Retatrutide and DKA in type 1 diabetes

A case report published in Cureus on 26 August 2026 describes what happened when a man in his mid-30s with 17 years of type 1 diabetes injected an online-sourced retatrutide product, then fell acutely ill. His ketonaemia peaked at 4.3 mmol/L. He spent five days in hospital. Stool cultures later grew Shigella flexneri. The authors, based at the Edinburgh Royal Infirmary, say the case illustrates a widening gap: clinicians do not routinely ask patients about peptides purchased online^[1].

Retatrutide has no marketing authorisation from the FDA, EMA, MHRA, or any other regulator. Its Phase 3 development targets obesity and type 2 diabetes. There are no published retatrutide trials in type 1 diabetes. For full regulatory status, see the retatrutide regulation overview.

What the patient did

The patient had an HbA1c of 74 mmol/mol (approximately 8.9%), a BMI of 28.7, and a background of retinopathy, raised urinary albumin-creatinine ratio, and prior diabetic ketoacidosis. He was on basal-bolus insulin. After seeing retatrutide promoted on social media as a weight-loss agent, he sourced a product online. It arrived as lyophilised powder labeled "not for human or veterinary consumption," which he reconstituted with bacteriostatic water and self-injected subcutaneously at an intended dose of 0.5 mg^[1].

Around three hours after the injection, nausea and vomiting began. Over the next 48 to 72 hours he had multiple episodes of large-volume vomiting, including coffee-ground emesis. His partner, who received the same preparation, also developed gastrointestinal symptoms. Both had eaten improperly defrosted chicken soup approximately 48 hours before symptoms started. The day before admission he omitted both basal and bolus insulin because he could not eat, a recognised trigger for ketoacidosis in type 1 diabetes^[1].

Laboratory findings and the impending DKA distinction

On admission: glucose 17.8 mmol/L, creatinine 154 micromol/L against a baseline of 117 (indicating acute kidney injury), ketonaemia peaking at 4.3 mmol/L, bicarbonate 26.5 mmol/L, pH 7.381, CRP 93 mg/L, heart rate 117 bpm, and blood pressure 162/110 mmHg^[1]. Despite the elevated ketones, the diagnostic criteria for diabetic ketoacidosis were not met: both pH and bicarbonate were preserved. The authors describe it as impending DKA, a clinically meaningful distinction because management and urgency differ from confirmed acidotic DKA.

Initial stool testing was negative for C. difficile, Cryptosporidium, Giardia, norovirus, adenovirus, and rotavirus. A stool culture subsequently grew Shigella flexneri, but the result came back after the patient had already been discharged. The clinicians therefore managed the acute episode without knowing whether the gastrointestinal illness was primarily infection-driven, retatrutide-driven, or both^[1].

Treatment and the glucagon question

Management included intravenous crystalloid fluids for dehydration and the acute kidney injury, a variable-rate intravenous insulin infusion alongside continued basal insulin (with diabetes team input), an intravenous proton pump inhibitor, and antiemetics. Gastroenterology attributed the coffee-ground vomiting to vomiting-induced oesophagitis or a Mallory-Weiss lesion^[1].

During recovery the patient had recurrent hypoglycaemia. The clinicians chose not to use glucagon rescue therapy, citing uncertainty about whether recent exposure to a glucagon receptor agonist might affect the response to exogenous glucagon. The authors are explicit that "there is no clinical evidence that recent retatrutide exposure alters the effectiveness of emergency glucagon." The withholding decision was precautionary rather than evidence-based^[1].

After five days the patient was discharged with an omeprazole course, sick-day management education, and follow-up renal and diabetes appointments. The Shigella result arrived post-discharge.

Why type 1 diabetes is higher-risk for this class of drug

People with type 1 diabetes have no endogenous insulin production. They depend on injected insulin to suppress ketogenesis even during reduced food intake. Any drug that causes nausea, vomiting, or appetite suppression can accelerate ketosis if the patient then omits insulin. That is the mechanism at work in this case, whether or not the retatrutide exposure was the primary driver^[1].

A 2026 consensus statement on GLP-1 and dual GLP-1/GIP receptor agonists as adjunctive therapy in type 1 diabetes, endorsed by six professional diabetes organisations, noted that these agents lack regulatory approval for this population and that safety risks include hypoglycaemia and hyperglycaemia-related ketosis^[2]. Retatrutide adds glucagon receptor agonism to that profile, which makes its pharmacological behaviour in type 1 diabetes even less predictable than agents already studied in this setting.

The product quality problem

The Edinburgh patient self-reconstituted the product. The report notes that "adverse effects may have reflected inaccurate dosing, contaminants, degradation products or substitution with another compound." Whether the intended 0.5 mg dose was what he actually injected is unknown. Products marketed as research chemicals and labeled not for human use are not subject to pharmaceutical manufacturing standards, sterility requirements, or dose-verification controls^[1].

What this case suggests for clinical practice

The authors conclude with a direct clinical recommendation: clinicians should routinely enquire about prescribed and non-prescribed agents when assessing unexplained illness or metabolic decompensation. They specifically call out unexplained gastrointestinal symptoms, ketosis, and acute kidney injury as presentations where asking about peptide use is relevant. A non-judgemental approach is recommended because patients may feel stigmatised for accessing compounds outside clinical channels^[1].

What we do not yet know

This is one case report. The incidence of adverse events in type 1 diabetes patients using online-sourced incretin-class peptides is unknown. Whether Shigella or retatrutide was the primary driver of the gastrointestinal illness cannot be determined. The unresolved glucagon rescue question will need prospective data. The TRIUMPH-1 Phase 3 readout (May 2026) reported 28.3% mean weight loss at 80 weeks in general obesity, a result that will increase public interest in retatrutide before any approval exists^[3]. For anyone with type 1 diabetes considering any incretin-class therapy, the appropriate route is through a diabetes care team, not an online vendor.