DexCom Hospital Study-CGM Directed Insulin Delivery

Overview

This is a pilot investigator initiated/sponsored clinical trial to evaluate the feasibility of using real-time DexCom G6 continuous glucose monitor (CGM or glucose sensor) trend data and an insulin-dosing algorithm to control the glucose levels of type 2 diabetic patients undergoing hip/knee orthopedic surgery at Thomas Jefferson University Hospital (TJUH). Blinded CGM glucose data will be recorded prior to hospital admission, during hospital admission, and after hospital discharge to evaluate glucose control in this diabetic patient population in relationship to clinical outcomes (20 days of recorded blinded CGM data maximum). Prior studies have associated hyperglycemia, hypoglycemia, and glycemic variability with an increased incidence of nosocomial infection, deep vein thrombosis, pulmonary embolism, cardiac arrhythmias, mental status changes, hospital length of stay, and cost. The 20 type 2 diabetic patients in the Control Group will be managed according to standard-of-care methods at TJUH using finger-stick blood glucose measurements in an attempt to maintain the patient's blood glucose levels in the desired target range (80 to 180 mg/dL). Three blinded CGM will be used to record the patient's glucose trend data for future download to a computer and analysis. The 40 type 2 diabetic patients in the Investigational Group will be managed by the orthopedic floor nurses using the real-time DexCom G6 CGM trend data to determine the appropriate therapy to maintain the patient's glucose levels in the desired target range (80 to 180 mg/dL). In addition, three blinded CGM will be used to record the patient's glucose trend data for future download to a computer and analysis. The primary objective of this pilot study is to observe the blood glucose values and glucose sensor (CGM) values of diabetic patients before, during and after hospitalization for hip and knee arthroplasty surgery to determine whether there is a correlation between glucose control and clinical outcomes. This pilot data will be used to plan a future clinical trial in orthopedic surgery patients using CGM to determine patterns of hyperglycemia, hypoglycemia, and glycemic variability that correlate with clinical complications, hospital length of stay, and cost.

The investigators hypothesize that real-time CGM trend data can be used by the hospital floor nurse to safely dose insulin to maintain the glucose concentration in the target range (80-180 mg/dl), minimize the risk for mild hypoglycemia (\< 70 mg/dL), and eliminate the risk for severe hypoglycemia (\<54 mg/dL). Subjects may be consented during their pre-operative visit to the TJUH Pre-Admissions Testing Center (PAT) or during their pre-operative visit to the Department of Orthopedic Surgery's outpatient Rothman Clinic. Twenty patients with type 2 diabetes will be assigned to a control group (n=20) and forty patients will be assigned to an investigational group (n=40). The skin surface around each CGM insertion site may be photographed; then prepped with alcohol/chlorhexidine and allowed to dry. Research personnel will insert 3 blinded Dexcom G6 CGM into the upper arm (n=1) and abdomen/flank (n=2) of each study subject using sterile gloves and aseptic technique. The 3 blinded CGM will measure and record the concentration of tissue fluid glucose every 5 minutes for up to 10 days. Control Group (Hospital Admission): Blinded CGM that were inserted more than 8 or 9 days ago will be removed by research personnel prior to surgery Three new blinded DexCom G6 CGM will be inserted into an alternate location of the study subject's upper arm (n=1) and abdomen/flank (n=2) using aseptic technique. Patients may be discharged from the hospital approximately 24 to 48 hours after the surgical procedure wearing 3 blinded CGM. Patients will be instructed to self-remove the 3 blinded CGM after a maximum of 10 days of wear-time, and place the blinded CGM into a plastic specimen cup and a stamped addressed envelope. Subjects will be instructed to mail the envelop to TJU research personnel. Investigational Group (Hospital Admission): Blinded CGM that were inserted more than 8 or 9 days ago will be removed by research personnel prior to surgery in the SDA Uni. Two new blinded DexCom G6 CGM will be inserted into an alternate location of the study subject's upper arm (n=1) and abdomen/flank (n=2) using aseptic technique. One new real-time Dexcom G6 CGM will be inserted into the subcutaneous tissue of the abdomen using aseptic technique The real-time CGM will be activated and coupled to a tablet computer for data display at the bedside. The iPad will be programmed to alarm if the CGM glucose falls below \< 90 mg/dL (threshold alarm) or if the CGM predicts the glucose concentration will fall below \< 55 mg/dL in the next 20 minutes (predictive alarm). Patients may be discharged from the hospital approximately 24 to 48 hours after the surgical procedure wearing 3 blinded CGM. They will be instructed to self-remove the 3 blinded CGM after a maximum of 10 days of wear-time, and place the blinded CGM in a plastic cup and stamped addressed envelope. Subjects will be instructed to mail the envelope to TJU research personnel. In-hospital Glucose Control Methods (Control Group): The control group of patients will be managed using routine standard-of-care TJUH methods. Study subjects will have fingerstick blood glucose (BG) measurements performed before breakfast, before lunch, before dinner, at 9:00pm and 2:00am (5 assessments per day) using a commercial meter and test strips. The floor nurse may obtain additional finger stick BG measurements at any time to detect or confirm hypoglycemia/hyperglycemia. Diabetic patients will be managed by their orthopedic surgeons (with input from the endocrinologists) using the oral anti-diabetic medications and/or long-acting and rapid-acting insulin injections routinely used at home. In-hospital Glucose Control Methods (Investigational Group): The blood glucose management of the investigational group of diabetic patients will be per the usual practice of the Thomas Jefferson University Hospital (TJUH) orthopedic surgeons. Some of the investigational study subjects will be managed with oral anti-diabetic medications as their baseline diabetes therapy in the hospital, while other subjects will be managed with intermittent injections of long-acting insulin and/or rapid acting insulin with meals. This baseline diabetes therapy will be adjusted based upon the standard-of-care TJUH Correction Dose Insulin Therapy protocol. The orthopedic floor nurses will make an assessment using the real-time CGM trend data before each meal and 2-hours after each meal. More or less correction dose insulin will be injected into the patient's subcutaneous tissue, based upon the real-time CGM glucose measurement (mg/dL), direction of change, and rate of change. An additional nursing assessment will be made at 9 pm and 2 am using the real-time CGM glucose measurement to determine if the glucose concentration is greater than 300 mg/dL. The orthopedic nurses will use the displayed real-time CGM trend data (absolute glucose measurement, direction of change, and rate of change) to modify the correction dose of rapid acting insulin according to the TJUH Correction Dose Insulin (CDI) Protocol. The orthopedic floor nurse will assess the real-time displayed CGM trend data before breakfast, 2-hours after breakfast, before lunch, 2-hours after lunch, before dinner, 2-hours after dinner, at 9:00 pm and at 2:00 am (8 nursing assessments per day). Pre-meal Assessment: The orthopedic surgery floor nurse will make an assessment using the real-time CGM absolute glucose value (mg/dL) and the CGM direction/rate-of-change glucose trend data (↑-↓ arrows) prior to each meal to determine the correction dose of rapid acting insulin. The correction dose of rapid acting insulin (lispro) will be added or subtracted from the dose of insulin routinely ordered by the orthopedic surgeon to cover the meal. Several examples of pre-meal assessments and insulin dose calculations are provided below. Post-meal Assessments: The orthopedic floor nurse will utilize the displayed real-time CGM trend data to determine the correction dose of insulin to be injected 2-hours after each meal. (2-hours after the pre-meal insulin injection). This additional correction insulin dose of insulin may be given two hours after the meal if the GCM data displays hyperglycemia (CGM \> 150 mg/dL) with an increasing rate of change. Several examples of post-meal assessments and insulin dose calculations are provided below. An additional correction insulin dose of insulin may be given at 9 pm and/or 2 am only if the CGM displays a glucose concentration \> 300 mg/dL. CGM Alarms For Hypoglycemia: The orthopedic floor nurse will have the study subject consume 15 grams of oral glucose (tablets or gel) if the real-time CGM alarms for predicted hypoglycemia (CGM \< 55 mg/dl predicted 20 minutes in the future) or hypoglycemia below a threshold value (CGM \< 80 mg/dL). The orthopedic nurse will make an assessment using the real-time CGM data display approximately 15-20 minutes after the hypoglycemia treatment to confirm the CGM glucose measurements remains \> 80 mg/dL. An addition 15 grams of oral glucose will be given approximately 15-20 minutes after the initial oral glucose treatment if the CGM measurement does not remain \> 80 mg/dL. If unable to consume oral glucose or altered consciousness, the nurse will treat the hypoglycemia with an intravenous injection of 25 ml dextrose 50% (12.5 grams) or 1 mg glucagon IM/SC based upon the standard-of-care methods used at TJUH to treat hypoglycemia (Table # 3). Finger Stick BG Measurements: Study subjects in the Investigational Group will have finger stick capillary blood glucose (BG) measurements performed using a hospital meter/test strips before breakfast, before lunch, before dinner, at 9:00pm, and at 2:00am (5 per day). The nurse may obtain additional finger stick BG measurements at any time to detect or confirm accuracy of the CGM measure and to confirm hypoglycemia/hyperglycemia. All finger stick BG measurement will be recorded in the electronic medical record (EPIC). Hospital Discharge: Study subjects will wear the 3 blinded CGM at home during their daily activities for a maximum of 10 days (from date of insertion). Subjects will be instructed to self-remove the blinded 3 CGM after a maximum of 10 days of wear-time, and place the blinded CGM in a stamped addressed envelope. Subjects will be instructed to mail the addressed envelope to TJU research personnel. Research Personnel will contact each study subject by telephone after CGM removal to confirm no serious device related adverse events. Orthopedic Surgery Medical Record Review: The hospital and orthopedic surgeon's medical record of each patient will be evaluated following the post-operative visit at the Department of Orthopedic Surgery's Rothman Institute. The medical record review will determine whether there were any adverse events or complications related to the hip/knee surgical procedure (surgical site infection, urinary tract infection, lower respiratory tract infection, myocardial infarction, deep vein thrombosis, pulmonary embolism, stroke, kidney failure, death, emergency room visit, and hospital re-admission).