May 26 , 2005

In This Issue . . .

Arterial Blood Gas Analysis and the Management of Diabetic Ketoacidosis
By Richard A. Pizzi

The potential for a full-blown hyperglycemic crisis substantiates diabetic ketoacidosis’ (DKA) reputation as perhaps the most serious acute complication of diabetes. Even if managed properly by clinicians, a prompt and accurate DKA diagnosis is imperative, since delayed treatment can lead to coma and even death. Many clinicians refer to the series of essential diagnostic tests in the American Diabetes Association’s (ADA) emergency care standards for evaluation of potential DKA patients. One of the recommended assessment tools is arterial blood gas (ABG) analysis, which measures blood pH. Recently, however, new research suggests that ABG analysis may not be as critical as once thought for the management of DKA, which could mean a change in the ADA’s DKA standards. In this issue, Strategies examines some recent research and assesses its meaning for the future of ABG analysis in the emergency room.

Most common in patients with type 1 diabetes, DKA is caused by an increase in blood glucose levels, which stimulates dehydration and prompts the body to burn excessive amounts of fat for energy. This fat metabolism produces ketones that accumulate in the blood and also appear in the urine, leading to a condition known as acidemia. Without intervention, the condition can be fatal, as acidemia leads to cell damage.

According to the ADA, an initial laboratory evaluation of patients with suspected DKA should include tests for plasma glucose, serum and urine ketones, electrolytes, blood urea nitrogen/creatinine, complete blood count, and ABG analysis. Because the severity of acidemia is one of the diagnostic criteria essential to categorizing the degree of DKA, blood pH levels help determine how a patient will be treated. In addition to determining blood pH, ABG analysis also measures the amounts of oxygen and carbon dioxide in the blood, indicating how effectively the lungs are delivering oxygen to the blood and how efficiently they are eliminating carbon dioxide. For the purposes of confirming DKA, however, the acid-base component of the ABG analysis is critical. When the body’s carbonic acid-bicarbonate buffer system is in equilibrium, blood pH is maintained at a level of 7.35–7.45. When blood pH falls below 7.30, however, a diagnosis of DKA may be suspected or confirmed, depending upon other factors.

Finding a Substitute for ABG Analysis

While ABG analysis undoubtedly gives an accurate assessment of blood pH levels, a recent paper by researchers with the Department of Emergency Medicine at the University of Pennsylvania (Philadelphia) published in the Annals of Emergency Medicine questioned the necessity of the test in the management of patients with suspected DKA.

After reviewing four research studies of ABG analysis, the researchers found that the values obtained from an ABG pH test might not alter patient management and can be replaced with the results of a venous pH, a diagnostic test that the ADA also considers to be a reliable indicator of acidemia. In fact, the authors suggested that the ABG test may be completely unnecessary.

One of the authors of the paper, Esther H. Chen, MD, Assistant Professor of Emergency Medicine, University of Pennsylvania, said that her research was first inspired by discussions with members of the admitting medical team at her hospital, who questioned why an initial ABG analysis was not completed on some admitted patients. Through managing many diabetic patients and its complications daily, she said that “arterial blood sampling is another invasive procedure that the patient has to endure in addition to IV insertion,” adding that arterial blood sampling is “often more painful than venipunctures, arterial bleeding is more profuse, and hematomas can more easily develop post-procedure.”

Chen wondered why ABG analysis could not be eliminated in favor of venous pH testing, which can be performed during IV insertion, pointing out that her review of the medical literature supports this assertion. She discovered four peer-reviewed research articles written in the last 20 years that dealt specifically with ABG analysis. All four suggested that the blood pH results obtained through ABG testing were not substantially different from the results of venous pH analysis, or in one case finger capillary blood pH analysis. One 2003 study found that ABG results rarely influenced the final diagnosis, treatment plan, or final disposition in patients with suspected DKA.

In all the studies Chen examined, venous pH correlated extremely well with pH determined by ABG. Venous pH results tended to be slightly lower than ABG results, but not by much—the largest average difference in blood pH in all the studies was approximately 0.05—and these small differences are clinically insignificant, Chen said. Because ABG values might not alter patient management, and venous pH appears to be as reliable as ABG pH in determining acidemia, Chen concluded that “it would be within the standard of care to use venous blood gas sampling in place of ABG sampling for the emergency management of DKA.”

A Role for Stringent Standards

However, some experts believe that ADA’s Standards of Care need to remain stringent to cover the worst case scenarios. “While there probably wouldn’t be any worse outcome most of the time in DKA without an initial ABG…it is somewhat ingrained into clinical practice,” said Sue Kirkman, MD, Associate Professor of Medicine, Indiana University (Indianapolis), adding that most physicians might not be familiar with measuring venous pH. There are recommendations for giving bicarbonate that are based on the ABG, not on venous pH, and while “these cases are pretty rare, one could not be 100% sure that venous pH would correlate with arterial pH in the very rare but more deadly cases of severe acidosis,” she explained.

While Chen’s conclusions may conflict with the ADA’s endorsement of an initial ABG analysis for patients with suspected DKA, the ADA itself is not opposed to changing its position statement on the subject. Nathaniel Clark, MD, ADA’s Vice President of Clinical Affairs, explained that the association’s Standards of Care are reviewed annually. Chen’s survey of the literature on ABG analysis will be scrutinized carefully when the standards for this specific topic come up for review later this year. “It is an excellent article and raises a very important point,” Clark said. “We will take the article into consideration in our updating of the Standards of Care.”

For more information:

This Annals of Emergency Medicine study is available online to subscribers at

Richard A. Pizzi is a freelance writer based in Portland, Maine.

Page Access: