Alarming blood-sugar levels seen among dialysis patients

The largest study to date of real-time blood-sugar patterns among kidney-dialysis patients reveals a surprising disconnect between what standard medical tests show and what's happening with patients' glucose levels throughout the day. 

Led by University of Washington School of Medicine researchers, the study followed 420 dialysis patients wearing continuous glucose monitors. These small sensors, typically attached to the abdomen or upper arm, track users’ blood sugar every five minutes for 10 days. The monitors’ readings exposed an unexpected volume of high and low blood-sugar episodes not indicated by the standard clinic-based blood draws that most doctors rely on. 

“It was eye-opening because there clearly are abnormalities going on that we are not aware of. We presume high and low glucose values matter in this population, but we can’t say for sure. In any case, if I’m a patient on dialysis, I want to get one of these [monitors] and see what’s going on,” said Dr. Ian de Boer.  

He’s a kidney specialist and professor at the University of Washington School of Medicine, and led the investigation whose results were published in September in the Journal of the American Society of Nephrology. 

Among dialysis patients generally, de Boer said, a little over half have diabetes. The condition is typically monitored by a blood test measuring hemoglobin A1C (HbA1c), which patients get every three to six months. That value indicates patients’ average blood-sugar level and whether their medication dose should be changed.   

In the study, among patients with diabetes who were taking blood-sugar medications, only 22% met contemporary treatment targets — even as they showed normal-range results on the HbA1c tests. Equally striking: Patients with untreated diabetes maintained consistently elevated blood-sugar readings despite having normal-range HbA1c levels. 

The researchers also documented 714 episodes of low blood sugar, including 179 characterized as “severe.” Hypoglycemia was most common in nondiabetic dialysis patients receiving hemodialysis. 

“We expected glycemic excursions out of range, but the outlier glucose levels we saw with continuous monitoring, both high and low, were much farther out of range than an HBA1c would have predicted,” de Boer added. 

The findings suggest that continuous glucose monitors are appropriate and potentially beneficial for people on dialysis, especially those being treated for diabetes, he said.  

De Boer clarified, though, that the results offer no evidence about what specific measures a clinician should take with dialysis patients whose continuous glucose monitor levels are significantly unaligned with their HbA1c values, or how effective those measures might be.  

“We don't know for sure whether controlling glucose with this population is going to improve wellbeing. More research is needed to address that question,” he said.  

“There might be multiple benefits of better glucose control, for example it might prevent heart disease or infections or make someone a healthier candidate for kidney transplant. But we don’t yet know how glucose control stacks up in the list of priorities for patients on dialysis, who already have many demands on their health.” 

For details about UW Medicine, please visit https://uwmedicine.org/about.