Comparison of Capillary and Venous Glucose in Diabetic Patient in a Peripheral Hospital

Methods: This was cross-sectional analytic study done in United Mission Hospital, Palpa, Tansen. Calculated sample size was 92, convenience sampling technique was used. During 5 month of duration in 92 diabetic patients, where fasting capillary and venous glucose were performed consecutively. Confounding was ruled out with matching approach, adjustment tests were also used like X2 Mantel -Haenszel and logistic regression. Reporting guideline of this observational study was done with the help of SROBE guidelines.


INTRODUCTION
According to the latest WHO data published in April 2011 Diabetes Mellitus deaths in Nepal reached 3224, 2.17% of total death and the figure is expected to increase to 366 million in 2030 worldwide. [1][2] The Nepal Diabetic Association reported that the diabetes affects approximately 15% of the people over 20 years and 19% of the people over 40 years of the age in the urban areas of Nepal. 3 In addition, there have been many concern raised about the accuracy of capillary blood glucose estimation in the face of systemic illness, and it has been suggested that in such patients, venous sampling may be more accurate. 4 Many studies have concluded that glucometers should not be used for diagnosis, only for self-monitoring. 5 In the present study, ascorbic acid, uric acid, a maltose, aspirin, icodextrin and acetaminophen cause around 11% difference in glucose levels. 6 The objective of this study is to compare capillary and venous blood glucose in diabetic population.

METHODS
A cross-sectional analytic study was conducted to compare difference between capillary and venous blood glucose. United Mission Hospital was selected as the study area for the research. Data collection was done form July to December 2014. All patient attending diabetic clinic and emergency department who were in fasting state were selected as the study population. Approval from the Institutional Review Board was taken.
Hypothesis: There will be strong correlation between capillary and venous blood glucose.
Null hypothesis: There will be poor correlation between capillary and venous blood glucose.

Inclusion Criteria
Age above 14, diagnosed case of diabetes mellitus and patient who are ordered for for fasting blood glucose were included in this study.

Exclusion Criteria
Patient who are in terminally ill, under peritoneal dialosis and who have other endocrinology problems were excluded.
All the collected data were examined, compiled, organized and analyzed. Descriptive statistical method was used to analyze and interpret data. Confounding was ruled out with matching approach, adjustment tests were also used like X² Mantel -Haenszel and logistic regression. Reporting guideline of this observational study was done with the help of SROBE guidelines. All analyzed data were shown in various table by using SPSS software ver.16.

RESULTS
Male and female were almost equally included in this study. More than half of the study population was more than 65 years of age (58.70%). Three fourth participants of this study were diabetic since less than 10 years (77.20%).
A strong correlation was observed between venous and capillary blood glucose, with Pearson correlation coefficient of 0.94 (Table 1). Mean capillary glucose in male and female was 13.94% and 3.19% lower than venous blood glucose respectively.  Capillary blood glucose in patients who were under oral hypoglycemic agents was 8.10% lower and in population who were under insulin was 14.12% lower than the venous blood glucose (Table 4). Although the measurement of glucose is one of the oldest established tests in clinical medicine, it is extremely complex and often only an approximation of the "true" level. 10 Blood glucose levels measured from different sites, using different fractions of blood and different methods provide different results. Glucose measurements can be performed on whole-blood, plasma, and serum, and these may be native or deproteinized or hemolyzed in the case of capillary whole blood. Furthermore, the blood may be arterial, capillary, or venous in origin.
Glucose is dissolved only in the aqueous part of the drawn specimen and not in its entire volume. This is the major reason for differing glucose concentration in plasma and wholeblood samples. The glucose concentration is approximately 10% higher when measured in plasma as compared to whole blood. This difference is more marked in patients with higher Hematocrit. 11 A number of glucometers correct for this difference and give results as "plasma equivalent." Other blood glucose strips retain red blood cells through a filtering process and measure glucose content in plasma in their reaction zone (e.g., Accu-Chek Comfort Curve test, Roche Diagnostics).
Even the Yellow Springs Instrument's Blood Glucose analyzer, which is considered the reference gold standard, yields laboratory glucose results dependent on Hct when whole-blood samples are used. 12 Furthermore, if laboratory glucolysis is not inhibited in whole-blood specimens the laboratory glucose levels fall with delays in processing the specimen. 13 The method of glucose measurement used by glucometers in routine clinical use are based on either chromogenic or electrochemical reactions of the three enzymes glucose oxidase, dehydrogenase, and hexokinase. This gives rise to method-based specific interferences such as blood and ambient (altitude) oxygen tension, blood pH, and serum cholesterol and triglycerides levels and interfering drugs (maltose, d-xylose, icodextrin, dopamine, acetaminophen). 14