The blood gas analysis offers important data to guide the treatment and prognostication of critically ill patients. Providers are able to quickly assess acid-base status, tissue perfusion, and respiratory mechanics. Traditionally, providers relied on arterial blood gas (ABG) for these data; however, a relatively new consensus of data supports venous blood gas (VBG) pH, bicarbonate, and lactate as surrogates for arterial blood values1,2,3.
In a recent prospective study, Middleton et al matched 168 arterial and venous blood samples and determined that VBG and ABG could be used interchangeably (95% limits of agreement- This means that the difference did not have meaningful clinical variation) for pH, bicarbonate, base excess, and lactate1. This is good news for patients because arterial punctures are often painful, more difficult to obtain, and have associated complications (although rare) such as nerve injury and digital ischemia. However, we can’t relegate the ABG to the part of the bitter past we’d like to forget (much like I’ve already done to the Cruz-Fiorina campaign and life before the TSA PreCheck option) because there is little consensus on the VBG’s reliability as a surrogate to assess CO2 levels, particularly when the value is moderate to severely elevated.
The Studies
There are 5 recent studies comparing ABG to VBG CO2 inter-reliability that publish their 95% limits of agreement.
Author, year | No. Subjects | Mean difference (mmHg) | 95% limits of agreement (mmHg) | Conditions studied | Additional Findings |
Kelly, 20054 | 112 | 6 | -14 to 26 | COPD | PvCO2 > 45 mmHg 100% sensitive for detecting hypercarbia
29% reduction in need for ABG sample |
McCanny, 20125 | 84 | 8.6 | -7.8 to 25 | COPD | PvCO2 > 45 mmHg 100% sensitive for detecting hypercarbia
|
Toftegaard, 20086 | 103 | 4.5 | -12.8 to 3.9 | various | PvCO2 standard deviation outside the acceptable lab performance criteria, central PCO2 was within |
Ibrahim, 20087 | 122 | 3.3 | -17.4 to 23.9 | various | PvCO2 <30 mmHg 100% sensitive to rule out hypercarbia
|
Malatesha, 20078 | 95 | 8 | -7.6 to 6.8 | various | – |
In summation, these studies suggest peripheral venous CO2 (PvCO2) is a useful screening tool. Low levels of PvCO2 (<30 mmHg) may help rule out hypercarbia and high levels (>45 mm Hg) strongly correlate with presence of hypercarbia on the ABG. Determining the presence of hypercarbia based on the VBG may negate the need for ABG evaluation in some instances. However, PvCO2 cannot fully replace arterial peripheral arterial CO2 (PaCO2) because of the wide and unpredictable variation of the credibility interval. Moreover a “correction factor” cannot be used to calculate the PaCO2 from PvCO2 because the actual value may be higher or lower9. The average mean difference in ABG vs VBG CO2 is large enough that it may change patient management, such as the decision to start non-invasive ventilation (NIV) in hypercapneic acute respiratory failure.
High Stakes for COPD patients
COPD patients represent one particular population where a reliable CO2 is critical to your decision-making process. You may be able to rule in or out hypercarbia based on the VBG, but knowing with accuracy that your patient’s CO2 is worsened from baseline may trigger early NIV (however, I would like to emphasize most decisions to begin NIV should be made clinically, even before a VBG is resulted). Early appropriate NIV in patients with arterial pH < 7.35 and hypercapnea has demonstrated lower mortality, reduced rates of intubation, reduction in total number of hospital admission days, and fewer complications such as ventilator-induced lung injury and nosocomial lung infections10,11,12.
Trending PCO2 to Guide Duration of NIV
Monitoring CO2 during trials of self-ventilation may help clinicians appropriately titrate down NIV use. The 2016 British Thoracic Society guidelines recommend trending transcutaneous, arterial, or capillary pCO2 along with pH. Once these lab values have normalized and the patient is clinically improved, NIV can be discontinued13. While their recommendation does not specifically address venous blood sampling, a low-normal value would effectively rule out hypercarbia.
Conclusions
-Draw the PvCO2 while you are getting your other labs. You may be able to avoid an arterial stick.
–PvCO2 > 45 is a useful screening test for the presence of hypercarbia and PvCO2 < 30 is a good “rule out” test.
-PvCO2 does not accurately predict PaCO2, and the true value may be significantly higher or lower.
-You may need to obtain an accurate CO2 (ABG) measurement in some patients because it might change your management. In patients with COPD, early NIV in moderate to severe COPD exacerbations has been shown to reduce morbidity and mortality
-Trending CO2 and pH to normal values can help guide discontinuation of NIV. BEWARE! VBGs have limited utility in trending CO2 until they reach a low-normal level.
Written by Wendy Chan MD
Peer-Review by Eden Kim DO and Raul Hernandez MD
Reviewed/Edited by Ian deSouza MD
Works Cited
wendyrollerblades
Latest posts by wendyrollerblades (see all)
- Should We Consider Mathematical Arterialization of VBGs in ED Patients? - January 16, 2017
- VBG vs ABG in Hypercarbia - September 6, 2016
- Preventing Iatrogenic Opioid Overdose in the ED - April 18, 2016
4 Comments
ablumenberg · September 6, 2016 at 3:03 pm
Great post Dr. Chan!
Ian deSouza · September 7, 2016 at 4:27 pm
In practice and if available and despite no direct, supporting evidence, I’ve used nasal ETCO2 to monitor the response to treatment of patients with type 2 respiratory insufficiency (opioid overdose, COPD). But apparently, there are large differences between ETCO2 and PaCO2 in spontaneously breathing nonintubated patients with acute respiratory failure.
Lermuzeaux M, et al. Superiority of transcutaneous CO2 over end-tidal CO2 measurement for monitoring respiratory failure in nonintubated patients: A pilot study. J Crit Care. 2016 Feb;31(1):150-6. doi: 10.1016/j.jcrc.2015.09.014. Epub 2015 Sep 25.
Stephen E Rees · October 7, 2016 at 12:37 pm
You may want to look at the other papers from our group (Rees SE et al) where we have developed a method for mathematically calculating ABG from venous values. There are about 5 additonal papers from our group which is that of the Toftegaard paper you cite. Just send me a mail if you want the details
Dr Mikey · November 12, 2017 at 2:09 pm
Awesome! I love using VBG and avoid ABG outside of obtunded patients as much as possible.