Garry S. Tobin

Prediction and Prevention of Treatment-Related Inpatient Hypoglycemia

Background: Prolonged severe hypoglycemia (SH) in hospitalized patients is associated with increased morbidity and mortality. This study was undertaken to identify risk factors for SH, to apply that knowledge to the development of a prediction algorithm, and to institute a prevention program at a tertiary medical center. Methods: We analyzed SH events for 172 patients and developed computer algorithms to predict SH that were tested on a population of 3028 inpatients who were found to have blood glucose (BG) <90 mg/dl during their hospital stay. Variables with significant bivariate associations were entered into partition analyses to identify interactions. Logistic regression was performed by calculating parameters related to the odds of hypoglycemia below each cut point. Sensitivity and specificity were determined at various cut points. The cut points resulting in 50% sensitivity for each hypoglycemia level were determined. These algorithms were tested against the initial 172 adjudicated patients. Results: Variables related to the BG <40 mg/dl cut off point were basal and adjustment scale insulin doses, weight, and creatinine clearance, while variables related to the 60 mg/dl and 70 mg/dl cut points were basal, prandial, and adjustment scale insulin doses, weight, creatinine clearance, and sulfonylurea use. The 50% sensitivity cut point developed using the <70 mg/dl algorithm correctly identified 71% of the adjudicated cases, while the <60 mg/dl and <40 mg/dl algorithms identified 70% and 55% respectively. Conclusions: A validated prediction algorithm for SH can aid in the identification of patients at risk for SH and may be useful in the development of prevention strategies.

Inpatient insulin orders: Are patients getting what is prescribed?†

BACKGROUND In-hospital insulin administration is associated with many medication errors, but the frequency and reasons for insulin administration errors are poorly described. To document types and frequency of errors related to insulin administration, an examination of 4 units was conducted. METHODS Using snapshot methodology, 4 non-intensive care unit (ICU) areas (medicine, cardiology, transplant, and surgery) were examined in an observational, prospective manner for 4 weeks. Each patient on insulin on the first day was followed for 7 days. Definitions and error categories were defined prior to data collection. Error types and numbers were collected and quantified on per-day or per-patient basis. RESULTS A total of 116 patient audit periods covering a total of 378 inpatient hospital days were examined. Inpatient insulin regimens on day 1 included correctional insulin only (51.7% of cases), neutral protamine Hagedorn ([NPH] 12%), and glargine (28.4%). A total of 199 administration errors occurred at a rate of 1.72 errors/patient-period and 0.53 errors/patient day. Missing documentation of doses (15.5% of all patients) and insulin being held without an order (25% of patients) were the most frequently occurring events. Other errors include transcription (7.5%), timing errors (22.7%), and lack of documentation of physician notification of hypoglycemia (12.6%). CONCLUSIONS Errors associated with insulin in the hospital are common and reveal a number of system errors that should be addressed. These data provide a foundation for future performance improvement.

Inpatient glycemic control on the vascular surgery service.

OBJECTIVE To describe a structured inpatient insulin management protocol and order set for glycemic control on a vascular surgery service. METHODS Patients admitted to the vascular surgery service with underlying diabetes were enrolled in a study of use of a preprinted basal-bolus insulin order set based on a total daily dose of 0.5 U/kg (0.25 U/kg of insulin glargine and 0.25 U/kg of insulin as part divided into 3 equal mealtime doses). Outcomes included the mean glycemic control at each of 5 established time intervals, the percentage of blood glucose measurements within the target range of 70 to 180 mg/dL, the incidence of hypoglycemia, and the insulin dosages. Historical control patients with diabetes from the same hospital service were used for comparison. RESULTS Both the study group and the control group consisted of 26 patients. The number of finger-stick blood glucose measurements performed was 871 in the control group and 896 in the intervention group. The mean blood glucose level (+/- SD) for the intervention group was 149.4 +/- 50.7 mg/dL, in comparison with 165.2 +/- 64.4 mg/dL for the control group. The incidence of hypoglycemia decreased 50% in the intervention group-from 32 (4% of the finger-stick assessments in the control group) to 19 (2% of the finger-stick blood glucose measurements in the study group). The blood glucose target range of 70 to 180 mg/dL was achieved in 75% of the measurements in the study group versus 61% in the control group. The basal insulin dose was unchanged in 65% of the patients, and of the 9 patients requiring a change in the dose, 5 had the dose decreased by 10% and 4 had the dose increased by 10%. CONCLUSION The use of a standardized basal-bolus weight-based insulin regimen was successful at achieving improved glycemic control as well as reducing the incidence of hypoglycemia in an inpatient population with diabetes.

Round Table Discussion on Inpatient Use of Continuous Glucose Monitoring at the International Hospital Diabetes Meeting

In May 2015 the Diabetes Technology Society convened a panel of 27 experts in hospital medicine and endocrinology to discuss the current and potential future roles of continuous glucose monitoring (CGM) in delivering optimum health care to hospitalized patients in the United States. The panel focused on 3 potential settings for CGM in the hospital, including (1) the intensive care unit (ICU), (2) non-ICU, and (3) continuation of use of home CGM in the hospital. The group reviewed barriers to use and solutions to overcome the barriers. They concluded that CGM has the potential to improve the quality of patient care and can provide useful information to help health care providers learn more about glucose management. Widespread adoption of CGM by hospitals, however, has been limited by added costs and insufficient outcome data.In May 2015 the Diabetes Technology Society convened a panel of 27 experts in hospital medicine and endocrinology to discuss the current and potential future roles of continuous glucose monitoring (CGM) in delivering optimum health care to hospitalized patients in the United States. The panel focused on 3 potential settings for CGM in the hospital, including (1) the intensive care unit (ICU), (2) non-ICU, and (3) continuation of use of home CGM in the hospital. The group reviewed barriers to use and solutions to overcome the barriers. They concluded that CGM has the potential to improve the quality of patient care and can provide useful information to help health care providers learn more about glucose management. Widespread adoption of CGM by hospitals, however, has been limited by added costs and insufficient outcome data.

Prevention of inpatient hypoglycemia with a real-time informatics alert.

BACKGROUND Severe hypoglycemia (SH), defined as a blood glucose (BG) <40 mg/dL, is associated with an increased risk of adverse clinical outcomes in inpatients. OBJECTIVE To determine whether a predictive informatics hypoglycemia risk-alert supported by trained nurse responders would reduce the incidence of SH in our hospital. DESIGN A 5-month prospective cohort intervention study. SETTING Acute care medical floors in a tertiary care academic hospital in St. Louis, Missouri. PATIENTS From 655 inpatients on designated medical floors with a BG of <90 mg/dL, 390 were identified as high risk for hypoglycemia by the alert system. MEASUREMENTS The primary outcome was the incidence of SH occurring in high-risk intervention versus high-risk control patients. Secondary outcomes included: number of episodes of SH in all study patients, incidence of BG < 60 mg/dL and severe hyperglycemia with a BG >299 mg/dL, length of stay, transfer to a higher level of care, the frequency that high-risk patients orders were changed in response to the alert-intervention process, and mortality. RESULTS The alert process, when augmented by nurse-physician collaboration, resulted in a significant decrease by 68% in the rate of SH in alerted high-risk patients versus nonalerted high-risk patients (3.1% vs 9.7%, P = 0.012). Rates of hyperglycemia were similar on intervention and control floors at 28% each. There was no difference in mortality, length of stay, or patients requiring transfer to a higher level of care. CONCLUSION A real-time predictive informatics-generated alert, when supported by trained nurse responders, significantly reduced inpatient SH.

Weight‐based insulin dosing for acute hyperkalemia results in less hypoglycemia

Hyperkalemia treatment with intravenous insulin has been associated with hypoglycemia. This single-center, retrospective study compared the effects on hypoglycemia between weight-based insulin dosing (0.1 U/kg of body weight up to a maximum of 10 U) compared to standard flat doses of 10 U among patients weighing less than 95 kg. Of the 132 charts randomly selected for review, hypoglycemic events (blood glucose <70 mg/dL) were reduced from 27.3% in the 10-U group to 12.1% in the weight-based group (P = 0.05). The number of affected patients was reduced with 19.7% in the 10-U group and 10.6% in the weight-based group (P = 0.22). The potassium-lowering effects of these 2 strategies were similar between groups. Female patients and those with baseline glucose values <140 mg/dL were at increased risk for hypoglycemia. Weight-based insulin dosing (0.1 U/kg) for acute hyperkalemia therapy resulted in less hypoglycemia without impacting potassium lowering. Journal of Hospital Medicine 2016;11:355-357.

A CAsE Of sEVERE NEuROPAThy AssOCIATED WITh hyPERTRIglyCERIDEmIA

OBJECTIVE To describe a case of severe neuropathy associated with hypertriglyceridemia. METHODS We describe the clinical and laboratory findings of the study patient and review the relevant literature. RESULTS A 45-year-old woman presented to the emergency department with recurrent abdominal pain and severe peripheral neuropathy. Her laboratory data revealed elevated lipase and a very high triglyceride concentration (>10,000 mg/dL), consistent with a diagnosis of recurrent hypertriglyceridemia-induced pancreatitis. Workup for peripheral neuropathy showed normal concentrations of thyrotropin, fasting blood glucose, vitamin B(12), and creatinine, as well as a normal hemoglobin A(1c) level, serum protein electrophoresis, and urine protein electrophoresis. Rapid plasma reagin antibodies, antinuclear antibodies, and lyme antibodies were not detected. In the absence of other identifiable causes, hypertriglyceridemia was deemed the likely etiology of severe neuropathy in this patient. CONCLUSIONS Peripheral nerve conduction abnormalities can be identified in patients with mild hypertriglyceridemia in the absence of symptoms. Early recognition and aggressive management of hypertriglyceridemia may prevent the complications of severe peripheral neuropathy.

Effect of a glucagon receptor antibody (REMD-477) in type 1 diabetes: A randomized controlled trial

The aim of the current study (Clinical trial reg. no. NCT02715193, clinicaltrials.gov) was to study the efficacy and safety of REMD‐477, a glucagon receptor antagonist, in type 1 diabetes. This was a randomized controlled trial in which 21 patients with type 1 diabetes were enrolled. Glycaemic control and insulin use were evaluated in outpatient and inpatient settings, before and after a single 70‐mg dose of REMD‐477 (half‐life 7‐10 days) or placebo. Inpatient insulin use was 26% (95% CI, 47%, 4%) lower 1 day after dosing with REMD‐477 than with placebo (P = .02). Continuous glucose monitoring during post‐treatment days 6 to 12 showed that average daily glucose was 27 mg/dL lower (P < .001), percent time‐in‐target‐range (70‐180 mg/dL) was ~25% greater (~3.5 h/d) (P = .001), and percent time‐in‐hyperglycaemic‐range (> 180 mg/dL) was ~40% lower (~4 h/d) (P = .001) in the REMD‐477 group than in the placebo group, without a difference in percent time‐in‐hypoglycaemic‐range (<70 mg/dL). No serious adverse events were reported. Glucagon receptor antagonism decreases insulin requirements and improves glycaemic control in patients with type 1 diabetes.

Late-Onset T1DM and Older Age Predict Risk of Additional Autoimmune Disease

OBJECTIVE Type 1 diabetes (T1DM) is associated with other autoimmune diseases (AIDs), which may have serious health consequences. The epidemiology of AIDs in T1DM is not well defined in adults with T1DM. In this cross-sectional cohort study, we sought to characterize the incident ages and prevalence of AIDs in adults with T1DM across a wide age spectrum. RESEARCH DESIGN AND METHODS A total of 1,212 adults seen at the Washington University Diabetes Center from 2011 to 2018 provided informed consent for the collection of their age, sex, race, and disease onset data. We performed paired association analyses based on age at onset of T1DM. Multivariate logistic regression was used to evaluate the independent effects of sex, race, T1DM age of onset, and T1DM duration on the prevalence of an additional AID. RESULTS Mean ± SD age of T1DM onset was 21.2 ± 14.4 years. AID incidence and prevalence increased with age. Female sex strongly predicted AID risk. The most prevalent T1DM-associated AIDs were thyroid disease, collagen vascular diseases, and pernicious anemia. T1DM age of onset and T1DM duration predicted AID risk. Patients with late-onset T1DM after 30 years of age had higher risks of developing additional AIDs compared with patients with younger T1DM onset. CONCLUSIONS The prevalence of AIDs in patients with T1DM increases with age and female sex. Later onset of T1DM is an independent and significant risk factor for developing additional AIDs. Individuals who are diagnosed with T1DM at older ages, particularly women, should be monitored for other autoimmune conditions.

HIGH-RISK GASTRIC PATHOLOGY AND PREVALENT AUTOIMMUNE DISEASES IN PATIENTS WITH PERNICIOUS ANEMIA

OBJECTIVE Pernicious anemia (PA) develops from atrophic gastritis due to autoimmune destruction of parietal cells and results in achlorhydria, vitamin B12 and iron deficiencies, anemia, neurologic deficits, and premalignant and malignant stomach lesions. We report the presentation, diagnosis and gastric complications of PA in patients from an endocrinology practice. METHODS Thirty-four patients (31 female, 3 male) with PA who underwent esophagogastroduodenoscopy (EGD) or gastrectomy were identified. Pertinent clinical, laboratory, and pathology findings were reviewed and summarized. RESULTS The mean age of patients was 58.6 ± 14.2 years; the onset of PA was age 50.2 ± 15.3 years. Anemia reflected vitamin B12 and/or iron deficiencies. Parietal cell antibodies (PCA) were detected in 97% of patients, and intrinsic factor blocking antibody (IFBA) was found in 52%. Fasting gastrin and chromogranin A levels were elevated (1,518.0 ± 1,588.3 pg/mL, and 504.9.1 ± 1,524.9 ng/mL respectively). Autoimmune or immunologic diseases (AIDs) were present in 32/34 patients. Stomach pathology showed premalignant or malignant lesions in 26 patients, including gastric neuroendocrine tumors (GNETs) in 6 and adenocarcinoma in 1. One patient presented with neurologic symptoms and subacute combined degeneration of the posterior column of the spinal cord. CONCLUSION PA should be suspected in patients with unexplained anemia or neurologic symptoms. The diagnosis of PA relies on fasting gastrin and gastric auto-antibody testing, in addition to hematologic evaluation. EGD with measurement of gastric pH and biopsies of the fundus and antrum identifies patients with achlorhydria, atrophic gastritis, and premalignant and malignant stomach lesions. EGD surveillance of patients with high-risk stomach lesions is recommended. ABBREVIATIONS AID = autoimmune or immunologic disease; EGD = esophagogastroduodenoscopy; GNET = gastric neuroendocrine tumor; IFBA = intrinsic factor blocking antibody; PA = pernicious anemia; PCA = parietal cell antibody; T1D = type 1 diabetes.