KDIGO 2024: staging CKD with GFR, albuminuria and risk

A 66-year-old woman with type 2 diabetes and hypertension comes to a routine visit. She had an above-knee amputation a few years ago and walks with a prosthesis. Her serum creatinine came back low and the estimated filtration rate looks reassuring. No one has ordered the urine albumin-to-creatinine ratio in recent years. The question is simple and often ignored: does this patient have chronic kidney disease, and if so, what is her real risk of reaching dialysis?

The 2024 Kidney Disease: Improving Global Outcomes (KDIGO) guideline for the evaluation and management of chronic kidney disease (CKD) answers this question with three ideas. The first is that CKD is classified by cause, glomerular filtration rate (GFR) and albuminuria. The second is that creatinine alone can mislead, and cystatin C helps in these cases. The third is that a validated risk equation should guide referral to the nephrologist (KDIGO, 2024).

This text follows the guideline through the reasoning of staging, without becoming a summary of every recommendation. At the end, a section compares KDIGO with the first guideline from the European Society of Cardiology dedicated to cardiovascular disease and CKD, published in 2026 with the European Renal Association (ESC/ERA, 2026).

Key points

• CKD is defined by an alteration of kidney structure or function present for at least three months, with implications for health, and classified by cause, GFR and albuminuria (KDIGO, 2024).
• GFR estimated from creatinine is the starting point, and combining it with cystatin C is recommended when creatinine is less accurate and GFR changes the clinical decision.
• The use of race in GFR calculation should be avoided.
• Albuminuria enters staging alongside GFR, and the test is rarely ordered in practice.
• In CKD G3 to G5, a validated risk equation estimates the absolute risk of kidney failure and helps decide on referral, multidisciplinary care and preparation for kidney replacement therapy.
• The ESC/ERA 2026 guideline recommends screening for CKD with GFR and albuminuria in every patient with cardiovascular disease.

Summary

• Why creatinine alone misleads
• Definition and classification: cause, GFR and albuminuria
• Cystatin C and race-free equations
• Albuminuria: the missing test
• Monitor and predict: the kidney failure risk equation
• Treat according to stage
• What ESC/ERA 2026 adds for those who care for the heart
• How NEXO helps train this reasoning
• Frequently asked questions
• Take-home messages

Why creatinine alone misleads

Serum creatinine is measured routinely in almost the whole world, and that is why KDIGO keeps creatinine with an estimating equation as the initial assessment of GFR (KDIGO, 2024). The guideline recommends using a validated equation to derive GFR from serum markers, instead of interpreting the marker value in isolation (recommendation 1.2.4.1, level 1D) (KDIGO, 2024).

The problem is that creatinine depends on factors that have nothing to do with filtration. KDIGO calls these factors non-GFR determinants and recalls that interpreting creatinine requires considering diet (KDIGO, 2024). Reduced muscle mass, amputation, spinal cord injury, eating disorders, high-protein diets and creatine supplements appear in the table of indications for cystatin C precisely because they change creatinine without changing kidney function (KDIGO, 2024).

What muscle mass does to the estimate

A Japanese study of 86 adults with CKD not on dialysis compared GFR estimated from creatinine with a reference based on 24-hour urine creatinine clearance (Mae et al., 2026). The estimate exceeded the reference in a clinically relevant way in nearly a quarter of the patients. The bioimpedance phase angle, linked to muscle quality, explained this discordance better than appendicular muscle mass (Mae et al., 2026). The authors call for studies with measured GFR to confirm the finding, but the practical message is clear: little muscle or muscle of poorer quality makes creatinine look better than the kidney actually is.

Back to the case

The amputated patient is exactly the scenario the guideline describes. In above-knee amputation, KDIGO considers that eGFR based only on cystatin C may be appropriate when there are no other diseases, and suggests combining creatinine and cystatin C when comorbidities are present (KDIGO, 2024). Because she has diabetes and hypertension, the combined estimate is the suggested choice.

Definition and classification: cause, eGFR and albuminuria

KDIGO defines CKD as an abnormality of kidney structure or function present for at least three months, with implications for health, and classifies the disease by cause, eGFR category (G1 to G5) and albuminuria category (A1 to A3), which the guideline abbreviates as CGA (KDIGO, 2024). The criteria include markers of kidney damage, such as albuminuria from 30 mg/g, urinary sediment abnormalities, persistent hematuria, tubular abnormalities, histologic or imaging findings and a history of kidney transplant, or else eGFR below 60 (KDIGO, 2024).

The guideline calls attention to a nuance that, according to the text itself, often escapes professionals and students: the definition includes many markers of damage, but the classification relies on two dimensions, eGFR and albuminuria (KDIGO, 2024). In the absence of evidence of kidney damage, neither G1 nor G2 meets the criterion for CKD (KDIGO, 2024).

Table: KDIGO 2024 categories

Category	Range	Term used by KDIGO
G1	eGFR of 90 or more	Normal or high
G2	eGFR of 60 to 89	Mildly decreased
G3a	eGFR of 45 to 59	Mildly to moderately decreased
G3b	eGFR of 30 to 44	Moderately to severely decreased
G4	eGFR of 15 to 29	Severely decreased
G5	eGFR below 15	Kidney failure
A1	Albumin/creatinine below 30 mg/g	Normal to mildly increased
A2	Albumin/creatinine of 30 to 300 mg/g	Moderately increased
A3	Albumin/creatinine above 300 mg/g	Severely increased

eGFR ranges are expressed in mL/min per 1.73 m² of body surface area, as in tables 2 and 3 of the guideline (KDIGO, 2024).

Cause comes first in the abbreviation

Cause opens the classification because it changes management. KDIGO asks that the cause be established through the clinical context, personal and family history, social and environmental factors, medications, physical examination, laboratory and imaging tests and, when indicated, genetic and pathologic diagnosis (KDIGO, 2024). Kidney biopsy is suggested as an acceptable and safe diagnostic test to assess cause and guide treatment when clinically appropriate (recommendation 1.1.4.1) (KDIGO, 2024). An example of a cause that requires its own reasoning is chronic kidney disease of unknown cause in heat workers.

Three months, not a single test

Chronicity requires a minimum duration of three months. KDIGO lists as evidence the review of previous eGFR and albuminuria, imaging findings such as a small kidney and thin cortex, biopsy findings, clinical history and repeat testing within and beyond the three-month mark (KDIGO, 2024). A single abnormal eGFR or albuminuria value may reflect acute kidney injury, such as that of crush syndrome, and the guideline asks that chronicity be confirmed (KDIGO, 2024). At the same time, KDIGO allows starting treatments for CKD at the first presentation when other clinical indicators make CKD likely (KDIGO, 2024).

Cystatin C and race-free equations

KDIGO recommends, in adults at risk of CKD, using eGFR based on creatinine and, when cystatin C is available, estimating the eGFR category from the combination of the two markers (recommendation 1.1.2.1, level 1B) (KDIGO, 2024). The guideline also recommends the combined estimate in situations where creatinine is less accurate and eGFR affects the clinical decision (recommendation 1.2.2.1, level 1C) (KDIGO, 2024).

When to order cystatin C

Table 8 of the guideline organizes the indications by domain (KDIGO, 2024). Body habitus includes eating disorders, athletes and bodybuilders, above-knee amputation, spinal cord injury with paraplegia or tetraplegia, and class III obesity, in which the combined estimate proved the most accurate. Diet includes low-protein diets, ketogenic diets, vegetarian diets, high-protein diets and creatine supplements. Diseases other than CKD include malnutrition, cancer, heart failure, cirrhosis, wasting diseases and diseases with muscle loss (KDIGO, 2024).

Cystatin C also has limits. Smoking, for example, changes cystatin C through factors unrelated to filtration, and in malnutrition the combined estimate may lose accuracy because inflammation coexists (KDIGO, 2024). When accuracy matters for treatment, the guideline suggests measuring GFR with an exogenous marker, if available (KDIGO, 2024).

The difference between the two estimates has meaning

KDIGO asks clinicians to understand the implications of the difference between the creatinine-based estimate and the cystatin C-based estimate, because the direction and size of that difference can be informative (KDIGO, 2024). For medication adjustment, creatinine with a validated equation serves in most situations, and the guideline reserves more accurate estimates for drugs with a narrow therapeutic window, toxicity, or scenarios in which creatinine is unreliable (KDIGO, 2024). This care connects with deprescribing, since CKD calls for periodic review of the medication list.

Why race left the equation

KDIGO states that the use of race in GFR calculation should be avoided (practice point 1.2.4.2) and asks that the same equation be used within each geographic region (KDIGO, 2024). The practical doubt was whether switching equations would disrupt patient classification. In the HELIUS study, with 21,617 adults from six ethnic groups in Amsterdam, the 2021 CKD-EPI equation, without race, gave lower GFR values in participants of African ancestry and slightly higher values in the other groups (Huisman et al., 2026). Even so, CKD classification and cardiovascular risk classification were practically identical to those of the 2009 equation, and the authors advocate adopting the race-free equation (Huisman et al., 2026).

Albuminuria: the missing test

KDIGO asks that people at risk of CKD or with CKD be tested with urine albumin and with GFR (practice point 1.1.1.1) (KDIGO, 2024). For the initial test, the order of preference starts with the albumin-to-creatinine ratio in a single sample, preferably the first morning midstream urine, followed by the reagent strip with automated reading (KDIGO, 2024). When albuminuria is detected by a less accurate method, the guideline asks for confirmation by a more accurate method (KDIGO, 2024).

The size of the gap

Recent studies show that albuminuria is the weakest link in staging. In the SEARCH study, with more than 14 thousand adults with diabetes or hypertension in Southern Brazil and no prior CKD diagnosis, almost one in five met KDIGO criteria for CKD (Sa, 2026). Before screening, fewer than 5 in 100 participants had undergone the albumin-to-creatinine ratio, and even after abnormal results testing remained below 10 in 100 (Sa, 2026). The editorial describes the paradox of a system with universal access to laboratory tests that coexists with underdiagnosis and poor risk stratification (Sa, 2026).

In Norway, among 13,480 adults with CKD G3 to G5 confirmed by tests ordered in primary care, about 85 in 100 had no recorded diagnosis and no nephrology follow-up, and proteinuria quantification was among the least used tests (Abedini et al., 2026). In Madrid, only 19 in 100 adults aged 75 or older had the complete screening recommended by KDIGO, mainly because of missing albuminuria, and albuminuria added prognostic information beyond GFR (Uriarte-Ayestarán et al., 2026).

What this teaches about diagnostic error

This pattern is a reminder that many diagnostic errors in primary care stem from an exam that was never ordered. In a Brazilian cohort of 211 patients with CKD G3b referred to the nephrologist by the SUS, follow-up was regular, but adherence to KDIGO exam recommendations was low in both primary and specialized care (Macedo et al., 2026). The authors call for better communication between the two levels, the same weak point as in transitions between hospital discharge and primary care (Macedo et al., 2026).

Monitoring and predicting: the kidney failure risk equation

KDIGO asks that albuminuria and GFR be assessed at least once a year in people with CKD, and more often when the risk of progression is higher and the measurement changes management (KDIGO, 2024). The guideline also defines what a real change is: a GFR drop above 20 per 100 on a subsequent exam exceeds expected variability and deserves evaluation, as does a doubling of the albumin/creatinine ratio (KDIGO, 2024). After starting therapies with hemodynamic effects, the drop threshold that calls for evaluation becomes above 30 per 100 (KDIGO, 2024).

What KDIGO recommends

In CKD G3 to G5, KDIGO recommends using an externally validated risk equation to estimate the absolute risk of kidney failure (recommendation 2.2.1, level 1A) (KDIGO, 2024). Practice points translate risk into decisions. A five-year risk of 3 to 5 per 100 may define the need for referral to a nephrologist, in addition to GFR and albuminuria criteria (KDIGO, 2024). A two-year risk above 10 per 100 may guide the timing of multidisciplinary care, and above 40 per 100 may guide education on modalities, preparation for kidney replacement therapy, vascular access and referral for transplantation (KDIGO, 2024).

The guideline notes that these equations, developed for G3 to G5, may not apply to G1 and G2, and that IgA nephropathy and autosomal dominant polycystic kidney disease require their own equations (KDIGO, 2024).

The equation works outside its country of origin

The kidney failure risk equation uses age, sex, estimated GFR and albumin/creatinine ratio, and is the most validated prognostic model in CKD (Walker and Lees, 2026). In a multicenter Spanish cohort of 503 patients with advanced CKD, the four-variable two-year version discriminated well who would start kidney replacement therapy, including in older age groups, with acceptable calibration (Temprado Collado et al., 2026). The authors conclude that the equation helps plan kidney replacement therapy in older adults with advanced CKD (Temprado Collado et al., 2026).

When albuminuria is missing

Dependence on albuminuria is the Achilles heel of the equation. In the SCREAM project in Stockholm, 77 per 100 people with CKD G3 to G4 did not have an albumin/creatinine ratio available near diagnosis to calculate risk (Walker and Lees, 2026). A new model without albuminuria performed well, but the editorial insists that the albumin test matters beyond predicting kidney failure, because it also guides treatment and cardiovascular risk (Walker and Lees, 2026).

When to refer

Figure 48 of KDIGO lists the situations for referral to a nephrologist (KDIGO, 2024). These include uncertain cause, hereditary disease and extensive recurrent nephrolithiasis; five-year risk above 3 to 5 per 100, GFR below 30 or sustained GFR decline; albuminuria of 300 mg/g or more associated with hematuria, doubling of already significant albuminuria or ratio above 700 mg/g; red blood cell casts; hypertension refractory to four or more antihypertensive drugs; and persistent abnormalities in potassium, acidosis, anemia, bone disease or malnutrition (KDIGO, 2024). The blog has already covered the kidney in multiple myeloma.

Treat according to stage

KDIGO calls for a broad treatment strategy to reduce the risk of CKD progression and its complications (KDIGO, 2024). The main pharmacological pillars depend precisely on GFR and albuminuria, which shows why complete staging matters.

Renin-angiotensin system blockade

The guideline recommends starting an angiotensin-converting enzyme inhibitor or an angiotensin receptor blocker in CKD with severely increased albuminuria without diabetes (level 1B) and in CKD with moderately to severely increased albuminuria with diabetes (level 1B), and suggests the same in moderate albuminuria without diabetes (level 2C) (KDIGO, 2024). The combination of an angiotensin-converting enzyme inhibitor, a receptor blocker and a direct renin inhibitor should be avoided (level 1B) (KDIGO, 2024). KDIGO advises using the highest approved tolerated dose, checking blood pressure, creatinine and potassium in two to four weeks, and continuing treatment unless creatinine rises by more than 30 per 100 in four weeks (KDIGO, 2024).

SGLT2 inhibitors

KDIGO recommends an SGLT2 inhibitor in people with type 2 diabetes, CKD and an eGFR of 20 or more (recommendation 3.7.1, level 1A) (KDIGO, 2024). In adults with CKD, the recommendation also applies to an eGFR of 20 or more with an albumin-to-creatinine ratio of 200 mg/g or more, or to heart failure at any level of albuminuria (level 1A), and is suggested for an eGFR of 20 to 45 with a ratio below 200 mg/g (level 2B) (KDIGO, 2024). Once started, it is reasonable to continue even if the eGFR falls below 20, and the reversible early drop in eGFR is usually a reason to keep it (KDIGO, 2024).

Nonsteroidal mineralocorticoid antagonist

KDIGO suggests a nonsteroidal mineralocorticoid antagonist with proven kidney or cardiovascular benefit in adults with type 2 diabetes, an eGFR above 25, normal potassium and albuminuria above 30 mg/g despite maximum tolerated blockade of the renin-angiotensin system (level 2A) (KDIGO, 2024). The drug can be added to renin-angiotensin system blockade and to an SGLT2 inhibitor, with potassium monitored regularly (KDIGO, 2024). A 2026 review notes that, in the CONFIDENCE trial, starting finerenone and empagliflozin together reduced albuminuria more in the short term than either alone, but that this gain in a marker does not yet prove benefit in hard outcomes (Su et al., 2026). The blog has already discussed finerenone in heart failure with preserved ejection fraction.

Blood pressure, statin and the rest of care

KDIGO suggests a target systolic pressure below 120 mmHg, when tolerated, measured in a standardized way in the office (level 2B), with less intensive targets in frailty, fall risk and other situations (KDIGO, 2024). In adults aged 50 or older with an eGFR below 60 and not on dialysis, the guideline recommends a statin or a statin with ezetimibe (level 1A) (KDIGO, 2024). In atrial fibrillation with CKD G1 to G4, direct oral anticoagulants are preferred over vitamin K antagonists (level 1C), with adjustment for eGFR (KDIGO, 2024). The suggested protein intake in adults with CKD G3 to G5 is 0.8 g/kg per day (KDIGO, 2024).

What ESC/ERA 2026 adds for those who care for the heart

In August 2026, ESC published with the European Renal Association its first guideline dedicated to cardiovascular disease and CKD, organized under the acronym STAMP on CKD: screen, triage, address kidney risk, modify cardiovascular management and plan services (ESC/ERA, 2026). Screening for CKD with eGFR and albuminuria is recommended in every patient with cardiovascular disease, with two measurements over at least three months to confirm chronicity (ESC/ERA, 2026).

Where the two guidelines agree

ESC/ERA recommends the kidney failure risk equation in CKD G3 to G5 to refer to the nephrologist in time, as does KDIGO (ESC/ERA, 2026). Cystatin C should be considered when there are likely sources of error in creatinine and the decision depends on an accurate eGFR (ESC/ERA, 2026). An SGLT2 inhibitor is recommended with an eGFR of 20 or more in CKD with type 2 diabetes and in CKD without diabetes with albuminuria of 200 mg/g or more, and the combination of a receptor blocker with an angiotensin-converting enzyme inhibitor is contraindicated (class III) (ESC/ERA, 2026).

Where ESC/ERA goes further or differs

The blood pressure target is different. ESC/ERA recommends a systolic of 120 to 129 mmHg, if tolerated, in CKD with eGFR of 30 or more (class I), and individualized targets below 30 (ESC/ERA, 2026), while KDIGO suggests less than 120 mmHg (KDIGO, 2024). Finerenone gains strength: ESC/ERA recommends it (class I, level A) in type 2 diabetes with eGFR of 25 or more and albuminuria of 30 mg/g or more, and also recommends subcutaneous semaglutide in defined ranges of eGFR and albuminuria (ESC/ERA, 2026). ESC/ERA also advises against routine antiplatelet therapy in CKD with eGFR below 60 without symptomatic atherosclerotic disease, due to the risk of major bleeding (ESC/ERA, 2026).

Cardiovascular risk changes with the kidney

ESC/ERA recommends risk scores that include eGFR and albuminuria, such as SCORE2 with the CKD add-on, and classifies severe CKD as very high cardiovascular risk for primary prevention (ESC/ERA, 2026). For the patient in the case, with diabetes and possible CKD, this means that the same pair of tests, combined eGFR and albuminuria, decides both kidney and cardiovascular risk at the same time. This joint view is one of the attributes of primary care.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by looking for the code in ICD-10.

The limits of use of the game are described on the NEXO economy page. If you research teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information on curation and on creators, and institutions can learn about the proposal on the partnerships page. Looking for the right ICD-10 code completes this training.

Frequently asked questions

Does a normal creatinine rule out chronic kidney disease?

Normal creatinine leaves out albuminuria and other markers of damage, which also define CKD according to KDIGO (KDIGO, 2024). In people with low muscle mass, the estimate from creatinine may still overestimate function (Mae et al., 2026).

When should cystatin C be requested?

KDIGO recommends the combined estimate of creatinine and cystatin C when creatinine is less accurate and eGFR changes the decision, as in amputation, spinal cord injury, class III obesity, malnutrition, cancer, cirrhosis or heart failure (KDIGO, 2024).

Does the race-free equation change the diagnosis?

KDIGO asks that race not enter the calculation (KDIGO, 2024). In the HELIUS study, the 2021 equation changed individual values, but the classification of CKD and cardiovascular risk remained practically the same as in 2009 (Huisman et al., 2026).

What is the kidney failure risk equation?

It is a validated model that uses age, sex, eGFR and albumin/creatinine ratio to estimate the risk of kidney failure (Walker and Lees, 2026). KDIGO recommends using it in CKD G3 to G5 to guide referral and preparation for kidney replacement therapy (KDIGO, 2024).

At what risk should referral to a nephrologist be made?

KDIGO considers that a five-year risk of 3 to 5 in 100 may indicate referral, together with eGFR below 30, significant albuminuria and other situations in figure 48 (KDIGO, 2024).

Is the blood pressure target in CKD below 120 or below 130?

KDIGO suggests a systolic below 120 mmHg when tolerated (KDIGO, 2024). ESC/ERA 2026 recommends 120 to 129 mmHg with eGFR of 30 or more and individualized targets below that (ESC/ERA, 2026).

Take-home messages

• KDIGO 2024 classifies CKD by cause, eGFR and albuminuria, and chronicity requires at least three months.
• Creatinine-estimated eGFR is the starting point, and cystatin C comes in when creatinine misleads and the decision depends on eGFR.
• Race is out of the eGFR calculation, with little change in classification.
• Albuminuria is part of staging and is still rarely requested, in Brazil and elsewhere.
• In CKD G3 to G5, the kidney failure risk equation guides referral, multiprofessional care and preparation for kidney replacement therapy.
• ESC/ERA 2026 extends CKD screening to every patient with cardiovascular disease and differs from KDIGO on the blood pressure target.

References

1. Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int. 2024;105(4S):S117-S314. PMID 38490803. DOI 10.1016/j.kint.2023.10.018.
2. European Society of Cardiology. 2026 ESC Guidelines for the management of cardiovascular disease and chronic kidney disease, in collaboration with the European Renal Association (ERA): official slide set. Eur Heart J. 2026. Official slides. DOI 10.1093/eurheartj/ehag098.
3. Mae Y, Takata T, Kawata K, et al. Low Muscle Quality Predicts Creatinine-Based eGFR Overestimation Better than Skeletal Muscle Mass in Patients with Chronic Kidney Disease. Biomedicines. 2026;14(10):2225. PMID 42940031. DOI 10.3390/biomedicines14102225.
4. Huisman BJMV, Vosters TG, Agyemang C, et al. Impact of Adopting the Race/Ethnicity-Free 2021 CKD-EPI Equation on Cardiovascular Risk Stratification in a Multiethnic European Population: Results from the HELIUS Study. Glob Heart. 2026;21(1):62. PMID 42631246. DOI 10.5334/gh.1579.
5. Sa JR. Chronic kidney disease in high-risk populations: the urgent need to close the diagnostic gap through systematic screening. J Bras Nefrol. 2026;48(2):e2026E008. PMID 42330548. DOI 10.1590/2175-8239-jbn-2026-e008en.
6. Abedini S, Lindberg M, Bjune T, et al. Chronic kidney disease in Norwegian primary care: underdiagnosis and undertreatment - findings from the VISTAKID (Vestfold Investigation of Stage 3-5 Kidney Disease) study. BMC Nephrol. 2026;27(1):170. PMID 41709175. DOI 10.1186/s12882-026-04796-2.
7. Uriarte-Ayestarán MI, Gutiérrez-Misis A, Castell-Alcalá MV, et al. Implementation of KDIGO CKD Screening and Its Prognostic Implications in Community-Dwelling Adults Aged 75 Years or Older: A Population-Based Cohort Study. J Clin Med. 2026;15(16):6151. PMID 42652556. DOI 10.3390/jcm15166151.
8. Macedo JG, Almeida AM, Sanders-Pinheiro H, et al. Patients with chronic kidney disease stage 3b: do general practitioners and nephrologists follow KDIGO guidelines? J Bras Nefrol. 2026;48(3):e20250229. PMID 41996668. DOI 10.1590/2175-8239-jbn-2025-0229en.
9. Walker HY, Lees JS. Predicting kidney failure risk without albuminuria: implications in chronic kidney disease. BMJ Med. 2026;5(1):e002559. PMID 41658048. DOI 10.1136/bmjmed-2026-002559.
10. Temprado Collado A, Toapanta N, Román Cabezas M, et al. Planning of kidney replacement therapy in advanced CKD using the KFRE formula in a Spanish multicenter cohort. Clin Kidney J. 2026;19(7):sfag192. PMID 42453745. DOI 10.1093/ckj/sfag192.
11. Su C, Li X, Wang S, et al. eGFR- and Potassium-Guided Stewardship of SGLT2 Inhibitors and Finerenone in Albuminuric Chronic Kidney Disease Associated with Type 2 Diabetes. Diabetes Metab Syndr Obes. 2026;19:621958. PMID 42741732. DOI 10.2147/dmso.s621958.