Unexplained severe hypoglycaemia in hospital: a difficult diagnostic challenge

We report a case of severe unexplained hypoglycaemia occurring in a surgical patient during a routine hospital admission. This case caused significant diagnostic problems and highlights a number of important learning points. A 66-year-old male inpatient on a surgical ward was found unrousable by a member of staff after a visit from a friend. Until this episode he had been making an uneventful recovery from a complicated open cholecystectomy 13 days earlier. He had Type 2 diabetes that was treated with diet alone, and during the admission his blood sugars were in the range of 4.7–7.7 mmol l−1. He was known to have chronic renal insufficiency secondary to hypertension and had had a renal transplant 5 years earlier. A recent serum creatinine concentration was 372 µmol l−1 (normal range 70–150). His blood glucose concentration was found to be 1.7 mmol l−1. He was treated initially with 50 ml of 50% intravenous glucose, but became symptomatically hypoglycaemic again within 1 h. He was normotensive throughout. The endocrinology team was contacted and they recommended that insulin, pro insulin, C-peptide and glucose concentrations should be measured and the presence of a sulphonylurea excluded by analysis. They also advised that all current infusions should be changed. It soon became apparent that the patient required repeated injections of 50% glucose to maintain euglycaemia. A continuous infusion of 20% glucose was started after he had received 300 ml of 50% glucose. His blood glucose concentration stabilized only after treatment with 650 g of i.v. glucose over 41 h. Results showed raised insulin (364 pmol l−1) and C-peptide (7.67 nmol l−1) concentrations in the presence of a low blood glucose concentration (1.1 mmol l−1). The sulphonylurea screen was strongly positive. Using selective and sensitive high-performance liquid chromatogrphy mass spectrometry, the Regional Laboratory for Toxicology, Birmingham, confirmed the presence of gliclazide. The plasma concentration 20 h after the initial hypoglycaemic event was 3.44 mg l−1. As our patient weighed 80 kg and the apparent volume of distribution of gliclazide is 0.28 l kg−1, there was approximately 77 mg gliclazide in his body at the time the sample was taken. Although it is difficult to assess accurately the half-life of gliclazide in renal impairment, extrapolation of this plasma concentration using a half-life of approximately 10 h [1] showed that the presenting hypoglycaemia could have been the result of a single oral dose of approximately 320 mg gliclazide given at 14.00 h. Although the cause of the hypoglycaemia was now clear, the circumstances leading to the ingestion of the gliclazide were not. As the patient had not been prescribed gliclazide during this admission, the possible causes were considered to be a medication dispensing error, a medication administration error, deliberate self-harm or malicious intent. This prompted a thorough investigation by a multidisciplinary team, which included a clinical pharmacologist, hospital pharmacist and senior members of the medical team. Initially it was felt the most likely cause was an administration error; however, gliclazide was not a stock item on the ward. When the ward drug cupboard was examined a gliclazide packet was found next to furosemide (Figure 1). After significant investigation it became clear that 1 month earlier a ward patient had received gliclazide. This should have been kept in the Patient Own Drugs locker beside the bed and then removed on discharge, but contrary to hospital policy it had been placed in the ward drug cupboard. Figure 1 Gliclazide and furosemide in the ward drug cupboard Our patient was due to receive furosemide 160 mg (4 × 40-mg tablets) at 14.00 h. As well as the alphabetical proximity of the two drugs, the packaging of gliclazide (Figure 1) and size of the tablet (Figure 2) are very similar and could easily be mistaken. It seems likely that the four tablets that our patient received were four tablets of gliclazide 80 mg (320 mg) rather than four tablets of furosemide 40 mg. The nurse involved with the administration of the medicines was unaware that an error had occurred. The patient made a full recovery, with no long-term consequences. A thorough investigation has been undertaken and corrective action has been implemented. Figure 2 Similarities between the tablet size of gliclazide and furosemide This case raises several important learning points. First, it illustrates the diagnostic challenges faced in identifying the cause of unexplained hypoglycaemia and the need to make prompt and appropriate investigations if the situation arises. It also highlights that involving a multidisciplinary team, including help from the National Poisons Information Service and the Regional Laboratory for Toxicology, can be extremely valuable. Third, it demonstrates the intractable hypoglycaemia that can ensue following administration of a sulphonylurea to patients with renal impairment. Finally, this case is a sobering reminder of how easily medication errors can occur. Up to 33% of all medication errors are to due to packaging and/or labelling confusion [2]. Approximately 9% of patients staying in hospital will experience medication-related harm – many of these incidents are preventable [3].