Validity of venous blood gas analysis for

diagnosis of acid-base imbalance in children

admitted to pediatric intensive care unit

Nemat Bilan, Afshin G. Behbahan, A. J. Khosroshahi

Tabriz, Iran

Author Affiliations: Department of Pediatrics, Tabriz Children's Hospital, Tabriz University of Medical Sciences, Tabriz, I. R. Iran (Bilan N, Behbahan AG, Khosroshahi AJ)

Corresponding Author: Nemat Bilan, MD, Division of Respiratory and Critical Care Medicine, Department of Pediatrics, Tabriz Children's Hospital, Tabriz University of Medical Sciences, Tabriz, I. R. Iran (Tel/Fax: +98-411-5262280; Email: bilan@tbzmed.ac.ir)

Background: Arterial blood gas (ABG) analysis is the major tool for proper diagnosis and treatment of acid-base imbalance, but the invasive nature of arterial puncture and its possible hazards such as arterial spasm have resulted in a worldwide trend toward less-invasive diagnostic methods including venous blood gas (VBG) analysis. This study aimed to evaluate the validity of VBG and its clinical agreement with ABG in the 10 most common diseases in pediatric intensive care unit (PICU), and to answer how far it can replace the ABG test. 

Methods: In a cross-sectional analytical study from September 2004 to September 2005, 200 patients in 10 disease categories received blood gas analysis. Results of blood-gas tests such as pH, PCO₂ and HCO₃ of both arterial and venous blood samples (simultaneously taken from each patient) were recorded and compared by statistical analysis (kappa statistics) to determine their validity and clinical agreement.

Results: In some diseases such as respiratory distress syndrome, neonatal sepsis, renal failure, pneumonia, diabetic ketoacidosis and status epilepticus, VBG analysis showed a good validity (high sensitivity and specificity) accompanied by a suitable clinical agreement (over 40%), but in other diseases such as neonatal seizure, shock, congestive heart failure and congenital heart disease, there was either an inappropriately low validity or a weak clinical agreement (under 20%).

Conclusions: VBG can be used instead of ABG in some diseases such as respiratory distress syndrome, neonatal sepsis, renal failure, pneumonia, diabetic ketoacidosis and status epilepticus, but in other diseases such as neonatal seizure, shock, congestive heart failure and congenital heart diseases, ABG is preferable and must not be replaced by VBG. These results may be used for the formulation of future guidelines for PICU.

Key words: acid-base imbalance; arterial blood gas; intensive care unit; pediatric intensive care unit; venous blood gas

                  World J Pediatr 2008;4(2):114-117

Introduction

Assessment of acid-base imbalances is necessary for most patients who are admitted to pediatric intensive care unit (PICU) every day and this need is routinely accomplished by arterial blood gas (ABG) analysis, but the invasive nature of arterial puncture and its possible hazards such as bleeding, arterial spasm and thrombosis have changed the general trend toward less invasive methods such as venous blood gas (VBG) analysis in all hospitals worldwide. According to a study by Kirubakaran et al^[1] in India, there was a good correlation between pH of venous and arterial blood samples while their CO₂-concentrations were less correlative. Another study by Kelly and coworkers^[2] in Australia showed that pH of venous and arterial blood samples correlate with each other by a high degree of agreement (about 90%). The study of Chu and coworkers^[3] on patients under mechanical ventilation due to acute respiratory failure in Taiwan revealed that VBG analysis could accurately predict the expected values of pH, PCO₂ and HCO₃ from ABG analysis. Two separate studies were carried out by Brandenburg et al^[4] (on patients with diabetic ketoacidosis in USA) and Gokel et al^[5] (on uremic patients in Turkey), and both revealed that VBG analysis could determine the patient's acid-base status as accurately as ABG analysis. Yildizdas et al^[6] reported a study on a wide spectrum of diseases including 19 different diagnoses showed that pH, PCO₂ and HCO₃ of both arterial and venous blood samples had a meaningful correlation with each other. Although this study covered a relatively large number of patients, practically its results could not be generalized to every disease category because of the inappropriate distribution of studied patients and their small number in each disease group. A study by Ak and coworkers^[7] on adult patients with acute exacerbation of chronic obstructive pulmonary disease in Turkey showed that there was a significant correlation between ABG and VBG values of pH, PCO₂ and HCO₃. But another study by Razi and coworkers^[8] in Iran on the same disease as Ak's study showed that there was a relatively good correlation between ABG and VBG values of pH and PCO₂ although this correlation is not so close, so they concluded that ABG analysis should not be replaced by VBG analysis in these patients. Eizadi-Mood and coworkers^[9] studied patients poisoned by tri-cyclic antidepressants in Iran and found that pH was the only parameter measured by VBG which was a valid and reliable substitute for ABG. Kelly and coworkers^[10] in Australia also found that venous bicarbonate estimation has a high level of agreement with the arterial value and suggested that only venous values of HCO₃ may be an acceptable substitute for arterial measurements.

In the above mentioned studies, findings were controversial about the reliability of VBG values for predicting ABG values. Besides, they were predominantly limited to adult patients or a number of diseases. Therefore we attempted to assess the validity of VBG values in determination of acid-base status, specifically in children admitted to PICU and to focus on 10 diseases that consist of the major load of PICU turnover.

Methods

In a cross-sectional and analytical study from September 2004 to September 2005, we analyzed blood gas of 200 patients in 10 disease categories (20 patients in each group) that formed the largest job burden of PICU, including: (1) respiratory distress syndrome, (2) neonatal sepsis, (3) neonatal seizure, (4) status epilepticus, (5) pneumonia, (6) diabetic ketoacidosis, (7) renal failure, (8) congestive heart failure, (9) shock, and (10) congenital heart disease. We also classified these diseases according to their severity, e.g., if a patient with sepsis was in shock at admission, we put him/her in a shock group. The final number of 20 patients was considered for completion and termination of sampling in each group of this study. Venous and arterial blood samplings were done simultaneously for each patient. Arterial blood was drawn from radial or brachial arteries and venous blood from veins on the dorsum of hands or antecubital veins.

All laboratory standard measures adhered to the international protocols for blood gas sampling, sample transport and their analysis for determination and recording of pH, PCO₂ and HCO₃. Each ABG or VBG report sheet was interpreted as a separate diagnostic test by PICU physicians who were blind to this study and their final diagnosis was recorded as one of the four single acid-base imbalances (consisting of main diagnoses of blood gas analysis), including metabolic acidosis, metabolic alkalosis, respiratory acidosis and respiratory alkalosis. Two statistical methods were used to determine the accuracy of VBG-based clinical diagnosis in comparison with the acid-base imbalance revealed by ABG analysis of the simultaneous arterial blood sample from the respective patient. In other words, ABG analysis was considered as the gold standard test for determining if VBG-based diagnoses were compatible and precise enough to be trustful:

1) Validity included calculation of sensitivity, specificity, and positive and negative predictive values.

2) Clinical agreement was determined by the results of "kappa statistics", considering the following scale:^[11,12] κ≤20% meant negligible or few agreement; 20%<κ≤40% meant minimal agreement; 40%<κ≤60% meant medium agreement; 60%<κ≤80% meant good agreement; κ>80% meant excellent agreement.

Results

This study consisted of 200 patients, 136 (68%) boys and 64 (32%) girls. Among them, 107 patients (53.5%) were under 2 months of age, 51 (25.5%) were 2-12 months, 21 (10.5%) were 12-60 months, and 21 (10.5%) over 60 months.

Fifty-nine patients (29.5%) were subjected to mechanical ventilation. The components of validity for VBG-based clinical diagnosis (VBG analysis was considered as a diagnostic test) such as sensitivity, specificity, positive and negative predictive values and also its clinical agreement with ABG-based clinical diagnosis (kappa statistics) were acceptable in 6 of 10 study groups (Table 1).

We also assessed the validity of VBG analysis and its clinical agreement with ABG analysis for detection of acidosis and alkalosis regardless of their metabolic or respiratory nature. A good sensitivity and clinical agreement was shown for detection of acidosis unlike a low sensitivity and medium clinical agreement for detection of alkalosis (Table 2).

Table 1. Validity and clinical agreement of acid-base imbalance diagnosed by VBG analysis as compared with ABG in 10 diseases

PPV: positive predictive value; NPV: negative predictive value; κ: kappa statistics.

Table 2. Validity and clinical agreement for detection of acidosis and alkalosis by VBG analysis as compared with ABG (regardless of their metabolic or respiratory nature) in 10 diseases

PPV: positive predictive value; NPV: negative predictive value; κ: kappa statistics.

Discussion

As mentioned in the results section, VBG analysis can replace ABG analysis in 6 of 10 diseases (respiratory distress syndrome, neonatal sepsis, renal failure, pneumonia, diabetic ketoacidosis and status epilepticus) which form the major job burden of PICU. Similar results have been reported by many studies. The study of Chu and coworkers^[3] in Taiwan showed that VBG analysis is an acceptable alternative for ABG analysis to assess pH, PCO₂ and HCO₃ in patients with acute respiratory failure. Brandenburg et al^[4] studied 38 patients with diabetic ketoacidosis, and found that VBG analysis could determine the severity of acidosis as accurately as ABG analysis. Such a substitution was also been recommended by Gokel et al^[5] on diabetic and uremic patients. The similarity of all VBG parameters to those of ABG has been confirmed by Yildizdas et al.^[6] In another study, Malinoski et al^[13] in USA found that central venous and arterial PCO₂, pH and base excess values correlate well in mechanically ventilated trauma patients; however, they suggested that clinically reliable conclusions can be reached with VBG analysis rather than simple substitution of VBG for ABG in mechanically ventilated trauma patients. Ma and coworkers^[14] declared that venous pH correlates well and is precise enough with arterial pH to serve as a substitute in patients with diabetic ketoacidosis. Malatesha et al^[15] claimed that VBG analysis for pH, PCO₂ and HCO₃ may be a reliable substitute for ABG analysis in the initial evaluation of adult patients presenting to the emergency department. Middleton et al^[16] found a high level of agreement between central venous and arterial values for pH, HCO₃ and base excess in ICU patients, therefore central venous values may be an acceptable substitute for arterial measurements in ICU setting.

In our study, VBG values in patients who have disturbed tissue perfusion (hemodynamic instability) showed a very low level of clinical agreement with the arterial values and therefore no validity. As declared by McGillivray,^[17] there was a meaningful correlation between results of ABG and VBG analysis only in patients with perfect tissue perfusion; however, Kelly^[18] found that venous and arterial pH had sufficient agreement as to be clinically interchangeable in patients with diabetic ketoacidosis who were hemodynamically stable and without respiratory failure. Adrogue et al^[19] reported little difference between arterial and venous blood in pH and PCO₂ (0.03 and 5.7 mmHg, respectively) in patients with normal cardiac output, but this difference was higher in patients with moderate heart failure. The mean values of pH in VBG and ABG were 7.21 and 7.31, and their mean PCO₂ were 68 and 44 mmHg, respectively.^[19] Our results in congestive heart failure and congenital heart diseases are similar to the results mentioned above, and can justify the related lower levels of validity and clinical agreement.

In conclusion, VBG analysis can replace ABG analysis at least in 6 of 10 diseases with a perfect validity and proper clinical agreement. This can be used for the regulation of guidelines especially in PICU.

Acknowledgements

We wish to thank our colleagues: Mrs. Jigary and Miss Shafaei, for their labors in accomplishment of this study.

Funding: This study was supported by grants from the vice chancellor for research, Tabriz University of medical sciences.

Ethical approval: Approved by the Regional Committee for Medical Research Ethics.

Competing interest: None declared.

Contributors: Bilan N wrote the first draft of this paper. All authors contributed to the intellectual content and approved the final version.

References

1  Kirubakaran C, Gnananayagam JE, Sundaravalli EK. Comparison of blood gas values in arterial and venous blood. Indian J Pediatr 2003;70:781-785.

2  Kelly AM, McAlpine R, Kyle E. Venous pH can safely replace arterial pH in the initial evaluation of patients in the emergency department. Emerg Med J 2001;18:340-342.

3  Chu YC, Chen CZ, Lee CH, Chen CW, Chang HY, Hsiue TR. Prediction of arterial blood gas values from venous blood gas values in patients with acute respiratory failure receiving mechanical ventilation. J Formos Med Assoc 2003;102:539-543.

4  Brandenburg MA, Dire DJ. Comparison of arterial and venous blood gas values in the initial emergency department evaluation of patients with diabetic ketoacidosis. Ann Emerg Med 1998;31:459-465.

5  Gokel Y, Paydas S, Koseoglu Z, Alparslan N, Seydaoglu G. Comparison of blood gas and acid-base measurements in arterial and venous blood samples in patients with uremic acidosis and diabetic ketoacidosis in the emergency room. Am J Nephrol 2000;20:319-323.

6  Yildizdaş D, Yapicioğlu H, Yilmaz HL, Sertdemir Y. Correlation of simultaneously obtained capillary, venous, and arterial blood gases of patients in a paediatric intensive care unit. Arch Dis Child 2004;89:176-180.

7  Ak A, Ogun CO, Bayir A, Kayis SA, Koylu R. Prediction of arterial blood gas values form venous blood gas values in patients with acute exacerbation of chronic obstructive pulmonary disease. Tohoku J Exp Med 2006;210:285-290.

8  Razi E, Moosavi GA. Comparison of arterial and venous blood gases analysis in patients with acute exacerbation of chronic obstructive pulmonary disease. Saudi Med J 2007;28:862-865.

9  Eizadi-Mood N, Moein N, Saghaei M. Evaluation of relationship between arterial and venous blood gas values in the patients with tricyclic antidepressant poisoning. Clin Toxicol (Phila) 2005;43:357-360.

10 Kelly AM, McAlpine R, Kyle E. Agreement between bicarbonate measured on arterial and venous blood gases. Emerg Med Australas 2004;16:407-409.

11 Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistics. Fam Med 2005;37:360-363.

12 McGinn T, Wyer PC, Newman TB, Keitz S, Leipzig R, For GG; Evidence-Based Medicine Teaching Tips Working Group. Tips for learners of evidence-based medicine: 3. Measures of observer variability (kappa statistic). CMAJ 2004;171:1369-1373.

13 Malinoski DJ, Todd SR, Slone S, Mullins RJ, Schreiber MA. Correlation of central venous and arterial blood gas measurements in mechanically ventilated trauma patients. Arch Surg 2005;140:1122-1125.

14 Ma OJ, Rush MD, Godfrey MM, Gaddis G. Arterial blood gas results rarely influence emergency physician management of patients with suspected diabetic ketoacidosis. Acad Emerg Med 2003;10:836-841.

15 Malatesha G, Singh NK, Bharija A, Rehani B, Goel A. Comparison of arterial and venous PH, bicarbonate, PCO₂ and PO₂ in initial emergency department assessment. Emerg Med J 2007;24:569-571.

16 Middleton P, Kelly AM, Brown J, Robertson M. Agreement between arterial and central venous values for pH, bicarbonate, base excess, and lactate. Emerg Med J 2006;23:622-624.

17 McGillivray D, Ducharme FM, Charron Y, Mattimoe C, Treherne S. Clinical decisionmaking based on venous versus capillary blood gas values in the well-perfused child. Ann Emerg Med 1999;34:58-63.

18 Kelly AM. The case for venous rather than arterial blood gases in diabetic ketoacidosis. Emerg Med Australas 2006;18:64-67.

19 Adrogué HJ, Rashad MN, Gorin AB, Yacoub J, Madias NE. Assessing acid-base status in circulatory failure. Differences between arterial and central venous blood. N Engl J Med 1989;320:1312-1316.

Received November 5, 2007 Accepted after revision January 7, 2008