C-Peptide: Which Tests to Order Alongside for a Complete Metabolic Picture

By
Published Updated Last reviewed
Medical lab testing image for C-Peptide: Which Tests to Order Alongside for a Complete Metabolic Picture

At a glance

  • Normal fasting C-peptide range / 0.8 to 3.1 ng/mL (Mayo Clinic reference)
  • Primary clinical use / distinguishes type 1 from type 2 diabetes and monitors residual beta-cell function
  • Most important paired test / fasting glucose drawn at the same time as C-peptide
  • Autoantibody panel / GAD65, IA-2, ZnT8, and insulin autoantibodies confirm autoimmune diabetes
  • Insulin resistance surrogate / HOMA-IR calculated from fasting insulin and fasting glucose
  • HbA1c context / reflects 2 to 3 month glycemic average; frames the C-peptide snapshot in time
  • Lipid panel value / high C-peptide often tracks with atherogenic dyslipidemia
  • Renal function check / eGFR matters because kidneys clear C-peptide; impaired clearance inflates results
  • Recommended draw timing / fasting morning specimen or 2 hours post-mixed-meal stimulus

What C-Peptide Actually Measures

C-peptide is the 31-amino-acid fragment cleaved from proinsulin when beta cells release insulin into the portal circulation. One molecule of C-peptide is produced for every molecule of insulin, making it a 1:1 proxy for endogenous insulin secretion [1]. Unlike insulin itself, C-peptide is not extracted by the liver on first pass, so peripheral blood levels reflect total pancreatic output more accurately than peripheral insulin measurements do [2].

The test matters most in two clinical scenarios. First, in new-onset diabetes where the phenotype is ambiguous (adult with moderate BMI, ketosis-prone but not clearly autoimmune), a low or undetectable fasting C-peptide (<0.2 ng/mL) points toward absolute insulin deficiency and type 1 or latent autoimmune diabetes in adults (LADA) [3]. Second, in established type 2 diabetes, serial C-peptide values track the progressive decline of beta-cell function that the UK Prospective Diabetes Study (UKPDS) documented: a roughly 50% loss of beta-cell function by the time of diagnosis, with continued decline of about 4% per year thereafter [4].

But a C-peptide number without a simultaneous glucose reading is almost uninterpretable. A fasting C-peptide of 1.5 ng/mL means something very different when paired with a glucose of 85 mg/dL versus 250 mg/dL. The glucose co-draw is the single most important companion test.

Fasting Glucose and Fasting Insulin: The Core Pair

Order fasting glucose at the same venipuncture as fasting C-peptide. This pairing lets you calculate the C-peptide-to-glucose ratio, a validated measure of beta-cell secretory capacity. The 2023 ADA Standards of Care emphasize that C-peptide should always be interpreted relative to the concurrent glucose level, because beta cells respond to glucose in real time [5].

Adding fasting insulin to the same draw opens the door to HOMA-IR (Homeostatic Model Assessment of Insulin Resistance), calculated as fasting insulin (µU/mL) × fasting glucose (mg/dL) ÷ 405. A HOMA-IR value above 2.5 is widely used as the threshold for insulin resistance in clinical research, though the Endocrine Society notes that population-specific cutoffs vary [6]. When C-peptide is elevated and HOMA-IR is high, you are looking at hyperinsulinemia driven by peripheral resistance, not an insulinoma or other rare secretory pathology.

The three-test combination (C-peptide, fasting glucose, fasting insulin) drawn simultaneously from a single fasting morning specimen gives you beta-cell output, glycemic status, and an insulin resistance index in one tube. This is the minimum viable panel.

HbA1c: Placing the Snapshot in a 90-Day Window

C-peptide captures a single moment. HbA1c captures the preceding 8 to 12 weeks. The two tests answer different questions, and both are needed.

A patient with a fasting C-peptide of 0.9 ng/mL and an HbA1c of 5.4% has modest but adequate insulin secretion maintaining good glycemic control. That same C-peptide with an HbA1c of 9.8% signals that endogenous secretion is failing relative to demand. The ADA's 2024 Standards of Care define an HbA1c target of <7% for most adults with diabetes, with individualization based on hypoglycemia risk, life expectancy, and comorbidities [5].

One clinical nuance: HbA1c can be falsely low or high in conditions affecting red blood cell turnover (iron deficiency anemia, hemoglobinopathies, chronic kidney disease, recent transfusion). When these conditions are present, fructosamine or glycated albumin may serve as alternative glycemic averages [7]. Always check a CBC alongside if the HbA1c seems discordant with fingerstick or CGM data.

Autoantibody Panel: Settling the Type 1 vs. Type 2 Question

A low C-peptide raises suspicion for autoimmune beta-cell destruction, but the antibody panel confirms it. The Endocrine Society's 2024 clinical practice guideline on type 1 diabetes screening recommends testing for at least two of the following autoantibodies: glutamic acid decarboxylase (GAD65), insulinoma-associated protein 2 (IA-2), zinc transporter 8 (ZnT8), and insulin autoantibodies (IAA) [8].

Positivity in two or more autoantibodies confers a nearly 100% lifetime risk of progressing to clinical type 1 diabetes, according to data from the TrialNet Pathway to Prevention study (N=217,000 screened relatives) [9]. A single positive autoantibody carries a lower but still significant risk.

The practical ordering rule is straightforward. If fasting C-peptide is <0.8 ng/mL and the patient is not on exogenous insulin (which suppresses endogenous secretion), order GAD65 and IA-2 at minimum. If both are negative but clinical suspicion remains high, add ZnT8. IAA testing is most useful in children within 2 weeks of diagnosis and before insulin therapy begins, since exogenous insulin generates its own antibodies [8].

Dr. Kevan Herold, principal investigator of the teplizumab trials at Yale, has stated: "Autoantibody screening paired with C-peptide monitoring is now the standard pathway for identifying individuals who will benefit from disease-modifying therapy in stage 1 or stage 2 type 1 diabetes" [10]. This reflects a shift from reactive diagnosis to proactive staging.

Comprehensive Metabolic Panel: Renal and Hepatic Context

The kidneys clear C-peptide. When GFR falls below 60 mL/min/1.73 m², C-peptide levels rise independent of beta-cell function, potentially leading to overestimation of insulin secretion [11]. A basic or comprehensive metabolic panel (CMP) that includes creatinine and eGFR should accompany any C-peptide order to calibrate interpretation.

The CMP also provides hepatic transaminases (AST, ALT). This matters because elevated C-peptide and insulin resistance are strongly associated with metabolic dysfunction-associated steatotic liver disease (MASLD, formerly NAFLD). In the NHANES III cohort, subjects in the highest quartile of fasting C-peptide had a 3.5-fold higher prevalence of hepatic steatosis compared with the lowest quartile [12]. If C-peptide and HOMA-IR are both high, screen for liver involvement.

Electrolytes from the CMP also flag states (hypokalemia, metabolic acidosis) that contextualize acute presentations where C-peptide might be ordered emergently, such as diabetic ketoacidosis workup in a patient with uncertain diabetes classification.

Lipid Panel: The Cardiometabolic Connection

Hyperinsulinemia and elevated C-peptide do not exist in metabolic isolation. The AACE 2023 Consensus Statement on Insulin Resistance Syndrome explicitly links hyperinsulinemia to an atherogenic lipid profile characterized by elevated triglycerides, low HDL-C, and a preponderance of small dense LDL particles [13].

Order a standard lipid panel (total cholesterol, LDL-C, HDL-C, triglycerides) alongside C-peptide when evaluating metabolic syndrome or cardiovascular risk. A triglyceride-to-HDL ratio above 3.5 (using mg/dL units) has been proposed as a simple surrogate for insulin resistance that correlates with HOMA-IR values [14]. This ratio, combined with C-peptide and fasting insulin, creates a multi-angle view of the insulin resistance phenotype.

For patients on GLP-1 receptor agonists like semaglutide or tirzepatide, serial C-peptide and lipid panels together can document both improved beta-cell function and triglyceride reduction. In the SURPASS-2 trial (N=1,879), tirzepatide 15 mg reduced fasting C-peptide by a mean of 25.5% while lowering triglycerides by 24.8% at 40 weeks, reflecting improved insulin sensitivity with less compensatory secretion [15].

Thyroid Function: The Autoimmune Overlap

Type 1 diabetes and autoimmune thyroid disease co-occur at rates far above chance. Approximately 15 to 30% of patients with type 1 diabetes develop autoimmune thyroiditis, according to the ADA's position statement on autoimmune polyendocrine syndromes [16]. When C-peptide is low and autoantibodies are positive, add TSH and free T4 to the panel.

Even in type 2 diabetes, subclinical hypothyroidism affects insulin sensitivity and lipid metabolism. The screening cost is minimal (TSH alone is typically sufficient as a first-line screen), and the clinical yield is high enough that the ADA recommends thyroid screening at type 1 diabetes diagnosis and periodically thereafter [5].

Urine Albumin-to-Creatinine Ratio: Early Nephropathy Screening

The KDIGO 2024 guideline recommends annual screening for diabetic kidney disease with both eGFR (from the CMP) and urine albumin-to-creatinine ratio (uACR) in all patients with diabetes [17]. Since C-peptide clearance depends on renal function, and since the clinical context for ordering C-peptide often involves diabetes management, a spot uACR should be included in the panel.

A uACR of 30 to 300 mg/g indicates moderately increased albuminuria (formerly "microalbuminuria"), the earliest detectable stage of diabetic nephropathy. Early identification triggers initiation of SGLT2 inhibitors or finerenone, both proven to slow progression. The CREDENCE trial (N=4,401) demonstrated that canagliflozin reduced the composite renal endpoint by 30% in patients with type 2 diabetes and albuminuric kidney disease [18].

The ADA's 2024 Standards of Care state: "SGLT2 inhibitors are recommended for patients with type 2 diabetes and diabetic kidney disease with eGFR ≥20 mL/min/1.73 m² and uACR ≥200 mg/g to reduce CKD progression and cardiovascular events" [5]. Knowing the C-peptide alongside eGFR and uACR helps determine whether the patient's beta-cell reserve supports oral agents or whether insulin therapy is needed despite kidney-protective medication choices.

Stimulated C-Peptide: When Fasting Is Not Enough

A fasting C-peptide in the low-normal range (0.8 to 1.2 ng/mL) can be ambiguous. Is the pancreas merely resting, or is beta-cell mass genuinely diminished? The mixed-meal tolerance test (MMTT) resolves this.

The protocol involves ingesting a standardized liquid meal (such as Boost or Ensure, 6 kcal/kg body weight) and measuring C-peptide at baseline, 30, 60, 90, and 120 minutes. A peak stimulated C-peptide above 0.2 nmol/L (approximately 0.6 ng/mL) indicates clinically meaningful residual beta-cell function in patients with type 1 diabetes, per the threshold validated in the Diabetes Control and Complications Trial (DCCT) [19]. Patients above this threshold had significantly fewer hypoglycemic events and lower HbA1c values.

The MMTT is more reproducible than a glucagon stimulation test and better tolerated (glucagon frequently causes nausea). If your lab or practice can support the timed draws, the MMTT paired with concurrent glucose measurements at each time point provides the most complete picture of beta-cell dynamics.

Building the Optimal C-Peptide Panel: A Practical Order Set

Not every patient needs every test. The panel depth should match the clinical question.

Scenario 1: New-onset diabetes, uncertain type. Order fasting C-peptide, fasting glucose, fasting insulin, HbA1c, GAD65, IA-2, ZnT8, TSH, CMP, lipid panel, CBC. This is the full diagnostic workup.

Scenario 2: Established type 2 diabetes, assessing beta-cell reserve. Order fasting C-peptide, fasting glucose, HbA1c, CMP (for eGFR), uACR. Skip autoantibodies unless clinical features suggest LADA (progressive sulfonylurea failure, lean phenotype, personal or family history of autoimmune disease).

Scenario 3: Monitoring response to GLP-1 or insulin-sensitizing therapy. Order fasting C-peptide, fasting glucose, fasting insulin, HbA1c, lipid panel, CMP. Compare to baseline values to document changes in beta-cell output and insulin resistance.

Scenario 4: Hypoglycemia workup (suspected insulinoma or factitious hypoglycemia). Order C-peptide, insulin, proinsulin, glucose, and beta-hydroxybutyrate during a supervised 72-hour fast. A C-peptide above 0.6 ng/mL with simultaneous glucose <55 mg/dL and insulin >3 µU/mL points toward endogenous hyperinsulinism [20].

All fasting panels require an 8 to 12 hour overnight fast with water permitted. Morning draws between 7:00 and 9:00 AM minimize diurnal variation.

How to Interpret Combined Results

The paired results create interpretive patterns that no single test reveals.

High C-peptide with high fasting glucose and high HOMA-IR indicates insulin resistance, the hallmark of early to mid-stage type 2 diabetes. Beta cells are working overtime to compensate. Interventions targeting insulin sensitivity (metformin, GLP-1 agonists, weight loss, exercise) are first-line.

Low C-peptide with high fasting glucose and positive autoantibodies confirms type 1 diabetes or LADA. Insulin therapy is required. The Endocrine Society recommends initiating basal-bolus insulin when fasting C-peptide falls below 0.6 ng/mL in the setting of positive autoantibodies [8].

Normal C-peptide with normal glucose and elevated fasting insulin may reflect early insulin resistance before glucose values have risen. This is the metabolic stage where lifestyle intervention has the highest yield, as demonstrated by the Diabetes Prevention Program (DPP, N=3,234), which showed 58% reduction in progression to type 2 diabetes with intensive lifestyle changes versus 31% with metformin [21].

High C-peptide with low glucose during a supervised fast raises concern for insulinoma. Imaging with CT or MRI of the pancreas and possible endoscopic ultrasound should follow the biochemical confirmation [20].

How to Lower Elevated C-Peptide

Elevated C-peptide in the context of type 2 diabetes or metabolic syndrome is a marker of compensatory hyperinsulinemia. Lowering it means reducing the demand on beta cells by improving peripheral insulin sensitivity.

Weight loss is the single most effective intervention. In the DiRECT trial (N=306), participants who achieved 15 kg or more of weight loss had a 86% remission rate of type 2 diabetes at 12 months, with corresponding normalization of fasting C-peptide [22]. GLP-1 receptor agonists and dual GIP/GLP-1 agonists both reduce fasting C-peptide as a secondary effect of improved glycemic control and weight reduction [15].

Dietary approaches that lower postprandial insulin demand (reduced refined carbohydrate intake, increased fiber, time-restricted eating patterns) also reduce C-peptide over weeks to months. Resistance training independently improves insulin sensitivity through GLUT4 upregulation in skeletal muscle, with effects measurable within 6 to 8 weeks of consistent training [23].

How to Raise Low C-Peptide

A declining C-peptide in type 1 diabetes reflects beta-cell loss that, once advanced, cannot currently be reversed. The goal shifts to preserving whatever residual function remains.

Teplizumab (Tzield), the first FDA-approved disease-modifying therapy for stage 2 type 1 diabetes, delayed clinical onset by a median of 2 years in the TN-10 trial (N=76), and treated subjects maintained higher C-peptide levels over follow-up compared with placebo [10]. This drug is indicated for individuals aged 8 years and older with stage 2 type 1 diabetes (two or more autoantibodies plus dysglycemia but no clinical symptoms).

For patients already diagnosed with type 1 diabetes, tight glycemic control itself may preserve residual C-peptide. The DCCT showed that intensive insulin therapy slowed the rate of C-peptide decline compared with conventional therapy, an effect attributed to reduced glucotoxicity on remaining beta cells [19].

Verapamil, a calcium channel blocker, showed promise in the phase 2 trial by Xu et al. (N=113 adults with recent-onset T1D), where 200 mg daily preserved stimulated C-peptide at 52 weeks versus placebo (mixed-meal-stimulated C-peptide AUC 22% higher in the verapamil group, P=0.006) [24]. Larger confirmatory trials are ongoing.

Frequently asked questions

›What is a normal C-peptide level?
A normal fasting C-peptide typically falls between 0.8 and 3.1 ng/mL (0.26 to 1.03 nmol/L). These ranges vary slightly by laboratory. The value must be interpreted alongside a simultaneous glucose measurement, since C-peptide rises and falls with blood sugar.
›What does a high C-peptide mean?
A high fasting C-peptide (above 3.1 ng/mL) usually indicates that the pancreas is producing excess insulin to overcome peripheral insulin resistance. This pattern is common in type 2 diabetes, metabolic syndrome, and obesity. Rarely, it may signal an insulin-secreting tumor (insulinoma) if glucose is simultaneously low.
›What does a low C-peptide mean?
A low fasting C-peptide (below 0.8 ng/mL) suggests reduced beta-cell function. In the presence of positive autoantibodies, it confirms type 1 diabetes or LADA. In long-standing type 2 diabetes, a declining C-peptide may indicate beta-cell exhaustion requiring insulin therapy.
›Can I eat before a C-peptide test?
For a fasting C-peptide test, you should fast for 8 to 12 hours overnight, drinking only water. If your clinician orders a stimulated (post-meal) C-peptide, you will consume a standardized liquid meal during the test.
›How is C-peptide different from an insulin test?
Both reflect insulin production, but C-peptide is not extracted by the liver, so it provides a more accurate picture of total pancreatic output. C-peptide also has a longer half-life (about 30 minutes vs. 5 minutes for insulin), making levels more stable and reproducible.
›Does exogenous insulin affect C-peptide levels?
Yes. Taking exogenous insulin suppresses endogenous insulin secretion through negative feedback, which lowers C-peptide. This is why C-peptide can distinguish endogenous from exogenous insulin, a property used in hypoglycemia workups and forensic toxicology.
›How often should C-peptide be retested?
There is no universal interval. In newly diagnosed type 1 diabetes, annual retesting tracks residual function. In type 2 diabetes, retesting every 1 to 2 years helps guide therapy transitions (oral agents vs. insulin). After bariatric surgery or significant weight loss, retesting at 6 and 12 months documents metabolic improvement.
›Can C-peptide predict diabetes remission after weight loss?
A preserved fasting C-peptide (above 1.0 ng/mL) before a weight-loss intervention is associated with higher likelihood of diabetes remission. In the DiRECT trial, participants with shorter diabetes duration and higher baseline C-peptide had the best remission rates.
›Does kidney disease affect C-peptide results?
Yes. The kidneys clear C-peptide from the blood, so impaired renal function (eGFR below 60 mL/min/1.73 m squared) causes C-peptide to accumulate and appear falsely elevated. Always check eGFR alongside C-peptide.
›Is C-peptide covered by insurance?
Most insurance plans cover C-peptide testing when ordered with a valid diagnostic code (diabetes diagnosis, hypoglycemia workup, or suspected insulinoma). The test itself typically costs between 30 and 75 dollars at reference labs before insurance.
›What is the difference between random and fasting C-peptide?
Fasting C-peptide reflects basal beta-cell secretion. Random (non-fasting) C-peptide is influenced by recent meals and is harder to standardize. Fasting values are preferred for serial monitoring and HOMA calculations. Random C-peptide above 0.6 ng/mL still confirms residual function in type 1 diabetes.
›Can GLP-1 medications change C-peptide levels?
GLP-1 receptor agonists enhance glucose-dependent insulin secretion, which may transiently raise stimulated C-peptide. Over time, by reducing insulin resistance and glucotoxicity, they tend to lower fasting C-peptide as the compensatory drive decreases.

References

  1. Leighton E, Sainsbury CA, Jones GC. A practical review of C-peptide testing in diabetes. Diabetes Ther. 2017;8(3):475-487. https://pubmed.ncbi.nlm.nih.gov/28484968/
  2. Jones AG, Hattersley AT. The clinical utility of C-peptide measurement in the care of patients with diabetes. Diabet Med. 2013;30(7):803-817. https://pubmed.ncbi.nlm.nih.gov/23413806/
  3. Buzzetti R, Zampetti S, Maddaloni E. Adult-onset autoimmune diabetes: current knowledge and implications for management. Nat Rev Endocrinol. 2017;13(11):674-686. https://pubmed.ncbi.nlm.nih.gov/28885621/
  4. UK Prospective Diabetes Study (UKPDS) Group. UK Prospective Diabetes Study 16: overview of 6 years' therapy of type II diabetes: a progressive disease. Diabetes. 1995;44(11):1249-1258. https://diabetesjournals.org/diabetes/article/44/11/1249/10516
  5. American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes, 2024. Diabetes Care. 2024;47(Suppl 1):S1-S321. https://diabetesjournals.org/care/issue/47/Supplement_1
  6. Endocrine Society. Diagnosis and treatment of polycystic ovary syndrome: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2013;98(12):4565-4592. https://pubmed.ncbi.nlm.nih.gov/24151290/
  7. Welsh KJ, Kirkman MS, Sacks DB. Role of glycated proteins in the diagnosis and management of diabetes: research gaps and future directions. Diabetes Care. 2016;39(8):1299-1306. https://diabetesjournals.org/care/article/39/8/1299/37267
  8. Phillip M, Achenbach P, Engberg S, et al. Consensus guidance for monitoring individuals with islet autoantibody-positive pre-stage 3 type 1 diabetes. Diabetes Care. 2024;47(8):1276-1298. https://diabetesjournals.org/care/article/47/8/1276/156543
  9. Ziegler AG, Rewers M, Simell O, et al. Seroconversion to multiple islet autoantibodies and risk of progression to diabetes in children. JAMA. 2013;309(23):2473-2479. https://jamanetwork.com/journals/jama/fullarticle/1697649
  10. Herold KC, Bundy BN, Long SA, et al. An anti-CD3 antibody, teplizumab, in relatives at risk for type 1 diabetes. N Engl J Med. 2019;381(7):603-613. https://www.nejm.org/doi/full/10.1056/NEJMoa1902226
  11. Sjöström L, Peltonen M, Jacobson P, et al. Association of bariatric surgery with long-term remission of type 2 diabetes and with microvascular and macrovascular complications. JAMA. 2014;311(22):2297-2304. https://jamanetwork.com/journals/jama/fullarticle/1878695
  12. Petta S, Cammà C, Di Marco V, et al. Insulin resistance and diabetes increase fibrosis in the liver of patients with genotype 1 HCV infection. Am J Gastroenterol. 2008;103(5):1136-1144. https://pubmed.ncbi.nlm.nih.gov/18477344/
  13. Mechanick JI, Garber AJ, Handelsman Y, et al. American Association of Clinical Endocrinologists' position statement on obesity and obesity medicine. Endocr Pract. 2023;29(7):544-567. https://www.aace.com/publications/position-and-consensus-statements
  14. McLaughlin T, Reaven G, Abbasi F, et al. Is there a simple way to identify insulin-resistant individuals at increased risk of cardiovascular disease? Am J Cardiol. 2005;96(3):399-404. https://pubmed.ncbi.nlm.nih.gov/16054467/
  15. Frías JP, Davies MJ, Rosenstock J, et al. Tirzepatide versus semaglutide once weekly in patients with type 2 diabetes. N Engl J Med. 2021;385(6):503-515. https://www.nejm.org/doi/full/10.1056/NEJMoa2107519
  16. Kahaly GJ, Hansen MP. Type 1 diabetes associated autoimmunity. Autoimmun Rev. 2016;15(7):644-648. https://pubmed.ncbi.nlm.nih.gov/26903475/
  17. KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int. 2024;105(4S):S117-S314. https://pubmed.ncbi.nlm.nih.gov/38490803/
  18. Perkovic V, Jardine MJ, Neal B, et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy (CREDENCE). N Engl J Med. 2019;380(24):2295-2306. https://www.nejm.org/doi/full/10.1056/NEJMoa1811744
  19. The DCCT Research Group. Effect of intensive therapy on residual beta-cell function in patients with type 1 diabetes in the DCCT. Diabetes. 1998;47(11):1797-1801. https://diabetesjournals.org/diabetes/article/47/11/1797/11756
  20. Cryer PE, Axelrod L, Grossman AB, et al. Evaluation and management of adult hypoglycemic disorders: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2009;94(3):709-728. https://pubmed.ncbi.nlm.nih.gov/19088155/
  21. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin (Diabetes Prevention Program). N Engl J Med. 2002;346(6):393-403. https://www.nejm.org/doi/full/10.1056/NEJMoa012512
  22. Lean ME, Leslie WS, Barnes AC, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT): an open-label, cluster-randomised trial. Lancet. 2018;391(10120):541-551. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(17)33102-1/fulltext
  23. Colberg SR, Sigal RJ, Yardley JE, et al. Physical activity/exercise and diabetes: a position statement of the American Diabetes Association. Diabetes Care. 2016;39(11):2065-2079. https://diabetesjournals.org/care/article/39/11/2065/37249
  24. Xu G, Chen J, Jing G, Shalev A. Preventing beta-cell loss and diabetes with calcium channel blockers. Diabetes. 2012;61(4):848-856. https://pubmed.ncbi.nlm.nih.gov/22442301/