/l (IQR, 6.6 to 14.9 mU/l), respectively, in patients without elevated IC. The BMI of patients with elevated IC was 22.9 kg/m2 (IQR, 20.7 to 24.2 kg/m2) compared with 27.3 kg/m2 (IQR, 25.2 to 29.4 kg/m2) in patients who did not have elevated IC. There were no clinically significant differences in renal or hepatic function test results.


Our data suggest that there is a group of patients with T2DM with elevated IC, and that they are nonobese and have decreased blood insulin concentrations. If confirmed, this novel form of T2DM could affect the treatment of such patients. (UMIN Clinical Trials Registry number, UMIN000032014.)


In type 2 diabetes mellitus (T2DM), hyperglycemia develops clinically because of impaired insulin action attributable to heterogeneous pathogenic causes. The balance between blood insulin levels and whole-body insulin sensitivity determines the effects of insulin on glucose metabolism.1 Japanese patients with T2DM are usually not severely obese and usually do not exhibit hyperinsulinemia when hyperglycemic.2,3 It has been speculated that pancreatic β-cell dysfunction could play a primary, central pathogenic role in hyperglycemia in these patients.2,3 The pathologic mechanisms leading to lower blood insulin concentrations and subsequently to impaired glucose metabolism in these patients are thought to include not only decreased insulin secretion attributable to pancreatic β-cell deterioration but also elevated insulin clearance (IC) and increased insulin degradation.4-12 The importance of increased IC in clinical practice has not been widely studied. Recently, we reported a patient with T2DM who had decreased fasting blood insulin concentrations and elevated IC. We proposed that this patient represents a novel subtype of T2DM, which we refer to as “type 2 Japanese diabetes mellitus” (T2JDM).13
Data derived from a hyperinsulinemic-euglycemic clamp examination are the established gold standard for objective and quantitative evaluation of insulin action in clinical practice.14,15 In this clamp test, insulin is infused at a rate calculated to bring the steady-state blood insulin concentration to approximately 100 mU/l on the basis of data from the healthy population. However, in practice, it is recognized that individual differences exist in blood insulin concentrations that are actually achieved at the steady-state. These differences are thought to be attributable to variations in the metabolic clearance rate of insulin (MCRI). A decreased MCRI was found in obese patients with T2DM,4 whereas an increased MCRI was associated with glycemic deterioration.5
In this study, we identified patients with elevated IC as defined by an upper limit of normal case definition for MCRI of 700 ml/min/m2. We performed hyperinsulinemic-euglycemic clamp testing of patients with newly diagnosed, treatment-naïve T2DM. We showed that, in Japan, these patients accounted for just over 40% of such patients with T2DM. In our case series, we also defined a number of clinical characteristics of patients with elevated IC compared with Japanese patients without elevated IC.


Study Patients and Protocol

We invited consecutive patients with newly diagnosed and drug-treatment–naïve T2DM who visited the Diabetes Care Center of Jinnouchi Hospital in Kumamoto, Japan, between September 2019 and September 2021, to participate in the study. Patients with ketoacidosis at presentation, pregnancy, cancer, thyroid disease, adrenal tumor, renal failure, active infection, inflammatory disease, autoimmune disease, liver cirrhosis, viral hepatitis, dementia, or schizophrenia and those who did not agree to undergo hyperinsulinemic-euglycemic clamp examination were excluded. Among the 161 such consecutive patients invited (76.4% were men), 115 agreed to join the study and were admitted to Jinnouchi Hospital. Over a period of 2 days, daily urinary C-peptide excretion was measured, and a hyperinsulinemic-euglycemic clamp examination was performed to evaluate IC as assessed by MCRI. We chose MCRI values higher than 700 ml/min/m2 as “elevated” on the basis of two studies in healthy, nonobese Japanese men, in which the upper value of the interquartile range (IQR) of MCRI in the study with the highest measured values was 663 ml/min/m2.16,17
The primary goal of our study was to identify patients with T2DM who met our case definition of elevated IC and to identify their clinical characteristics compared with the patients in our case series who did not meet the case definition of elevated IC. All tests were conducted at our hospital within 1 week of patient admission, during which time patients consumed our standard in-hospital diet for diabetes (30 kcal/kg of ideal body weight, consisting of carbohydrates, 60%; protein, 1.2 g/kg; lipids, ≤300 mg/day cholesterol; nonsaturated fatty acids, <10%; saturated fatty acids, <7%; no alcohol; and sodium chloride, 6 g/day). We also examined the presence of elevated IC in patients with or without increased levels of hepatic transaminases; in a post hoc analysis, we examined imaging studies to determine whether a patient had data consistent with fatty liver (Supplementary Appendix, pages 3 to 4).
In another post hoc analysis, we limited our analysis to patients without diabetic nephropathy18 (Supplementary Appendix, page 3). Written informed consent was obtained from all patients. This study was conducted in accordance with the Declaration of Helsinki, and the study protocol was approved by the Human Ethics Review Committee of Jinnouchi Hospital (2018-3-4).

Hyperinsulinemic-Euglycemic Clamp

The systemic glucose-lowering action of insulin was evaluated with a hyperinsulinemic-euglycemic clamp using an artificial pancreas (STG-55; Nikkiso), as reported previously.13,14,19 An intravenous loading dose of insulin (regular human insulin, recombinant DNA origin; Novo Nordisk) was administered over 10 minutes (starting at 3.56 mU/kg of body weight per minute, hereafter referred to as mU/kg/min, and gradually decreasing to 1.25 mU/kg/min), followed by a continuous infusion at 1.25 mU/kg/min. To achieve the steady-state, we continued the clamp examination for more than 120 minutes (mean, 226 minutes; 95% confidence interval [95% CI], 222 to 230 minutes). Plasma glucose concentrations were monitored and maintained at 5.5 mmol/l(100 mg/dl) by variable infusion of 10% glucose. The glucose infusion rate at steady-state (termed the M value in mg glucose/kg/min) was measured as the index of whole-body insulin resistance. The blood insulin concentration at steady-state was measured at the end of the clamp examination (Fig. S1). The M value divided by the steady-state insulin value (M/steady-state insulin), adjusted by the steady-state serum insulin concentration, was calculated to yield the insulin sensitivity index. MCRI was calculated according to the following formula:
MCRI (ml/min/m2) = (Insulin infusion rate at steady-state ) / (steady-state serum insulin)

Blood Sampling and Measurement of Clinical Parameters

Fasting blood samples were collected from the antecubital vein in the morning. Blood and urinary analyses were conducted in our hospital laboratory to measure fasting plasma glucose, lipids, glycated hemoglobin A1c (HbA1c), C-peptide, serum insulin, high-sensitivity C-reactive protein, and other biochemical parameters. We examined the presence of fatty liver as determined by a history of fatty liver, abdominal ultrasonography, or an abdominal computed tomographic scan (Supplementary Appendix, pages 3 to 4). We measured urinary levels of albumin and creatinine. Diabetic nephropathy was defined as the presence of albuminuria (≥30 mg/g creatinine)18 (Supplementary Appendix, page 3).

Statistical Analysis

To identify patients with T2DM with elevated IC and to determine their clinical characteristics by comparing two groups including patients with and without elevated IC (power, 0.80; type I error α, 0.05; effect size d, 0.80), we required more than 26 cases in each group. We planned to include more than 30 cases in each group. On the basis of our preliminary observation of the frequency of patients with T2DM with elevated IC in our hospital (about 40%), we set a sample size of at least 100 patients. The Shapiro–Wilk test was used to assess the normal distribution of continuous data. Normally distributed data are expressed as the mean and standard deviation, whereas data with skewed distributions are expressed as the median value with IQR. Categorical data are presented as frequencies with percentages. Differences between the two groups were subjected to Fisher’s exact test for categorical variables. Differences in continuous variables were analyzed with the unpaired t-test or Mann–Whitney U test, as appropriate. Correlations between clinical parameters of interest and MCRI were analyzed using Spearman’s correlation coefficient. Simple logistic regression analysis was used to evaluate the association between the presence of elevated IC and the clinical parameters and laboratory data. The Nagelkerke R2 method was used to select the most suitable equation for multivariate logistic regression analysis to prevent multicollinearity among the body measurement variables (body weight, BMI, and waist circumference), fasting blood insulin-related variables (fasting blood insulin concentrations, fasting blood C-peptide, homeostasis model assessment-insulin resistance [HOMA-IR], and molar ratio of C-peptide to insulin), and blood chemistry variables. Multivariate logistic regression analysis was conducted using the selected suitable outcomes by the Nagelkerke R2 test, and the Hosmer–Lemeshow goodness-of-fit statistic was calculated. We used receiver operating characteristic (ROC) curve analysis to calculate the area under the curve (AUC) and the cutoff value of variables for elevated IC. The optimal cutoff value was selected with the maximum Youden’s index. No multiplicity adjustments for the secondary and exploratory end points were defined. Therefore, only point estimates and 95% CIs are provided. The CIs have not been adjusted for multiple comparisons and should not be used to infer definitive differences between the groups. Statistical analyses were performed using SPSS version 23 (SPSS Inc.).


Clinical Characteristics of Treatment-Naïve Patients with T2DM Enrolled in This Study

A total of 161 Japanese patients with newly diagnosed, treatment-naïve T2DM were recruited initially. Most of these patients (84.2%) had learned of their diagnosis at a routine examination; 46 declined to undergo hyperinsulinemic-euglycemic clamp examination. In addition, 14 patients were excluded for the following reasons: 3 had ketoacidosis at presentation, 3 had cancer, 1 was pregnant, 2 had inflammatory disease, 1 had thyroid disease, 1 had liver cirrhosis, 1 had viral hepatitis, 1 had an active infection, and 1 agreed to but could not complete the clamp test. This left 101 patients whose clinical characteristics are shown in Table 1 and Table S2. All patients were confirmed as having T2DM on the basis of a clinical course of diabetes onset, undetectable plasma glutamic acid decarboxylase antibody, and urinary C-peptide excretion greater than 20 μg/day. The mean age was 54.1 years, and 78.2% were men. The median BMI was 25.1 kg/m2 (IQR, 22.9 to 28.4 kg/m2), and the median HbA1c value was 10.0% (IQR, 8.0 to 12.3%). More than half of the patients had a family history of diabetes. Using the case definitions shown in the legend to Table 1, hypertension (46 [45.5%]) and dyslipidemia (83 [82.2%]) were identified as noted. A history of cerebrovascular/cardiovascular disease was present in 6.9%. Of the patients, 48.5% had findings consistent with microvascular complications of diabetes. The median M value and M/steady-state insulin levels were decreased (4.84 mg glucose/kg/min and 61.0 mg glucose⋅l/U/kg/min, respectively) compared with values in healthy volunteers14 (7.00 mg glucose/kg/min and 67.4 mg glucose⋅l/U/kg/min, respectively). The MCRI values were not normally distributed (median MCRI, 654.8 ml/min/m2; Fig. 1 and Table 1). We describe the representativeness of our study participants in Table S1.
Figure 1
Distribution of the Metabolic Clearance Rate of Insulin Values in Patients with Treatment-Naïve Type 2 Diabetes Mellitus.
The histogram indicates the number of patients with each metabolic clearance rate of insulin value. The vertical dashed line indicates the cutoff value (700 ml/min/m2) of elevated insulin clearance.
Table 1
CharacteristicAll Patients (n=101)Patients with Elevated IC (n=44)Patients without Elevated IC (n=57)
Age — yr54.1 (11.7)54.3 (12.9)53.9 (10.8)
Men79 (78.2)34 (77.3)45 (78.9)
BMI — kg/m2§25.1 (22.9 to 28.4)22.9 (20.7 to 24.2)27.3 (25.2 to 29.4)
Hypertension46 (45.5)18 (40.9)28 (49.1)
Dyslipidemia83 (82.2)33 (75.0)50 (87.7)
History of CVD**7 (6.9)3 (6.8)4 (7.0)
Family history of diabetes53 (52.5)25 (56.8)28 (49.1)
Diabetic microvascular complication††   
Diabetic retinopathy17 (16.8)5 (11.4)12 (21.1)
Diabetic nephropathy33 (32.7)10 (22.7)23 (40.4)
Diabetic neuropathy24 (23.8)15 (34.1)9 (15.8)
Hemoglobin A1c — %§,‡‡10.0 (8.0 to 12.3)10.5 (8.7 to 12.2)9.6 (7.6 to 12.4)
Fasting plasma glucose — mg/dl§,‡‡207.0 (162.5 to 270.5)222.5 (173.5 to 284.0)181.0 (153.5 to 251.5)
Urinary C-peptide — μg/d§89.3 (62.2 to 125.9)85.6 (57.2 to 121.2)89.6 (63.9 to 129.4)
Serum creatinine — mg/dl§0.72 (0.63 to 0.82)0.68 (0.63 to 0.75)0.78 (0.64 to 0.87)
eGFR — ml/min/1.73 m2§83.0 (72.7 to 96.3)88.0 (76.3 to 97.7)75.7 (65.0 to 95.7)
Serum uric acid — mg/dl5.2 (1.4)4.6 (1.2)5.7 (1.3)
Albumin — mg/dl4.3 (0.3)4.3 (0.3)4.3 (0.3)
AST — U/ml§27 (20 to 37)21 (15 to 30)33 (23 to 50)
ALT — U/ml§28 (19 to 50)23 (15 to 29)40 (24 to 66)
γ-GTP — U/ml§51 (33 to 76)43 (26 to 61)57 (40 to 87)
LDH — U/ml§160 (140 to 187)145 (133 to 180)173 (149 to 192)
Glucose clamp data§   
Basal (fasting) glucose — mg/dl155.0 (131.0 to 197.0)177.5 (139.5 to 218.0)146.0 (127.5 to 188.0)
Basal (fasting) insulin — mU/l6.7 (4.2 to 10.7)4.2 (2.7 to 5.5)9.6 (6.6 to 14.9)
Basal (fasting) C-peptide — ng/ml2.06 (1.49 to 2.63)1.52 (1.17 to 1.96)2.46 (2.00 to 3.28)
Steady-state insulin — mU/l78.2 (57.4 to 95.8)55.6 (47.1 to 64.4)93.7 (82.6 to 107.9)
M value — mg glucose/kg/min4.84 (3.37 to 6.79)5.49 (4.04 to 8.11)4.38 (3.10 to 6.12)
M/steady-state insulin — mg glucose⋅l/U/kg/min61.0 (38.6 to 98.6)103.7 (67.0 to 153.7)45.6 (30.8 to 64.2)
MCRI — ml/min/m2654.8 (561.9 to 810.2)842.1 (746.7 to 975.4)570.0 (499.8 to 628.1)
HOMA-IR§2.58 (1.75 to 4.84)1.85 (1.09 to 2.57)3.90 (2.40 to 5.69)
Molar ratio of C-peptide to insulin§14.2 (11.4 to 17.4)17.0 (13.8 to 22.2)13.0 (10.2 to 15.3)
Clinical Characteristics of Patients.*
Values are presented as No. (%) unless otherwise indicated. ALT denotes alanine aminotransferase, AST aspartate aminotransferase, BMI, body-mass index, CVD cerebrovascular/cardiovascular disease, eGFR estimated glomerular filtration rate, IC insulin clearance, γ-GTP γ-glutamyl transpeptidase, HOMA-IR homeostasis model assessment-insulin resistance, LDH lactate dehydrogenase, and MCRI metabolic clearance rate of insulin.
Elevated IC is defined as an MCRI greater than 700 ml/min/m2.
Values are presented as mean (standard deviation).
Values are presented as median (interquartile range).
Hypertension was defined as systolic blood pressure of 140 mm Hg or less and/or diastolic blood pressure of 90 mm Hg or greater or taking blood pressure–lowering medications.
Dyslipidemia was defined as fasting blood levels of low-density lipoprotein cholesterol of 140 mg/dl or greater, triglycerides of 150 mg/dl or greater, or high-density lipoprotein cholesterol less than 40 mg/dl or taking lipid-lowering medications.
CVD was defined as ischemic or bleeding stroke, myocardial infarction, or coronary artery revascularization.
Diabetic microvascular complications included the presence of diabetic retinopathy (ophthalmologist checked the presence of microaneurysms, leaking fluid or bleeding into the retina, and new blood vessels), nephropathy (presence of albuminuria; ≥30 mg/g creatinine18), or neuropathy (defined according to the American Diabetes Association20).
Values were measured at the time of the first visit to Jinnouchi Hospital.

Comparison of Clinical Characteristics and Parameters in Patients with or without Elevated IC

Forty-four of the patients (43.6%) had elevated IC values greater than 700 ml/min/m2 (Fig. 1 and Table 1). Patients with elevated IC were nonobese, with a median BMI of 22.9 kg/m2 (IQR, 20.7 to 24.2 kg/m2). No differences existed in age, gender, HbA1c values, daily urinary C-peptide excretion, duration of diabetes, percentage of family history of diabetes, diabetic retinopathy, nephropathy, or blood pressure between patients with or without elevated IC. Median fasting blood glucose and insulin level were 178 mg/dl (IQR, 140 to 218 mg/dl) and 4.2 mU/l (IQR, 2.7 to 5.5 mU/l) in patients with elevated IC compared with 146 mg/dl (IQR, 128 to 188 mg/dl) and 9.6 mU/l (IQR, 6.6 to 14.9 mU/l) in patients without elevated IC. Other data comparing patients with elevated IC with those without elevated IC are presented in Table 1. We include data on the correlations among the various measured parameters in Table S3.

Hepatic Architecture, Hepatic Enzymes, and Elevated IC

Because insulin metabolism may be affected by fatty liver, we examined hepatic architecture and function in our cohort. Fatty liver was found in 54 (53.5%) of the 101 patients in our entire cohort; of these, 15 (27.8%) met our case definition of elevated IC. In the 47 patients in our cohort without fatty liver, 29 (61.7%) had elevated IC. Levels of hepatic transaminases above the upper limit of normal in our laboratory were found in 53 (52.5%) of the 101 patients in our cohort; among these, 19 (35.8%) met our case definition of elevated IC. In the 48 patients in our cohort without hepatic transaminase levels above the upper limit of normal, 25 (52.1%) had elevated IC. In the patients in our cohort with elevated IC, more than half had neither increased levels of hepatic transaminases nor fatty liver.

Logistic Regression Analysis for the Presence of Elevated IC

The Nagelkerke R2 model fit test identified that BMI, aspartate aminotransferase, serum uric acid, and fasting blood insulin were suitable variables (Supplementary Appendix, Results). Multivariate logistic regression analysis of these four factors showed that BMI, serum uric acid, and fasting blood insulin were correlated with elevated IC (Table S4).

ROC Curve Analysis

We used the same data to perform a ROC curve analysis. The AUC of fasting blood insulin for the presence of elevated IC was 0.887 (95% CI, 0.823 to 0.951; Fig. 2). If the fasting blood insulin cutoff value was set at 5.55 mU/l, the sensitivity was 77.3% and the specificity was 87.7%. The AUC of BMI for the presence of elevated IC was 0.862 (95% CI, 0.788 to 0.937). The cutoff value for BMI was 24.4 kg/m2 (sensitivity, 79.5%; specificity, 84.2%). These estimates may be optimistic because they were measured in the same data set as the model was fitted.
Figure 2
Receiver Operating Characteristic Curve of Fasting Insulin Concentrations for Elevated Insulin Clearance.
The cutoff value of fasting insulin concentrations for elevated insulin clearance, defined as the optimal sensitivity (0.773) and specificity (0.877) of the receiver operating characteristic curve, was 5.55 mU/l (area under the curve, 0.887; 95% confidence interval, 0.823 to 0.951).

Post Hoc Subgroup Analyses in Patients without Diabetic Nephropathy

A post hoc subgroup analysis was performed on the 68 patients in our cohort without diabetic nephropathy.18 Of those patients, 34 met the definition for elevated IC. In these patients without diabetic nephropathy, fasting insulin levels and BMI values were lower in patients with elevated IC than in those without elevated IC. These results were similar to the results in our whole study population (Table S5).

Post Hoc Correlation Analyses

MCRI values showed a negative correlation with fasting insulin levels in patients without diabetic nephropathy (Table S6). Multivariate logistic regression analysis showed that in patients without diabetic nephropathy, serum uric acid and fasting blood insulin were correlated with elevated IC (Table S7). Multivariate logistic regression analysis with BMI and fasting blood insulin adjusted by age and gender suggested that only fasting blood insulin was correlated with elevated IC (fasting blood insulin concentrations: odds ratio, 0.696; 95% CI, 0.535 to 0.906).


We found elevated IC in just under half of the 101 patients of Japanese descent with newly diagnosed, treatment-naïve T2DM included in our study. The patients were not obese (median BMI, 22.9 kg/m2) and had lower fasting blood insulin concentrations than our patients with T2DM without elevated IC. The median fasting serum insulin of our patients with elevated IC was 4.2 mU/l compared with 9.6 mU/l in our patients without elevated IC. Patients with elevated IC had good insulin sensitivity (median M/steady-state insulin, 103.7 mg glucose⋅l/U/kg/min) and preserved pancreatic β-cell insulin secretion (median urinary C-peptide, 85.6 μg/day). We speculate that the pathophysiology of the hyperglycemia seen in these patients is caused by elevated IC rather than by the insulin resistance typically associated with T2DM.
Blood insulin concentrations play a major role in mediating the effects of insulin on glucose metabolism. The clearance of insulin is a crucial component of the physiologic mechanisms that determine its blood levels.4,15,21 It seems possible, on the basis of our data, that enhanced IC may contribute to abnormal glucose metabolism.4,6,21 Elevated IC would presumably be involved in impaired glucose metabolism by reducing blood insulin concentrations,6-13,22-24 but sufficient clinical studies have not been conducted to confirm this. Recently, we reported a patient with T2DM with decreased blood insulin concentrations and elevated IC, which we termed T2JDM13; we undertook this study to better understand the prevalence of this endotype in Japan and to provide a more robust understanding of the clinical aspects of this condition.
In the clinical practice setting, whole-body IC can be quantitatively and objectively evaluated by MCRI using the hyperinsulinemic-euglycemic clamp test.14,15 Previous studies in healthy nonobese Japanese men have reported normal MCRIs in the range of 460 to 660 ml/min/m2.16,17 On the basis of these data, we developed a case definition for elevated IC as an MCRI of greater than 700 ml/min/m2. In this report, we examined the clinical features of patients with newly diagnosed, treatment-naïve T2DM with this characteristic. We chose these patients because their relatively recent clinical onset of disease provides some reassurance that they did not suffer from advanced, severe pancreatic β-cell dysfunction.1 We found that just under half of such patients with T2DM included in our study had elevated IC. They were nonobese, had lower HOMA-IR values, and exhibited higher insulin sensitivity. The central pathogenesis of T2DM is widely considered to be impaired glucose metabolism owing to attenuated insulin action. In this report, we identify elevated IC as a possible pathogenic factor involved in the mechanism of attenuated insulin action in T2DM. If our supposition is correct, this mechanism of abnormal glucose metabolism is the opposite of that observed in the forms of T2DM commonly identified in patients in the United States and Europe. In those cases, the predominant mechanism of pathogenesis of T2DM is hyperinsulinemia as a result of reduced IC and impaired insulin sensitivity associated with obesity.1,2,4
The half-life of insulin in the blood is approximately 4 minutes.22 Insulin acts on tissues such as the liver, kidneys, and muscles and is then degraded.22 Hepatic IC, a physiologic process that, in response to nutritional cues, degrades more than half of the insulin secreted from pancreatic β-cells into the portal vein, is thought to be an important factor in T2DM pathophysiology.22-24 Extrahepatic IC occurs in the kidney, muscle, and other organs.23,24 A previous study suggested that IC may be genetically regulated, based on inherited differences.25 For example, it has been reported that hepatic IC is higher in Japanese people with T2DM than in Caucasian people with T2DM.26 In our study, we found that the MCRI of Japanese patients with T2DM (median 654.8 ml/min/m2) was higher than the previously reported MCRIs of patients with T2DM whose ethnicity was Caucasian and Mexican American (400 to 550 ml/min/m2).27,28
Merovci et al.29 demonstrated that sustained hyperglycemia significantly reduced IC in healthy individuals. In our study patients, it is possible that glucose toxicity may have had some effect on IC, but this would not vitiate our finding that nearly half of the patients we studied had increased IC. Because our study was performed in patients who had recently received a diagnosis of T2DM, further studies are needed to determine whether some of our patients who did not meet the case definition we propose for elevated IC will have enhanced IC when they are in a stable treatment phase after the glucose toxicity that could have been present at the time of initial diagnosis has resolved.
Hepatocellular injury, as indicated by elevated circulating levels of hepatic transaminases, may attenuate hepatic IC.30 In our study, we found elevated IC present in just over one of three patients with elevated hepatic transaminases; this observation indicates that even in the presence of hepatocellular injury, patients could have findings consistent with enhanced IC. We also confirmed the presence of elevated IC even in patients with T2DM with ultrasonographic or computed tomography findings consistent with fatty liver, who are expected to have diminished hepatic IC.31 On the basis of these data, we speculate that even if hepatic IC is decreased, the presence of elevated IC in extrahepatic organs may be partly involved in the state of elevated systemic IC assessed by higher MCRI values.
Our study had several limitations, including the small sample size and possible selection bias attributable to being conducted at single center. In addition, we recruited only treatment-naïve patients with newly diagnosed T2DM. Because we did not measure or evaluate insulin secretion from pancreatic β cells (e.g., by means of a hyperglycemic clamp) or by physiological stimulation (e.g., by an oral glucose tolerance test), we do not know whether the insulin secretory response in patients with elevated IC in the current study was normal. Further studies are needed on the details of the insulin secretory response in patients with T2DM with elevated IC, including studies on hepatic glucose production and hepatic insulin sensitivity. Because the sample size of our study was small, we did not have substantial statistical power in our training and testing data sets. Therefore, we acknowledge that the sensitivity and specificity calculated from ROC analysis may be overestimated. Finally, our data do not define the causal mechanisms leading to elevated IC in T2DM.
In conclusion, we have identified a subpopulation of patients with T2DM in Japan who have data consistent with enhanced IC as a physiological mechanism leading to their disease. These patients are nonobese, have decreased fasting blood insulin concentrations, and exhibit increased insulin sensitivity. The pathobiology of T2DM in these patients is in sharp contrast to that observed in obese patients with Western-type T2DM. Our data suggest that impaired insulin action as a result of enhanced IC plays a role in the pathogenesis of this novel variant of T2DM. Our findings need to replicated and extended to determine the extent to which genetic ancestry, environmental factors, and comorbidities affect the expression of this endotype of T2DM.


A data sharing statement provided by the authors is available with the full text of this article.
Disclosure forms provided by the authors are available with the full text of this article.
We thank Takashi Umabayashi and Yuuko Shiiba for their assistance. We thank Emily Crow, PhD, from Edanz ( for editing a draft of this manuscript. We thank Satista Co., Ltd. for supporting the statistical analyses.

Supplementary Material

Supplementary Appendix (evidoa2100052_appendix.pdf)
Disclosure Forms (evidoa2100052_disclosures.pdf)
Data Sharing Statement (evidoa2100052_data-sharing.pdf)


DeFronzo RA. Pathogenesis of type 2 diabetes mellitus. Med Clin North Am 2004;88:787-835, ix.
Møller JB, Pedersen M, Tanaka H, et al. Body composition is the main determinant for the difference in type 2 diabetes pathophysiology between Japanese and Caucasians. Diabetes Care 2014;37:796-804.
Yabe D, Seino Y, Fukushima M, Seino S. β cell dysfunction versus insulin resistance in the pathogenesis of type 2 diabetes in East Asians. Curr Diab Rep 2015;15:602.
Kim SH, Reaven GM. Insulin clearance: an underappreciated modulator of plasma insulin concentration. J Investig Med 2016;64:1162-1165.
Bizzotto R, Jennison C, Jones AG, et al. Processes underlying glycemic deterioration in type 2 diabetes: an IMI DIRECT study. Diabetes Care 2021;44:511-518.
Okura T, Fujioka Y, Nakamura R, et al. Hepatic insulin clearance is increased in patients with high HbA1c type 2 diabetes: a preliminary report. BMJ Open Diabetes Res Care 2020;8:e001149.
Ohashi K, Fujii M, Uda S, et al. Increase in hepatic and decrease in peripheral insulin clearance characterize abnormal temporal patterns of serum insulin in diabetic subjects. NPJ Syst Biol Appl 2018;4:14.
Adler GK, Murray GR, Turcu AF, et al. Primary aldosteronism decreases insulin secretion and increases insulin clearance in humans. Hypertension 2020;75:1251-1259.
Nijs HG, Radder JK, Frölich M, Krans HM. Increased insulin action and clearance in hyperthyroid newly diagnosed IDDM patient. Restoration to normal with antithyroid treatment. Diabetes Care 1989;12:319-324.
Ohguni S, Notsu K, Kato Y. Correlation of plasma free thyroxine levels with insulin sensitivity and metabolic clearance rate of insulin in patients with hyperthyroid Graves’ disease. Intern Med 1995;34:339-341.
Tamaki M, Fujitani Y, Hara A, et al. The diabetes-susceptible gene SLC30A8/ZnT8 regulates hepatic insulin clearance. J Clin Invest 2013;123:4513-4524.
Lanng S, Thorsteinsson B, Røder ME, Nerup J, Koch C. Insulin sensitivity and insulin clearance in cystic fibrosis patients with normal and diabetic glucose tolerance. Clin Endocrinol (Oxf) 1994;41:217-223.
Sugiyama S, Jinnouchi H, Hieshima K, Kurinami N, Jinnouchi K. A non-obese, treatment-naive Japanese diabetic patient with elevated insulin clearance and hyperglycemia under enhanced insulin sensitivity and increased insulin secretion: elevated insulin clearance as type 2 Japanese diabetes mellitus (T2JDM). Cureus 2021;13:e14354.
DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;237:E214-E223.
Piccinini F, Bergman RN. The measurement of insulin clearance. Diabetes Care 2020;43:2296-2302.
Kaga H, Tamura Y, Takeno K, et al. Correlates of insulin clearance in apparently healthy non-obese Japanese men. Sci Rep 2017;7:1462.
Suzuki R, Tamura Y, Takeno K, et al. Three days of a eucaloric, low-carbohydrate/high-fat diet increases insulin clearance in healthy non-obese Japanese men. Sci Rep 2019;9:3857.
Haneda M, Utsunomiya K, Koya D, et al. A new classification of diabetic nephropathy 2014: a report from Joint Committee on Diabetic Nephropathy. J Diabetes Investig 2015;6:242-246.
Jinnouchi H, Sugiyama S, Yoshida A, et al. Liraglutide, a glucagon-like peptide-1 analog, increased insulin sensitivity assessed by hyperinsulinemic-euglycemic clamp examination in patients with uncontrolled type 2 diabetes mellitus. J Diabetes Res 2015;2015:706416.
Pop-Busui R, Boulton AJ, Feldman EL, et al. Diabetic neuropathy: a position statement by the American Diabetes Association. Diabetes Care 2017;40:136-154.
Watada H, Tamura Y. Impaired insulin clearance as a cause rather than a consequence of insulin resistance. J Diabetes Investig 2017;8:723-725.
Najjar SM, Perdomo G. Hepatic insulin clearance: mechanism and physiology. Physiology (Bethesda) 2019;34:198-215.
Duckworth WC, Bennett RG, Hamel FG. Insulin degradation: progress and potential. Endocr Rev 1998;19:608-624.
Polidori DC, Bergman RN, Chung ST, Sumner AE. Hepatic and extrahepatic insulin clearance are differentially regulated: results from a novel model-based analysis of intravenous glucose tolerance data. Diabetes 2016;65:1556-1564.
Goodarzi MO, Guo X, Cui J, et al. Systematic evaluation of validated type 2 diabetes and glycaemic trait loci for association with insulin clearance. Diabetologia 2013;56:1282-1290.
Møller JB, Dalla Man C, Overgaard RV, et al. Ethnic differences in insulin sensitivity, β-cell function, and hepatic extraction between Japanese and Caucasians: a minimal model analysis. J Clin Endocrinol Metab 2014;99:4273-4280.
Navalesi R, Pilo A, Ferrannini E. Kinetic analysis of plasma insulin disappearance in nonketotic diabetic patients and in normal subjects. A tracer study with 125I-insulin. J Clin Invest 1978;61:197-208.
Goodarzi MO, Palmer ND, Cui J, et al. Classification of type 2 diabetes genetic variants and a novel genetic risk score association with insulin clearance. J Clin Endocrinol Metab 2020;105:1251-1260.
Merovci A, Tripathy D, Chen X, et al. Effect of mild physiologic hyperglycemia on insulin secretion, insulin clearance, and insulin sensitivity in healthy glucose-tolerant subjects. Diabetes 2021;70:204-213.
Bonnet F, Ducluzeau P-H, Gastaldelli A, et al. Liver enzymes are associated with hepatic insulin resistance, insulin secretion, and glucagon concentration in healthy men and women. Diabetes 2011;60:1660-1667.
London A, Lundsgaard AM, Kiens B, Bojsen-Møller KN. The role of hepatic fat accumulation in glucose and insulin homeostasis-dysregulation by the liver. J Clin Med 2021;10:390.

Information & Authors


Published In


Published online: March 15, 2022
Published in issue: March 22, 2022




Seigo Sugiyama, M.D., Ph.D. [email protected]
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Division of Cardiovascular Medicine, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Hideaki Jinnouchi, M.D., Ph.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Division of Cardiovascular Medicine, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Division of Preventive Cardiology, Department of Cardiovascular Medicine, Kumamoto University Hospital, Kumamoto, Japan
Kunio Hieshima, M.D., Ph.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Infectious Disease Division, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Noboru Kurinami, M.D., Ph.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Obesity Treatment Division, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Katsunori Jinnouchi, M.D., Ph.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Division of Gastroenterology and Nephrology, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Akira Yoshida, Ph.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Pharmacology Division, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Tomoko Suzuki, M.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Division of Cardiovascular Medicine, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Keizo Kajiwara, M.D., Ph.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Division of Cardiovascular Medicine, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Fumio Miyamoto, M.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Ophthalmology Division, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Tomio Jinnouchi, M.D., Ph.D.
Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan
Division of Cardiovascular Medicine, Diabetes Care Center, Jinnouchi Hospital, Kumamoto, Japan


Dr. Sugiyama can be contacted at [email protected] or at Division of Cardiovascular Medicine, Diabetes Care Center, Jinnouchi Hospital, 6-2-3 Kuhonji, Chuo-ku, Kumamoto City 862-0976, Japan.
Dr. S. Sugiyama and Dr. H. Jinnouchi contributed equally to this work.

Metrics & Citations




Export citation

Select the format you want to export the citation of this publication.

Cited by

  1. Precision Diagnostics for Type 2 Diabetes Mellitus — Have We Arrived?, NEJM Evidence, 1, 4, (2022)./doi/10.1056/EVIDe2200039

View Options

View options


View PDF









An essential resource for clinical evidence, trial design, and clinical decision-making.