The study required more participants than anticipated for statistical significance.

This study is for people with diabetes who have either healthy kidneys or who have end stage renal disease (ESRD) and are on hemodialysis that are having surgery. This research will help increase our knowledge about how having diseased kidneys and being on hemodialysis changes how the body responds to insulin. In the time period directly before, during, and after surgery, maintenance of blood sugar within normal physiological range is essential. When patients come for surgery and have an elevated blood sugar, it is up to each individual physician to decide on the right amount of insulin to treat them. Currently, there are many different protocols across the country to treat elevated blood sugar; however, there is no one specific protocol to treat elevated blood sugar at SUNY Downstate Medical Center at this time. In this study, the investigators will evaluate the change in blood sugar following insulin administration to determine whether there is a significantly different response between ESRD and non-ESRD patients. The results of the study will help develop safer practice guidelines to patients with ESRD with an elevated blood sugar while they are having surgery.

The objective of our study is to compare perioperative changes in blood glucose levels following insulin administration (Novolin R human insulin, Novo Nordisk Inc. Plainsboro, NJ) for preoperative hyperglycemia (as defined by blood glucose over 150 mg/dL) in diabetic ESRD patients on hemodialysis vs. diabetic non-ESRD patients without CKD. The investigators hypothesize that diabetic ESRD patients on hemodialysis with preoperative hyperglycemia will demonstrate a more rapid decrease in blood glucose following insulin administration compared to diabetic non-ESRD patients without CKD.

The study will take place at SUNY Downstate Medical Center at the University Hospital of Brooklyn. Following enrollment of eligible patients, the following data will be collected: age, gender, height, weight, BMI, indicated procedure, medical record number, past medical history (including liver dysfunction and EtOH abuse), ASA, and initial blood glucose. All patient data associated with renal function and dialysis will be obtained and recorded for ESRD patients. Relevant kidney function data, such as GFR, will also be obtained, if available. The surgical procedures will include but are not limited to the following elective surgeries: major abdominal surgery, major GU/GYN surgeries, vascular surgery, renal transplantation, neurosurgery, vascular and orthopedic surgery . The investigators expect patients in both groups to undergo similar surgical procedures. All medications prescribed to the patient will be noted, as will the type of diabetes the patient has, type 1 or type 2. If the patient takes insulin or oral diabetic medications, the time and amount of their last dose will be recorded. The interval between the last dialysis and insulin testing will not be similar and the extent of uremia will not be equivalent among participants.

All consented patients with hyperglycemia (blood glucose > 150 mg/dL) will be administered an IV bolus of Novolin R human insulin based on a sliding scale according to the table below. The anesthesiologist that is taking care of the patient for that specific surgery will administer the insulin based on the sliding scale, will measure and record blood glucose at 30 minute intervals, and will be responsible to adjust treatment as per the study's sliding scale. Prior to initiating the protocol, the particular anesthesiologist that is responsible for the case will be explained exactly how to follow the protocol. He/she will also be given a supplemental form with written instructions. He/she will also be given contact information for the principle investigator if they should have any questions or issues. The investigators developed a dosing protocol based on guidelines in previous studies, determined as safe and effective for perioperative hyperglycemic patients. The treatment protocol will be administered as follows:

1. Patients with preoperative blood glucose > 150 and < 200 mg/dL will be given a bolus of 2 insulin units.
2. Patients with preoperative blood glucose > 200 and < 250 mg/dL will be given a bolus of 3 insulin units.
3. Patients with preoperative blood glucose > 250 and < 350 mg/dL will be given a bolus of 4 insulin units.

Point-of-care blood glucose testing will be measured at 30-minute intervals using a glucometer (Nova Biomedical, Waltham, MA. USA), and the investigators will adjust treatment to maintain a safe glucose window. Plasma insulin levels will not be tested. ESRD patients will be administered normal saline and non-ESRD patients will be administered lactate ringers during study protocol. Glucose readings collected at each interval will be recorded, as will changes in insulin administration. Treatment will be adjusted as per the sliding scale given above. For the purpose of the study and statistical analysis, blood glucose sampling will be performed 3 times, one as baseline, second after ½ hour after the first dose of IV insulin, and the second after 30 additional minutes. Descriptive statistics (mean, standard deviation) for each group at each time point will be used to determine the difference between the ESRD and non-ESRD groups in the pilot study. Any additional blood glucose estimations will be strictly for clinical reasons to treat hyperglycemia and ascertain that no hypoglycemia has occurred. For patient safety, the investigators will not give insulin for blood glucose levels less than 150 mg/dL, but will continue to monitor blood glucose at 30-minute intervals for 2 hours after the last dose of insulin and will not administer insulin if the blood glucose levels fall below 150 mg/dL. Regular insulin when given intravenously has a rapid onset and a short duration of action; intravenously the half- life of regular insulin is 0.5 to 1 hour and the peak effect is 0.8 to 2 hours. The investigators will therefore monitor blood glucose levels until the expected peak effect of 2 hours where they expect the lowest blood glucose levels. Hypoglycemia is highly unlikely in renal patients at discharge. It is customary for patients to stay in the PACU for 2 hours post-surgery. During the 2-hour window from the patient's last dose of insulin, the patients will be required to remain NPO except water.

If blood glucose drops below 100 mg/dL, a 25 cc bolus of 50% dextrose in water (Hospira Inc., Lake Forest, IL, USA) will be administered and blood glucose will be checked after 15 minutes. The investigators will give another dextrose bolus if necessary to restore blood glucose above 100 mg/dL. Once blood glucose has reached a level above 100 mg/dL, blood glucose will continue to be checked every 30 minutes for 2 hours after the last dose of insulin given.

Of note, this treatment protocol is based on standard practice and designed for patient safety. If the patient is discharged before 2 hours, the investigators will terminate the study at that point. The patients will not be at increased risk of harm as insulin peak effect occurs within 30 minutes. With study termination, the patient will return to their normal anti-hyperglycemic regimen. No additional patient participation will be required after this point. Blood samples collected for glucose readings will be discarded immediately after use. Data will be entered into a password protected Microsoft Excel data sheet.

This will be pilot study to compare the response to insulin in diabetic patients with ESRD vs. diabetic patients without renal disease. The investigators will generate insulin response curves for each patient, with blood glucose level as a function of time. Then they will compute average drop in glucose in the two groups. Using these data the investigators will be able to observe the obtained effect size, determine whether it is clinically significant, and if so, determine sample size that would be required to obtain statistical significance for a larger study analysis. Based on the outcome of the pilot study, they will determine the sample size and the analysis plan for the main study.

Inclusion criteria:

• Adult patients over 18 years old with diabetes (both sexes) presenting without any kidney disease or with ESRD on hemodialysis
• Undergoing major elective surgery procedures such as abdominal, pulmonary, GU/GYN, vascular, renal, neurologic, and orthopedic surgery
• Signed informed consent for the study.
• Patients with elevated preoperative blood glucose of a value > 150 mg/dL or < 350 mg/dL
• English-speaking
• Baseline chemistry with a potassium at or above 3.5

Exclusion criteria:

• Patients under the age of 18.
• Pregnant women. (Pregnant women have altered glucose metabolism; specific insulin regimen used for pregnant women)
• Patient does not give informed consent for the study.
• Patient does not have decisional capacity or a surrogate.
• Patient cannot understand or read English.
• Patients with a preoperative blood glucose < 150 mg/dL or >350 mg/dL
• Patients without diabetes
• Patients with kidney disease that are not on hemodialysis
• A history of frequent hypoglycemia in the month prior to surgery by history or a change in insulin dose for hypoglycemia in the month prior to surgery
• Patients taking steroids or patients that will receive steroids during their surgery
• Any patients who refuse to remain NPO except water in the two hour window after their last insulin dose
• Patients scheduled for cardiac, thoracic and emergency surgeries
• Patients who have taken or received insulin or oral hypoglycemic agents after 12 am on the day of surgery.
• Patients have not adhered to the NPO status.