Continuous Glucose Monitoring in Patients With Diabetes on Peritoneal Dialysis

Background: Patients with chronic kidney disease (CKD) caused by diabetes mellitus (DM) experience metabolic alterations characterized by elevated blood glucose levels and associated cardiovascular complications, which increase morbidity and mortality. Therefore, adequate metabolic control is necessary to reduce the incidence of these complications. This task is challenging without icodextrin because insulin adjustment is complicated by the additional glucose supplied through dialysis bags and absorbed through the peritoneum. Under this premise, continuous glucose monitoring is important for assessing patients' metabolic status. This approach will provide clinically relevant data to improve care, minimize healthcare system expenses, and support decision-making regarding adjustments to dialysis treatment. Objective: To use continuous glucose monitoring to determine whether the type, dose, route of administration, and timing of insulin application are associated with glucose patterns identified through continuous glucose monitoring, including the magnitude and duration of periods of hyperglycemia and hypoglycemia during 24-hour periods of interstitial glucose monitoring. Material and Methods: This is a cross-sectional, non-interventional study involving adult patients with type 2 diabetes mellitus receiving automated peritoneal dialysis who exhibit high or high-average peritoneal transport characteristics. The inclusion criteria are patients older than 40 years, of any sex, diagnosed with diabetic nephropathy, who are insulin-dependent for metabolic control and have received peritoneal dialysis treatment for at least three months. The study will continuously evaluate patients' glycemic control, with automatic scanning and data recording every six hours during automated peritoneal dialysis infusion. According to the sample size calculation, 110 patients are required. The Guardian 3 sensor will be placed subcutaneously in the upper posterior region of the patient's non-dominant arm using the One-press serter. This is a minimally invasive procedure that does not require surgical protocols. The sensor will be connected to the Guardian Connect transmitter for continuous communication with the Guardian Connect application. The sensor has an approximate lifespan of seven days, corresponding to the duration of the enzymatic reaction that allows adequate measurement. Data will be transmitted every five minutes, 24 hours a day, for seven consecutive days. At the end of this period, participants will be scheduled for placement of a second sensor to complete 14 days of follow-up. On day seven, participants will undergo removal of the first sensor and placement of a new sensor to complete the 14-day follow-up period. On day 14, the total 24-hour peritoneal dialysis drainage volume will be collected to measure glucose, urea, and creatinine concentrations. Peritoneal glucose absorption, the dialysate-to-plasma creatinine ratio (D/P creatinine), and Kt/V will be calculated. Dietary information will be collected to calculate caloric intake and meal timing. This information, together with subcutaneous insulin administration, will be recorded in the Guardian Connect application. Statistical Analysis: The databases will be audited monthly through random sampling of blocks comprising 5% of their content. Semiannual reports will be prepared to monitor the records collected and the outcomes obtained up to each reporting date. These reports will contain basic descriptive information, including measures of central tendency and dispersion, according to the characteristics of the variables. Patients will be classified according to the percentage of time that glucose measurements remain within the predefined range of 70-180 mg/dL. Patients who spend at least 70% of the time within this range will be classified as having "Adequate" glycemic control, whereas those who do not reach this target will be classified as having "Not Adequate" glycemic control. Results will be reported using measures of central tendency and dispersion appropriate to the characteristics of the variables. Depending on the type of variable, the chi-square test, Student's t-test, or Mann-Whitney U test will be used to detect differences between the adequate and inadequate glycemic control groups. Logistic regression analysis will be used to identify variables significantly associated with glycemic control within the recommended ranges, as assessed by continuous glucose monitoring. In the first stage, each independent variable will be analyzed separately. In the second stage, multivariable analysis will be performed, considering insulin type, route of administration, dose, and administration schedules. Potential confounding variables, including obesity, treatment adherence, diet, and physical activity prescribed by the treating physician, will also be included.