How to Interpret Venous Blood Gas (VBG) Results

A venous blood gas gives you reliable information about acid-base status and metabolism, but it cannot tell you how well the lungs are oxygenating blood. That single distinction is the foundation of VBG interpretation. Venous pH and bicarbonate track their arterial counterparts closely enough to guide many clinical decisions, while venous pO2 is essentially useless as a stand-in for arterial oxygenation. Understanding which numbers you can trust on a VBG, which need adjustment, and which you should ignore entirely is what separates useful interpretation from misguided confidence.

pH and Bicarbonate Are the Strong Suit

The parameters most people want from a blood gas are often acid-base values, and this is where a VBG performs well. Venous pH runs slightly lower than arterial pH, typically by about 0.03 to 0.05 units, because tissues have already added CO2 to the blood by the time it reaches the veins. Despite that small offset, the correlation between arterial and venous pH is strong. In emergency department patients, a study using Bland-Altman analysis found the 95% limits of agreement for pH fell within a narrow range of about 0.13 to −0.10.¹ An ICU-based study reported a correlation coefficient of 0.783 between arterial and venous pH, with the mean arterial-venous difference sitting at just 0.03 units.²

Bicarbonate shows even tighter agreement. The same ICU study found a correlation coefficient of 0.846 for bicarbonate, and the mean difference between arterial and venous values was about 1 mEq/L.² In patients with renal failure, where acid-base disturbances are the whole reason you are drawing the sample, the Pearson correlation for both pH and bicarbonate was above 0.84.³ In practical terms, if you see a venous pH of 7.30, the arterial pH is very likely somewhere near 7.33 to 7.35. If you see a venous bicarbonate of 18, the arterial value is probably in the same ballpark. You can confidently use these numbers to identify metabolic acidosis, metabolic alkalosis, and to monitor trends over time.

Venous pCO2 Needs More Caution

Carbon dioxide is a trickier parameter. Venous pCO2 runs higher than arterial, usually by about 3 to 8 mmHg, because the veins carry CO2 that tissues have just produced. The correlation between arterial and venous pCO2 is decent but not as tight as pH. One study of emergency department patients reported a correlation coefficient of 0.838, but the Bland-Altman limits of agreement were wide: −17.4 to +23.9 mmHg around a mean difference of 3.3 mmHg.⁴ That range is broad enough that a venous pCO2 of 50 could correspond to an arterial value anywhere from roughly 26 to 67 in unusual cases, which is a clinically meaningful spread.

Where venous pCO2 shines is as a screening tool. If the venous pCO2 is normal or low, you can be fairly confident the patient does not have dangerous arterial CO2 retention. A study of patients with chronic obstructive pulmonary disease found that a venous pCO2 below 45 mmHg had a 100% sensitivity and 100% negative predictive value for ruling out arterial hypercarbia.⁵ Similarly, when venous pCO2 was below 30 mmHg, sensitivity and negative predictive value for excluding an arterial pCO2 above 45 were both 100%.⁴ The practical takeaway: a low venous pCO2 is reassuring. A high venous pCO2 tells you the patient might have CO2 retention, but you’ll need an arterial sample to know for certain how bad it is.

One ICU study from central India did report much tighter correlation for pCO2, with a coefficient of 0.992, suggesting that for every unit increase in venous pCO2, the arterial value changed by about 0.98 units.⁶ This is tighter than most other studies report, and the discrepancy likely reflects differences in patient populations and clinical settings. The safest interpretation is that venous pCO2 is a useful ballpark number but should not be treated as interchangeable with arterial pCO2 when precise CO2 measurement matters.

Venous pO2 Tells You Almost Nothing About Oxygenation

This is the parameter that trips people up. Venous pO2 does not reflect how well the lungs are doing their job, and it cannot be used as a substitute for arterial pO2. The reason is straightforward: by the time blood returns to the veins, tissues have already extracted oxygen from it. The amount extracted depends on local metabolic rate, blood flow to that particular limb, temperature, and other factors that have nothing to do with lung function. A narrative review on the role of venous blood gas analysis in critical care makes this point explicitly: pO2 values differ significantly between arterial and venous blood due to oxygen consumption gradients.⁷

The data backs this up consistently. A study comparing arterial and venous gases in COPD patients found the mean arterial pO2 was about 55 mmHg while the mean venous pO2 was about 43 mmHg, a gap of roughly 12 mmHg, with the correlation weaker than for pH or pCO2.⁸ The ICU study from central India that found excellent correlation for pH, pCO2, and bicarbonate reported a pO2 correlation coefficient of just 0.010, which is essentially no correlation at all.⁶ A pediatric study of congenital heart disease patients reached the same conclusion, finding poor correlation for pO2, oxygen content, and oxygen saturation between venous and arterial samples.⁹

If you need to know how well the patient is oxygenating, use a pulse oximeter for a quick screening or draw an arterial sample. The venous pO2 on your printout is a number about the tissue bed where the blood was drawn, not about the lungs.

Lactate on a VBG

Lactate is one of the most clinically useful values on a VBG, and the good news is that venous lactate correlates well with arterial lactate, with one consistent caveat: venous values tend to run a bit higher. A systematic review of multiple studies found that peripheral venous lactate levels exceeded arterial levels by a mean difference ranging from 0.18 to 1.06 mmol/L across the literature.¹⁰ That same review found that at a venous lactate cutoff of about 2 mmol/L, sensitivity for ruling out arterial hyperlactatemia was between 97% and 100%. So a normal venous lactate is very reassuring.

An emergency department study found even more granular thresholds: when venous lactate was below 2 mmol/L, the sensitivity for arterial lactate being below 2 mmol/L was 100%. When venous lactate was above 3 mmol/L, specificity for arterial hyperlactatemia was also 100%.¹¹ The gray zone sits between about 2 and 3.5 mmol/L, where a venous value might overestimate the arterial level enough to change clinical decisions. In sepsis patients specifically, venous lactate correlated with arterial lactate and was consistently higher by about 0.68 mg/dL, with correlation improving over time during resuscitation.¹²

The bottom line for lactate: a low venous lactate is a reliable negative screen. A very high venous lactate means the arterial level is also elevated. Borderline values deserve a closer look, especially if the clinical picture is unclear, but in most emergency settings the venous lactate gives you enough information to make initial decisions about resuscitation.

Where You Draw the Blood Matters

Not all venous blood is the same. Most VBGs come from a peripheral vein, usually in the arm. But in critical care, blood can also be drawn from a central venous catheter sitting in the superior vena cava, or from a pulmonary artery catheter. These three sources give different information, and the differences matter.

Peripheral venous blood reflects local tissue metabolism at the sampling site. If the arm is cold, if a tourniquet has been on too long, or if the limb has poor perfusion, the gas values will be skewed by conditions in that specific limb rather than reflecting the body as a whole. Central venous blood, drawn from the superior vena cava via a central line, captures venous return primarily from the upper body and offers a broader view. Mixed venous blood, available only from a pulmonary artery catheter, represents the most complete picture because it includes venous return from the entire body after all tributaries have mixed.¹³

This distinction becomes especially relevant when measuring venous oxygen saturation. Central venous oxygen saturation (ScvO2) is typically about 2 to 3 percentage points lower than true mixed venous oxygen saturation (SvO2), but it serves as a practical substitute in most ICU settings because pulmonary artery catheters are not placed routinely.¹³ For acid-base parameters like pH and bicarbonate, the distinction between peripheral and central samples is less dramatic, but for oxygenation and perfusion assessment, the source of the venous sample is something you need to know.

Central Venous Oxygen Saturation in Shock

ScvO2 is a parameter unique to central or mixed venous blood, and it deserves its own discussion because it tells you something an arterial blood gas cannot: whether your patient’s tissues are getting enough oxygen to meet their metabolic demands. A normal ScvO2 sits around 65 to 75%. When it drops below that, the body is extracting more oxygen from each unit of blood than usual, which signals that either oxygen delivery is inadequate (low cardiac output, anemia, hypoxemia) or oxygen demand is dangerously high.

In the management of septic shock, ScvO2 has been used alongside lactate as a guide to resuscitation. Each measure has limitations on its own. Lactate can be elevated for reasons other than poor perfusion, and ScvO2 can be misleadingly normal if cells are too sick to extract oxygen. Used together, they give clinicians a more nuanced view of the patient’s hemodynamic state and allow targeted therapy aimed at restoring oxygen delivery.¹⁴ ScvO2 is easily measured from any existing central line and provides a window into cardiac output and oxygen consumption that no peripheral VBG can offer.¹⁵

VBG in Diabetic Ketoacidosis

Diabetic ketoacidosis is one of the clinical scenarios where VBG has essentially replaced ABG in many emergency departments. The diagnosis depends on demonstrating acidosis, elevated blood sugar, and ketones. You do not need to know arterial pO2 to make that diagnosis, and the pH and bicarbonate values you need are reliably captured by a venous sample.

A study of hyperglycemic emergency department patients found that VBG electrolytes were nearly 98% sensitive and 100% specific for diagnosing DKA, supporting the use of VBG in place of both ABG and separate serum chemistry panels.¹⁶ A separate study comparing arterial and venous blood gas values in DKA patients specifically found excellent correlation and agreement for pH, with good agreement for pCO2 and bicarbonate, and concluded that venous sampling could serve as a less invasive alternative with fewer complications.¹⁷ For DKA management, the serial monitoring of pH and bicarbonate to track resolution of acidosis is something a VBG handles well, sparing the patient repeated arterial punctures.

VBG in COPD Exacerbations

Chronic obstructive pulmonary disease is where VBG interpretation gets more nuanced. Patients presenting with an acute exacerbation often need assessment of CO2 retention and acid-base status. Research confirms that venous pH and pCO2 correlate well with arterial values in these patients, supporting VBG as a feasible first-line assessment of acid-base status.¹⁸ But venous pO2 does not reliably reflect arterial oxygenation in this population, which means that the question most specific to COPD management — how hypoxemic is this patient? — still requires either an ABG or reliance on pulse oximetry.

A systematic review and meta-analysis looking at VBG in acute COPD exacerbations found that while differences exist between ABG and VBG values, the evidence base was small enough that further verification was needed.¹⁹ In practice, many clinicians use a VBG as the initial screen in COPD: if the venous pCO2 is normal and the pH is not acidotic, the patient is unlikely to have dangerous CO2 retention. If the venous pCO2 is elevated or the pH is low, an ABG can follow to quantify the problem precisely. This stepwise approach can reduce the number of arterial punctures by roughly a third in some estimates.⁵

When VBG Values Diverge Dramatically From Arterial Values

In severe shock and during cardiopulmonary resuscitation, the usual relationship between arterial and venous blood gases can break down entirely. During CPR, cardiac output is a fraction of normal, and blood moves sluggishly through the tissues. The result is a striking paradox: arterial blood may appear well-oxygenated with low pCO2 (because the small volume reaching the lungs gets efficiently ventilated), while venous blood simultaneously shows severe acidosis, high pCO2, and very low oxygen content. An experimental study of CPR found that arterial blood was well oxygenated with pCO2 as low as 5 to 9 mmHg, while venous blood had oxygen content below 25% and a progressively worsening metabolic acidosis, with base deficit climbing from about 2 mEq/L before CPR to nearly 17 mEq/L after 30 minutes.²⁰

In this situation, a venous blood gas actually tells you more about the patient’s true metabolic state than an arterial sample would. The arterial values look deceptively reassuring because they reflect what the lungs are doing, while the venous values expose what the tissues are experiencing. This is a scenario where the two samples tell fundamentally different stories, and clinicians managing cardiac arrest should be aware that arterial pH can lag far behind the tissue-level acidosis that a venous sample reveals.

Sample Handling and Common Errors

Interpreting a VBG assumes the sample was collected and processed properly. Pre-analytical errors can distort results in ways that mimic real pathology. Air bubbles in the syringe are a common culprit: they cause gas exchange between the air and the blood sample, artificially raising pO2 and lowering pCO2. A study examining pre-analytical stability found that air bubbles significantly interfered with pO2 regardless of whether the sample was stored at room temperature or chilled, and that pCO2 increased significantly at room temperature after just 30 minutes.²¹

Temperature and processing delay also matter. At room temperature, the study found pH and glucose remained stable for about 90 minutes, lactate and pCO2 for about 60 minutes, and pO2 was the least stable parameter when chilled. Chilling the sample kept glucose and lactate stable longer but did not fully protect other values.²¹ The practical advice is to run the sample as quickly as possible and to expel any visible air bubbles from the syringe immediately after collection. If the numbers on a VBG look unexpectedly abnormal and the clinical picture does not match, a handling error is worth considering before acting on the result.

Tourniquet time is another often-overlooked factor. A tourniquet left on for too long causes venous stasis in the arm, and the pooled blood accumulates CO2 and lactic acid while becoming progressively deoxygenated. This can produce a falsely low pH, falsely elevated pCO2, and elevated lactate on the VBG that have nothing to do with the patient’s systemic acid-base status. Releasing the tourniquet and allowing blood to flow for a few seconds before drawing can reduce this artifact.

Pediatric and Neonatal Considerations

Arterial punctures are technically difficult and distressing in newborns and small children, making VBG and capillary blood gases the workhorses of pediatric acid-base assessment. The question of whether venous values are interchangeable with other sample types has been studied in neonates. A study of non-critically ill neonates defined clinically acceptable differences for each parameter and then measured how often venous and capillary values fell within those thresholds. Interchangeability was highest for bicarbonate, at 94%, and for base excess at 90%. For pH it was 88%, and for pCO2 it was 72%. For pO2, interchangeability dropped to just 55%.²²

The pattern mirrors what we see in adults: acid-base parameters are reasonably interchangeable, while pO2 is not. In critically ill neonates, the agreement may be less reliable because hemodynamic instability and poor peripheral perfusion affect venous sampling sites disproportionately. Most neonatal and pediatric units treat venous pH and bicarbonate as acceptable for routine monitoring, but rely on arterial or arterialized capillary samples when precise oxygenation data is needed, particularly in infants with respiratory failure or congenital heart disease.⁹

A Quick Framework for Reading the VBG Printout

When a VBG lands in front of you, a structured approach keeps you from over-interpreting or under-interpreting the results:

pH: Subtract roughly 0.03 to 0.05 from the arterial value you would expect. A venous pH of 7.32 suggests an arterial pH around 7.35 to 7.37. A venous pH below 7.25 almost certainly reflects true arterial acidemia.
Bicarbonate: Treat the venous value as a close approximation. Differences of 1 to 2 mEq/L from the arterial value are typical and usually do not change management.
pCO2: Add about 3 to 8 mmHg to convert roughly to an arterial estimate, but recognize the spread can be wider. A normal venous pCO2 reliably rules out dangerous CO2 retention. A high venous pCO2 warrants an ABG if precise quantification matters.
pO2: Ignore it for the purpose of assessing oxygenation. Use pulse oximetry or an ABG instead.
Lactate: Venous lactate runs slightly higher than arterial. A normal value is reassuring. A value above 3.5 to 4 mmol/L almost certainly reflects true arterial hyperlactatemia. Values in the 2 to 3.5 range may benefit from a second look if the clinical picture is ambiguous.

Context always matters more than any single number. The VBG sits alongside the patient’s clinical presentation, vital signs, and other lab values. A venous pH of 7.28 in a patient who looks well and has chronic kidney disease means something very different from the same pH in a febrile patient with hypotension. The gas tells you the chemistry; you have to supply the clinical reasoning.

1
PubMed Central. Comparison of arterial and venous pH, bicarbonate, PCO2 and PO2 in initial emergency department assessment
2
PubMed Central. Correlation between peripheral venous and arterial blood gas measurements in patients admitted to the intensive care unit
3
PubMed Central. Correlation of Arterial and Venous pH and Bicarbonate in Patients With Renal Failure
4
PubMed. Point-of-care bedside gas analyzer: limited use of venous pCO2 in emergency patients
5
PubMed. Validation of venous pCO2 to screen for arterial hypercarbia in patients with chronic obstructive airways disease
6
Vidarbha Journal of Internal Medicine. Study of Correlation of Arterial Blood Gas Measurements with Peripheral Venous Blood Gas Values in Adult Patients Admitted in ICU in Tertiary Care Hospital in Central India
7
PubMed Central. The Role of Venous Blood Gas Analysis in Critical Care: A Narrative Review
8
Saudi Medical Journal. Comparison of arterial and venous blood gases analysis in patients with exacerbation of chronic obstructive pulmonary disease
9
PubMed Central. Usefulness of the VBG analysis as an option of ABG analysis in Congenital Heart Disease in pediatric age group
10
The American Journal of Emergency Medicine. Agreement between arterial and peripheral venous lactate levels in the ED: A systematic review
11
PubMed Central. Peripheral venous lactate levels substitute arterial lactate levels in the emergency department
12
PubMed. Arterial vs venous lactate: Correlation and predictive value of mortality of patients with sepsis during early resuscitation phase
13
Acute and Critical Care. Clinical applications of blood gas analysis: a comparative review of arterial and venous blood gas monitoring in critical care
14
PubMed. Optimizing oxygen delivery in the critically ill: the utility of lactate and central venous oxygen saturation (ScvO2) as a roadmap of resuscitation in shock
15
PubMed. Central venous oxygen saturation: analysis, clinical use and effects on mortality
16
PubMed. Diagnostic accuracy of venous blood gas electrolytes for identifying diabetic ketoacidosis in the emergency department
17
Biomedicine and Pharmacotherapy Journal. Comparison of Blood Gas and Acid–Base Measurement in Arterial and Venous Blood Samples in Patients with Diabetic Ketoacidosis
18
European Journal of Cardiovascular Medicine. Correlation of venous blood gas and pulse oximetry with arterial blood gas in patients with acute exacerbation of chronic obstructive pulmonary disease
19
PubMed. Systematic review and meta-analysis: value of venous blood gas in the diagnosis of acute exacerbation of chronic obstructive pulmonary disease in Emergency Department
20
PubMed. Blood gas and acid-base balance during cardiopulmonary resuscitation by intrathoracic and abdominal pressure variations
21
PubMed Central. Detection of preanalytical errors in arterial blood gas analysis
22
PubMed Central. Correlation and Interchangeability of Venous and Capillary Blood Gases in Non-Critically Ill Neonates