302-OR: Causes of Blinded CGM Failure during Clinical Trials in Patients with T2D

Abstract

Blinded CGM is useful in clinical trials to preserve the validity of outcomes. However, lack of real-time feedback about performance of the system generates a significant number of failed sessions. We analyzed causes of such failed CGM sessions in 453 T2D subjects (mean age 55.2±9.6 years, 51.9% males) using the Medtronic iPro2 to obtain a 6-day baseline CGM during two randomized clinical trials of oral agents and insulin. At the end of the CGM session, non-calibrated data (ISIG - electric signals) were immediately reviewed by a third party to determine the completeness and quality of the dataset (at least 72 continuous hours of ISIG between 10-200 nA) and the need for a new CGM session. If data were not acceptable, clinical sites were promptly contacted to ascertain the conditions of the study subject, insertion technique and technical problems that could cause the failed session, and to recommend other approaches to CGM reinsertion. Subjects who failed the first CGM sessions were reinserted a second or third time when possible. Eighty-two of 453 subjects failed their first CGM insertion, 9 of 56 failed the second insertion, and one of four subjects failed all three insertions (cumulative success rate = 99.8%). The causes of failure were discernible in 38 (41.3%) of the 92 failed cases, including sensor detachment (19.6%), insufficient BG for calibration (17.4%), user error (3.3%), and intentional removal (1.1%). For the remaining 58.7% of cases, the non-calibrated ISIG patterns were analyzed and most likely causes found to be undetermined (25.0%), disconnection (20.7%), sensor failure (8.7%) and device failure (4.3%). Upon clinical review, most failed sessions of undetermined cause or probable disconnection were attributable to patient characteristics (morbid obesity, profuse sweating) or improper insertion technique. Conclusions: 1) Most identifiable causes of failed CGM can be potentially corrected by proper user/subject training, and 2) only a minimal portion of failed sessions are secondary to faulty devices. Disclosure S. Saldivar-Salazar: Speaker's Bureau; Self; Medtronic MiniMed, Inc. M.A. Testa: Stock/Shareholder; Spouse/Partner; GI Windows, Inc. Stock/Shareholder; Self; Phase V Technologies, Inc. J.F. Hayes: None. D.C. Simonson: Stock/Shareholder; Self; GI Windows, Inc. Stock/Shareholder; Spouse/Partner; Phase V Technologies, Inc.