Resuscitation and arterial blood gas abnormalities during prolonged cardiopulmonary resuscitation

Arthur B. Sanders, Gordon A. Ewy, Tracy V. Taft

Research output: Contribution to journalArticlepeer-review

43 Scopus citations


A study was undertaken to determine the pattern of arterial blood gas (ABG) concentration in the canine model undergoing prolonged cardiopulmonary resuscitation (CPR) from fibrillatory arrest, and to determine the importance of acid base abnormalities in predicting resuscitation. Ventricular fibrillation was induced electrically in 12 dogs. CPR was begun at 3 minutes and continued for 27 minutes, at which time the dogs were defibrillated. ABG samples were taken at 0, 8, 18, and 28 minutes of ventricular fibrillation. Seven of the 12 dogs were resuscitated successfully. There was no difference in pH, PCO2, or PO2 between the survivors and nonsurvivors at any of the points measured. A pattern of pH and PCO2 abnormalities was noted in each dog over 30 minutes. Each developed a respiratory alkalosis that peaked at 8 minutes. During the next 22 minutes the pH gradually declined. This combination of respiratory alkalosis and metabolic acidosis resulted in normalization of the pH at about 18 minutes of fibrillation. We concluded that when adequate ventilation is provided in the canine model undergoing CPR, significant arterial acidemia does not occur for at least 18 minutes. Further, acid base abnormalities did not correlate with successful resuscitation.

Original languageEnglish (US)
Pages (from-to)676-679
Number of pages4
JournalAnnals of emergency medicine
Issue number9 PART 1
StatePublished - Sep 1984


  • CPR, arterial blood gases
  • acidosis, in cardiac arrest
  • arterial blood gases, in CPR
  • cardiac arrest, acidosis
  • resuscitation, from cardiac arrest, arterial blood gases

ASJC Scopus subject areas

  • Emergency Medicine


Dive into the research topics of 'Resuscitation and arterial blood gas abnormalities during prolonged cardiopulmonary resuscitation'. Together they form a unique fingerprint.

Cite this