Classification of diabetes mellitus and genetic diabetic syndromes

INTRODUCTION — Traditionally, diabetes has been classified into type 2 diabetes (which accounts for approximately 90 percent of cases of diabetes in the United States, Canada, and Europe) and type 1 diabetes (which accounts for another 5 to 10 percent of cases). The remainder are due to specific etiologic or pathophysiologic factors (table 1) [1]. Other forms of diabetes with variable pathophysiology are emerging that do not fit the typical phenotypes of type 1 or type 2 diabetes and are collectively termed "atypical diabetes." Known monogenic causes of diabetes (eg, those causing maturity onset diabetes of the young or neonatal diabetes) represent a fraction of these cases of atypical diabetes.

The genetic bases of common forms of type 1 and type 2 diabetes remain complex, with common gene variants individually contributing only small degrees of risk or protection. Furthermore, the worldwide epidemic of overweight and obesity has superimposed some aspects of the pathophysiology of type 2 diabetes (eg, those associated with increased adiposity) across all other types.

The current classification of diabetes mellitus will be reviewed here, together with brief descriptions of some emerging forms of atypical diabetes. The definition and diagnostic criteria for diabetes mellitus are discussed separately. (See "Clinical presentation, diagnosis, and initial evaluation of diabetes mellitus in adults".)

TYPE 1 DIABETES

Classic phenotype — Type 1 diabetes is characterized by destruction of the pancreatic beta cells, leading to absolute insulin deficiency. This is usually due to autoimmune destruction of the beta cells. Testing for islet cell antibodies (ICA) or other islet autoantibodies (antibodies to glutamic acid decarboxylase 65 [GAD65]; insulin; the tyrosine phosphatases, insulinoma-associated protein 2 [IA-2] and IA-2 beta; and zinc transporter 8 [ZnT8]) in serum is important; positive autoantibody titers (particularly if two or more autoantibodies are positive) indicate autoimmune or "type 1A" diabetes [2]. However, some patients with absolute insulin deficiency have no evidence of autoimmunity and have no other known cause for beta cell destruction. The classification system of the American Diabetes Association (ADA) applies the term "idiopathic" or "type 1B" diabetes to refer to these patients; these terms could encompass patients who may have islet autoimmunity that is not detected by the currently measured autoantibodies, as well as an array of nonautoimmune pathophysiologic processes leading to near-complete loss of beta cell function. (See "Pathogenesis of type 1 diabetes mellitus".)

The ADA classification of diabetes mellitus does not fully capture the recognized clinical heterogeneity of patients with diabetes. Other classification schemes have been proposed, accounting for beta cell autoimmunity, beta cell function, clinical features, and body weight. The high prevalence of overweight/obesity in the population has further complicated classification systems with an added element of insulin resistance even in type 1 diabetes. (See "Syndromes of ketosis-prone diabetes mellitus", section on 'Classification of KPD'.)

The emergence of atypical phenotypes of both type 1 and type 2 diabetes emphasizes the need for a better classification of diabetes, ideally on an etiologic basis.

Latent autoimmune diabetes in adults (LADA)

●Diagnosis – A diagnosis of LADA may be made in individuals with adult-onset diabetes who are positive for at least one diabetes autoantibody but exhibit prolonged preservation of insulin secretion and a more gradual onset of clinical diabetes than typical type 1 diabetes in children and adolescents. LADA may be considered a slowly progressive variant of type 1 diabetes. Patients with LADA are a heterogeneous group with variable titers of antibodies, body mass index (BMI), and rates of progression to insulin dependence [3]. Adults with LADA may not require insulin treatment at diagnosis but typically progress to insulin dependence after several months to years [4-6].

The clinical utility of the diagnosis lies in the identification of patients with a clinical course that will differ from the usual course of patients with type 2 diabetes [1,7]. Patients with LADA may differ from patients with classic type 2 diabetes and often present with a lower BMI, an earlier age of diabetes onset, and a personal or family history of autoimmune disease. A screening tool utilizing these clinical features has been developed to prompt islet autoantibody testing in adults with newly diagnosed type 2 diabetes in whom LADA should be suspected [8]. The presence and degree of elevation of anti-GAD or anti-ICA antibodies can help predict accelerated disease progression, an earlier requirement for insulin therapy, subtherapeutic responses to oral hypoglycemic medications, and greater risk of ketoacidosis [3,4,9-12]. A diagnosis of LADA therefore helps establish expectations for the disease course and informs monitoring and management strategies. (See 'Distinguishing type 1 from type 2 diabetes' below.)

●Prevalence – Older studies in predominantly Scandinavian populations that have a relatively high prevalence of type 1 diabetes have suggested that as many as 7.5 to 10 percent of adults with apparent type 2 diabetes, based on older age of onset, may have circulating autoantibodies directed against pancreatic beta cell antigens (ICA or GAD65) [9,13,14]. The prevalence of LADA is lower among patients diagnosed with type 2 diabetes in the more diverse United States population.

●Genetics – In genotyping analyses, LADA shares genetic features of both type 1 and type 2 diabetes [15-17]. As an example, one analysis showed that relative to healthy individuals, patients with LADA shared an increased frequency of an HLA-DQB1 genotype with patients with type 1 diabetes and of a variant in the transcription factor 7-like 2 (TCF7L2) gene with patients with type 2 diabetes [15]. The variant in TCF7L2 has been shown to increase the risk for type 2 diabetes in several populations, and the effect size was similar for LADA and type 2 diabetes [16]. (See "Pathogenesis of type 2 diabetes mellitus", section on 'Genetic susceptibility' and "Pathogenesis of type 1 diabetes mellitus".)

Currently unclassified forms of autoimmune diabetes in adults — The definition of LADA (see 'Latent autoimmune diabetes in adults (LADA)' above) is based on the presence of circulating type 1 diabetes-associated islet autoantibodies in adults. This definition does not include adult patients diagnosed with type 2 diabetes who have islet autoimmunity marked either by novel, non-type 1 diabetes-associated islet autoantibodies [18] or by cellular (eg, T cell) autoimmune biomarkers rather than humoral (autoantibody) biomarkers. Increasing evidence suggests that a substantial proportion of patients with an apparent phenotype of type 2 diabetes have T cell-mediated islet autoimmunity that contributes to progressive beta cell dysfunction and defective insulin secretion [19]. This form of autoimmune diabetes currently remains unclassified, in part because standardized clinical tests for T cell-mediated islet autoimmunity are not widely available.

TYPE 2 DIABETES

Classic phenotype — Type 2 diabetes is the most common phenotypic form of diabetes in adults and is characterized by hyperglycemia and variable degrees of insulin resistance and deficiency. Its prevalence rises markedly with increasing degrees of obesity. Insulin resistance and insulin deficiency can arise through genetic or environmental influences, making it difficult to determine the exact cause in an individual patient. In addition, hyperglycemia itself can impair pancreatic beta cell function and exacerbate insulin resistance, through a mechanism termed "glucotoxicity." (See "Pathogenesis of type 2 diabetes mellitus".)

DKA in type 2 diabetes — Diabetic ketoacidosis (DKA) was thought to be rare in patients with an apparent phenotype of type 2 diabetes, but it can occur in specific situations:

●DKA may occur due to severe stresses such as acute myocardial infarction, major trauma, or sepsis that provoke secretion of counterregulatory hormones that worsen insulin resistance in the face of markedly impaired insulin secretion; these patients are unable to respond to the increased insulin demand, potentially leading to DKA.

●DKA in the absence of a clinically apparent stressor may be the initial presentation of diabetes in patients with an apparent phenotype of type 2 diabetes who have a subtype of ketosis-prone diabetes. (See "Syndromes of ketosis-prone diabetes mellitus".)

Heterogeneity of type 2 diabetes — Machine-learning approaches such as cluster analysis of large, comprehensive datasets of patients with type 2 diabetes have revealed clinically relevant subtypes of type 2 diabetes. Based on clinical parameters, five subtypes were defined: severe autoimmune diabetes (SAID), severe insulin-deficient diabetes (SIDD), severe insulin-resistant diabetes (SIRD), mild obesity-related diabetes (MOD), and mild age-related diabetes (MARD) [20]. Originally derived from data in a Scandinavian cohort, these phenotypic subtypes have been described in other populations using similar clustering methods. Genomic sequence data from cohorts of patients with type 2 diabetes have yielded a range of genetic subtypes as well, each with associated genetic loci [21]. Subtypes derived from these analyses include those with decreased pancreatic beta cell function (with either increased or decreased proinsulin levels) and those with increased insulin resistance (associated with features of obesity, lipodystrophy, or abnormal hepatic lipid metabolism). (See "Pathogenesis of type 2 diabetes mellitus", section on 'Subtypes of diabetes based on pathogenesis'.)

DISTINGUISHING TYPE 1 FROM TYPE 2 DIABETES — With the explosion of diabetes worldwide and the emergence of variant phenotypes, it has become increasingly difficult to distinguish type 1 from atypical presentations of type 2 diabetes (table 2). Patients with type 1 diabetes may have at presentation or rapidly develop an absolute requirement for insulin therapy. However, many patients with type 2 diabetes lose beta cell function over time and require insulin for glucose management. Thus, need for insulin therapy per se does not distinguish type 1 from type 2 diabetes. As noted above, diabetic ketoacidosis (DKA) cannot be relied upon as an absolute indicator that the patient has type 1 diabetes or that long-term insulin therapy will be required. (See "Syndromes of ketosis-prone diabetes mellitus".)

Patients with type 1 diabetes may coincidentally have pathophysiologic elements of type 2 diabetes. In the past, poor glycemic management prevented most individuals with type 1 diabetes from gaining weight. Intensive therapy now commonly used to manage type 1 diabetes has contributed to a prevalence of overweight and obesity in the type 1 diabetes population that is similar to the prevalence in the general population. Insulin resistance and other features of type 2 diabetes may be evident in such patients with type 1 diabetes, especially those who also have a family history of type 2 diabetes [22].

●When to perform islet autoantibody testing – Given the overlap between type 1 and type 2 diabetes, the practical utility of identifying any underlying islet autoimmune process, and the emergence of many atypical forms of diabetes, clinicians should consider islet autoantibody testing in all patients who lack a classic phenotype of overweight-associated type 2 diabetes.

We measure islet autoantibodies in patients with any clinical characteristics or presentations that are atypical for type 2 diabetes, including the following:

•Patients who have a subtherapeutic response to initial therapy with sulfonylureas or metformin

•Adults without overweight or obesity (body mass index [BMI] <25 kg/m² or <23 kg/m² for persons of Asian descent)

•Individuals with a personal or family history of autoimmune disease

•Children and young adults (age <30 years)

•Adults who present with unintentional weight loss or ketoacidosis at the time of diagnosis

●Which islet autoantibodies should be measured? – At least two autoantibodies (ie, islet cell antibodies [ICA] and GAD65) or an autoantibody panel (insulin-associated antibodies [IAA], GAD65, insulinoma-associated protein 2 [IA-2], and zinc transporter 8 [ZnT8]) can be measured. Measuring more than one autoantibody will increase the likelihood of a positive value, but it is also more costly.

Insulin antibodies generally should not be measured for diagnostic purposes if the patient has received insulin therapy because some insulins can generate anti-insulin antibody formation. Nonetheless, this phenomenon is less common with human or analog insulins, which are the predominant insulins in clinical use.

If two or more of the autoantibodies are present, the patient should be presumed to have type 1 diabetes and should be treated promptly with insulin replacement therapy, as these patients respond poorly to diet and oral hypoglycemic drug therapy. During early stages in the development of type 1 diabetes, intensive insulin therapy may prolong beta cell function [23], although this benefit was not found in some studies in children and adolescents [24,25]. (See "Type 1 diabetes mellitus: Prevention and disease-modifying therapy", section on 'Preserving insulin secretion in clinical disease'.)

The presence of antibodies to GAD, islet cell, insulin, the tyrosine phosphatases (IA-2 and IA-2 beta), and ZnT8 in patients with presumed type 2 diabetes can identify patients who may have latent autoimmune diabetes in adults [LADA]) and are more likely to require insulin [4,9-11,26]. (See 'Latent autoimmune diabetes in adults (LADA)' above.)

Given the risk of DKA, insulin should also be started in any patient (regardless of whether they are thought to have type 1 or type 2 diabetes) who is catabolic (weight loss or dehydration in the setting of hyperglycemia), or who has evidence of increased ketogenesis (ketonuria or acidosis). Indications for insulin therapy are reviewed in detail separately. (See "Management of blood glucose in adults with type 1 diabetes mellitus" and "Insulin therapy in type 2 diabetes mellitus", section on 'Indications for insulin'.)

GENETIC VARIANTS — As the human genome is further explored, it is likely that multiple genetic variations at different loci will be found that confer varying degrees of predisposition to type 1 and type 2 diabetes. Polymorphisms of multiple genes are reported to influence the risk of type 1A diabetes, including genes in both the major histocompatibility complex (MHC) and elsewhere in the genome, but only human leukocyte antigen (HLA) alleles have a large effect, followed by insulin gene polymorphisms, and PTPN22. (See "Pathogenesis of type 1 diabetes mellitus", section on 'Genetic susceptibility'.)

Numerous common polymorphisms weakly contribute to the risk for or protection from type 2 diabetes [27]. The genes encode proteins that affect several pathways leading to diabetes, including pancreatic development; insulin synthesis, processing, and secretion; amyloid deposition in beta cells; cellular insulin resistance; adipose tissue dysfunction; and impaired regulation of gluconeogenesis. Monogenic causes of type 2 diabetes represent only a small fraction of cases and commonly inherited polymorphisms individually contribute only small degrees of risk for, or protection from, diabetes. Most of the genetic risk for type 2 diabetes results from complex polygenic risk factors. (See "Pathogenesis of type 2 diabetes mellitus", section on 'Genetic susceptibility'.)

Monogenic diabetes (formerly called maturity onset diabetes of the young) — Monogenic diabetes or maturity onset diabetes of the young (MODY) is a clinically heterogeneous disorder characterized by diabetes diagnosed at a young age (<25 years) with autosomal dominant transmission and lack of autoantibodies [28]. MODY is the most common form of monogenic diabetes, accounting for 2 to 5 percent of diabetes [29,30]. The population prevalence of MODY in the United Kingdom is estimated to be 68 to 108 cases per million [31]. These patients are quite heterogeneous, and clinical characteristics may not be reliable in predicting the underlying pathogenesis [32,33]. Many patients are misclassified as having either type 1 or 2 diabetes.

Several different genetic abnormalities have been identified, each leading to a different type of disease. The original MODY nomenclature ("MODY1," "MODY2," "MODY3," etc) has been superseded by the term "monogenic diabetes" with the name of the gene associated with the trait. Well-understood, known monogenic gene variants and the associated syndromes are described below and in the table (table 3). The genes involved control pancreatic beta cell development, function, and regulation, and the variants in these genes cause impaired glucose sensing and insulin secretion with minimal or no defect in insulin action [34]. Variants in hepatocyte nuclear factor-1-alpha (HNF1A) and the glucokinase (GCK) gene are most commonly identified, occurring in 52 to 65 and 15 to 32 percent of MODY cases, respectively [33,35]. Pathogenic variants in hepatocyte nuclear factor-4-alpha (HNF4A) account for approximately 10 percent of MODY cases. Some members of a family have the genetic defect but do not develop diabetes; the reason for this is unclear. Other patients may have the MODY phenotype but do not have a known pathogenic variant in any of the known MODY genes [34]. Finally, the increasing prevalence of obesity in the overall population can superimpose insulin resistance on MODY pathophysiology.

Hepatocyte nuclear factor-4-alpha — Pathogenic variants in the HNF4A gene on chromosome 20 cause the condition formerly called MODY1 [36]. HNF4A is expressed both in the liver and in pancreatic beta cells. The precise mechanism by which a defect in HNF4A causes hyperglycemia is not clear, but it has been associated with reduced insulin secretory response to glucose, suggesting a primary genetic defect in insulin secretion [37-39]. The secretory defect is progressive, and patients typically present with hyperglycemia in adolescence or early childhood. Patients respond to sulfonylureas but may require insulin as the secretory defect progresses. These patients are at risk for the hyperglycemia-associated microvascular and macrovascular complications of diabetes.

Although HNF4A plays a central role in the hepatic synthesis of lipoprotein and coagulation proteins, these functions are largely maintained in HNF4A diabetes, suggesting that this disorder is primarily one of impaired pancreatic beta cell function [38].

Glucokinase — More than 30 pathogenic variants in the GCK gene on chromosome 7 have been described and were formerly called MODY2 [40]. Defects in the expression of GCK, which phosphorylates glucose to glucose-6-phosphate and acts as a glucose sensor, result in a higher threshold for glucose-stimulated insulin secretion. On occasion, the expressed enzyme is functional but unstable, leading to an insulin secretory deficit [41]. (See "Pancreatic beta cell function".)

The resulting hyperglycemia is often stable and mild (eg, fasting glucose 97 to 150 mg/dL [5.4 to 8.3 mmol/L] and glycated hemoglobin [A1C] 5.8 to 7.6 percent) [42], and is therefore not associated with the vascular complications common in other types of diabetes [43]. Patients with a pathogenic variant in the GCK gene can often be managed with diet alone.

In pregnant individuals with GCK-MODY, management depends on the fetal genotype. If the fetus inherits the maternal GCK variant (which will prevent fetal hyperinsulinemia and excessive growth despite maternal hyperglycemia), maternal hyperglycemia does not require treatment. However, if the fetus does not inherit the pathogenic variant, maternal insulin therapy is indicated to prevent excessive fetal growth. Fetal ultrasound with measurement of abdominal circumference has been used to predict fetal genotype but has limited diagnostic utility. In an international cohort of 38 pregnant women with GCK-MODY, fetal genetic testing using cell-free DNA in maternal blood had higher sensitivity (100 versus 53 percent) and specificity (96 versus 61 percent) than measurement of fetal abdominal circumference for prenatal diagnosis of GCK-MODY [44]. Thus, when available, noninvasive prenatal genotyping should be used to guide management of GCK-MODY during pregnancy.

Hepatocyte nuclear factor-1-alpha — One of several pathogenic variants in the HNF1A gene on chromosome 12 was formerly called MODY3 [45]. This form of diabetes is more common among Europeans [46,47]. HNF1A is a weak transactivator of the insulin gene in beta cells. Variants of HNF1A can lead to abnormal insulin secretion; whether this or some other action is defective enough to cause diabetes mellitus is unclear [47]. Genetic variants also result in a low renal threshold for glucose. Thus, prior to onset of diabetes, mutation carriers have detectable glycosuria provoked by glucose loading [48]. Testing for glycosuria two hours after a glucose load could be used to screen children of variant carriers and guide the need for further evaluation.

Patients with HNF1A diabetes have a similar clinical phenotype as patients with HNF4A diabetes, perhaps since HNF1A is regulated positively by HNF4A. Patients exhibit increased insulin sensitivity and marked sensitivity to the hypoglycemic effects of sulfonylureas compared with metformin and compared with patients with type 2 diabetes (3.9-fold greater reduction in fasting plasma glucose) [49]. Thus, patients with pathogenic variants in the HNF1A gene can be successfully treated with sulfonylurea monotherapy, and in one clinical study, approximately 70 percent of patients previously treated with insulin successfully switched to sulfonylureas once an HNF1A mutation was identified [50]. These patients are at risk for micro- and macrovascular complications of diabetes. In addition, patients with diabetes caused by a mutation in HNF1A appear to have an increased risk of cardiovascular mortality compared with unaffected family members [51].

Insulin promoter factor 1 — Pathogenic variants in the insulin promoter factor 1 (IPF1) gene can lead to what was called MODY4 by reduced binding of the protein to the insulin gene promoter [52,53] and perhaps by altering fibroblast growth factor signaling in beta cells [54]. Less severe variants in IPF1 may predispose to late-onset type 2 diabetes [53,55]. In addition, carriers without diabetes have higher blood glucose concentrations and lower insulin-to-glucose ratios than family members without a pathogenic variant (and without diabetes) [56].

Hepatocyte nuclear factor-1-beta — Pathogenic variants in the hepatocyte nuclear factor-1-beta (HNF1B) gene produce a syndrome formerly called MODY5 [57-60]. Affected patients can develop a variety of manifestations in addition to early-onset diabetes. These include pancreatic atrophy (on computed tomography [CT] scan), abnormal renal development (renal dysplasia that can be detected on ultrasonography in the fetus, single or multiple renal cysts, glomerulocystic disease, oligomeganephronia [a form of renal hypoplasia]), slowly progressive renal insufficiency, hypomagnesemia, elevated serum aminotransferases, and genital abnormalities (epididymal cysts, atresia of vas deferens, and bicornuate uterus) [58]. (See "Kidney cystic diseases in children", section on 'Genetic disorders' and "Hypomagnesemia: Causes of hypomagnesemia", section on 'Hepatocyte nuclear factor-1-beta gene mutations' and "Hypomagnesemia: Causes of hypomagnesemia" and "Renal hypodysplasia", section on 'Genetic disorders'.)

In addition, some patients have a phenotype consistent with autosomal dominant tubulointerstitial kidney disease. (See "Autosomal dominant tubulointerstitial kidney disease", section on 'Other types of ADTKD'.)

One of the functions of HNF1B is the regulation of tissue-specific gene expression. In the kidney, the proximal promoter of the PKHD1 gene has a binding site for HNF1B. Pathogenic variants in HNF1B inhibit the expression of PKHD1 and lead to cyst formation [60]. This is not surprising, since pathogenic variants in PKHD1 are responsible for the autosomal recessive form of polycystic kidney disease. (See "Autosomal recessive polycystic kidney disease in children", section on 'Pathogenesis'.)

Neurogenic differentiation factor 1 — Pathogenic variants in the gene for neurogenic differentiation factor 1 (also called NEUROD1 or BETA2) can lead to what was called MODY6 [61,62]. NEUROD1 normally functions as a regulatory switch for endocrine pancreatic development.

Other genes — Pathogenic variants in carboxyl ester lipase (CEL) (see 'Genetically linked disorders of both exocrine and endocrine pancreatic lineages' below); insulin (INS); ATP-binding cassette, subfamily C, member 8 (ABCC8); potassium channel, inwardly rectifying, subfamily J, member 11 (KCNJ11); and paternal uniparental isodisomy of chromosome 6q24 (UPD6) genes have been associated with the MODY phenotype [34]. Pathogenic variants in INS, ABCC8, and KCNJ11 are more commonly associated with neonatal diabetes mellitus. (See "Neonatal hyperglycemia", section on 'Neonatal diabetes mellitus'.)

Monogenic diabetes diagnosed after infancy also has been attributed to pathogenic variants in tRNA methyltransferase 10A (TRMT10A), DNA heat shock protein family member C3 (DNAJC3), and deoxyuridine triphosphatase (DUT) [63]. These and other emerging and established monogenic diabetes genes are listed in the University of Exeter panel of diabetes genes.

Diagnosis — A suspected diagnosis of monogenic diabetes should be confirmed by diagnostic genetic testing. When available, comprehensive, panel-based, next-generation sequencing is preferred to single gene analysis, unless the clinical phenotype or family history strongly implicates a specific underlying genetic variant [63]. Laboratories in several countries offer clinical testing, primarily for variants in HNF4A, HNF1A, and GCK. A list of laboratories that provide genetic testing is available at Genetic Testing Registry. Only CLIA (Clinical Laboratory Improvement Amendments)-certified labs should be used. Genetic testing should only be performed after informed consent and genetic counseling. (See "Genetic testing", section on 'Practical issues'.)

It is important to distinguish monogenic diabetes from type 1 and type 2 diabetes because the optimal treatment and risk for diabetes complications varies with the underlying genetic defect (table 2). As an example, patients with monogenic diabetes due to HNF1A or HNF4A variants are frequently misdiagnosed as having insulin-requiring type 1 diabetes because they present at an early age and in the absence of obesity. However, many of these patients can be successfully managed with sulfonylurea monotherapy. In addition, distinguishing monogenic diabetes from type 1 and type 2 diabetes allows earlier identification of at-risk family members.

Indications for genetic testing — In all patients, it is important to obtain a detailed history of diabetes at diagnosis, including age, body mass index (BMI), and presenting symptoms [64]. It is also important to ascertain insulin dependency and the presence or absence of family history of diabetes. Genetic testing for monogenic diabetes should be performed when there is a high index of suspicion, as indicated by any of the following [31,34,65,66]:

●Multigenerational family history of diabetes (eg, ≥3 generations or history in one parent and at least one other first-degree relative of that parent) with other clinical characteristics described above and in the table (table 2) and negative autoantibodies to glutamic acid decarboxylase 65 (GAD65), insulinoma-associated protein 2 (IA-2), and zinc transporter (ZnT8) [63]. In a patient with presumed type 1 diabetes, measurement of serum autoantibodies (islet cell antibodies [ICA], GAD65, insulin, tyrosine phosphatases IA-2 and IA-2 beta) should be performed prior to consideration of genetic testing for MODY. The presence of type 1 diabetes-associated islet autoantibodies makes monogenic diabetes unlikely [30].

●A high probability of monogenic diabetes (eg, >25 percent in people not treated with insulin) using the MODY Clinical Risk Calculator.

●In individuals with presumed type 1 diabetes, preserved fasting C-peptide (>0.6 ng/mL [0.2 nmol/L] when glucose is >72 mg/dL [4 mmol/L]) three to five years after initial presentation.

Differentiation between monogenic diabetes and other atypical forms of diabetes or type 2 diabetes may be difficult. For patients with presumed type 2 diabetes, the presence of a simple (non-multigenerational) family history does not discriminate between monogenic and type 2 diabetes. Further, pathogenic variants may arise de novo, so the absence of a family history of diabetes does not exclude the possibility of monogenic diabetes. Insulin resistance is not a common feature of monogenic diabetes. Thus, diabetes in the absence of obesity is suspicious for monogenic diabetes, particularly in adolescents with presumed type 2 diabetes. However, the absence of obesity or surrogate markers of insulin resistance is, in general, a poor discriminator between monogenic diabetes and type 2 diabetes in adults [31,64]. There are no biochemical tests that reliably differentiate between the two diseases.

For family members of pathogenic variant carriers, biochemical testing to confirm diabetes should be performed before genetic testing is considered [65]. If the biochemical tests are consistent with a diagnosis of diabetes, genetic testing can be performed to confirm the diagnosis of a monogenic diabetes subtype. (See "Clinical presentation, diagnosis, and initial evaluation of diabetes mellitus in adults", section on 'Diagnostic criteria'.)

Detailed current guidance on genetic testing and counseling for monogenic diabetes may be obtained at the National Society of Guidance Counselors practice resource [67].

Other beta cell gene defects — Other rare genetic defects in beta cell function can cause diabetes. One type results from a dominantly inherited missense mutation in the sulfonylurea 1 receptor subunit (SUR1) that causes hyperinsulinemia in childhood, but beta cell dysfunction and diabetes in adulthood [68,69] (see "Pathogenesis, clinical presentation, and diagnosis of congenital hyperinsulinism"). Other examples include point variants in mitochondrial DNA [70], genetic abnormalities that result in the inability to convert proinsulin to insulin [71], and production of mutant insulin molecules [72].

Genetic defects in insulin action — Rare abnormalities in the insulin receptor (due to a genetic defect) or in the structure of insulin itself can cause impaired insulin signaling and diabetes. (See "Insulin resistance: Definition and clinical spectrum".)

Genetic defects in mitochondrial DNA — A pathogenic variant in position 3243 of mitochondrial DNA (m.3243 A>G) causes a range of disease manifestations, including diabetes. Approximately 10 percent of patients with this variant have a syndrome termed MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes). The syndrome of maternally inherited diabetes and deafness (MIDD) is more common. MIDD has significant phenotypic variation depending on the degree of heteroplasmy (ie, the ratio of the variant mitochondrial DNA to wild-type mitochondrial DNA in the cells), but patients with clinically defined MIDD have both a defect in insulin secretion, which progresses to insulin dependence, and sensorineural hearing loss. The mean age of onset of diabetes and hearing loss is between 30 and 40 years [73]. Other manifestations include cardiac conduction defects, gestational diabetes, proteinuria, and neuropathy.

Although subjects may be treated with insulin secretagogues until insulin dependence develops, metformin is less effective and carries a higher risk of lactic acidosis in this population [73,74]. Supplementation with CoQ10 may be of some benefit. (See "Metformin in the treatment of adults with type 2 diabetes mellitus", section on 'Lactic acidosis'.)

Wolfram syndrome — Another example of a rare genetic syndrome associated with diabetes is the Wolfram or DIDMOAD (arginine vasopressin deficiency [AVP-D; formerly diabetes insipidus], diabetes mellitus, optic atrophy, and deafness) syndrome [75,76]. This disorder is inherited as an autosomal recessive trait with incomplete penetrance. Two forms of Wolfram syndrome (WFS1 and WFS2) have been identified with different causative genes but overlapping phenotypes.

The first gene discovered to cause Wolfram syndrome, named WFS1, encodes an endoplasmic reticulum membrane-embedded protein called wolframin that is expressed in pancreatic beta cells and neurons [77-79]. The second, named WFS2 (also called CDSGH iron sulfur domain 2, or CISD2), encodes a protein in the outer mitochondrial membrane [80]. The estimated prevalence of Wolfram syndrome is 1 in 770,000, and it is believed to occur in 1 of 150 patients with a phenotype of type 1 diabetes [79].

Affected patients usually develop insulin-requiring diabetes and optic atrophy in early childhood. WFS1 patients also develop AVP-D as teenagers or young adults [81], whereas WFS2 patients may not develop AVP-D and may present with bleeding due to platelet dysfunction. In WFS1, AVP-D is due to loss of vasopressin-secreting neurons in the supraoptic nucleus and impaired processing of vasopressin precursors [82] (see "Arginine vasopressin deficiency (central diabetes insipidus): Etiology, clinical manifestations, and postdiagnostic evaluation"). Anterior pituitary dysfunction has also been reported [79].

Other manifestations of Wolfram syndrome include progressive sensorineural deafness, hydronephrosis in WFS1 (due in part to the high urine flow in AVP-D), and neurologic dysfunction [83]. Why severe insulin-requiring diabetes develops is not known; immunologic factors do not appear to be important [76].

Fulminant diabetes — This distinctive form of diabetes, termed a "subtype" of type 1 diabetes, was originally described in Japan and has been reported predominantly in a series of case reports in adults of Far Eastern descent [84]. It is characterized by presentation in diabetic ketoacidosis (DKA) of persons with no prior history of diabetes, who remain completely insulin-dependent following this abrupt onset. Acute beta cell destruction is the cause, presumably due to either a hitherto uncharacterized viral infection or a novel form of islet autoimmunity against a background of genetic susceptibility. The Japan Diabetes Society has proposed the following diagnostic criteria [85]:

●Occurrence of diabetic ketosis or DKA within approximately seven days after onset of hyperglycemic symptoms

●Plasma glucose >288 mg/dL and A1C <8.5 percent

●Fasting serum C‐peptide <0.3 ng/mL (or <0.5 ng/mL after glucagon stimulation)

Intensive insulin therapy is required lifelong following recovery from DKA.

Other emerging forms of diabetes with acute, severe presentation include checkpoint inhibitor-associated diabetes and SARS-CoV-2 (COVID-19) associated diabetes. (See "Pathogenesis of type 2 diabetes mellitus", section on 'Drug-induced hyperglycemia' and "COVID-19: Issues related to diabetes mellitus in adults", section on 'Clinical presentations'.)

DISEASES OF THE EXOCRINE PANCREAS — Any disease that damages the pancreas, or surgical removal of pancreatic tissue (eg, for treatment of congenital hyperinsulinism or for pancreatic cancer), can result in diabetes. There is wide variability in the frequency with which this occurs, primarily determined by the degree of pancreatic insufficiency [86]. In general, in the absence of other risk factors for diabetes, removal of as much as one-half to three-quarters of the pancreas may not result in clinical diabetes if the remaining pancreas functions normally. Among patients with pancreatic exocrine disease, diabetes is more likely to occur in those with a family history of type 1 or type 2 diabetes. This observation suggests a role for an underlying decrease in pancreatic reserve or in insulin responsiveness that makes overt diabetes more likely in patients with pancreatic insufficiency. (See "Overview of the complications of chronic pancreatitis", section on 'Pancreatic diabetes' and "Surgical resection of lesions of the head of the pancreas", section on 'New-onset diabetes'.)

Diabetes that occurs in patients with pancreatic disease is usually insulin requiring. However, it is different from type 1 diabetes in that the glucagon-producing alpha cells are also affected. As a result, risk of insulin-induced hypoglycemia is higher.

Cystic fibrosis — The mechanisms of cystic fibrosis-related diabetes are unique and share features with both type 1 and type 2 diabetes, with both decreased insulin production and insulin resistance [87,88]. Patients with no pancreatic exocrine deficiency have normal insulin secretion and responsiveness. In comparison, patients with exocrine deficiency have decreased insulin secretion, but often glucose tolerance remains normal despite increased hepatic glucose production [89] because increased energy expenditure (glucose utilization) occurs concomitantly. Patients with exocrine deficiency and either impaired glucose tolerance or overt diabetes have reductions in both peripheral glucose utilization and hepatic insulin sensitivity. (See "Cystic fibrosis: Nutritional issues", section on 'Cystic fibrosis-related diabetes mellitus' and "Cystic fibrosis: Overview of gastrointestinal disease", section on 'Cystic fibrosis-related diabetes'.)

Hereditary hemochromatosis — Diabetes is common in patients with hereditary hemochromatosis, being present at diagnosis in up to 50 percent of symptomatic patients. (See "Clinical manifestations and diagnosis of hereditary hemochromatosis".)

Chronic pancreatitis — Glucose intolerance occurs with some frequency in chronic pancreatitis, but overt diabetes mellitus usually occurs late in the course of disease. (See "Chronic pancreatitis: Clinical manifestations and diagnosis in adults" and "Clinical manifestations and diagnosis of chronic and acute recurrent pancreatitis in children".)

Fibrocalculous pancreatic diabetes — Fibrocalculous pancreatic diabetes is a unique form of diabetes secondary to tropical pancreatitis that is endemic in certain parts of the world (eg, southern India). In a prospective evaluation of 370 patients, all of the macro- and microvascular complications typically associated with diabetes were found. Pancreatic cancer and complications of chronic pancreatitis also contribute to the mortality associated with this disease [90]. (See "Etiology and pathogenesis of chronic pancreatitis in adults".)

Genetically linked disorders of both exocrine and endocrine pancreatic lineages — Both exocrine and endocrine pancreatic cells originate from the same endodermal pool. A genetic factor common to both endocrine and exocrine pancreatic development may account for some cases of pancreatic exocrine dysfunction with diabetes. Norwegian kindreds with an autosomal dominant inheritance pattern for diabetes and exocrine pancreatic dysfunction have been identified with a single-base deletion in the carboxyl ester lipase (CEL) gene [91]. The mechanism by which carboxyl ester lipase deficiency in pancreatic acinar cells is linked to beta cell failure is not known [92].

ENDOCRINOPATHIES — Several hormones, such as epinephrine, glucagon, cortisol, and growth hormone, antagonize the action of insulin. Release of these hormones constitutes the protective counterregulatory response to hypoglycemia. On the other hand, primary oversecretion of these hormones can result in impaired fasting glucose or overt diabetes. (See "Physiologic response to hypoglycemia in healthy individuals and patients with diabetes mellitus".)

Endocrine abnormalities that can lead to abnormalities in glucose regulation include:

●Cushing syndrome, due to pituitary or adrenal disease or to exogenous glucocorticoid administration. (See "Epidemiology and clinical manifestations of Cushing syndrome".)

●Acromegaly. (See "Causes and clinical manifestations of acromegaly".)

●Catecholamine excess in pheochromocytoma. (See "Clinical presentation and diagnosis of pheochromocytoma".)

●Glucagon-secreting tumors (glucagonomas), associated with an unusual constellation of other clinical features, including skin rash, weight loss, anemia, and thromboembolic problems. (See "Glucagonoma and the glucagonoma syndrome".)

●Somatostatin-secreting tumors (somatostatinomas), typically associated with the triad of diabetes mellitus, cholelithiasis, and diarrhea with steatorrhea. (See "Somatostatinoma: Clinical manifestations, diagnosis, and management".)

●Hyperthyroidism, which can interfere with glucose metabolism, although overt diabetes is unusual. (See "Overview of the clinical manifestations of hyperthyroidism in adults".)

DRUG-INDUCED DIABETES — A large number of drugs can impair glucose tolerance; they act by decreasing insulin secretion, increasing hepatic glucose production, or causing resistance to the action of insulin. This topic is discussed separately. (See "Pathogenesis of type 2 diabetes mellitus", section on 'Drug-induced hyperglycemia'.)

VIRAL INFECTIONS — Certain viruses (eg, Coxsackie, mumps, SARS-CoV-2) can cause diabetes, either through direct beta cell destruction or, hypothetically, by inducing autoimmune damage (see "Pathogenesis of type 1 diabetes mellitus"). Chronic hepatitis C virus infection has been associated with an increased incidence of diabetes, but it is uncertain if there is a cause-and-effect relationship. (See "Extrahepatic manifestations of hepatitis C virus infection", section on 'Diabetes mellitus'.)

COVID-19 associated diabetes — SARS-CoV-2 (COVID-19) appears to precipitate severe manifestations of diabetes in patients with or without a prior history of diabetes, including diabetic ketoacidosis (DKA), hyperosmolar hyperglycemic state (HHS), and severe insulin resistance. This topic is reviewed in more detail elsewhere. (See "COVID-19: Issues related to diabetes mellitus in adults", section on 'Clinical presentations'.)

GESTATIONAL DIABETES MELLITUS — Gestational diabetes occurs when a woman's insulin secretory capacity is not sufficient to overcome both the insulin resistance created by the anti-insulin hormones secreted by the placenta during pregnancy (eg, estrogen, prolactin, human placental lactogen, cortisol, and progesterone) and the increased fuel consumption necessary to provide for the growing mother and fetus. It is estimated to occur in approximately 2.1 percent of pregnant women in the United States, usually developing in the second or third trimester. (See "Gestational diabetes mellitus: Screening, diagnosis, and prevention".)

UNCOMMON IMMUNE-MEDIATED DIABETES — Several uncommon forms of immune-mediated diabetes have been identified.

Stiff-person syndrome — The stiff-person syndrome (formerly called stiff-man syndrome) is an autoimmune disorder of the central nervous system, which is characterized by progressive muscle stiffness, rigidity, and spasm involving the axial muscles, with severe impairment of ambulation. Patients usually have high titers of anti-glutamic acid decarboxylase (GAD) antibodies, and diabetes occurs in approximately one-third of cases. (See "Stiff-person syndrome".)

Anti-insulin receptor antibodies — Anti-insulin receptor antibodies can bind to insulin receptors and either act as an agonist, leading to hypoglycemia, or block the binding of insulin and cause diabetes [93].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diabetes mellitus in adults" and "Society guideline links: Diabetes mellitus in children".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Type 1 diabetes (The Basics)" and "Patient education: Type 2 diabetes (The Basics)")

●Beyond the Basics topics (see "Patient education: Type 1 diabetes: Overview (Beyond the Basics)" and "Patient education: Type 2 diabetes: Overview (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Type 1 diabetes – Type 1 diabetes is characterized by destruction of the pancreatic beta cells, leading to absolute insulin deficiency. This is usually due to autoimmune destruction of the pancreatic beta cells. (See 'Type 1 diabetes' above.)

●Type 2 diabetes – Type 2 diabetes is by far the most common type of diabetes and is characterized by variable degrees of insulin resistance and deficiency. (See 'Type 2 diabetes' above.)

●Difficulty distinguishing between type 1 diabetes and atypical presentation of type 2 diabetes – When the diagnosis of type 1 or type 2 diabetes is uncertain by clinical presentation (eg, patient with poor response to initial therapy with sulfonylureas or metformin, personal or family history of autoimmune disease), we measure at least two autoantibodies (glutamic acid decarboxylase 65 [GAD65], insulin, tyrosine phosphatases [insulinoma-associated protein 2 (IA-2) and IA-2 beta], islet cell, or zinc transporter 8 [ZnT8]). If two or more of the autoantibodies are present, the patient should be presumed to have type 1 diabetes and should be treated with insulin replacement therapy, as these patients respond poorly to diet and oral hypoglycemic drug therapy. (See 'Distinguishing type 1 from type 2 diabetes' above.)

●Genetic variants – As the human genome is further explored, multiple genetic variants at different loci likely will be found that confer varying degrees of predisposition to type 1 and type 2 diabetes. Monogenic causes of type 2 diabetes represent a fraction of cases, and commonly inherited polymorphisms individually contribute only small degrees of risk for, or protection from, diabetes. Most of the genetic risk for type 2 diabetes results from complex polygenic risk factors. (See 'Genetic variants' above and "Pathogenesis of type 2 diabetes mellitus", section on 'Genetic susceptibility'.)

●Monogenic diabetes – Monogenic diabetes, formerly referred to as maturity onset diabetes of the young (MODY), is a clinically heterogeneous disorder characterized by onset of diabetes at a young age (<25 years) with autosomal dominant transmission and absence of diabetes-related autoantibodies. It is classified by the underlying genetic defect (table 3). Many patients with monogenic diabetes are misclassified as having either type 1 or 2 diabetes (table 2). (See 'Monogenic diabetes (formerly called maturity onset diabetes of the young)' above.)

•Diagnosis – The diagnosis of monogenic diabetes (MODY) is made by performing diagnostic genetic testing by either next-generation, gene panel-based sequencing or sequencing of a specific gene when strongly implicated by clinical phenotype. Laboratories in several countries offer clinical testing, including panels for the most common genetic variants: hepatocyte nuclear factor-4-alpha (HNF4A), hepatocyte nuclear factor-1-alpha (HNF1A), and glucokinase (GCK). (See 'Diagnosis' above.)

We typically perform genetic testing for monogenic diabetes (MODY) when there is a high index of suspicion (familial diabetes with autosomal dominant pattern of inheritance [≥3 generations or an affected parent who has at least one other first-degree relative with diabetes], onset <25 years, negative islet autoantibodies) (table 2). For family members of mutation carriers, biochemical testing to confirm diabetes should be performed before genetic testing is considered. If the biochemical tests are consistent with a diagnosis of diabetes, genetic testing can be performed to confirm the diagnosis of a monogenic diabetes subtype. (See 'Indications for genetic testing' above.)

●Diseases of the exocrine pancreas – Diseases that damage the pancreas, or surgical removal of pancreatic tissue, can result in diabetes. There is wide variability in the frequency with which this occurs, primarily determined by the degree of pancreatic insufficiency. (See 'Diseases of the exocrine pancreas' above.)

●Endocrinopathies – Several endocrinopathies, including Cushing syndrome, acromegaly, and pheochromocytoma, can lead to abnormalities in glucose regulation. (See 'Endocrinopathies' above.)

●Drug-induced diabetes – A large number of drugs can impair glucose tolerance; they act by decreasing insulin secretion, increasing hepatic glucose production, or causing resistance to the action of insulin. (See "Pathogenesis of type 2 diabetes mellitus", section on 'Drug-induced hyperglycemia'.)

ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledges David McCulloch, MD, who contributed to an earlier version of this topic review.