Recognized and unrecognized hyperglycemia in cancer inpatients

Abstract

9079 Background: Diabetes mellitus (DM) is a recognized co-morbidity with cancers. Prevalence of DM in cancer inpatients and occurrence of acute hyperglycemic crises in cancer patients have not been reported. Methods: Chart, laboratory, and pharmacy data were evaluated retrospectively over 3 months in 2006 and cases of acute hyperglycemic crises were evaluated from 2001–2004 in a big cancer hospital. Results: Of the 2,636 unique patient admissions identified through lab and pharmacy data over a 3 month period in 2006, 1,735 (65%) met the epidemiological (fasting glucose (FG)>=126 mg/dL and/ or receiving DM therapy (Rx)) and 1,306 (50%) met the clinical criteria for a diagnosis of DM (FG>=126 mg/dL and/ or any glucose>=200 mg/dL on two different days, and/ or receiving DM Rx). Of the 463 treated patients 224 (48%) were on one therapy (excluding 38 (8.2%) on short acting insulin only); 13.4% were on combination/ long acting insulin therapy; 16% were on insulin + oral antidiabetics (OADs), and 18% were on combination OADs. On these respective therapies during hospitalization 68%, 49%, and 31% had severe hyperglycemia (glucose >300 mg/dL); and 19%, 14%, and 0% had glucose >400 mg/dL. Eighty eight with a glucose >300 mg/dL and 13 with glucose >400 mg/dL remained untreated during hospitalization. In the case study severe hyperglycemia (>300 mg/dL) predated by 54±18 (Mean±SEM) days (7 known DM on Rx) in 12 out of 22 patients admitted with acute hyperglycemic crises to cancer hospital emergency services over 4 years (14 diabetic ketoacidosis, 8 hyperglycemic hyperosmolar state). Subsequent to the hyperglycemic crises 17/22 died in 14±4 months (median survival: 12 months). In patients not known to be DM 23±6 days elapsed between starting known diabetogenic drugs and testing plasma glucose (10±4 days in known DM). Conclusions: A large number of cancer patients with diabetes remain unrecognized, untreated or inadequately treated even during hospitalizations. Our data reinforces the need for testing, recognizing and appropriately managing diabetes with an aim to prevent acute hyperglycemic complications in cancer patients. No significant financial relationships to disclose.