Management of cardiopulmonary bypass

INTRODUCTION — Cardiopulmonary bypass (CPB) is a form of extracorporeal circulation in which the patient's blood is diverted from the heart and lungs and rerouted outside of the body. The normal physiologic functions of the heart and lungs, including circulation of blood, oxygenation, and ventilation, are temporarily taken over by the CPB machine. In most cases, the heart is also separated from the circulation (eg, aortic cross-clamping) and cardioplegia solution is administered to allow the cardiac surgeon to operate on a nonbeating heart in a field largely devoid of blood, while other end organs remain adequately oxygenated and perfused.

This topic will discuss routine management of CPB. Certain types of patients who require unique management techniques during CPB are discussed separately. (See "Management of special populations during cardiac surgery with cardiopulmonary bypass".)

Some surgical procedures require temporary interruption of cerebral and/or systemic blood flow (eg, repair of portions of the ascending aorta or aortic arch). Elective circulatory arrest is accomplished during a period of deep hypothermic circulatory arrest (DHCA) after cooling with the aid of CPB, traditionally to 16 to 18°C. Alternatives to DHCA include moderate hypothermic circulatory arrest (eg, 22 to 24°C) with antegrade cerebral perfusion through axillary artery cannulation. This technique reduces operative time and appears to have equivalent safety [1]. Anesthetic management during and after DHCA is discussed separately. (See "Anesthesia for aortic surgery with hypothermia and elective circulatory arrest in adult patients", section on 'Cardiopulmonary bypass with DHCA'.)

Preparations for initiation of CPB, the process of weaning from CPB, and problems that are commonly encountered in the post-bypass period are discussed separately:

●(See "Initiation of cardiopulmonary bypass".)

●(See "Weaning from cardiopulmonary bypass".)

●(See "Intraoperative problems after cardiopulmonary bypass".)

GENERAL PRINCIPLES

Equipment and physiology — Components of the CPB machine include pumps, tubing, filters, reservoir, and gas (oxygenator) and heat exchange units (figure 1A) [2-5]. Modern CPB machines are also equipped with systems that continuously monitor line or circuit pressure, temperature, and blood parameters (eg, oxygen saturation and estimated delivery, blood gases, hemoglobin, potassium), as well as safety features such as air and fluid level detection systems and a blood filter in the arterial line [3].

During CPB, venous blood is drained from the right atrium (RA; or both superior and inferior vena cavae) and is diverted through the venous line of the CPB circuit into a venous reservoir (figure 1A-B). CPB machines are typically equipped with vacuum-assisted technology that facilitates drainage to maintain a bloodless surgical field and allow use of smaller venous cannulae and reduced CPB circuit volumes. The arterial pump functions as an artificial heart by withdrawing blood from this reservoir and propelling it through a heat exchanger, an artificial lung (oxygenator or gas exchanger), and finally an arterial line filter. The blood is then returned to the patient via an arterial cannula positioned in the ascending aorta or other major artery. Additional CPB circuit pumps or other components are employed as needed to suction blood from the surgical field, deliver cardioplegia solution to produce cardiac electromechanical silence, decompress heart chambers via a vent, and remove fluid (ultrafiltration). Thus, the cardiopulmonary machine temporarily takes over the functions of the heart, lungs, and, to a lesser extent, the kidneys.

Blood contact with nonendothelial surfaces of the CPB circuit induces an intense inflammatory response [2,6]. This results in platelet activation, initiation of the coagulation cascade, and decreased levels of circulating coagulation factors. Endothelial cells and leukocytes are activated, releasing mediators that may contribute to capillary leakage and tissue edema. Many of the challenges encountered during weaning from CPB and the postbypass period (eg, myocardial dysfunction, vasodilation, bleeding) are thought to be partially attributable to this inflammatory process [7-10]. Also, hemodilution due to the volume of the priming solution for the CPB circuit (typically 1 to 2 L of a balanced crystalloid solution) may result in temporary or persistent anemia and dilutional coagulopathy [11-13].

Protocols and standards — Surgical procedures requiring CPB follow a standard sequence of events that includes priming and testing of the CPB circuit, anticoagulation, vascular cannulation, initiation and maintenance of CPB, and finally weaning and termination of CPB. Cardiac arrest (standstill) with myocardial protection followed by myocardial reperfusion is employed when a stationary heart and bloodless field are desired.

Preparations for initiation of CPB and checklists to ensure readiness for initiation are described separately (table 1) [3]. (See "Initiation of cardiopulmonary bypass".)

Established protocols for management of CPB use parameters approximating normal physiology (table 2) [3,14-16]:

●In adults, the target flow rate during CPB is 2.2 to 2.4 L/minute/m² for normothermic patients to approximate a normal cardiac index; cardiac index is appropriately decreased if hypothermia is induced. (See 'Management during cooling and hypothermia' below.)

●Mean arterial pressure (MAP) is generally targeted at ≥65 mmHg, but the target may be higher in older patients and those with cerebrovascular disease [4,14,15,17]. MAP should not exceed 100 mmHg.

●Adequacy of end-organ perfusion is determined by arterial blood gas analysis and the mixed venous oxygen saturation (SvO₂), which is continuously monitored and maintained ≥75 percent throughout CPB. Arterial blood gases, base deficit, and lactate levels are intermittently checked (approximately every 30 minutes).

Preparations for separation from CPB and checklists to ensure readiness for weaning of circulatory support are described separately (table 3) [3]. (See "Weaning from cardiopulmonary bypass", section on 'Preparation for weaning' and "Weaning from cardiopulmonary bypass", section on 'The weaning process'.)

Goals — Goals during CPB include maintenance of general anesthesia, anticoagulation, and parameters that approximate normal physiology for optimal end-organ function (table 2) [3,14,15].

MONITORING

Standard monitoring — We agree with recommendations made in the 2019 European Association for Cardio-Thoracic Surgery (EACTS)/European Association of Cardiothoracic Anaesthesiology (EACTA)/European Board of Cardiovascular Perfusion (EBCP) guidelines regarding standards of monitoring during CPB in adult cardiac surgical patients [3]. These include confirmation of adequate pump flow with continuous flow measurements on the arterial line, as well as continuous monitoring of arterial line pressure in the CPB circuit (pre-oxygenator and post-oxygenator), arterial blood gases, mixed venous oxygen saturation (SvO₂), oxygenator arterial outlet temperature, and, when available, exhaust concentrations of volatile anesthetic agents. Other necessary monitoring includes intermittent or continuous monitoring of hemoglobin or hematocrit and potassium, as well as frequent (approximately every 30 minutes) monitoring of glucose, electrolytes, lactate, arterial blood gases, and pH, as well as urine output. To ensure safety of institutional CPB programs, these EACTS/EACTA/EBCP guidelines also recommend recording and timely analysis of all adverse events related to CPB practice [3].

Neuromonitoring modalities — Some centers use one or more neuromonitoring modalities for patients at high risk for adverse neurologic outcomes due to patient-related factors such as significant cerebrovascular disease or procedure-related factors such as the need for deep hypothermia and elective circulatory arrest. (See "Management of special populations during cardiac surgery with cardiopulmonary bypass", section on 'Cerebrovascular disease' and "Anesthesia for aortic surgery with hypothermia and elective circulatory arrest in adult patients", section on 'Brain monitors'.)

●Cerebral oximetry – The use of intraoperative near-infrared spectroscopy (NIRS) cerebral oximetry monitoring and maintenance of regional cerebral oxygen saturation (rSO₂) within 20 percent of baseline has been advocated for patients at high risk for adverse neurologic outcomes, including those with known cerebrovascular disease [18-29]. Notably, baseline preoperative rSO₂ varies considerably, but cerebral oximetry can be used as a trend monitor [28]. In a 2017 meta-analysis that included 953 patients undergoing cardiac surgery, the overall pooled mean baseline rSO₂ was 66.4±7.8 percent [25].

When available, NIRS is used to detect and mitigate consequences of acute decreases in rSO₂ during CPB [26,27]. Interventions to treat decreases in rSO₂ that are >20 percent below baseline include (algorithm 1) [18,20,26,29]:

•Increasing cardiac output by increasing pump flow.

•Increasing mean arterial pressure (MAP) by administering a vasopressor. However, a 2019 randomized trial in 197 patients did not demonstrate improvement in rSO₂ when patients were managed to achieve a higher target MAP. In this study, when vasopressors were used during fixed CPB flow rate to increase MAP to 70 to 80 mmHg as compared with a lower MAP target of 40 to 50 mmHg, patients in the higher MAP group had lower rSO₂ as well as more frequent and pronounced episodes of cerebral desaturation [30].

•Increasing arterial oxygen tension (PaO₂) by increasing the fraction of inspired oxygen (FiO₂) in the gas delivered to the oxygenator. Although there have been theoretical concerns that high oxygen concentrations might worsen ischemia-reperfusion injury, a randomized trial in 100 older patients undergoing cardiac surgery noted similar postoperative neurocognitive scores in patients exposed to hyperoxia versus normoxia during CPB [31].

•Increasing arterial carbon dioxide tension (PaCO₂) by decreasing the oxygenator fresh gas flow "sweep speed."

•Decreasing the cerebral metabolic rate of oxygen consumption (CMRO₂) by deepening anesthesia.

•Increasing blood oxygen-carrying capacity with red blood cell (RBC) transfusion (if anemic eg, if hemoglobin is <7.5 g/dL).

No studies have demonstrated that monitoring rSO₂ values can prevent stroke by increasing or restoring low rSO₂ values in patients with or without significant cerebrovascular disease [27,29,32-37]. Evidence that monitoring and manipulating rSO₂ values improved other neurologic or general postoperative outcomes is also inconsistent [26,27,29,38-42]. A 2018 meta-analysis of randomized trials noted that use of intraoperative NIRS in 312 patients undergoing cardiac surgery resulted in less postoperative cognitive dysfunction (POCD) at one week compared with 297 patients who did not receive NIRS-guided interventions (risk ratio [RR] 0.55; 95% CI 0.36-0.86; four trials) [39]. However, a subsequently published study noted that neither the duration nor the extent of intraoperative decreases in rSO₂ values affected incidence of POCD at hospital discharge or delayed neurocognitive recovery after three postoperative months [40]. Another subsequently published randomized trial in 134 patients noted that targeted therapy to optimize cerebral oxygenation was associated with less change in postoperative memory scores [43].

Use of cerebral oximetry to monitor for adequate cerebral perfusion during deep hypothermia with elective circulatory arrest (DHCA) alone or DHCA with retrograde cerebral perfusion or selective antegrade cerebral perfusion (SACP) is discussed in a separate topic. (See "Anesthesia for aortic surgery with hypothermia and elective circulatory arrest in adult patients", section on 'Cerebral oximetry'.)

●Other neuromonitors – Some studies have employed other neuromonitoring techniques during CPB. For example, raw waveform or processed electroencephalography (EEG) monitoring has been used to detect acute development of hypoxemia or hypercapnia [44], or to monitor EEG burst suppression [45]. A 2022 retrospective study of 14,086/ 42 ,932 patients in the Society of Thoracic Surgeons Adult Cardiac Surgery Database noted that the 14,086 who had processed EEG monitoring did not have better or worse neurologic outcomes including stroke or postoperative delirium [46]. However, in a retrospective single-institution study of 2454 consecutive patients in a single institution who had processed EEG monitoring of bispectral index (BIS) values, postoperative stroke in 58 patients (2.4 percent) was associated with cumulative duration of BIS values <25 (odds ratio [OR] 1.45. 95% ci 1.12-1.87) and with duration of MAP <60 mmHg (OR 1.52, 95% CI 1.02-2.27 [42].

Also, transcranial Doppler (TCD) technology may be used to improve real-time feedback regarding cerebral autoregulation and detection of cerebral embolism [47-49].

Urine output — Maintenance of renal blood flow is achieved by maintaining adequate CPB pump flow throughout CPB to minimize the risk of acute kidney injury (AKI). (See "Management of special populations during cardiac surgery with cardiopulmonary bypass", section on 'Chronic kidney disease and renal risk mitigation'.)

If oliguria develops during CPB (ie, urine output <0.5 mL/kg per hour), we check the bladder catheter for kinking or disconnection, as well as check the bladder itself by palpation or ultrasound performed on the surgical field. Aortic dissection as a possible cause of oliguria is ruled out with transesophageal echocardiography (TEE). Adequacy of pump flow, MAP, SvO₂, and arterial blood gases are checked and closely monitored (table 2).

There are no data supporting the efficacy of adding mannitol or other pharmacologic agents (eg, furosemide, or low "renal dose" dopamine infusion) to the CPB prime to prevent AKI after CPB [50,51].

OXYGENATION, VENTILATION, AND ARTERIAL BLOOD GASES — Arterial pO₂ is maintained at 150 to 250 mmHg during CPB [14,15]. A system to continuously monitor trends in arterial blood gas parameters is located in the arterial line or on a small shunt attached to the arterial line of the CPB circuit, and a continuous venous oximeter is located in the venous return line. Arterial blood gas values are checked in the laboratory or by point-of-care testing approximately every 30 minutes, which also allows intermittent recalibration of the continuous blood gas monitor in the arterial line. The algorithm shows details regarding managing inadequate oxygenation during CPB (algorithm 2).

More specific oxygenation strategies (eg, targeting hyperoxia) have not been shown to be clinically beneficial, and can lead to alveolar collapse and generation of oxygen radicals which might exacerbate ischemia-reperfusion injury after CPB [3,52]. A 2018 systematic review of 12 randomized trials noted scant evidence of differences in outcome when a hyperoxic rather than a normoxic oxygenation strategy was used during cardiac surgery, but these and subsequent studies were small and heterogenous [53,54].

Alpha-stat management of arterial blood gases without temperature correction is employed to maintain a normal range for pCO₂ (35 to 45 mmHg [4.7 to 6 kPa]) and pH (7.35 to 7.45) [3,14,15,55]. Maintaining PaCO₂ and pH within this physiologic range during CPB is important to preserve cerebral autoregulation because hypocarbia decreases cerebral blood flow. There is no evidence that induced hypercapnia improves cerebral oxygenation during bypass, but it may increase pulmonary artery pressure (PAP) in the post-bypass period [56].

Ventilation of the lungs during CPB has not been demonstrated to improve pulmonary function and may increase technical difficulty for the surgeon [57-60]. Meta-analyses of trials examining the use of mechanical ventilation or maintenance of 5 to 15 cmH₂O continuous positive airway pressure (CPAP) during CPB have noted lower alveolar to arterial oxygen tension differences (AaDO₂) in the immediate post-bypass period compared with no ventilator or CPAP use, but no effect on postoperative pulmonary complications [58,59].

PUMP FLOW AND MIXED VENOUS OXYGEN SATURATION — CPB flow rates are set at 2.2 to 2.4 L/minute per m² in a normothermic patient to provide adequate blood flow for optimal perfusion of the brain and other end organs. These rates may be slightly decreased if hypothermia is employed. Conversely, especially with a lower hemoglobin value, it is important that the rate of oxygen delivery (DO₂) is ≥280 mL/min/m² is achieved to lower incidence of acute kidney injury (AKI) (table 4) [61]. (See "Management of special populations during cardiac surgery with cardiopulmonary bypass", section on 'Preexisting chronic kidney disease'.)

Acute decreases in mean arterial pressure (MAP) or increases in central venous pressure (CVP) may indicate acute reduction in venous return as the surgeon lifts the heart (causing a reduction in flow), malpositioning or kinking of the arterial or venous cannula, obstruction to blood flow by an air lock, or acute blood loss. With severe reduction in venous return, the perfusionist may need to administer volume into the CPB reservoir or to reduce arterial flow. Persistent reductions in arterial line flow due to venous return must be urgently addressed by identifying and correcting the cause.

Mixed venous oxygen saturation (SvO₂) is maintained ≥75 percent throughout CPB as a monitor of adequacy of global tissue perfusion. Persistent SvO₂ values <75 percent may indicate inadequate oxygen delivery and are associated with worse outcomes including postoperative delirium and decreased long-term survival [62,63]. Also, base deficit and lactate values, if available, are measured when arterial blood gases are obtained approximately every 30 minutes. Although there are multiple reasons for rising lactate values during bypass, anaerobic metabolism at the cellular level due to poor tissue perfusion and inadequate tissue oxygen delivery is a possibility, particularly if elevated lactate levels are sustained or associated with SvO₂ <70 percent [64-66].

During CPB, oxygen delivery may be improved by increasing pump flow or by increasing hemoglobin to treat SvO₂ <75 percent, base deficit <–5, or lactate levels >4 mEq/L. Although it is common practice to administer sodium bicarbonate if the base deficit is <–5 or lactate level is >4 mEq/L, excessive sodium bicarbonate administration can cause postoperative hypernatremia [66,67]. Persistent lactic acidosis in the postoperative period may be due to other factors (eg, the beta-adrenergic metabolic effects of epinephrine infusion) [68].

MEAN ARTERIAL PRESSURE

Mean arterial pressure targets — If CPB flow rates are acceptable (see 'Pump flow and mixed venous oxygen saturation' above), mean arterial pressure (MAP) is generally targeted at 50 to 80 mmHg [3,4,14,15,17,69,70]. A higher target may be selected for older patients, particularly those with cerebrovascular or peripheral vascular disease (see "Management of special populations during cardiac surgery with cardiopulmonary bypass", section on 'Cerebrovascular disease'). However, a 2022 systematic review noted that use of a MAP target ≥65 mmHg during CPB resulted in little to no difference in patient outcomes including acute kidney injury (AKI), cognitive deterioration, and mortality compared with a lower MAP target ≤65 mmHg (three trials; 737 participants) [71]. Of note, professional society guidelines recommend against using vasopressors to increase MAP to values >80 mmHg during CPB, and MAP should never exceed 100 mmHg [3].

Episodes of low or high blood pressure during CPB have been associated with risk for cerebral and other adverse outcomes, although a causative relationship has not been established [72,73].

●A randomized controlled trial in 197 patients did not show reduction in the number or volume of cerebral infarcts in those who received vasopressor therapy to maintain a MAP target at near physiological values (70 to 80 mmHg), compared with those who had a lower target MAP (40 to 50 mmHg) [69].

●In one large retrospective study in nearly 7500 patients, postoperative stroke was associated with sustained periods with MAP <65 mmHg, with an adjusted odds ratio (OR) of 1.13 for every 10-minute period that MAP was 55 to 64 mmHg during and after CPB (95% CI 1.05-1.21), and an adjusted OR of 1.16 for every 10-minute period that MAP was <55 mmHg (95% CI 1.08-1.23) [70]. Other factors associated with stroke in that study were older age, history of hypertension, combined coronary artery bypass grafting (CABG) plus valve procedures, prolonged duration of CPB, emergency surgery, and new onset of postoperative atrial fibrillation.

●In another large retrospective study in nearly 5000 patients, episodes of hypotension <65 mmHg occurring for >10 minutes before, during, or after CPB were associated with a composite outcome that included stroke, AKI, or mortality after cardiac surgery (OR 1.06, 95% CI 1.03-1.10) [74].

Impaired cerebral autoregulation is common, occurring in nearly one-third of patients during CPB in one retrospective study [75]. Impaired autoregulation was associated with small cerebral vessel disease (identified with brain magnetic resonance imaging) in this study, but was not associated with large vessel disease (identified by transcranial Doppler). Other studies have noted that identifying an individual patient's lower limit of cerebral autoregulation with either near-infrared spectroscopy (NIRS) or transcranial Doppler (TCD) technology, then maintaining MAP above this lower limit may decrease the incidence of postoperative delirium or other adverse neurologic outcomes [30,47,76,77]. However, use of these technologies to assess cerebral autoregulation is not widely practiced.

During attempts to increase MAP, it is particularly important to ensure that pump flow is adequate and that there is evidence of satisfactory end-organ perfusion (see 'Pump flow and mixed venous oxygen saturation' above). The perfusionist may increase or decrease the arterial pressure by increasing or decreasing CPB pump flow. It is particularly important to maintain clear communication between the anesthesiologist and perfusionist to avoid concurrent administration of vasopressors to increase the arterial pressure by both clinicians. It is also common that MAP measured in a radial artery catheter underestimates the true aortic pressure during CPB. If it is suspected that the radial artery pressure is providing a false measurement, measuring the arterial pressure from the brachial artery, femoral artery, or even directly in the aorta can determine the true arterial pressure.

Management of hypotension

●Moderate hypotension – In the context of acceptable CPB flow rates, if MAP falls below the target range, the perfusionist may increase the pump flow (equivalent to increasing cardiac output), particularly if it is <2.4 L/minute/m². If hypotension persists after increasing pump flow, a vasopressor can be administered as an intravenous (IV) bolus or via continuous infusion. In many institutions, small bolus doses of phenylephrine (eg, 40 to 100 mcg) are administered directly into the CPB reservoir to treat hypotension. Infusions of phenylephrine at 10 to 200 mcg/minute, vasopressin at 0.04 units/minute, or norepinephrine at 0.02 to 0.06 mcg/kg/minute are also commonly employed (table 5).

•Vasoplegia – In rare instances, severe systemic vasodilation develops that is unresponsive to potent vasopressors and is characterized by markedly decreased systemic vascular resistance (SVR) and low MAP during and after CPB (vasoplegia syndrome) [3,78-83]. Risk factors include preoperative use of agents such as angiotensin-converting enzyme (ACE) inhibitors, heparin, or calcium channel blockers, as well as prebypass hemodynamic instability [3,79,80,82-84]. (See "Postoperative complications among patients undergoing cardiac surgery", section on 'Vasodilatory shock' and "Intraoperative problems after cardiopulmonary bypass", section on 'Vasoplegia'.)

Prior to treating low blood pressure near the end of the period on CPB, it is important to verify that the radial arterial pressure is not markedly underestimating the central aortic pressure [85-87]. A significant central to peripheral pressure gradient associated with rewarming at the end of CPB is often present during cardiac surgery (figure 2). Connection of a pressure transducer to the side port of the aortic cannula after termination of CPB, or use of a femoral arterial catheter inserted by the surgeon on the field will provide a more accurate measurement of central aortic pressure.

Although no optimal approach has been established for selection of vasopressors to address vasoplegia during CPB, observations in the setting of distributive shock suggest that administration of vasopressin or a combination of vasopressin with norepinephrine or phenylephrine is associated with lower rates of atrial fibrillation compared to administration of norepinephrine alone (table 6) [88,89]. If these agents are ineffective, vasoplegic syndrome is confirmed, and methylene blue 1 to 2 mg/kg IV over 20 minutes may be administered to reduce resistance vessel responsiveness to nitric oxide [3,83,90]. Notably, methylene blue should be avoided in patients receiving chronic serotonergic therapy (eg, fluoxetine) due to the risk of serotonin syndrome [91], and may interfere with monitors that employ oximetry to measure oxygen saturation (eg, pulse oximetry and cerebral oximetry) [78,92]. Other agents used for treatment of refractory vasoplegia include angiotensin II, vitamin B12 (hydroxocobalamin), and vitamin C [3,83]. Refractory vasoplegic syndrome during the postbypass period, persistent to the end of surgery, and extending into the postoperative period, is associated with longer hospital stays, and higher mortality [79,93,94].

Management of hypertension — If MAP increases to >90 mmHg during CPB, initial treatment is to ensure adequate anesthetic depth by increasing volatile anesthetic concentration administered via the CPB circuit and/or by administering additional IV anesthetic. In some cases, administration of a vasodilator may be necessary (table 7) [3]. For brief periods, pump flow may be reduced while implementing these pharmacologic interventions.

MAINTENANCE OF ANTICOAGULATION — Adequacy of heparin anticoagulation is measured with point-of-care tests such as activated whole blood clotting time (ACT) every 30 minutes to maintain a targeted value throughout CPB (typically above 480 seconds) [3,95]. If available, plasma heparin concentrations may also be determined by point-of care assays such as heparin-protamine titration, with target heparin concentration ≥4 units/mL [3,95,96]. Protocols in some institutions emphasize treatment of heparin concentrations <4 units/mL, even if ACT values are adequate. (See "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Systemic anticoagulation'.)

Anticoagulation in patients with heparin-induced thrombocytopenia (HIT) is discussed separately. (See "Management of heparin-induced thrombocytopenia (HIT) during cardiac or vascular surgery".)

MAINTENANCE OF ANESTHESIA AND NEUROMUSCULAR BLOCKADE

Anesthetic agents — During CPB, continued maintenance of general anesthesia is typically with a volatile anesthetic agent administered via the CPB circuit [97]. Potential benefits of volatile anesthetics include reduced incidence of myocardial infarction, although data from large meta-analyses are inconsistent [98-100] (see "Anesthesia for cardiac surgery: General principles", section on 'Maintenance techniques'). Potential disadvantages of volatile anesthetic-based anesthesia during CPB include variable uptake of volatile anesthetic agent with different types of oxygenators, difficulties with maintenance of a steady state plasma concentration of volatile anesthetic agent with variable fresh gas flow rates during different phases of CPB, poor correlation between oxygenator exhaust concentrations of volatile anesthetic agent compared with bispectral index values and similar monitors, and inefficient scavenging of waste gases that may pollute the operating room [101].

Use of a total intravenous anesthetic (TIVA) technique during CPB, or combinations of volatile and intravenous agents are reasonable alternatives. Potential disadvantages of TIVA-based anesthesia during CPB include changes in plasma protein binding capacity with variations in the free fraction of propofol in plasma due to hemodilution, effects of hypothermia on hepatic clearance of propofol, and absorption of propofol in the CPB circuit [101].

Anesthetic depth — Inadequate anesthetic depth is treated by increasing the volatile anesthetic concentration administered via the CPB circuit or by administering additional intravenous (IV) anesthetic agents [102]. It is reasonable to monitor processed EEG indices (eg, bispectral index) or an unprocessed EEG to provide data that may detect inadequate anesthesia during CPB [2,3,103-108]. Many institutions measure concentration of the volatile anesthetic agent in the expiratory gas of the oxygenator [2,3,104,105]. Notably, there is poor correlation between these two standard approaches to assessment of anesthetic depth during CPB [109,110]. Furthermore, despite use of these monitoring techniques, anesthesia awareness may occur during cardiac surgery [111]. (See "Accidental awareness during general anesthesia", section on 'Monitoring'.)

If deliberate hypothermia is employed during CPB, anesthetic requirements for either volatile or intravenous agents is decreased at the colder temperatures [101,104,105,112]. Furthermore, uptake and plasma solubility of volatile anesthetics increase during hypothermia [101]. Subsequently, to avoid inadequate anesthesia during rewarming, the dose of volatile anesthetic agent should be increased to compensate for decreasing blood/gas solubility of volatile anesthetics, or additional IV agents such as propofol should be administered since blood concentration of propofol rapidly decreases during rewarming (see 'Temperature' below). Lighter depth of anesthesia is sometimes heralded by new onset of venous desaturation (eg, mixed venous oxygen saturation [SvO₂] <75 percent). Use of a benzodiazepine may reduce risk of awareness during rewarming, but administration of large doses that may delay postoperative awakening is avoided, particularly in older patients [2,113]. (See "Accidental awareness during general anesthesia", section on 'Risk factors'.)

Neuromuscular blocking agents — Neuromuscular blockade is employed during CPB to facilitate surgical operating conditions [114]. Decreased dosing of the selected neuromuscular blocking agent (NMBA) may be adequate during the hypothermic period of CPB because hypothermia directly reduces muscle strength (up to 10 percent per degree Celsius) and enhances NMBA action [2,115]. However, additional NMBA is typically required during rewarming.

Neuromuscular function can be assessed with a peripheral nerve stimulator (PNS) throughout CPB to maintain an appropriate degree of neuromuscular blockade [2]. The PNS electrodes are placed along the course of facial nerve supplying the orbicularis oculi muscle. Since movement of the diaphragm or other slight movement provides an indication of inadequate anesthetic depth, complete paralysis during CPB is avoided in some institutions. Complete neuromuscular blockade with absence of twitches on the PNS may increase risk of awareness. (See "Accidental awareness during general anesthesia", section on 'Neuromuscular blockade'.)

TEMPERATURE — Mild (32 to 35°C), moderate (28 to 32°C), or deep (<28°C) hypothermia is used as a protective strategy for the brain and vital organs during CPB for many cardiac surgical procedures [116-118]. The cerebral metabolic rate of oxygen consumption (CMRO₂) decreases approximately 7 percent per degree Celsius reduction in temperature [119]. Mild hypothermia (approximately 34°C) is typically selected for coronary artery bypass grafting (CABG) surgery [83,120]. Moderate hypothermic temperatures may be selected for cardiac valve repair or replacement surgery due to the length and complexity of these procedure. Moderate reductions in temperature confer the same neuroprotective benefits as deeper levels of hypothermia during focal ischemia [118,121]. For procedures requiring a temporary period of elective circulatory arrest (eg, repair of portions of the ascending aorta or aortic arch), a deep hypothermic temperature (eg, 20°C) may be selected to achieve EEG isoelectricity. (See "Anesthesia for aortic surgery with hypothermia and elective circulatory arrest in adult patients", section on 'Cooling strategies'.)

Management during cooling and hypothermia

●Management during cooling – During cooling, the temperature gradient between the venous inflow and arterial outlet on the oxygenator is maintained at <10°C, similar to the recommendations of the Society of Thoracic Surgeons (STS), Society of Cardiovascular Anesthesiologists (SCA), and American Society of ExtraCorporeal Technology (AmSECT) [3,116,117]. Any peripheral body warming devices (eg, forced-air warming blankets, insulation water mattresses, devices for warming intravenous [IV] fluids) are turned off during cooling and hypothermia.

●Management during hypothermia – During hypothermic CPB, the oxygenator arterial outlet temperature should be used as the best measure of cerebral temperature [3,116,117]. However, other temperature sites that are also monitored include the oxygenator venous inlet and specific sites in the patient's body (eg, nasopharyngeal, bladder [or the rectal site for patients who do not make urine], pulmonary arterial blood temperature if a pulmonary artery catheter [PAC] is in place). Predictable discrepancies between temperatures measured at each site reflect differences in perfusion. For example, the arterial inlet temperature decreases first during active cooling, followed by the nasopharyngeal temperature, while bladder temperature is usually the slowest to change. Temperature during CPB measured from the pulmonary artery catheter may be affected by ice in the pericardium used for topical cooling of the heart in the absence of pulmonary blood flow.

Management during rewarming and weaning

●Management during rewarming – Similar to the recommendations of the STS, SCA, and AmSECT [116,117], the rate for gradual rewarming at temperatures >30°C is limited to ≤0.5°C/minute, and the temperature gradient between the venous inflow and arterial outlet on the oxygenator is maintained at ≤4°C. This slow rewarming may require 60 to 90 minutes or more in a moderately or deeply hypothermic patient [117,122,123].

Although the final target temperature for separation for CPB is 37°C at the nasopharyngeal site, hyperthermia is to be avoided. The nasopharyngeal temperature site serves as the in vivo monitor of brain temperature and is typically higher than other sites due to the proximity of the aortic cannula to the great vessels and head. However, all monitored temperature sites may underestimate the true cerebral temperature during rewarming [124]. To minimize risk of cerebral and systemic hyperthermia, which is associated with worsened neurologic and neurocognitive outcomes [116,117,122,123,125], acute kidney injury (AKI) [126], and mediastinitis [127], the arterial outlet temperature should never exceed 37°C.

The oxygenator venous blood inflow temperature typically lags behind the arterial outlet and nasopharyngeal temperatures. Target temperature will be achieved more rapidly at highly perfused sites (eg, nasopharyngeal tissue) that receive the majority of the systemic blood flow throughout rewarming. The peripheral ("core" or "shell") sites required a considerably longer period to equilibrate. For example, the bladder temperature site (which is used to estimate "core temperature" in most cases) will only be approximately 35.5°C when nasopharyngeal temperature is stable at 37°C.

●Management during weaning from CPB – The nasopharyngeal and/or PAC sites are used for temperature monitoring during weaning from CPB and in the immediate postbypass period. The temperature gradient between these highly perfused sites and the periphery (eg, the bladder site) produces volume and heat redistribution. Thus, mild systemic hypothermia may develop before the patient is admitted to the intensive care unit (ICU). Depending on the urine output, bladder temperature may remain a poor indicator of core temperature until complete equilibration of tissue temperature has occurred, typically several hours after CPB [128].

LABORATORY PARAMETERS

Hemoglobin/hematocrit — For hemoglobin <7.5 g/dL (or hematocrit <22 percent), initial treatment during CPB is removal of fluid by ultrafiltration (hemoconcentration) when possible. Although decisions to transfuse packed red blood cells (RBCs) are individualized, transfusion is reasonable if hemoglobin remains <7.5 g/dL when hemoconcentration is not possible or is ineffective. Available salvaged blood from autotransfusion is returned first, followed by reinfusion of blood units harvested via normovolemic hemodilution, before allogenic RBCs are transfused. (See "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Management of anemia' and "Surgical blood conservation: Intraoperative blood salvage".)

Glucose — We agree with the Society of Thoracic Surgeons (STS) and European Association for Cardio-Thoracic Surgery (EACTS)/European Association of Cardiothoracic Anaesthesiology (EACTA)/European Board of Cardiovascular Perfusion (EBCP) guidelines, which recommend maintaining blood glucose levels <180 mg/dL (10 mmol/L) during CPB and the postbypass period, with a single dose or intermittent doses of insulin if effective or with a continuous insulin drip if glucose levels are persistently elevated (>180 mg/dL [10 mmol/L]) [3,129]. Blood glucose and potassium are monitored frequently (approximately every 30 minutes) to prevent hypoglycemia and hypokalemia in response to treatment. (See "Glycemic control in critically ill adult and pediatric patients".)

Electrolytes and lactate — Metabolic abnormalities (eg, hyperglycemia, hypocalcemia, hyperkalemia, hypokalemia, hypomagnesemia) are common during CPB [3]. Frequent laboratory testing during CPB (approximately every 30 minutes) is helpful for recognition and treatment. Management of persistent abnormalities during the postbypass period is addressed separately. (See "Intraoperative problems after cardiopulmonary bypass", section on 'Metabolic abnormalities'.)

Hypocalcemia — Hypocalcemia (measured as the ionized fraction of total calcium concentration) is common during CPB and is typically corrected (eg, with administration of calcium chloride 5 to 10 mg/kg intravenous [IV]) [3]. Administration of calcium is avoided while the aortic cross-clamp is in place and for at least 10 to 15 minutes after removal of the aortic cross-clamp, thus allowing a period of myocardial reperfusion [130]. (See "Intraoperative problems after cardiopulmonary bypass", section on 'Hypocalcemia'.)

Hyperkalemia — Transient hyperkalemia is common during CPB, particularly just after administration of high-potassium cardioplegia solution (ie, before systemic redistribution of cardioplegia solution has occurred) [3].

Persistent hyperkalemia during CPB may be managed by administration of combinations of glucose and insulin, or the perfusionist may employ zero-balance ultrafiltration (Z-BUF), which allows removal of potassium from the blood [131,132]. As plasma water is removed, an equal amount of buffered potassium-free solution is added. Normal saline solution is typically used, but monitoring to avoid hypernatremia or hyperchloremia is necessary. Alternative treatments include administration of furosemide to eliminate potassium via diuresis or administration of calcium chloride, (which is more typically administered to treat hyperkalemia during the postbypass period). (See "Treatment and prevention of hyperkalemia in adults", section on 'Patients with a hyperkalemic emergency'.)

Hypokalemia — Hypokalemia as a consequence of the administration of insulin and beta-adrenergic agonists such as epinephrine is common during CPB and weaning from CPB, and in the postbypass period immediately after weaning. Typically, potassium chloride is administered in increments of 10 to 20 mEq IV by slow infusion over 30 to 60 minutes into a central venous catheter. Further management after weaning from CPB is discussed separately. (See "Intraoperative problems after cardiopulmonary bypass", section on 'Hypokalemia'.)

Hypomagnesemia — Studies have demonstrated that hypomagnesemia is common during and after CPB due to diuresis and hemodilution with magnesium-free fluids during CPB [133]. Postoperative hypomagnesemia is associated with dysrhythmias, myocardial ischemia, and ventricular dysfunction [133,134]. Thus, magnesium sulfate 2 g is often administered near the conclusion of CPB [3]. (See "Intraoperative problems after cardiopulmonary bypass", section on 'Hypomagnesemia'.)

Lactate — Lactate is often measured periodically during bypass to identify occult tissue hypoperfusion.

ARRHYTHMIAS — After removal of the aortic cross-clamp, cardiac dysrhythmias are initially common (eg, heart block, ventricular fibrillation, junctional rhythm), although normal sinus rhythm is eventually restored in most patients who had a normal preoperative rhythm. A lidocaine bolus of 100 mg is often administered immediately prior to aortic cross-clamp removal to decrease the risk of ventricular fibrillation [135-137]. Until recovery to normal sinus rhythm, temporary external cardiac pacing may be necessary for patients with heart block, intraventricular conduction delay, or bradycardia.

Defibrillation with internal paddles applied directly to the heart using 10 to 20 joules is usually effective for the treatment of ventricular fibrillation immediately after aortic cross-clamp removal if blood temperature is in the range of 32° to 34°C, mean arterial pressure (MAP) is adequate, and there are no significant electrolyte abnormalities such as hyperkalemia. For persistent or recurrent ventricular fibrillation, it is important to identify and treat potential underlying causes (eg, hypothermia, hypokalemia, hypomagnesemia, compromised coronary graft anastomosis with inadequate coronary flow, air embolism into a coronary artery, left ventricular [LV] distention). After correcting potential causes, persistent ventricular arrhythmias are typically treated with boluses of antiarrhythmic drugs such as lidocaine (100 mg x 2) or amiodarone (300 mg). In some cases, a continuous infusion of amiodarone may be necessary, as described separately. (See "Amiodarone: Clinical uses", section on 'Amiodarone for ventricular arrhythmias'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Management of cardiopulmonary bypass".)

SUMMARY AND RECOMMENDATIONS

●General principles – Cardiopulmonary bypass (CPB) is a form of extracorporeal circulation in which the patient's blood is diverted from the heart and lungs and rerouted outside of the body (figure 1A and figure 1B). The normal physiologic functions of the heart and lungs, including circulation of blood, oxygenation, and ventilation, are temporarily taken over by the CPB machine. (See 'General principles' above.)

●Physiologic targets – Typical parameters during CPB in adults include:

•Flow rate controlled at 2.2 to 2.4 L/minute/m², maintenance of mean arterial pressure (MAP) ≥65 mmHg, and mixed venous oxygen saturation (SvO₂) ≥75 percent (table 2). (See 'Pump flow and mixed venous oxygen saturation' above and 'Mean arterial pressure' above.)

•Arterial pO₂ is maintained at 150 to 250 mmHg during CPB alpha-stat management of arterial blood gases without temperature correction to maintain a normal range for pCO₂ (35 to 45 mmHg [4.7 to 6 kPa]) and pH (7.35 to 7.45). (See 'Oxygenation, ventilation, and arterial blood gases' above.)

•For Hgb <7.5 g/dL (or Hct <22 percent), initial treatment during CPB is removal of a limited volume of fluid (<30 ml/kg) by ultrafiltration (hemoconcentration) when feasible. Transfusion of packed red blood cells (RBCs) is reasonable if hemoglobin remains <7.5 g/dL when hemoconcentration is not possible or is ineffective. (See 'Hemoglobin/hematocrit' above.)

•Blood glucose levels <180 mg/dL (10 mmol/L) during CPB and the postbypass period, with a single dose or intermittent doses of insulin if effective or with a continuous insulin drip if glucose levels are persistently elevated (>180 mg/dL [10 mmol/L]). (See 'Glucose' above.)

•Metabolic abnormalities (eg, hyperglycemia, hypocalcemia, hyperkalemia, hypokalemia, hypomagnesemia) are common during CPB. Frequent laboratory testing during CPB (approximately every 30 minutes) is helpful for recognition and treatment. (See 'Electrolytes and lactate' above.)

•If near-infrared spectroscopy (NIRS) is available, regional cerebral oxygen saturation (rSO₂) is maintained within 20 percent of baseline (algorithm 1). (See 'Neuromonitoring modalities' above.)

●Anticoagulation – Adequacy of heparinization is measured with point-of-care tests such as activated whole blood clotting time (ACT) every 30 minutes to maintain a targeted value throughout CPB (typically 400 to 480 seconds), or by point-of care assays such as heparin-protamine titration, with target heparin concentration typically ≥4 units/mL. (See 'Maintenance of anticoagulation' above.)

●Monitoring during CPB – Standard monitoring during CPB includes (see 'Monitoring' above and 'Laboratory parameters' above):

•Continuous measurement

-Pump flow

-Arterial line pressure

-Arterial blood gas trending

-SvO₂

-Temperature

-Hemoglobin or hematocrit

-Exhaust concentrations of volatile anesthetic agents, if available

•Frequent measurements (approximately every 30 minutes)

-Potassium and other electrolytes

-Glucose

-Urine output

-ACT

-Lactate

•Neuromonitoring modalities – Cerebral oximetry, processed EEG, or transcranial Doppler technology is used in some centers for patient-related or procedure-related risk factors for adverse neurologic outcomes. Also, it is reasonable to monitor processed EEG indices (eg, bispectral index) or the unprocessed EEG to provide data that may detect inadequate anesthesia during CPB. (See 'Neuromonitoring modalities' above.)

●Maintenance of anesthesia and neuromuscular blockade

•Anesthetic agent selection – We suggest maintenance of general anesthesia with a volatile anesthetic agent administered via the CPB circuit (Grade 2C). Use of a total intravenous anesthetic (TIVA) technique or combinations of volatile and intravenous agents are reasonable alternatives. (See 'Anesthetic agents' above.)

•Anesthetic depth – Inadequate anesthetic depth is treated by increasing the volatile anesthetic concentration administered via the CPB circuit or by administering additional intravenous (IV) anesthetic agents.

•Neuromuscular blocking agents (NMBAs) – Decreased dosing of the selected NMBA may be adequate during the hypothermic period of CPB; however, additional NMBA is typically required during rewarming. (See 'Neuromuscular blocking agents' above.)

●Temperature management – Deliberate hypothermia is employed in selected patients undergoing CPB. During cooling, the temperature gradient between the venous inflow and arterial outlet on the oxygenator is maintained at <10°C. Rewarming is gradual (≤0.5°C/minute), and the temperature gradient between the venous inflow and the oxygenator arterial outlet is maintained at ≤4°C. To minimize risk of cerebral hyperthermia, the arterial outlet and nasopharyngeal temperatures are maintained ≤37°C. (See 'Temperature' above.)

●Arrhythmia management – After removal of the aortic cross-clamp, management of cardiac dysrhythmias (eg, heart block, ventricular fibrillation, junctional rhythm) is often necessary. (See 'Arrhythmias' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Ryan Konoske, MD, who contributed to an earlier version of this topic review.