This article is more than 5 years old. Information may no longer be current.
Sweet ups and downs: Investigating the cause of recurrent severe hyperglycemia in a hospitalized patient
Click Here to Manage Email Alerts
The inpatient diabetes service was called to help with the blood glucose management of a 54-year-old male with a history of acute myeloid leukemia and Burkitt’s lymphoma admitted for a cycle of chemotherapy. The patient had been diagnosed with type 2 diabetes 3 years prior and had been suffering from glycemia excursions whenever he was being treated with glucocorticoids as part of his chemotherapy regimen. His home diabetes regimen consisted of metformin and glipizide ER, with an HbA1c of 7.9%.
Ronald Tamler
At the time of visit, the patient had no longer received glucocorticoid treatment for 2 days and was eating a carbohydrate-consistent diet.
Other medical history was remarkable for hypertension, ventriculopleural shunt insertion and myocardial infarct.
Medications in the hospital:
- Sodium chloride 0.9% 1,000 mL infusion IV, continuous;
- Folic acid (Folvite, Wyeth Pharmaceuticals) tablet 1 mg oral daily;
- Dextrose 50% injection, 25-mL IV push as-needed for severe hypoglycemia <50mg/dL
- Heparin (porcine) injection 5,000 units/mL (5,000 units subcutaneous every 12 hours);
- Insulin glargine (Lantus, Sanofi) 100 unit/mL injection (9 units subcutaneous before sleep);
- Insulin lispro (Humalog, Eli Lilly) 100 unit/mL injection (2-12 units, subcutaneous, three times daily with meals);
- Dextrose (Glutose, Paddock Laboratories) oral gel 15 g (as-needed for hypoglycemia <70 mg/dL)
- Rituximab (Rituxan; Genentech, Biogen Idec) 653 mg in sodium chloride 0.9% (653 mL chemo infusion, 375 mg/m2, IV, every 7 days);
-
Dexamethasone (Decadron, Merck) 0.1% ophthalmic suspension (two drops in both eyes every 6 hours);
- Ondansetron (Zofran, GlaxoSmithKline) injection 8 mg (IV push every 12 hours);
- Filgrastim (Neupogen, Amgen) injection 600 mcg (subcutaneous, once daily);
- Acyclovir (Zovirax, Valeant) tablet 800 mg (every 12 hours, SCH 6 and 6);
- Atovaquone (Mepron, GlaxoSmithKline) 750 mg/5 mL suspension (1,500 mg, oral, daily);
- Carvedilol (Coreg, GlaxoSmithKline) tablet 3.125 mg (orally twice a day with meals);
- Docusate sodium (Colace, Purdue Pharma) capsule 100 mg (orally three times per day);
- Glipizide ER (Glucotrol XL, Pfizer) 24-hour tablet (5 mg orally each day with breakfast);
- Polyethylene glycol 3350 (Miralax, Merck) packet (17 g orally each day);
- Sennosides (Senokot, Purdue Pharma) tablet (17.2 mg orally at bedtime);
- Grade 0 mouthwash (30 mL, swish and spit every 2 hours while awake);
- Famotidine (Pepcid, McNeil) 20 mg in iso-osmotic solution with 50 mL IV piggyback (20 mg, intravenously every 12 hours).
At the time of visit, the patient was afebrile with normal heart rate and blood pressure and a weight of 145 lb at 5’6” (BMI, 23.5). He was looking surprisingly well for his condition, and his physical exam was unremarkable, except for a port that facilitated access for chemotherapy. There were no active IV infusions.
The patient had received eight units of insulin glargine the previous night and four units of insulin lispro at 7 a.m.
A tabulation of his blood glucose levels shows these results:
What is the most likely explanation for this patient’s erratic blood glucose levels?
A.) Glucocorticoid-related exacerbation of hyperglycemia
B.) Inappropriate acquisition of samples for glucose measurement
C.) Extreme sensitivity to rapid-acting insulin, responding with a blood glucose drop of around 150 mg/dL per unit given
D.) Consumption of a large mug of coffee with three pieces of sugar at 5 a.m.
E.) A new intern in the lab handling specimens incorrectly
Case Discussion
Answer: B
This case was a real head-scratcher for us. While we had been suspicious of a blood sample having been obtained from a location upstream of a running dextrose infusion, the patient had no peripheral IV access. Moreover, both finger stick (point of care) and phlebotomy (central lab) samples showed unusually high blood glucose levels, several hours apart. The much more moderate hyperglycemia 4 minutes after the very elevated finger-stick glucose just added to the confusion.
We started our investigation by asking the patient, who denied any consumption of food or drink at 5 a.m. (D), but did recall being woken up for a finger stick. He never had experienced extreme insulin sensitivity during prior hospitalizations (C). He also had not received any glucocorticoids for more than 2 days, and during prior hospitalizations had gone back to his baseline glycemia about 2 to 3 days after his last dexamethasone dose (A). The fact that both point of care and lab data were affected makes mishandling of the sample in the lab rather unlikely (E).
In the end, we asked the patient’s nurse about the way she acquired the samples. It turned out that nurses on this ward frequently obtained blood samples from the chemotherapy port (B). The nurse had placed a droplet of blood from the port on the blood glucose test strip, and when the result was unreasonably high, verified it by waking the patient up with a finger stick. This was repeated a few hours later, when the lab sample was not obtained by phlebotomy, but rather by withdrawing a sample from the port, which had previously been the site of a dextrose-containing infusion.