Saudi Journal of Kidney Diseases and Transplantation

CASE REPORT
Year
: 2011  |  Volume : 22  |  Issue : 4  |  Page : 764--768

Severe hypoglycemia in peritoneal dialysis patients due to overestimation of blood glucose by the point-of-care glucometer


Hasan M Al-Dorzi1, Hythem Al-Sum1, Salem Alqurashi2, Saleh J Aljaser3, Yaseen M Arabi1,  
1 Intensive Care Department, King Saud Bin Abdulaziz University for Health Sciences, King Abdulaziz Medical City, Riyadh, Kingdom of Saudi Arabia
2 Department of Nephrology, King Saud Bin Abdulaziz University for Health Sciences, King Abdulaziz Medical City, Riyadh, Kingdom of Saudi Arabia
3 Department of Endocrinology, King Saud Bin Abdulaziz University for Health Sciences, King Abdulaziz Medical City, Riyadh, Kingdom of Saudi Arabia

Correspondence Address:
Yaseen M Arabi
FCCM Chairman - Intensive Care Department, Associate Professor - College of Medicine, King Saud Bin Abdulaziz, University for Health Sciences, King Abdulaziz Medical City, P.O. Box 22490, Riyadh, 11426
Kingdom of Saudi Arabia

Abstract

Although overestimation of blood glucose (BG) by certain glucometers in peritoneal dialysis (PD) patients has been reported, awareness of this problem by healthcare providers of multiple disciplines and different specialties is probably insufficient. This is a case series of four patients who had severe symptomatic hypoglycemia and normal BG by point-of-care glucometer at a tertiary care center from December 2007 to September 2008. We report four insulin-treated diabetic patients (age = 58.2 ± 16.2 years, 3 men and 1 woman) on PD, who had acute decrease in level of consciousness in the emergency department (n = 1) and the hospital ward (n = 3). While they had their symptoms, they all had normal BG measured by the Accuchek glucometer (7.1 ± 3.3 mmol/L); nonetheless, simultaneous or near-simultaneous laboratory-measured BG levels were very low (2.0 ± 1.3 mmol/L). The mean difference in BG was 5.8 mmol/L (12 simultaneous or near-simultaneous measurements). Three patients had icodextrin-based PD in the night before symptomatic hypoglycemia. The first two patients, whose treatment for hypoglycemia was delayed, remained comatose and died later. The latter two patients were promptly treated with intravenous dextrose and did not have any neurologic sequelae. One of them died later from multiple organ failure. Over-estimation of BG in peritoneal dialysis patients by certain point-of-care glucometers is a serious problem and can be fatal. Increased awareness of this problem for all healthcare providers and use of appropriate glucometers are required.



How to cite this article:
Al-Dorzi HM, Al-Sum H, Alqurashi S, Aljaser SJ, Arabi YM. Severe hypoglycemia in peritoneal dialysis patients due to overestimation of blood glucose by the point-of-care glucometer.Saudi J Kidney Dis Transpl 2011;22:764-768


How to cite this URL:
Al-Dorzi HM, Al-Sum H, Alqurashi S, Aljaser SJ, Arabi YM. Severe hypoglycemia in peritoneal dialysis patients due to overestimation of blood glucose by the point-of-care glucometer. Saudi J Kidney Dis Transpl [serial online] 2011 [cited 2019 Dec 8 ];22:764-768
Available from: http://www.sjkdt.org/text.asp?2011/22/4/764/82686


Full Text

 Introduction



Peritoneal dialysis (PD) has been used as a modality of renal replacement therapy for patients with end-stage renal disease (ESRD) for more than three decades. [1] It is based on instilling dialyzate solutions in the peritoneal space to enhance fluid exchange. The commonly used dialyzate solutions contain osmotic agents such as glucose and icodextrin. [2] Although there have been increasing reports and warnings of hypoglycemia occurring in PD patients using icodextrin-containing dialyzate, we believe that this problem is still under-recognized by many healthcare providers. We report four cases of severe symptomatic hypoglycemia in PD patients with significant discrepancy in blood glucose (BG) between the point-of-care glucometer and the laboratory.

 Case Reports



Between late December 2007 and September 2008, our critical care team managed four patients on PD who developed severe symptomatic hypoglycemia. The institutional review board of our hospital approved the review of these cases, whose characteristics and outcomes are summarized in [Table 1]. The following is a description of the four encounters.{Table 1}

Patient 1 was a 65-year-old diabetic man who was switched from hemodialyis to PD 10 days before presenting to the emergency department with unresponsiveness after taking insulin at home. He was afebrile, normotensive and his BG by the Accuchek glucometer was 7.8 mmol/L.

He was intubated, mechanically ventilated and started on broad-spectrum antibiotics for suspected sepsis. Brain computed tomography did not show any acute neurologic event. Laboratory-based BG, taken near-simultaneously with the glucometer measurement, was 0.6 mmol/L. He was given 50% dextrose (D50) intravenously, almost 90 minutes after presentation. Within the next 24 hours, laboratory-based measurements continued to show hypoglycemia (1.4 and 2.1 mmol/L). Near-simultaneous BG measurements by the Accuchek glucometer were within normal (7.6 and 8.7 mmol/L, respectively). Upon transfer to the intensive care unit (ICU), he was given intravenous dextrose and nasogastric tube feeding. Simultaneous BG measurements were ordered and continued to show discrepancy for almost 24 hours, coinciding with changing the peritoneal dialyzate fluid. The patient continued to have deep coma and died several weeks later.

Patient 2 was a 56-year-old diabetic man admitted to the hospital because of gangrene of multiple fingers. After surgical debridement and during his stay in the ward, he developed severe hypoglycemia (1.6 and 1.5 mmol/L) on laboratory-measured BG on two consecutive mornings. This was not treated as he had normal near-simultaneous BG on the Accuchek glucometer (7.9 and 7.5 mmol/L, respectively). Later, he developed seizures and respiratory failure for which he required antiepileptic management, intubation and ICU admission. Hypoglycemia was not recognized during the seizures as the point-of-care glucometer showed normal BG. He suffered severe persistent encephalopathy and died several weeks later.

Patient 3 was a 75-year-old diabetic lady admitted to the hospital for hip fracture. In the ward, she had sudden unresponsiveness around three hours after taking subcutaneous insulin as per treatment regimen. At that time, the Accuchek glucometer showed a BG reading of 5.8 mmol/L. However, BG checked in the laboratory an hour earlier was 2.0 mmol/L. The hospital's critical care response team immediately gave her D50 intravenously after which her mental status quickly returned to normal. She was discharged home later.

Patient 4 was a 37-year-old man who had endocarditis and septic foot gangrene requiring amputation. Postoperatively, he developed decreased level of consciousness and seizures. Simultaneous BG levels by the Accuchek glucometer and laboratory measurement were 7.2 and 2.3 mmol/L, respectively. When given D50 intravenously, he had immediate improvement in mental status. This patient was on Dianeal-based PD and he received 6 units of regular insulin two hours before his symptomatic hypoglycemia for BG of 13.7 mmol/L by the Accuchek glucometer. The patient required admission to ICU and later died from multi-organ failure.

In summary, the four patients had normal BG by the Accuchek glucometer (7.07 ± 3.26 mmol/L) while they were clinically symptomatic for hypoglycemia. [Figure 1] shows the simultaneous or near-simultaneous measurements of BG by glucometer and laboratory in the four patients (12 measurements). The mean BG by glucometer was 7.8 ± 2.8 mmol/L and by laboratory was 2.0 ± 1.3 mmol/L. The mean difference was 5.8 mmol/L.{Figure 1}

 Discussion



We reported here four cases of severe hypo-glycemia that was not detected by the point-of-care glucometer. All patients were insulin treated diabetics on PD for ESRD. Extraneal (Baxter Healthcare, Chicago, IL, USA) was used as a dialyzate for PD before the occurrence of hypoglycemia in three patients. Interestingly, the fourth patient was not on Extraneal. Hypoglycemia contributed to death in two of these patients as it was not promptly and adequately treated due to overestimation of BG by the glucometer.

Our hospital uses Accuchek Inform glucometer (Roche Diagnostics, Mannheim, Germany) for the point-of-care BG measurements and the glucose hexokinase assay for laboratory-based glucose measurements. Many PD patients get alternating Dianeal (Baxter Healthcare, Chicago, IL, USA) and Extraneal during hospital stay. Dianeal contains anhydrous glucose, which when absorbed to systemic circulation causes hyperglycemia. Extraneal contains 7.5% icodextrin, a cornstarch-derived glucose polymer. After entering the peritoneal space, it is absorbed into systemic circulation and metabolized by plasma amylases into oligosaccharides such as maltose, maltotriose, and maltotetraose. [3] These metabolites accumulate in the circulation due to the absence of circulating maltase and remain in the circulation for a full seven days after the last dwell. [4] This leads to overestimation of BG level by glucometers whose test strips use glucose dehydrogenase enzyme with coenzyme pyrroloquinolinequinone (GDH-PQQ) system to measure BG as this system does not differentiate between glucose and other mono-saccharides. [5] Many point-of-care glucometers utilize this system, including the Accuchek Inform. Management difficulty occurs when using Extraneal alternating with Dianeal, which usually causes hyperglycemia, in insulin requiring patients.

The problem of spuriously elevated BG in PD patients using icodextrin was first described in 1998. [6] Since then, several cases have been reported of clinical hypoglycemia occurring in patients using icodextrin-containing dialyzate due to overestimation of BG. [7],[8],[9],[10] Most cases were not serious and patients recovered after treatment with dextrose. However, some patients had significant brain injury and death. [11] There have been increasing warnings by regulatory agencies in Australia, Canada, the United Kingdom and USA about this problem. Moreover, the manufacturers of glucometers utilizing GDH-PQQ system recommend against their use in PD patients using icodextrin-containing dialyzates.

Overestimation of BG can also occur when patients receive maltose-containing solutions such as Octagam (intravenous immunoglobulins) and HepaGam B (Hepatitis B immunoglobulin) or D-Xylose to evaluate malabsorption syndromes. Other substances or conditions reported to interfere with point-of-care glucometer measurements include anemia (hematocrit <20% or >55%), acetaminophen (>8 mg/dL), hyperuricemia (>10 mg/dL) and hyper-bilirubinemia (>20 mg/dL). [12] The fourth patient in our case series had hyperuricemia and hyperbilirubinemia, but not at the levels reported to interfere with glucometer measurement of BG. Improper BG measurement (e.g. inadequate blood sample) can give false reading. However, other unknown factors cannot be ruled out. Moreover, inaccuracy of glucometers has been described in critically ill patients and usually manifests as overestimation of BG. [13],[14] The degree of overestimation occurs more when there is actual hypoglycemia. [14]

To overcome the problem of BG overestimation for PD patients, multiple interventions were done in our hospital. First, physicians and nurses taking care of PD patients were educated about this problem early on and Medical Services Department was notified of the events. Second, an article was written and published in the hospital patient safety newsletter. The article described the events, the mechanisms of BG overestimation and the suggested solutions. Third, nursing staff were instructed not to use the Accuchek glucometer for PD patients. Fourth, a glucometer that uses a system other than the GDH-PQQ for BG measurement was brought into the hospital for use in PD patients. We have not encountered this problem again till date.

In conclusion, overestimation of BG in PD patients by certain point-of-care glucometers is a serious problem and is probably under-recognized. It can lead to significant morbidity and even death. The solution of this problem starts by its recognition. In addition, glucometers that utilize the GDH-PQQ technology should not be used for patients being treated with Extraneal or other compounds that contain mono-or oligosaccharides which may interfere with glucose measurements. Other glucometers that utilize different technologies, like glucose oxidase-based or glucose hexokinase-based tests, to measure BG are available and should be recommended for such patients.

References

1Krediet RT. 30 years of peritoneal dialysis development: the past and the future. Perit Dial Int 2007;27 Suppl 2:S35-41.
2Mistry CD, Gokal R. The use of glucose polymer (icodextrin) in peritoneal dialysis: an overview. Perit Dial Int 1994;14 Suppl 3: S158-61.
3Gokal R, Moberly J, Lindholm B, Mujais S. Metabolic and laboratory effects of icodextrin. Kidney Int Suppl 2002;81:S62-71.
4Hoftman N. Interference between Extraneal peritoneal dialysis and the Accu-Chek blood glucose monitor. Anesthesiology 2005;102(4): 871.
5Riley SG, Chess J, Donovan KL, Williams JD. Spurious hyperglycaemia and icodextrin in peritoneal dialysis fluid. BMJ 2003;327(7415): 608-9.
6Wens R, Taminne M, Devriendt J, et al. A previously undescribed side effect of icodextrin: overestimation of glycemia by glucose analyzer. Perit Dial Int 1998;18(6):603-9.
7Mehmet S, Quan G, Thomas S, Goldsmith D. Important causes of hypoglycaemia in patients with diabetes on peritoneal dialysis. Diabet Med 2001;18(8):679-82.
8Disse E, Thivolet C. Hypoglycemic coma in a diabetic patient on peritoneal dialysis due to interference of icodextrin metabolites with capillary blood glucose measurements. Diabetes Care 2004;27(9):2279.
9Charles RA, Goh SY. Occult hypoglycemia in a diabetic patient on peritoneal dialysis. Eur J Emerg Med 2005;12(6):322-3.
10Kroll HR, Maher TR. Significant hypoglycemia secondary to icodextrin peritoneal dialysate in a diabetic patient. Anesth Analg. 2007;104(6): 1473-4.
11Schleis TG. Interference of maltose, icodextrin, galactose, or xylose with some blood glucose monitoring systems. Pharmacotherapy 2007;27 (9):1313-21.
12Eastham JH, Mason D, Barnes DL, Kollins J. Prevalence of interfering substances with point-of-care glucose testing in a community hospital. Am J Health Syst Pharm 2009;66(2): 167-70.
13Kanji S, Buffie J, Hutton B, et al. Reliability of point-of-care testing for glucose measurement in critically ill adults. Crit Care Med 2005;33 (12):2778-85.
14Hoedemaekers CW, Klein Gunnewiek JM, Prinsen MA, et al. Accuracy of bedside glucose measurement from three glucometers in critically ill patients. Crit Care Med 2008; 36(11): 3062-6.