Background
Hypoglycaemia remains a prevalent and dangerous complication of diabetes management in hospitalised dialysis patients, contributing to increased morbidity, mortality, and healthcare burden. This study evaluates the diagnostic performance, clinical applicability, and user acceptability of continuous glucose monitoring (CGM) in this vulnerable inpatient population.
Methods
A prospective pilot study was conducted involving 30 adult patients with diabetes mellitus undergoing either haemodialysis or peritoneal dialysis in an inpatient renal ward. Participants were monitored with the Dexcom G6 CGM system in parallel with routine capillary blood glucose (CBG) testing. Hypoglycaemic detection was assessed via sensitivity, specificity, positive predictive values (PPV), negative predictive values (NPV), and ROC analysis. Clinical concordance was evaluated using Bland-Altman plots, linear regression, mean absolute relative difference (MARD) and Parkes (Consensus) Error Grid analysis. Nurse and patient feedback were captured via validated questionnaires.
Results
CGM demonstrated a sensitivity of 68.8% and specificity of 97.3% for hypoglycaemia detection, with a PPV of 42.3% and a NPV of 99.1%. Subgroup analysis revealed similar trends across dialysis modalities, with slightly higher sensitivity in peritoneal dialysis patients. ROC curve analysis showed high diagnostic accuracy (area under the curve >0.95), while Bland-Altman and regression analyses confirmed strong agreement with CBG. The estimated MARD was 11.1%. Parkes (Consensus) Error Grid analysis also revealed that 98.6% (570 of 578) of CGM readings in clinically acceptable Zones A and B. Both patient satisfaction and nursing acceptance were high, supporting real-world feasibility.
Conclusions
CGM is a safe, reliable, and well-accepted adjunct for detecting hypoglycaemia in hospitalised dialysis patients. Its high specificity and NPV make it particularly valuable for ruling out hypoglycaemia. Broader implementation may enhance safety and reduce nursing burden. Further research with larger cohorts is warranted.
扫码关注我们
求助内容:
应助结果提醒方式:
