Diabetes mellitus and blood glucose variability increases the 30-day readmission rate after kidney transplantation.

INTRODUCTION: Inpatient hyperglycemia is an established independent risk factor among several patient cohorts for hospital readmission. This has not been studied after kidney transplantation. Nearly one-third of patients who have undergone a kidney transplant reportedly experience 30-day readmission. METHODS: Data on first-time solitary kidney transplantations were retrieved between September 2015 and December 2018. Information was linked to the electronic health records to determine diagnosis of diabetes mellitus and extract glucometric and insulin therapy data. Univariate logistic regression analysis and the XGBoost algorithm were used to predict 30-day readmission. We report the average performance of the models on the testing set on bootstrapped partitions of the data to ensure statistical significance. RESULTS: The cohort included 1036 patients who received kidney transplantation; 224 (22%) experienced 30-day readmission. The machine learning algorithm was able to predict 30-day readmission with an average area under the receiver operator curve (AUC) of 78% with (76.1%, 79.9%) 95% confidence interval (CI). We observed statistically significant differences in the presence of pretransplant diabetes, inpatient-hyperglycemia, inpatient-hypoglycemia, minimum and maximum glucose values among those with higher 30-day readmission rates. The XGBoost model identified the index admission length of stay, presence of hyper- and hypoglycemia, the recipient and donor body mass index (BMI) values, presence of delayed graft function, and African American race as the most predictive risk factors of 30-day readmission. Additionally, significant variations in the therapeutic management of blood glucose by providers were observed. CONCLUSIONS: Suboptimal glucose metrics during hospitalization after kidney transplantation are associated with an increased risk for 30-day hospital readmission. Optimizing hospital blood glucose management, a modifiable factor, after kidney transplantation may reduce the risk of 30-day readmission.

Validation of automated data abstraction for SCCM discovery VIRUS COVID-19 registry: practical EHR export pathways (VIRUS-PEEP).

Background: The gold standard for gathering data from electronic health records (EHR) has been manual data extraction; however, this requires vast resources and personnel. Automation of this process reduces resource burdens and expands research opportunities. Objective: This study aimed to determine the feasibility and reliability of automated data extraction in a large registry of adult COVID-19 patients. Materials and methods: This observational study included data from sites participating in the SCCM Discovery VIRUS COVID-19 registry. Important demographic, comorbidity, and outcome variables were chosen for manual and automated extraction for the feasibility dataset. We quantified the degree of agreement with Cohen's kappa statistics for categorical variables. The sensitivity and specificity were also assessed. Correlations for continuous variables were assessed with Pearson's correlation coefficient and Bland-Altman plots. The strength of agreement was defined as almost perfect (0.81-1.00), substantial (0.61-0.80), and moderate (0.41-0.60) based on kappa statistics. Pearson correlations were classified as trivial (0.00-0.30), low (0.30-0.50), moderate (0.50-0.70), high (0.70-0.90), and extremely high (0.90-1.00). Measurements and main results: The cohort included 652 patients from 11 sites. The agreement between manual and automated extraction for categorical variables was almost perfect in 13 (72.2%) variables (Race, Ethnicity, Sex, Coronary Artery Disease, Hypertension, Congestive Heart Failure, Asthma, Diabetes Mellitus, ICU admission rate, IMV rate, HFNC rate, ICU and Hospital Discharge Status), and substantial in five (27.8%) (COPD, CKD, Dyslipidemia/Hyperlipidemia, NIMV, and ECMO rate). The correlations were extremely high in three (42.9%) variables (age, weight, and hospital LOS) and high in four (57.1%) of the continuous variables (Height, Days to ICU admission, ICU LOS, and IMV days). The average sensitivity and specificity for the categorical data were 90.7 and 96.9%. Conclusion and relevance: Our study confirms the feasibility and validity of an automated process to gather data from the EHR.

Hybrid Closed-Loop Insulin Pump Technology Can Be Safely Used in the Inpatient Setting.

BACKGROUND: Hybrid closed-loop (HCL) systems, also known as automated insulin delivery systems, are a rapidly growing technology in diabetes management. Because more patients are using these systems in the outpatient setting, it is important to also assess inpatient safety to determine whether HCL use can be continued when those patients become hospitalized. METHODS: The records of patients using HCL technology on admission to our hospital between June 1, 2020, and June 30, 2021, were analyzed. RESULTS: The final analysis included 71 patients divided into 3 categories based on their pump use as an inpatient: (1) HCL users; (2) manual pump users; and (3) pump removed. All cohorts were similar in age, sex, race, hemoglobin A1C at admission, and in Medicare Severity Diagnosis Related Group. Pairwise comparisons indicated that patient-stay mean glucose levels, frequency of patient-specific hyperglycemic measurements, and frequency of hypoglycemic events were similar between all groups. No adverse events, particularly occurrences of diabetic ketoacidosis, pump site complications or infection, or equipment malfunction, were reported. CONCLUSION: This preliminary case series review indicates that continued use of HCL technology in the hospital is safe. Moreover, glycemic control in HCL users was comparable with that in those using insulin pump with manual settings and those converted to basal-bolus insulin therapy.

Use of Cross-sectional and Perspective Mapping to Spatially and Statistically Represent Inpatient Glucose Control.

BACKGROUND: The use of inpatient location for the depiction of glycemic control is an alternative approach to the traditional analysis of hospital-derived glucometric data. Our aim was to develop a method of spatial representation and to test for corresponding statistical variation in inpatient glucose control data. METHODS: Point-of-care blood glucose data from inpatients with diabetes mellitus were extracted. Calculations included patient-day weighted means (PDWMs) and percentage of patient hospital days with hypoglycemia. Results were overlaid onto hospital floor plans, and room numbers were used as geolocators to generate cross-sectional (2-dimensional) and perspective (3-dimensional) views of the data. Linear mixed and mixed-effects logistic regression models were used to compare the location effect and to assess statistical variation in the data after adjusting for age, sex, and severity of illness. RESULTS: Visual inspection of cross-sectional and perspective maps demonstrated variation in glucometric outcomes across areas within the hospital. Statistical analysis confirmed significant variation between some hospital wings and floors. CONCLUSIONS: Spatial depiction of glucometric data within the hospital could yield insights into hot spots of poor glycemic control. Future studies on how to operationalize this approach, and whether this method of analysis can drive changes in glycemic management practices, need to be conducted.

Nurse adherence to post-hypoglycemic event monitoring for hospitalized patients with diabetes mellitus.

PURPOSE: The aim of this study was to investigate adherence to a posthypoglycemic event-monitoring policy for inpatients with diabetes mellitus receiving insulin therapy. METHODS: We analyzed point-of-care blood glucose data from noncritically ill inpatients receiving insulin therapy who had a hypoglycemic event (<70 mg/dL glucose) from January 3, 2017, through April 7, 2018. Blood glucose was measured until 2 sequential readings showed hypoglycemia resolution. An interval of 10 to 20 min between measurements was defined as compliant with policy. We calculated the median (IQR) time of each interval. RESULTS: We analyzed 896 episodes of hypoglycemia in 426 patients: 698 events had only 1 hypoglycemic measurement; 165 had 2 sequential hypoglycemic measurements; and 33 had 3 sequential hypoglycemic measurements. Median (IQR) times between measurements ranged from 18 (15-23) minutes to 28 (21-38) minutes. For patients with 1 hypoglycemic measurement, less than 50% of follow-up measurements were compliant. Similarly, for those with 2 sequential hypoglycemic values, less than 50% of measurements were compliant; for those with 3 sequential hypoglycemic values, less than 58%. CONCLUSION: Many instances of hypoglycemia had intervals between sequential glucose measurements that were longer than required by policy. These longer-than-expected intervals could place patients at undue risk for a prolonged hypoglycemic event. A better understanding of barriers to post-hypoglycemic event management in inpatients is required. Inpatient nurses, who are at the forefront of assessing and treating patients with hypoglycemia, should be key partners in assessing the algorithms for hypoglycemia care.

Impact of measurement error on predicting population-based inpatient glucose control.

AIM: Instrument measurement error (ME) may affect ability of damped trend analysis to forecast inpatient glycemic control. MATERIALS & METHODS: A statistical approach was developed to introduce ME into damped trend analysis algorithm. Point-of-care blood glucose device data were extracted from the laboratory system. Forecasts were generated from various inpatient subgroups and time intervals. RESULTS: ME produced differences in damped trend model during the forecast learning cycle. However, forecast trajectory stayed identical regardless of ME in 85% (119/140) of studied scenarios. Forecasts did not change with greater ME. CONCLUSION: ME inherent in the point-of-care blood glucose device had little effect on trajectory of damped trend exponential forecasts and apparently would not influence decision making in inpatient glycemic control algorithms.

Basal-bolus insulin therapy in postoperative inpatients with diabetes mellitus: directions for future quality-improvement initiatives.

AIM: To determine variables associated with hyperglycemia and insulin therapy in postoperative inpatients with diabetes mellitus following a quality-improvement initiative. MATERIALS & METHODS: Patients with diabetes mellitus following an elective surgical procedure (n = 782; 877 surgical procedures) were selected. RESULTS: Age, hemoglobin A1c corticosteroids, insulin therapy and year of surgery were associated (p < 0.01) with hyperglycemia. Hemoglobin A1c, hyperglycemia, case mix index and corticosteroids were associated (p ≤ 0.03) with insulin therapy. Hyperglycemia and use of insulin varied by surgical specialty. CONCLUSION: Data could be used to modify current treatment algorithms. Variations in hyperglycemia and insulin use by surgical specialty require further investigation.

How is the weather? Forecasting inpatient glycemic control.

AIM: Apply methods of damped trend analysis to forecast inpatient glycemic control. METHOD: Observed and calculated point-of-care blood glucose data trends were determined over 62 weeks. Mean absolute percent error was used to calculate differences between observed and forecasted values. Comparisons were drawn between model results and linear regression forecasting. RESULTS: The forecasted mean glucose trends observed during the first 24 and 48 weeks of projections compared favorably to the results provided by linear regression forecasting. However, in some scenarios, the damped trend method changed inferences compared with linear regression. In all scenarios, mean absolute percent error values remained below the 10% accepted by demand industries. CONCLUSION: Results indicate that forecasting methods historically applied within demand industries can project future inpatient glycemic control. Additional study is needed to determine if forecasting is useful in the analyses of other glucometric parameters and, if so, how to apply the techniques to quality improvement.

Association of Therapy for Autoimmune Disease With Myelodysplastic Syndromes and Acute Myeloid Leukemia.

IMPORTANCE: Therapy-related myeloid neoplasms are a potentially life-threatening consequence of treatment for autoimmune disease (AID) and an emerging clinical phenomenon. OBJECTIVE: To query the association of cytotoxic, anti-inflammatory, and immunomodulating agents to treat patients with AID with the risk for developing myeloid neoplasm. DESIGN, SETTING, AND PARTICIPANTS: This retrospective case-control study and medical record review included 40 011 patients with an International Classification of Diseases, Ninth Revision, coded diagnosis of primary AID who were seen at 2 centers from January 1, 2004, to December 31, 2014; of these, 311 patients had a concomitant coded diagnosis of myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). Eighty-six cases met strict inclusion criteria. A case-control match was performed at a 2:1 ratio. MAIN OUTCOMES AND MEASURES: Odds ratio (OR) assessment for AID-directed therapies. RESULTS: Among the 86 patients who met inclusion criteria (49 men [57%]; 37 women [43%]; mean [SD] age, 72.3 [15.6] years), 55 (64.0%) had MDS, 21 (24.4%) had de novo AML, and 10 (11.6%) had AML and a history of MDS. Rheumatoid arthritis (23 [26.7%]), psoriasis (18 [20.9%]), and systemic lupus erythematosus (12 [14.0%]) were the most common autoimmune profiles. Median time from onset of AID to diagnosis of myeloid neoplasm was 8 (interquartile range, 4-15) years. A total of 57 of 86 cases (66.3%) received a cytotoxic or an immunomodulating agent. In the comparison group of 172 controls (98 men [57.0%]; 74 women [43.0%]; mean [SD] age, 72.7 [13.8] years), 105 (61.0%) received either agent (P = .50). Azathioprine sodium use was observed more frequently in cases (odds ratio [OR], 7.05; 95% CI, 2.35- 21.13; P < .001). Notable but insignificant case cohort use among cytotoxic agents was found for exposure to cyclophosphamide (OR, 3.58; 95% CI, 0.91-14.11) followed by mitoxantrone hydrochloride (OR, 2.73; 95% CI, 0.23-33.0). Methotrexate sodium (OR, 0.60; 95% CI, 0.29-1.22), mercaptopurine (OR, 0.62; 95% CI, 0.15-2.53), and mycophenolate mofetil hydrochloride (OR, 0.66; 95% CI, 0.21-2.03) had favorable ORs that were not statistically significant. No significant association between a specific length of time of exposure to an agent and the drug's category was observed. CONCLUSIONS AND RELEVANCE: In a large population with primary AID, azathioprine exposure was associated with a 7-fold risk for myeloid neoplasm. The control and case cohorts had similar systemic exposures by agent category. No association was found for anti-tumor necrosis factor agents. Finally, no timeline was found for the association of drug exposure with the incidence in development of myeloid neoplasm.

Hyperglycemia during the immediate period following liver transplantation.

AIM: High blood glucose levels in the hospital are common among transplant recipients. METHODS: Retrospective analysis, stratified by diagnosis of pretransplant diabetes mellitus (DM). RESULTS: Of 346 patients, 96 had pretransplant DM (insulin, n = 60; no insulin, n = 36) and 250 did not. Patients with pretransplant DM had higher inpatient mean glucose levels and more hyperglycemia and hypoglycemia (all p < 0.01). For patients without pretransplant DM, the need for insulin at discharge increased 23% for every 5-year age increase (odds ratio: 1.23; 95% CI: 1.06-1.44; p = 0.007) and 51% for every five units of glucose measurements >180 mg/dl (OR: 1.51; 95% CI: 1.23-1.95; p < 0.01). CONCLUSION: Inpatient hyperglycemia was common in liver transplant recipients. Hospital practitioners must anticipate the need to teach self-management skills to liver transplant recipients.

GUIDELINES TO IMPROVE PERIOPERATIVE MANAGEMENT OF DIABETES MELLITUS: ASSESSMENT OF THE IMPACT OF CHANGE ACROSS TIME.

OBJECTIVE: Assess the impact of guidelines on the care of patients with diabetes undergoing elective surgery. METHODS: A multidisciplinary team developed perioperative guidelines. Overall changes in key measures were evaluated after guidelines were introduced and compared with a historical cohort. RESULTS: The historical cohort included 254 surgical procedures, and the post-guidelines implementation cohort comprised 1,387. Glucose monitoring was performed preoperatively in 93% of cases in the post-guidelines implementation cohort and in 88% in the historical cohort (P<.01), but the percentage of cases with measurements decreased over 12 months (from 95% to 91%, P = .044). Glucose was intraoperatively monitored in 67% of cases after guidelines were introduced and in 29% historically (P<.01); the post-guidelines implementation percentage decreased over 12 months from 67% to 55% (P<.01). The performance of glucose monitoring in the postanesthesia care unit (PACU) did not differ (86% vs. 87%, P = .57), but it decreased over 12 months, from 91% to 84% (P<.01). After introduction of the guidelines, insulin use increased in the preoperative, intraoperative, and PACU areas (all P≤.01) but decreased by the end of 12 months (all P<.01). Mean preoperative and PACU glucose levels in the post- guidelines implementation cohort were significantly lower than in the historical cohort (P<.01). CONCLUSION: Multidisciplinary management guidelines for diabetes patients undergoing surgery can improve the performance of key measures of care. Although adherence to recommendations generally remained higher after guideline implementation than in the historical period, the improvement in several measures began to decline over time.

Statistical transformation and the interpretation of inpatient glucose control data from the intensive care unit.

Glucose control can be problematic in critically ill patients. We evaluated the impact of statistical transformation on interpretation of intensive care unit inpatient glucose control data. Point-of-care blood glucose (POC-BG) data derived from patients in the intensive care unit for 2011 was obtained. Box-Cox transformation of POC-BG measurements was performed, and distribution of data was determined before and after transformation. Different data subsets were used to establish statistical upper and lower control limits. Exponentially weighted moving average (EWMA) control charts constructed from April, October, and November data determined whether out-of-control events could be identified differently in transformed versus nontransformed data. A total of 8679 POC-BG values were analyzed. POC-BG distributions in nontransformed data were skewed but approached normality after transformation. EWMA control charts revealed differences in projected detection of out-of-control events. In April, an out-of-control process resulting in the lower control limit being exceeded was identified at sample 116 in nontransformed data but not in transformed data. October transformed data detected an out-of-control process exceeding the upper control limit at sample 27 that was not detected in nontransformed data. Nontransformed November results remained in control, but transformation identified an out-of-control event less than 10 samples into the observation period. Using statistical methods to assess population-based glucose control in the intensive care unit could alter conclusions about the effectiveness of care processes for managing hyperglycemia. Further study is required to determine whether transformed versus nontransformed data change clinical decisions about the interpretation of care or intervention results.

Statistical transformation and the interpretation of inpatient glucose control data.

OBJECTIVE: To introduce a statistical method of assessing hospital-based non-intensive care unit (non-ICU) inpatient glucose control. METHODS: Point-of-care blood glucose (POC-BG) data from hospital non-ICUs were extracted for January 1 through December 31, 2011. Glucose data distribution was examined before and after Box-Cox transformations and compared to normality. Different subsets of data were used to establish upper and lower control limits, and exponentially weighted moving average (EWMA) control charts were constructed from June, July, and October data as examples to determine if out-of-control events were identified differently in nontransformed versus transformed data. RESULTS: A total of 36,381 POC-BG values were analyzed. In all 3 monthly test samples, glucose distributions in nontransformed data were skewed but approached a normal distribution once transformed. Interpretation of out-of-control events from EWMA control chart analyses also revealed differences. In the June test data, an out-of-control process was identified at sample 53 with nontransformed data, whereas the transformed data remained in control for the duration of the observed period. Analysis of July data demonstrated an out-of-control process sooner in the transformed (sample 55) than nontransformed (sample 111) data, whereas for October, transformed data remained in control longer than nontransformed data. CONCLUSION: Statistical transformations increase the normal behavior of inpatient non-ICU glycemic data sets. The decision to transform glucose data could influence the interpretation and conclusions about the status of inpatient glycemic control. Further study is required to determine whether transformed versus nontransformed data influence clinical decisions or evaluation of interventions.

Care directed by a specialty-trained nurse practioner or physician assistant can overcome clinical inertia in management of inpatient diabetes.

OBJECTIVE: The study's objective was to determine the impact of care directed by a specialty-trained nurse practitioner (NP) or physician assistant (PA) on use of basal-bolus insulin therapy and glycemic control in a population of noncritically ill patients with diabetes. METHODS: A retrospective review of diabetes patients evaluated between July 1, 2011 and December 31, 2011 was conducted. Patients cotreated by a specialty-trained NP/PA were compared with patients who did not receive such care. RESULTS: In total, 171 patients with 222 hospitalizations were cotreated by an NP/PA and 543 patients with 665 hospitalizations were not. Patients with NP/PA involvement were younger, and had more frequent hyperglycemia, and had greater corticosteroid use than patients without NP/PA involvement (P<.01 for all). Basal-bolus insulin therapy was administered to 80% of patients with NP/PA involvement and 34% of patients without it (P<.01). After adjustment for age, sex, hyperglycemia measures, and corticosteroid use, the odds of basal-bolus insulin therapy being administered were increased significantly through NP/PA care (odds ratio, 3.66; 95% confidence interval, 2.36-5.67; P<.01). After adjustment for these variables and insulin regimen, NP/PA care was significantly correlated with lower mean point-of-care glucose levels at 24 hours before discharge (P = .042). CONCLUSION: Diabetes care assisted by an NP/PA trained in inpatient diabetes management results in greater use of recommended basal-bolus insulin therapy and is correlated with lower mean glucose levels before discharge. Adapting this model for use outside an endocrinology consult service needs to be explored so that the expertise can be brought to a broader inpatient population with diabetes.

Overcoming clinical inertia in the management of postoperative patients with diabetes.

OBJECTIVE: To assess the impact of an intervention designed to increase basal-bolus insulin therapy administration in postoperative patients with diabetes mellitus. METHODS: Educational sessions and direct support for surgical services were provided by a nurse practitioner (NP). Outcome data from the intervention were compared to data from a historical (control) period. Changes in basal-bolus insulin use were assessed according to hyperglycemia severity as defined by the percentage of glucose measurements >180 mg/dL. RESULTS: Patient characteristics were comparable for the control and intervention periods (all P≥.15). Overall, administration of basal-bolus insulin occurred in 9% (8/93) of control and in 32% (94/293) of intervention cases (P<.01). During the control period, administration of basal-bolus insulin did not increase with more frequent hyperglycemia (P = .22). During the intervention period, administration increased from 8% (8/96) in patients with the fewest number of hyperglycemic measurements to 60% (57/95) in those with the highest frequency of hyperglycemia (P<.01). The mean glucose level was lower during the intervention period compared to the control period (149 mg/dL vs. 163 mg/dL, P<.01). The proportion of glucose values >180 mg/dL was lower during the intervention period than in the control period (21% vs. 31% of measurements, respectively, P<.01), whereas the hypoglycemia (glucose >70 mg/dL) frequencies were comparable (P = .21). CONCLUSION: An intervention to overcome clinical inertia in the management of postoperative patients with diabetes led to greater utilization of basal-bolus insulin therapy and improved glucose control without increasing hypoglycemia. These efforts are ongoing to ensure the delivery of effective inpatient diabetes care by all surgical services.

Clinical inertia during postoperative management of diabetes mellitus: relationship between hyperglycemia and insulin therapy intensification.

OBJECTIVE: Our objective was to assess the application of insulin regimens in surgical postoperative patients with diabetes. METHODS: A chart review was conducted of patients with diabetes who were hospitalized postoperatively between January 1 and April 30, 2011. Analysis was restricted to patients hospitalized for ≥3 days and excluded cases with an endocrinology consult. Insulin regimens were categorized as "basal plus short acting," "short acting only," or "none," and the pattern of use was evaluated by hyperglycemia severity according to tertiles of both mean glucose and the number of glucose measurements >180 mg/dl. RESULTS: Among cases selected for analysis (n = 119), examination of changes in insulin use based on tertiles of mean glucose showed that use of basal plus short-acting insulin increased from 10% in the lowest tertile (mean glucose, 120 mg/dl) to 18% in the highest tertile (mean glucose, 198 mg/dl; p < .01); however, 70% of patients in the highest tertile continued to receive short-acting insulin only, with 12% receiving no insulin. Intensification of insulin to a basal plus short-acting regimen was also seen when changes were evaluated by the number of measurements >180 mg/dl (p < .01), but 70% and 12% of patients in the highest tertile still remained only on short-acting insulin or received no insulin, respectively. CONCLUSIONS: Use of basal plus short-acting insulin therapy increased with worsening hyperglycemia, but many cases did not have therapy intensified to the recommended insulin regimen--evidence of clinical inertia. Strategies should be devised to overcome inpatient clinical inertia in the treatment of postoperative patients with diabetes.

Perioperative management of patients with diabetes undergoing ambulatory elective surgery.

OBJECTIVE: The objective was to assess processes of care for patients with diabetes undergoing elective surgery. METHODS: A retrospective review of medical records was conducted to determine frequency of perioperative glucose monitoring, changes in glucose control, and treatment of intraoperative hyperglycemia. RESULTS: A total of 268 patients underwent 287 elective procedures. Mean age was 67 years, 63% were men, 97% had type 2 diabetes, and most (57%) were treated with oral hypoglycemic agents. Average perioperative time was approximately 8 h. Mean preoperative hemoglobin A1c was 7.0%; however, this value was checked in only 52% of cases. A glucose measurement was obtained in 89% of cases in the preoperative area and in 87% in the postanesthesia care unit, but in only 33% of cases did a value get checked intraoperatively. Average glucose was 139 mg/dl preoperatively, increasing to 166 mg/dl postoperatively (p <.001). Glucose levels increased regardless of type of outpatient medical therapy used to treat hyperglycemia, except for those on combination oral agents plus insulin (p =.06). CONCLUSIONS: These data indicate suboptimal documentation of outpatient hemoglobin A1c. Intraoperative glucose monitoring seldom occurred, despite prolonged periods under anesthesia and perioperative deterioration of glycemic control. Standards need to be developed and interventions are needed to enhance management of diabetes patients undergoing elective procedures.

Transitioning insulin pump therapy from the outpatient to the inpatient setting: a review of 6 years' experience with 253 cases.

BACKGROUND: We reviewed the care of a large cohort of patients with diabetes mellitus on insulin pump therapy who required an inpatient stay. METHODS: Records were reviewed of patients hospitalized between January 1, 2006, and December 31, 2011. RESULTS: A total of 136 patients using insulin pumps had 253 hospitalizations. Mean (standard deviation) patient age was 55 (16) years, diabetes duration was 29 (15) years, and pump duration was 6 (5) years. Insulin pump therapy was continued in 164 (65%) hospitalizations. Adherence to core process measures improved over time: by 2011, 100% of cases had an endocrinology consultation, 100% had the required insulin pump order set completed, and 94% had documentation of the signed agreement specifying patient responsibilities for continued use of the technology while hospitalized. Documentation of the insulin pump flow sheet also increased but could still be located in only 64% of cases by the end of 2011. Mean glucose was not significantly different among patients who remained on insulin pump therapy compared to those for whom it was discontinued (p > .1), but episodes of severe hyperglycemia (>300 mg/dl) and hypoglycemia (<40 mg/dl) were significantly less common among pump users. No pump site infections, mechanical pump failures, or episodes of diabetic ketoacidosis were observed among patients remaining on therapy. CONCLUSIONS: With appropriate patient selection and usage guidelines, most patients using insulin pumps can safely have their therapy transitioned to the inpatient setting. Further study is needed to determine whether this approach can be translated to other hospital settings.

Outpatient-to-inpatient transition of insulin pump therapy: successes and continuing challenges.

BACKGROUND: Insulin pump therapy is a complex technology prone to errors when employed in the hospital setting. When patients on insulin pump therapy require hospitalization, practitioners caring for them must decide whether to allow continued pump use. We provide the largest review regarding transitioning insulin pump therapy from the outpatient to inpatient setting. METHOD: Records of inpatient insulin pump users were retrospectively analyzed at a metropolitan Phoenix hospital between January 2006 and December 2009. Adherence to institutional procedures on insulin pump use was assessed, glycemic control was determined, and adverse events were examined. RESULTS: We examined records on 65 patients with insulin pumps, totaling 125 hospitalizations. Mean (standard deviation) patient age was 55 (17) years, diabetes duration was 27 (14) years, pump duration was 6 (5) years, length of hospital stay was 4.7 (6.3) days, hemoglobin A1c was 7.3 (1.3)%, 85% had type 1 diabetes mellitus, 57% were women, and 97% were white. Admissions involving insulin pumps increased (23 in 2006, 17 in 2007, 40 in 2008, and 45 in 2009). Insulin pump therapy was continued in 83 (66%) hospitalizations. Among these hospitalizations, endocrinology consultations were obtained in 89%, consent agreements were found in 83%, insulin pump order sets were completed in 89%, admission glucose was checked in 100%, and nursing assessments of pump insertion sites were documented in 89%, but bedside insulin pump flow sheets were found in only 55%. Mean glucose of 175 (57) mg/dl was not significantly different than that in hospitalizations where insulin pumps were discontinued [175 (42) mg/dl] or used intermittently [177 (7) mg/dl]. There was one instance of a pump catheter kinking; however, no other adverse events (pump site infections, mechanical pump failure, diabetic ketoacidosis) were observed, and there were no use-related fatalities. CONCLUSIONS: Most patients using insulin pumps can safely have their therapy transitioned when hospitalized. A policy on inpatient continuous subcutaneous insulin infusion use can be successfully implemented. Compliance with required procedures can be achieved, although there was room to improve adherence with some process measures. Further study is needed to determine how to optimize glycemic control when pumps are allowed during hospitalization.