Abstract
We felt it was important to highlight a problem that we encountered with a bedside glucometer and a patient with end stage renal failure on continuous ambulatory peritoneal dialysis. Vast discrepancies between true venous glucose levels and those estimated by certain glucometers in this situation can be problematic and dangerous. Not all glucometers are the same: they differ in their ability to estimate glucose levels. The Advantage (Roche) and Accutrend Sensor (Boehringer MannHeim) glucometers both use the glucose dehydrogenase enzyme reaction to estimate glucose levels. Certain dialysates, including Extraneal (Icodextrin), contain significant amounts of long-chain non-glucose-reducing substances. Such substances are metabolized by glucose dehydrogenase and lead to falsely elevated glucose readings. Precision Q.I.D (Medisense) and Glucocard (Merarini Diagnostics) glucometers analyse glucose levels using a different enzyme, glucose oxidase. Long-chain non-glucose-reducing substances do not interfere with glucose calculations based on this enzyme.
We encountered a young patient on continuous ambulatory peritoneal dialysis in our hospital who was using a Roche glucometer at home. Compliance with treatment seemed poor. The patient had been admitted with severe hypertension and hyperglycaemia. His Extraneal dialysate solutions were adjusted appropriately and a insulin sliding scale was commenced. At this time he had a hypoglycaemic coma that was not diagnosed until after the venous glucose had returned from the laboratory. During this delay he had failed to respond to naloxone and also had a brain CT, which was normal. The Consultant Physician, Nephrology Team and myself were all perplexed by his sudden deterioration. The glucometer read 11 mmol/L. The nursing staff reported that the glucometer had been calibrated by our Biochemistry department and it was the same type used by the patient at home, and that they had also repeated the glucometer reading.
Discrepancies have been reported a couple of times in the medical literature, however it seems that a knowledge of this problem is not widespread. This near-miss led to a policy change in our hospital, and the episode emphases the importance of not taking a glucometer reading on face value. Dehydration, vomiting, diarrhoea, diuretics, peripheral vascular disease, hyperuricaemia, significant paracetamol overdose and severe jaundice can also lead to unreliable results.
Footnotes