Rapid Gas Change During ECMO Linked to
Mortality Risk
Pediatric study shows plunging carbon dioxide worsens odds of survival
by Jennifer Hanawald
The dramatic drop in carbon dioxide that occurs at the initiation of oxygen treatment in critically ill infants and children may decrease their chances for survival, new research suggests.
In a retrospective study of children undergoing extracorporeal membrane oxygenation (ECMO), researchers from the Johns Hopkins University School of Medicine, in Baltimore, found that those who experienced the most extreme fall in carbon dioxide at the start of the therapy were less likely to live through it than those with more modest drops.
Melania Bembea, MD, assistant professor of pediatric anesthesiology at Hopkins, and leader of the study, said the investigation was the first to pinpoint the association between plummeting carbon dioxide and increased mortality. She emphasized that although the change in blood gas concentrations is multifactorial, the speed at which patients eliminate carbon dioxide from the body is modifiable.
“We have made changes in our practice,” Dr. Bembea told Anesthesiology News. “We have adjusted the settings that control the rate of carbon dioxide removal so that in patients with low pH and high carbon dioxide, we remove carbon dioxide much more slowly. This has been done in operating rooms, but not so much in ECMO practices that we’re aware of.”
Dr. Bembea and her colleagues started with the premise that abrupt removal of arterial carbon dioxide at the commencement of ECMO may cause sudden changes in cerebral blood flow and volume. These changes, the researchers speculated, might contribute to neurologic injury and increased mortality.
The researchers examined data from 201 pediatric patients who underwent ECMO at Johns Hopkins Hospital between 2002 and 2010. Blood gas data before and after the initiation of ECMO were available for 169 (84%) patients. Reasons for use of ECMO involved reversible life-threatening conditions including respiratory failure (51%), cardiac failure (23%) and sepsis (5%), and cardiopulmonary resuscitation (21%). The patients’ ages ranged from newborn to 16 years, with a median of 10 days.
Dr. Bembea’s group found that after adjusting for potential confounders such as age, use of epinephrine, volume of fluid administered and reason for and duration of ECMO, the magnitude of the decrease in carbon dioxide partial pressure (pCO2) was significantly associated with increased mortality. Children in the highest quartile of decrease—27 to 108 mm Hg—were about 50% more likely to die during treatment than those in the second and third quartiles (odds ratio, 1.51; 95% confidence interval, 1.07-2.13; P=0.019) (Figure).
The study also found a significant association between poor survival and ECMO in the presence of CPR and increasing patient age, which the authors stated was consistent with previous research. The researchers presented their findings at Pediatric Anesthesiology 2011, a meeting cosponsored by the Society for Pediatric Anesthesia and the American Academy of Pediatrics (abstract 10).
Joseph Savino, MD, professor of anesthesiology and critical care at the University of Pennsylvania School of Medicine, in Philadelphia, said the association between the pCO2 changes and survival deserved further investigation. But he stressed that a larger study with more prospective data was essential before making conclusions about pCO2 and outcome.
Dr. Savino said other factors could affect the outcome and should be included in the analysis. These factors included the types of ECMO used (venovenous or venoarterial), the degree of anticoagulation, patients’ temperatures and the form of blood gas analysis (alpha-stat or pH-stat) used.
“The authors are scraping the surface of an important clinical paradigm that is poorly understood and needs more rigorous investigation,” said Dr. Savino, who was not involved in the research. “I hope they continue and pursue such an issue in a multicenter approach that would provide larger numbers of patients and the opportunity for protocol-based care plans for entry into a national, and possibly international, database.”
However, Dr. Savino acknowledged that such data on a wide scale are difficult to gather without the existence of a centralized ECMO system. “Questions such as this could be more easily answered if the United States had a centralized care system for ECMO patients, akin to our solid-organ transplant system,” the United Network for Organ Sharing (UNOS), he said.
Mortality Risk
Pediatric study shows plunging carbon dioxide worsens odds of survival
by Jennifer Hanawald
The dramatic drop in carbon dioxide that occurs at the initiation of oxygen treatment in critically ill infants and children may decrease their chances for survival, new research suggests.
In a retrospective study of children undergoing extracorporeal membrane oxygenation (ECMO), researchers from the Johns Hopkins University School of Medicine, in Baltimore, found that those who experienced the most extreme fall in carbon dioxide at the start of the therapy were less likely to live through it than those with more modest drops.
Melania Bembea, MD, assistant professor of pediatric anesthesiology at Hopkins, and leader of the study, said the investigation was the first to pinpoint the association between plummeting carbon dioxide and increased mortality. She emphasized that although the change in blood gas concentrations is multifactorial, the speed at which patients eliminate carbon dioxide from the body is modifiable.
“We have made changes in our practice,” Dr. Bembea told Anesthesiology News. “We have adjusted the settings that control the rate of carbon dioxide removal so that in patients with low pH and high carbon dioxide, we remove carbon dioxide much more slowly. This has been done in operating rooms, but not so much in ECMO practices that we’re aware of.”
Dr. Bembea and her colleagues started with the premise that abrupt removal of arterial carbon dioxide at the commencement of ECMO may cause sudden changes in cerebral blood flow and volume. These changes, the researchers speculated, might contribute to neurologic injury and increased mortality.
The researchers examined data from 201 pediatric patients who underwent ECMO at Johns Hopkins Hospital between 2002 and 2010. Blood gas data before and after the initiation of ECMO were available for 169 (84%) patients. Reasons for use of ECMO involved reversible life-threatening conditions including respiratory failure (51%), cardiac failure (23%) and sepsis (5%), and cardiopulmonary resuscitation (21%). The patients’ ages ranged from newborn to 16 years, with a median of 10 days.
Dr. Bembea’s group found that after adjusting for potential confounders such as age, use of epinephrine, volume of fluid administered and reason for and duration of ECMO, the magnitude of the decrease in carbon dioxide partial pressure (pCO2) was significantly associated with increased mortality. Children in the highest quartile of decrease—27 to 108 mm Hg—were about 50% more likely to die during treatment than those in the second and third quartiles (odds ratio, 1.51; 95% confidence interval, 1.07-2.13; P=0.019) (Figure).
The study also found a significant association between poor survival and ECMO in the presence of CPR and increasing patient age, which the authors stated was consistent with previous research. The researchers presented their findings at Pediatric Anesthesiology 2011, a meeting cosponsored by the Society for Pediatric Anesthesia and the American Academy of Pediatrics (abstract 10).
Joseph Savino, MD, professor of anesthesiology and critical care at the University of Pennsylvania School of Medicine, in Philadelphia, said the association between the pCO2 changes and survival deserved further investigation. But he stressed that a larger study with more prospective data was essential before making conclusions about pCO2 and outcome.
Dr. Savino said other factors could affect the outcome and should be included in the analysis. These factors included the types of ECMO used (venovenous or venoarterial), the degree of anticoagulation, patients’ temperatures and the form of blood gas analysis (alpha-stat or pH-stat) used.
“The authors are scraping the surface of an important clinical paradigm that is poorly understood and needs more rigorous investigation,” said Dr. Savino, who was not involved in the research. “I hope they continue and pursue such an issue in a multicenter approach that would provide larger numbers of patients and the opportunity for protocol-based care plans for entry into a national, and possibly international, database.”
However, Dr. Savino acknowledged that such data on a wide scale are difficult to gather without the existence of a centralized ECMO system. “Questions such as this could be more easily answered if the United States had a centralized care system for ECMO patients, akin to our solid-organ transplant system,” the United Network for Organ Sharing (UNOS), he said.