This patient should have a repeat measurement of fasting C-peptide and glucose levels. He has ketosis-prone type 2 diabetes mellitus. Patients with ketosis-prone type 2 diabetes do not fulfill the classic phenotype associated with autoimmune type 1 diabetes. These patients are often older, overweight or obese, and of black or Latino ethnicity. Patients with new-onset ketosis-prone diabetes require insulin therapy initially but might be able to be managed with oral agents in the future. Prior to switching from insulin to oral therapy, his pancreatic beta-cell function should be assessed with fasting C-peptide and glucose measurements. Ketosis-prone type 2 diabetes is heterogeneous condition in that the presence of autoantibodies is variable across the population, as is the degree of pancreatic beta cell function. His initial C-peptide level in the setting of hyperglycemia and diabetic ketoacidosis is not an accurate indication of his pancreatic function. Due to the toxic effects of prolonged hyperglycemia on the pancreatic beta cells, the fasting C-peptide and glucose or a glucagon-stimulated C-peptide should be measured 7 to 14 days after the correction of the acidosis in order to better assess function. If his repeat fasting C-peptide value is greater than or equal to 1.0 ng/mL (0.33 nmol/L) or his glucagon-stimulated C-peptide value is greater than or equal to 1.5 ng/mL (0.5 nmol/L), his beta cell function is preserved.

A sliding-scale insulin regimen that does not include basal insulin and does not begin insulin administration unless the blood glucose level is at or above a threshold level will cause wide swings from hyperglycemia to hypoglycemia, and this is inappropriate treatment.

Discontinuation of his insulin and switching to an oral agent such as metformin could be attempted with evidence of beta cell function preservation. Close follow-up would be necessary to monitor for worsening hyperglycemia or development of ketoacidosis, which would require restarting insulin therapy.

Determining autoimmunity, in conjunction with beta cell function, is helpful in assessing whether a patient has the potential to become insulin-independent in the future. His autoantibodies were negative at the time of his presentation, and it is unlikely that these would now be positive. It is not necessary to retest antibodies in this setting.