Comparative analysis of hemoglobin, potassium, sodium, and ...

This study was conducted to assess the concordance between ABG and standard venous laboratory analyzer estimations of hemoglobin, potassium, sodium, and glucose concentrations in patients in the respiratory and critical care unit. Since there was limited literature comparing the differences between the two cohorts, the aim of this study was to determine whether physicians could be confident in the results of these two modalities being interchangeable in the diagnosis and management of COPD patients.

The study demonstrated a significant difference in hemoglobin concentrations between venous and ABG samples. Of the 165 samples, 92% (152/165) fell within the 95% LOA (− 10.38, 27.62, p 1 show a mean difference of (8.62 ± 9.69) g·L−1 in hemoglobin concentrations between arterial and venous samples, with a clinical error variability of 6.6%. However, the mean biases did not exceed the acceptable biases determined by the US-CLIA (7%). Consequently, ABG hemoglobin concentrations could be used as a quick determination of venous blood hemoglobin concentrations by respiratory and critical care unit physicians for emergency COPD evaluation until venous hemoglobin measurements are available.

According to the findings presented in Table 1, the potassium concentration in venous blood was slightly higher than ABG, and showed statistically significant differences, which is consistent with Zhang's research22. 94% (156/165) values were within the 95% LOA (− 0.60, 0.21, p −1 but within the TEa range of ± 0.5 mmol·L−1. Therefore, while there was some variability between the measurements, it was still within an acceptable range. Therefore, it may be considered that ABG potassium concentration could be used for prompt assessment before the arrival of laboratory results.

Venous blood potassium concentrations were found to be slightly higher than ABG potassium concentrations due to several reasons. First, there are inherent differences between ABG and venous blood. Second, differences in specimen collection tubes could also contribute to this variance. Specifically, ABG analysis was performed with sodium heparin as the anticoagulant, which diluted the arterial blood specimen, resulting in lower arterial potassium levels compared to those of the venous blood. Lastly, the specimen types used for testing are different. ABG potassium analysis was performed on whole blood, while venous blood potassium testing required blood coagulation and centrifugation. As blood cells contain a large number of potassium ions, hemolysis during blood collection or other processes could cause these ions to enter the blood, resulting in elevated venous blood potassium levels. To improve accuracy, specimens should be sent for testing as quickly as possible after clinical collection.

This study found a statistically significant difference between the measurements of sodium concentration in venous and ABG. The study observed that 158 out of a total of 165 data points (96%) were within the 95% LOA of − 4.35 to 3.33 with a significance level of p = 0.001. Only 4% of the data points were outside the limits of agreement, indicating satisfactory agreement between the two methods of measurement. Moreover, the mean difference observed was − 0.5 mmol·L−1. According to the US-CLIA guidelines, a difference of ± 4 mmol·L−1 from the gold-standard calibration solution was acceptable for sodium concentration. Thus, using ABG sodium concentrations for estimating venous blood sodium concentrations during the acute assessment is plausible.

An accurate glucose value was vital in the identification of acute presentations such as COPD combined with diabetes, especially in the management of maintaining strict glucose control in the critically ill. The current investigation discovered a statistically significant mean difference between the two modalities. ABG glucose concentrations were higher than those of venous blood glucose. The percentage of data within the 95% LOA (− 1.71, 3.88, p 23. The mean difference between ABG and venous blood glucose concentrations was 1.08 mmol·L−1, surpassing the US-CLIA TEa for glucose of ± 0.33 mmol·L-1. Hence, ABG glucose concentration was unsuitable for a quick examination of venous glucose values. Deming regression showed a positive correlation between ABG and venous blood glucose concentrations, with a correlation coefficient R2 = 0.882.

In addition, an estimated venous blood glucose concentration could be obtained from the ABG glucose concentration, using the formula “venous blood glucose (mmol-L−1) =  − 0.487 + 0.923 × ABG glucose (mmol-L−1)”. Bland–Altman analysis was also performed to compare the standard venous blood glucose concentrations and the estimated concentrations. The 95% LOA (− 2.56, 2.55, p -1, the Bland Altman's 95% LOA (-2.56, 2.55) mmol·L-1 for glucose was wider than the US-CLIA TEa of ± 0.33 mmol·L-1, which was not clinically acceptable. Hence, the estimated mean concentration of venous blood glucose obtained according to the linear regression equation was only able to reflect the mean value of the standard venous blood glucose to a certain extent, which could help clinicians make a quick and preliminary judgment of the blood glucose concentrations for rapid assessment of COPD patients before standard venous blood glucose was available from the medical laboratory24. However, it was noteworthy that estimating blood glucose couldn't completely substitute for standard venous blood glucose, thus final results should be still subjected to the laboratory venous blood glucose concentrations.

The following are the causes for the large discrepancy in arterial and venous blood glucose levels. First of all, as the name says, ABG glucose is taken from arterial blood, and venous blood glucose is taken from peripheral venous blood. There is a natural distinction between the two methods. Moreover, carbohydrate is absorbed into the blood and first enters the artery. After digestion and absorption, it is gradually transported to the capillaries, and gradually flows back to the vein, causing the arterial glucose higher than the venous blood glucose level. In addition, due to the rapid arterial blood flow, arterial glucose fluctuates a lot and is not very stable, in contrast to venous blood glucose, which is more consistently stable. At the same time, this is also the reason why venous blood is commonly used in clinical glucose analysis.

In terms of limitations, the study did not assess whether or not the extent of the difference had an effect on patient outcomes. Therefore, it is suggested that this be taken into account in future research. Additionally, only one ABG machine and one standard venous analyzer were employed, thus subsequent research should evaluate the agreement across several machines and compare each one to standard venous analyzers separately.