Background: Glucose monitoring among critically-ill patients has been a topic of attention for long time to prevent both hyperglycemia and hypoglycemia, which is important in an intensive care unit. Use of point of care blood glucose meters within some critically ill patient populations has resulted in varying degrees of confusion about off-label use and potential discrepancies in results. The aim of this study was to compare the accuracy of point of care capillary and venous/arterial samples verse venous/arterial samples by laboratory testing in critically ill both shocked and non-shocked patients.

Methodology: This is prospective case control study that was conducted among 268 critical ill patients among whom, capillary, Venous and arterial Blood Sample during routine care which is requested by treating physician will be done by the same bedside nurse. The study was conducted in King Fahd Miliary Medical Complex Dhahran Saudi Arabia.

Results: In this study, we were able to collect data for 268 patients. The results showed no significant difference between POCT and lap results of both of venous and central blood (P=0.389 and 0.208) where POCT showed slightly higher results with venous glucose concentration of 10.18 and 10.05 (POCT and lab tests respectively) and 9.18 and 9.54 in POCT and lab tests respectively. Moreover, the mean difference between POC and lab analysis of venous, arterial and central glucose were 1.03, -1.8 and 0.7 mmol.

Conclusion: We did not find a significant difference between results of POCT and regular laboratory analysis of glucose concentrations in critically ill patients except for arterial blood glucose. Using of POCT is slightly accurate with no difference between shocked and non-shocked patients compared to lab blood analysis.

One of the most important aspects of the critical care is to maintain normoglycemia in patients with established diabetes or stress-induced hyperglycemia. Hyperglycemia has been found to have a significant detrimental effect in those patients.^1,2 Appropriate management of hyperglycemia had been found to have a significant impact on reducing mortality and admission duration as well as preventing of acute renal injury and providing faster weaning from mechanical ventilation.^1,3,4 Moreover, hypoglycemia has its associated side effects⁵ and has been found to be an independent predictor of mortality in several studies.^6,7 

Furthermore, several organizations have determined their own target blood glucose levels for ICU patients^8,9 and to achieve these goals, many hospitals use point-of-care (POC) glucose meters in order to monitor glycemic status. The using of POC glucose meters provides the advantage of portability, ease of use as well as readily available results which is significantly important in those patients who required tight glycemic control where waiting for central laboratory results make the rapid adjustments and management of glucose level in therapy impractical. Most of the used meters have found to be accurate in stable outpatients however, among critically ill patients, results showed that there are various confounding factors that could interfere with POC glucometer as for example, hematocrit, oxygenation, acid-base disturbance, temperature and shock states.^10,11

Considering presence of shock, several mechanisms had been discussed as possible explanations for its impact on accuracy of POC glucometers including that peripheral vasoconstriction in hypoperfusion states which could lead to increased the glucose extraction by tissues because of low capillary flow resulting in falsely underestimate of glucose measurement with capillary blood.^12,13 Several publications concerning measuring blood glucose level in critically ill patients showing acceptance variation with point of care results compared to laboratory results however they didn’t differentiate between shocked relative to non-shocked patient.^4,5 Previous studies that had been conducted among critically ill patients included only small number of measurements taken for patients in shock that resulted in a heterogeneous cohort.^1,14–19 Therefore, in this study, we aimed to compare the accuracy of point of care capillary and venous/arterial samples verse venous/arterial samples by laboratory testing in critically ill both shocked and non-shocked patients.

This is a prospective case control study that was conducted among 268 critical ill patients who admitted to King Fahd Miliary Medical Complex Dhahran Saudi Arabia. Inclusion criteria included all adult, 18 years old and older, non-pregnant patients who admitted to the hospital with and without diabetes millitus after obtaining informed consent from primary decision makers. Exclusion criteria included hypovolemic shock because of severe active bleeding, patients with bleeding disorders, the use of substance that could interfere with POC glucose meter technology including icodextrin-containing solutions, intravenous immunoglobulins, abatacept, maltose), and lack of consent. In this study, we collected data of patients and laboratory results at baseline from the chart review including demographic factors as age, gender and anthropometric measurements as well as historical data as chronic comorbidities. Baseline clinical characteristics of patients included initial vital signs, the Glasgow Coma Scale score with derived verbal scores for intubated patients, the need for ventilator or dialysis and presence of acute significant arrhythmias. Laboratory parameters recorded were arterial blood gas analysis, serum creatinine, albumin, and complete blood count. The laboratory values taken at the time most proximate to the blood glucose determination were recorded.

Data was collected about the results of POC and lab analysis of blood for glucose concentrations, three blood samples were taken from venous, arterial and central blood as well as capillary blood. Data about the results of POC and lab analysis were taken and used in our comparison. The tests were conducted after request by treating physician and were done by the same bedside nurse. All data was entered, cleaned using MS Excel and anlayzed using SPSS version 26. Frequency and percent were used for descripting of the categorical variables while mean and standard deviation were used for descriping of ongoing of the variables. t paired test was used to analysis of the possible difference between measures of glucose using both of POC and lab analysis of blood. All statement with P value lower or equal to 0.05 is considered significant.

In this study, we were able to collect data for 268 patients admitted in KING FAHD MILITARY MEDICAL COMPLEX, DHAHRAN, Saudi Arabia. Among these patients, 26.1% were males with shock and 18.3% shocked females. The mean age is 63.01 years old. Most of included shocked patients were in cardiology department (20%) and only 2.5% of the shocked patient were in the neuro-surgery department. Moreover, 23.5% of the shocked patients were diabetic and 17.5 % needed hemodialysis while 59.3 % were ventilated (Table 1).

Furthermore, 53.5 % of the patients had SR arrhythmia and 27.5 % had atrial fibrillation (AF) (Figure 1). Moreover, 78.5 % of the patients need insulin infusion during their admission. In addition, we found that 55.6 % of the patients were non shocked patients while 44.4 % were shocked. In consideration of baseline characteristics of the patients, all available data were presented in Table 2. The mean U. O in the last 24 hour was 2195.28 ml in the shocked patients as a mean of 2064.9 in the non-shocked patients group. The mean weight of patients was 74.9 Kg in the shocked patients group as the mean in the non-shocked is 77.3 Kg.  The mean temperature of patients during their stay in hospital was 36.82 C with mean heart rate of 91.2 for the shocked patients group. The mean score of SOFA and APACHE 2 were 9.41 and 26.3 for the shocked patients respectively.

Figure 1 Type of arrhythmia.

In Table 3, we compared between the results of POC analysis and Lap analysis considering venous/arterial glucose values in patients who critically ill. The results showed no significant difference between POCT and lap results of both of venous and central blood (P=0.389 and 0.208) where POCT showed slightly higher results with venous glucose concentration of 10.18 and 10.05 (POCT and lab tests respectively) and 9.18 and 9.54 in POCT and lab tests respectively. The only difference between POCT and lap analysis of glucose concentrations was reported in arterial glucose concentrations where lap analysis showed significant higher glucose concentrations with 11.41 compared with 9.66 in POCT analysis (P=0.029) (Table 3). Moreover, the mean difference between POC and lab analysis of venous, arterial and central glucose were 1.03, -1.8 and 0.7 mmol.

Moreover, in Table 4, we showed the difference between results of POCT and Lab analysis of glucose concentrations in shocked and non-shocked patients. The results showed that there is no significant difference between the results of POCT or lab analysis between shocked and non-shocked patients with slightly difference between POCT and lab analysis. POCT showed that venous glucose concentration was higher in non-shocked patients and lower in case of capillary however, lab analysis showed that venous glucose concentration were higher in shocked patients and slightly lower in case of central analysis.

The appropriate measurement of glucose level of patients is extremely important especially among patients who are on a tight glycemic protocol and at increased risk of hypoglycemic episodes. This is typically true in this study among our population since many of these patients are not able to communicate with either the physicians or nurses and their signs of hypoglycemia are not readily available. POCT provides some advantages for using instead for regular blood glucose test including availability of glucose value to nurse within 2 minutes and it is immediately visible in the hospital information system. Moreover, POCT devices need very small amount of blood and risk of spilling blood from the syringe, or the device is very low.²⁰

In order to apply glucose regulation protocol, it is a need for fast and accurate way to measure the blood glucose levels.^21,22 Applying these protocols increase the nurse workload thus it should be feasible as well.^12,13,23 Therefore, this means that not the most accurate device but the most feasible and having still fairly accurate glucose analyzed would be chosen for this process. On the other hand, in critical ill patients, hypoglycemia is critic, and its warning symptoms are absent, thus these devices should be very reliable in the low range too.^14–19 When we compared the results between POCT results with lab analysis of glucose concentrations, we found that there is no difference between the venous and central glucose while significant difference was found between arterial glucose concentrations. Moreover, we found that POCT overestimated the venous and central glucose concentration and underestimated the arterial glucose concentration. According to previous study conducted by Petersen J et al., showed that glucose meters overestimated blood glucose levels in arterial, central, venous and capillary samples when compared to reference standard concentrations²⁴ and these results also reported in study of Boyd et al.,²⁵ and Critchell et al.¹⁹ In another study conducted by Clarke et al, the authors found a good accuracy for the subcutaneous CGMS in the euglycemic range²⁶ and study of Goldberg et al, which found that POCT had pearson correlation coefficient of 0.88 with 98.7 % within the clinical acceptable zones.²⁷ In a study conducted by Cook a et.al., the authors found that lab glucose values for blood in critical ill patients from catheter was significantly different from POC value for blood from the catheter (p=0.001) and from the fingerstick (p=0.001).¹⁴ Moreover, another study showed that clinical agreement between POCT and lab analysis is better in central than arterial blood analysis and in case of hypoglycemia, clinical agreement was found only in 26.3 % of cases with capillary blood analysis.¹⁶

However, hypoperfusion that occurred during shock in patients is known to be a problem in the underestimation of glucose value with capillary sampling,^19,21,22 it is not found to be a major issue in this study. For example, no significant difference considering glucose concentrations between shocked and non-shocked patients considering POCT or lab analysis. This results are similar to the results of previous studies.^12,13,23 In this study, we depended on seriously ill patients because we aimed to test the reliability of POCT under specific conditions as shock. However, this causes low population availability for sample inclusion which lead to have little data points under extreme conditions of pH, temperature, electrolyte disturbances, and hypoglycemia to make statements about the reliability of specific analyzers under these circumstances.

In conclusion, in this study, we did not find a significant difference between results of POCT and regular laboratory analysis of glucose concentrations in critically ill patients except for arterial blood glucose. Using of POCT is slightly accurate with no difference between shocked and non-shocked patients compared to lab blood analysis.

None.

None.

1. Clement S, Braithwaite SS, Magee MF, et al. Management of Diabetes and Hyperglycemia in Hospitals. Diabetes Care. 2004;27(2):553–591.
2. Malmberg K, Norhammar A, Wedel H. Glycometabolic State at Admission: Important Risk Marker of Mortality in Conventionally Treated Patients With Diabetes Mellitus and Acute Myocardial Infarction. Circulation. 1999;99(20):2626–2632.
3. Van den Berghe G, Wouters P, Weekers F, et al. Intensive Insulin Therapy in Critically Ill Patients. N Engl J Med. 2001;345(19):1359–1367.
4. Van den Berghe G, Wilmer A, Milants I, et al. Intensive Insulin Therapy in Mixed Medical/Surgical Intensive Care Units. Diabetes. 2006;55(11):3151–3159.
5. S M, KS P, PR RL, HM. K. Williams’ Textbook of Endocrinology. 12th ed. Saunders; 2011.
6. Griesdale DEG, de Souza RJ, van Dam RM, et al. Intensive insulin therapy and mortality among critically ill patients: a meta–analysis including NICE–SUGAR study data. Can Med Assoc J. 2009;180(8):821–827.
7. Krinsley JS, Grover A. Severe hypoglycemia in critically ill patients: Risk factors and outcomes. Crit Care Med. 2007;35(10):2262–2267.
8. Association. AD. Standards of Medical Care of Diabetes 2012. 2012.
9. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med. 2008;36(1):296–327.
10. Heinemann L. Quality of Glucose Measurement with Blood Glucose Meters at the Point–of–Care: Relevance of Interfering Factors. Diabetes Technol Ther. 2010;12(11):847–857.
11. Dungan K, Chapman J, Braithwaite SS. Glucose Measurement: Confounding Issues in Setting Targets for Inpatient Management. Diabetes Care. 2007;30(2):403–409.
12. Juneja D, Pandey R, Singh O. Comparison between arterial and capillary blood glucose monitoring in patients with shock. Eur J Intern Med. 2011;22(3):241–244.
13. Atkin SH. Fingerstick Glucose Determination in Shock. Ann Intern Med. 1991;114(12):1020.
14. Cook A, Laughlin D, Moore M, et al. Differences in Glucose Values Obtained From Point–of–Care Glucose Meters and Laboratory Analysis in Critically Ill Patients. Am J Crit Care. 2009;18(1):65–72.
15. Desachy A, Vuagnat AC, Ghazali AD, et al. Accuracy of Bedside Glucometry in Critically Ill Patients: Influence of Clinical Characteristics and Perfusion Index. Mayo Clin Proc. 2008;83(4):400–405.
16. Kanji 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):2778–2785.
17. Lonjaret L, Claverie V, Berard E, et al. Relative accuracy of arterial and capillary glucose meter measurements in critically ill patients. Diabetes Metab. 2012;38(3):230–235.
18. Shearer A, Boehmer M, Closs M, et al. Comparison of Glucose Point–of–Care Values With Laboratory Values in Critically Ill Patients. Am J Crit Care. 2009;18(3):224–230.
19. Critchell CD, Savarese V, Callahan A, et al. Accuracy of bedside capillary blood glucose measurements in critically ill patients. Intensive Care Med. 2007;33(12):2079–2084.
20. Corstjens AM, Ligtenberg JJM, van der Horst ICC, et al. Accuracy and feasibility of point–of–care and continuous blood glucose analysis in critically ill ICU patients. Crit Care. 2006;10(5):R135.
21. Krinsley JS. Effect of an Intensive Glucose Management Protocol on the Mortality of Critically Ill Adult Patients. Mayo Clin Proc. 2004;79(8):992–1000.
22. MASER RE, BUTLER MA, DECHERNEY GS. Use of arterial blood with bedside glucose reflectance meters in an intensive care unit. Crit Care Med. 1994;22(4):595–599.
23. Pereira AJ, Corrêa TD, de Almeida FP, et al. Inaccuracy of Venous Point–of–Care Glucose Measurements in Critically Ill Patients: A Cross–Sectional Study. Hamel FG, ed. PLoS One. 2015;10(6):e0129568.
24. Petersen JR, Graves DF, Tacker DH, et al. Comparison of POCT and central laboratory blood glucose results using arterial, capillary, and venous samples from MICU patients on a tight glycemic protocol. Clin Chim Acta. 2008;396(1–2):10–13.
25. Boyd R. Capillary versus venous bedside blood glucose estimations. Emerg Med J. 2005;22(3):177–179.
26. Clarke WL, Anderson S, Farhy L, et al. Evaluating the Clinical Accuracy of Two Continuous Glucose Sensors Using Continuous Glucose–Error Grid Analysis. Diabetes Care. 2005;28(10):2412–2417.
27. Goldberg PA, Siegel MD, Russell RR, et al. Experience with the Continuous Glucose Monitoring System® in a Medical Intensive Care Unit. Diabetes Technol Ther. 2004;6(3):339–347.