Diabetic ketoacidosis in a prediabetic patient on ophthalmic steroids

The patient

A 29-year-old man with a history of hyperlipidemia, prediabetes, and childhood ocular trauma presented with fatigue, polydipsia, and polyuria for two weeks. He did not report regular use of any systemic medications or supplements. One month prior, he had undergone emergent surgery for spontaneous retinal detachment and since then had been using steroidal eye drops. He was using one drop of prednisolone acetate 1% ophthalmic suspension to the right eye four times daily.

Vital signs were normal. The patient's body mass index (BMI) was 22.3 kg/m² (reference range, 18.5 to 24.9 kg/m²), and he had dry mucous membranes. Laboratory studies revealed an elevated blood glucose level of 881 mg/dL (reference range, 70 to 99 mg/dL) with an anion gap metabolic acidosis and elevated beta-hydroxybutyrate level of 4.45 mmol/L (reference range <0.28 mmol/L). His HbA1c level had been recorded at 6.1% nine months prior (reference range <5.7%) and was 13.7% at admission. Anti-islet cell and anti-glutamic acid decarboxylase antibodies were negative. The C-peptide level was low at 0.55 ng/mL (reference range, 1.1 to 4.4 ng/mL). He did not report any use of alcohol or illicit drugs; no testing was performed.

He was treated with IV fluids and insulin infusion. Steroid eye drops were discontinued. He was discharged on basal-bolus insulin and metformin. Two months after discharge, he self-discontinued all diabetic medications. At six-month follow-up, his HbA1c level had improved to 6.4% and repeat C-peptide level was 5.3 ng/mL. Eighteen months following admission, his HbA1c level was 6.3%.

The diagnosis

The diagnosis is diabetic ketoacidosis (DKA) in a patient with a possible rare, atypical form of diabetes that may have been unmasked by his use of glucocorticoids. Other known predisposing factors for DKA such as infection, acute major illness, medications (thiazides, sympathomimetics, antipsychotics), and drug/alcohol use were absent. Following the cessation of the presumed offending agent, his HbA1c level returned to baseline without ongoing treatment with diabetic medications. The patient's BMI, young age, and recent prediabetic HbA1c level would make type 2 diabetes unlikely as a diagnosis at presentation, and laboratory workup for type 1 diabetes yielded negative autoantibody results, although the zinc transporter antibody level was never tested.

Topical steroids, via the skin or eye, can be systemically absorbed. They lead to hyperglycemia in rare cases, but there has yet to be a reported case of DKA associated with ophthalmic steroid use. Although a causal relationship cannot be drawn in our case, it is hypothesized that systemic absorption of ophthalmic steroids may lead to downstream effects similar to those seen with systemic glucocorticoid use. Further workup with genetic testing is required to identify potentially rarer types of diabetes, such as monogenic diabetes, which is a noninsulin-dependent autoimmune form of the disease. Monogenic diabetes is not typically thought to cause DKA, but case reports have described patients with monogenic diabetes developing DKA in times of physiologic stress.

Pearls

• Clinicians should be aware of the potential of ophthalmic steroids to cause systemic effects like those of glucocorticoids.
• In patients who develop DKA but do not fit the clinical phenotype of type 2 diabetes and do not test positive for the usual markers for type 1 diabetes, further investigation into rarer forms of diabetes is warranted.