Point-of-care (POC) glucose monitor is often used in hospitalized patients and is a valuable tool for deciding therapy in cases of abnormal glucose levels. However, it has limitations in conditions such as Raynaud’s phenomenon, monoclonal gammopathies, leukemia, and peripheral hypoperfusion. (1). Here, we describe a patient whose results of POC were unreliable, resulting in overtreatment of hypoglycemia. A 76-year-old African American woman with a history of scleroderma, diagnosed after a skin biopsy was admitted for hypoglycemia. Her symptoms of scleroderma include significant esophageal dysmotility but minimal skin involvement. She was initially treated with hydroxychloroquine which improved her symptoms but discontinued after a few months when it was suspected of causing hypoglycemia. While in rehab for a pubic rami fracture, she was diagnosed with recurrent hypoglycemia and started on Diazoxide. She was transferred to our facility because of persistent hypoglycemia requiring a dextrose drip and multiple infusions of 25g of 50 percent glucose. Despite being on Diazoxide, she had remained hypoglycemic with POC blood glucose readings as low as 30-50 mg/dl. A 72-hour fast was initiated, and Diazoxide was stopped. A fasting blood sugar test off Diazoxide for four days showed discordant fingerstick and plasma blood glucose levels of 59 mg/dl and 138 mg/dl, respectively. Plasma levels of insulin (<1 uU/ml), proinsulin (1.6 pmol/L), and C-peptide (0.4 ng/ml) were appropriately low in the setting of a fast. Records from inpatient rehab showed three episodes of hypoglycemia measured by POC. However, all values of low glucose seen on POC testing had a corresponding normal plasma glucose level. We could not find an episode of hypoglycemia confirmed by a plasma glucose test. She had insulin and C-peptide levels checked several times, and none of them elevated in the setting of normal plasma glucose. Insulin levels ranged from 2.2-13.0 [2.6-24.9 uU/ml], and C-peptide levels ranged from 1.2-2.0 [1.1-4.4 ng/ml], all within normal limits. Additional workup for other etiologies of hypoglycemia showed normal cortisol stimulation test, low IGF-2, and elevated chromogranin A. This patient was diagnosed with hypoglycemia based on POC glucose levels only. If plasma glucose had been used to confirm hypoglycemia this patient could have been diagnosed with pseudohypoglycemia sooner and avoided unnecessary testing and medications. She was also treated for hypoglycemia with dextrose-containing intravenous fluids and Diazoxide. Treating hypoglycemia reflexively is understandable, given the significant risks of withholding treatment. However, Diazoxide likely contributed to this patient’s symptoms of nausea and anorexia already caused by her esophageal dysmotility. It is likely that the minimal skin involvement of her disease minimized the suspicion of psudohypoglycemia. It is worth mentioning that hydroxychloroquine, used for scleroderma management, was stopped for the presumed hypoglycemia. This case highlights the importance of confirming POC testing and evaluating for potential causes of discordant results.
Read full abstract