Several recent studies have shown increased morbidity and mortality associated with hyperglycemia in patients hospitalized for cardiovascular illnesses and acute illness requiring intensive care (1). While many of these studies have utilized intensive insulin therapy (IIT) as a means of controlling hyperglycemia, to our knowledge the possible iatrogenic contribution of intravenous medications has been overlooked.
A 78-year-old man with type 2 diabetes was hospitalized for respiratory failure. Glycemic control was suboptimal despite continuous insulin infusion. The diabetes service observed infusion of Vancomycin mixed in 250 ml of 5% dextrose (D5W) while examining the patient. Medication review revealed four different intravenous medications in D5W amounting to >1 l daily. This prompted review of medication profiles for all patients followed by the diabetes service. Of 10 patients, 7 were receiving intravenous medications mixed in D5W (median 1 [range 1–4]). All of these patients required insulin therapy, and three (43%) required an insulin drip to achieve mean glucose values ranging from 129 to 223 mg/dl. Several of the intravenous medications could be given in normal saline rather than D5W (2) (Table 1). Five of these patients were hospitalized long enough for an intervention. After removing dextrose diluent from their intravenous medications, four of the five patients showed improved glycemic control with mean glucose values decreasing by 17–51 mg/dl (mean 39 mg/dl). We attempted to reduce subsequent iatrogenic hyperglycemia by communicating with the pharmacist to “provide all intravenous medications mixed in normal saline” for each diabetic patient we managed.
Hyperglycemia is common in hospitalized patients and has been shown to be an important marker of increased mortality and poor clinical outcome (1). A recent study found that a modest increase in mean glucose from 80–99 to 100–119 mg/dl in an intensive care unit (ICU) setting led to a 27% relative increase in mortality (3). A prospective, randomized, controlled trial found that the use of IIT to maintain blood glucose ≤110 mg/dl led to relative risk reduction of death in the ICU by 42% (4). The American Association of Clinical Endocrinologists (AACE) has recently published guidelines on the management of inpatient hyperglycemia (1). Protocols to prevent iatrogenic hyperglycemia induced by use of dextrose as a diluent would also decrease resource utilization by decreasing the frequency of blood glucose measurement and insulin administration.
Our preliminary observations suggest that inpatients with diabetes frequently receive intravenous medications mixed in dextrose. Review of medications used in intravenous lines while examining the patient and careful review of medications and diluents can improve glycemic status and clinical outcomes. Creating a system to correct and prevent this occurrence can improve patient care and decrease resource utilization. Future guidelines for inpatient diabetes management would be improved by including this recommendation.
Drugs compatible in normal saline automatically made in D5W*
Acetazolamide | Cimetidine | Doxycycline | Norepinephrine |
Alprostadil | Ciprofloxacin† | Esmolol† | Octreotide |
Aminophylline† | Clindamycin† | Heparin† | Penicillins*† |
Argatroban | Cyclosporine | Ibutilide | Steroids‡ |
Azithromycin | Dobutamine† | Lidocaine† | Valproate sodium |
Cephalosporins†§ | Dopamine† | Nesiritide | Vancomycin |
Acetazolamide | Cimetidine | Doxycycline | Norepinephrine |
Alprostadil | Ciprofloxacin† | Esmolol† | Octreotide |
Aminophylline† | Clindamycin† | Heparin† | Penicillins*† |
Argatroban | Cyclosporine | Ibutilide | Steroids‡ |
Azithromycin | Dobutamine† | Lidocaine† | Valproate sodium |
Cephalosporins†§ | Dopamine† | Nesiritide | Vancomycin |
Includes nafcillin, penicillin G potassium, and piperacillin/tazobactam.
Manufacturer premade item(s).
Includes dexamethasone, hydrocortisone, and methylprednisolone.
Includes cefazolin, cefepime, cefotaxime, cefoxitin, ceftazidime, ceftriaxone, and cefuroxime.