Monitoring Performance of Cardiopulmonary Resuscitation

Until recently, palpation of the carotid or femoral pulse and observation of papillary size were the standard and very indirect measures for assessing the apparent adequacy of CPR. Obviously, a palpable large-artery pulse indicates only the transmission of a pressure wave into the arterial tree during chest compression and provides no objective evidence of the effectiveness of cardiac output. Initial papillary size and changes during CPR are of some prognostic value.^[63] Pupils that are persistently contracted or initially dilated but subsequently contracting are associated with a greater likelihood of successful resuscitation and neurologic recovery than persistently dilated or subsequently dilating pupils are.^[65]

Monitoring systemic arterial pressure directly, as can usually be accomplished during in-hospital CPR, is helpful in optimizing the chest compression rate and depth.^[66] Aortic diastolic pressure in particular, an index of coronary perfusion pressure, should be monitored whenever possible and optimized with appropriate changes in manual compression technique and early and repeated injection of epinephrine and vasopressin before restoration of spontaneous circulation.

In 1978 Kalenda^[67] described the use of capnography as a guide to the effectiveness of external chest compressions. He demonstrated the value of PETCO₂ monitoring in three patients in cardiac arrest and confirmed changes in PETCO₂ with changes in external chest compression technique. He also observed a rapid increase in PETCO₂ with restoration of spontaneous circulation. Kalenda proposed that when ventilation is constant, as during controlled mechanical ventilation, "the expired CO₂ is a precise and continuous mirror of lung perfusion and hence of cardiac output."^[67]

Despite the potential clinical importance of Kalenda's well-documented observations and conclusion, it was several years before PETCO₂ monitoring during cardiac arrest and CPR was explored further. In 1984, using a porcine model of cardiac arrest, Grundler and colleagues^[68] demonstrated a rapid decrease in PETCO₂ with the onset of arrest and its immediate increase with resuscitation. Based on these preliminary experimental observations, this group suggested that PETCO₂ monitoring may be a relatively simple and noninvasive guide to advanced life support interventions during CPR.^{[68] [69]} Severe reductions in pulmonary blood flow during cardiac arrest and CPR and therefore an acute failure in CO₂ delivery to the lungs explained the observed very low PETCO₂ and the increased venous CO₂ pressure (PVCO₂ ) (see the discussion of sodium bicarbonate later).^[70] Subsequent studies in dogs provided confirmation of the possible utility of PETCO₂ in that expired PVCO₂ correlated very closely with coronary perfusion pressure, which itself was shown to be a determinant of survival from cardiac arrest in this model.^{[71] [72]} In a porcine model of VF, onset of VF was associated with a rapid decline in PETCO₂ from a mean (±SEM) of 4.0% ± 0.2% to less than 0.7% ± 0.2%.^[73] With external chest compression and constant-volume mechanical ventilation, PETCO₂ increased to 1.9% ± 0.3%, with an immediate increase to 4.9% ± 0.3% after successful defibrillation following 12 minutes of CPR. Close correlation was found between changes in cardiac output and PETCO₂ during both closed-chest and open-chest CPR.^[73]

PETCO₂ monitoring has been reported in arrested humans undergoing CPR.^{[74] [75] [76] [77]} In a study of 10 patients in whom spontaneous circulation returned, PETCO₂ increased immediately on restoration of spontaneous circulation, from a mean (±SD) of 1.7% ± 0.6% to 4.6% ± 1.4%. ^[74] The rapid increase in PETCO₂ was often the first clinical evidence of resumption of spontaneous circulation.

In 10 critically ill patients, 9 of whom were in septic or cardiogenic shock at the onset of cardiac arrest, PETCO₂ decreased from a mean (±SD) of 1.4% ± 0.9% to 0.4% ± 0.4% shortly after the onset of cardiac arrest. ^[76] The very low prearrest value is indicative of low cardiac output and therefore low pulmonary blood flow. During external chest compression, this value increased to 1.0% ± 0.5%. In seven patients in whom circulation was restored, PETCO₂ rapidly increased from 1.3% ± 0.5% to 3.7% ± 2.1% and then declined to a stable volume of 2.4% ± 1.8% 4 minutes after resumption of spontaneous circulation. In patients in whom resuscitation failed to restore spontaneous circulation, PETCO₂ remained at 0.7% ± 0.4%.^[76]

Evidence is increasing that PETCO₂ measurements obtained during cardiac arrest and CPR may have predictive value relative to the likelihood that spontaneous circulation will be restored. In one study, patients who

2928

The major determinants of PETCO₂ are CO₂ production, alveolar ventilation, and pulmonary blood flow.^{[81] [82] [83] [84]} In the presence of constant-volume ventilation and apparently unchanged CO₂ production immediately after cardiac arrest, the changes in PETCO₂ observed both experimentally and clinically during cardiac arrest and CPR and resumption of spontaneous circulatory function indicate that PETCO₂ reflects pulmonary blood flow and therefore cardiac output.^{[74] [75] [76] [77] [78]} Monitoring of both systemic arterial pressure by arterial catheter and PETCO₂ with controlled ventilation should provide optimal hemodynamic assessment of the adequacy of the resuscitative effort and response to interventions such as changes in depth, rate, and location of manual chest compressions, as well as response to drugs such as epinephrine and vasopressin.^[66]