Inguinal Hernia Repair in a Diabetic Patient

A 54-yr-old man is admitted for an inguinal hernia repair. He has been a type 1 diabetic for 18 years. Two years ago he developed end stage renal disease (ESRD), secondary to his diabetic kidney. He now requires daily peritoneal dialysis. The peritoneal fluid used consisted of a glucose-containing dialysis fluid in the daytime and an overnight dialysis fluid using one exchange of 2 liters peritoneal fluid containing icodextrin (7.5% wt/vol) (Extraneal; Baxter Healthcare, Castlebar, Ireland). Icodextrin is a cornstarch-derived glucose polymer. His insulin regimen consisted of Human Insulatard (Novo Nordisk) before bed and humalog (Lispro) before his main meals. On the morning (7:00 am) of his surgery, in the preoperative holding area, his capillary blood sugar value was 480 mg/dl. He had not taken any insulin for 12h and had been non per os since midnight. The blood sugar test was done with an AccuChek Active (Roche, Mannheim, Germany). There were no ketones in the urine, and the patient claimed his blood sugar had, in the past 2yr, been running higher. You speak to the surgeon and decide on monitored anesthesia care, which includes local anesthesia and sedation. You order 12 U of fast-acting insulin to be given IV stat (7:00 am). There is a delay in getting the patient to the operating room, but at 7:40 am you are given the go-ahead to pick up the patient. Much to your surprise, you find the patient somewhat incoherent and sweating. You take blood for blood glucose estimation while you call for 50% glucose. When the 50% glucose ampoule arrives, you don’t wait for the blood sugar result, but give the patient 50 ml of 50% glucose IV with good effect. You are wondering how only 12 U of the above insulin could have caused this hypoglycemia. The nurse now tells you that the AccuChek shows blood glucose of 320 mg/dl. You realize something is not right when the venous blood glucose results come back as 2 mmol/liter (normal fasting levels, 3.33-6.60 mmol/liter). You ask the lab to run the venous blood glucose estimation again. You get the same result. However, the patient was obviously hypoglycemic. Why is there such a discrepancy between the two blood sugar estimations?