A Practical Approach to Cardiac Anesthesia (Practical Approach Series) 5th Ed.

10 Postoperative Care of the Cardiac Surgical Patient

Breandan Sullivan and Michael H. Wall


 1. Transport from the operating room (OR) to the intensive care unit (ICU) is a critical period for patient monitoring or vigilance. Emergency drugs and airway equipment should be present, and adequate transportation personnel (typically three people) should accompany the patient during transport.

 2. Patient “hand-off” to the ICU should be consistent, careful, and structured, and should not distract caregivers from continuous assessment of hemodynamics, oxygenation, and ventilation.

 3. Early postoperative respiratory support ranges from full mechanical ventilation to immediate extubation in the OR, depending upon institutional practice patterns, anesthetic techniques, and patient stability. There is no “best” ventilation mode for cardiac surgery patients.

 4. Weaning from mechanical ventilation involves assessment of oxygenation adequacy (typically PaO2/FIO2 >100 on positive end-expiratory pressure [PEEP] 5 cm H2O or less), hemodynamic stability, patient responsiveness to commands, and measured ventilatory parameters such as vital capacity and the rapid shallow breathing index (RSBI).

 5. Fast-tracking protocols designed to extubate cardiac surgery patients within several hours of completion of surgery are common. With such protocols, early postoperative continuous infusions of propofol or dexmedetomidine may be helpful.

 6. Early postoperative differential diagnosis of hypotension is often challenging, and includes hypovolemia, heart valve dysfunction, left ventricular (LV) and/or right ventricular (RV) dysfunction, cardiac tamponade, cardiac dysrhythmia, and vasodilation. Once a diagnosis has been made, optimal therapy usually becomes clear.

 7. Hypertension is not uncommon and must be acutely and effectively managed to minimize bleeding and other complications such as LV failure and aortic dissection. The differential diagnosis includes pain, hypothermia, hypercarbia, hypoxemia, intravascular volume excess, anxiety, and pre-existing essential hypertension, among others.

 8. Acute poststernotomy pain most often is managed by administering intravenous opioids, but other potentially helpful modalities include nonsteroidal anti-inflammatory drugs, intrathecal opioids, and central neuraxial or peripheral nerve blocks.

 9. Early postoperative acid–base, electrolyte, and glucose disturbances are common. They should be diagnosed and treated promptly.

10. Postoperative bleeding may be surgical, coagulopathic, or both. Aggressive diagnosis and treatment of coagulation disturbances facilitates early diagnosis and treatment of surgical bleeding (i.e., return to OR for re-exploration) and avoidance of cardiac tamponade.

11. Discharge from the ICU typically occurs in 1 to 2 days. Criteria vary with cardiac surgical procedures and with institutional capabilities for post-ICU patient care (e.g., stepdown ICU beds vs. traditional floor nursing care).

12. Adequate communication with patients’ family members and adequate family visitation and support greatly facilitate postoperative recovery.

THE PURPOSE OF THIS CHAPTER is to briefly discuss the transport of the cardiac surgery patient from the OR to the ICU, the hand-off of care from the OR team to the ICU team, and an approach to common problems that occur in the first 24 hrs in the ICU. The reader is referred to standard critical care text books for discussion of more chronic ICU problems such as nutrition, infectious disease, sepsis, and multiple organ failure.

I. Transition from operating room to intensive care unit

   A. General principles

      1. Movement of a critically ill patient in the immediate postoperative period to the ICU or to an intermediate level post-cardiac surgical recovery area is a risky business. Inter- or intrahospital transport of critically ill patients is associated with increased morbidity and mortality [1].


      2. The American College of Critical Care Medicine (ACCM) guidelines state that “during transport, there is no hiatus in the monitoring or maintenance of a patient’s vital signs” [2].

      3. The guidelines state there are four major areas to optimize efficiency and safety of patient transport: Communication (or hand-offs), personnel, equipment, and monitoring. Each of these areas will be discussed.

   B. The transport process

     1. Prior to movement of the patient from OR table to ICU bed

        a. Airway/Breathing. If patients are suitable candidates for fast-tracking (see subsequent section) and meet standard extubation criteria, they can be extubated in the OR, or within 6 to 8 hrs of arrival in the ICU. If the patient is to remain intubated, the endotracheal tube should be checked for position and patency, and should be securely attached to the patient. In addition, all chest tubes and drains should be checked for ongoing bleeding to ensure that immediate transport from the OR is appropriate, and for proper functioning to avoid hemothorax or pneumothorax during transport.

        b. Circulation. The patient should be hemodynamically “stable” prior to transport. In general, if the patient requires frequent bolus doses, or increasing doses of vasoactive drugs, it is better to stabilize prior to transport.

           (1) Pacemaker. Proper settings and functioning of the pacemaker should be checked at this point (see Chapter 17).

        c. Coagulation. Bleeding should be controlled, and a plan for correction of ongoing coagulopathy should be made prior to transport.

        d. Metabolic. Metabolic abnormalities (glucose, electrolyte, and acid–base) should be identified and corrected as much as possible prior to the transport.

        e. Brief Telephone Report. A brief verbal report to the ICU team should be provided prior to transport (see hand-off section).

        f. Special Bed. Patients at high risk for development of pressure ulcers (pre-existing pressure ulcers, poor nutritional status, elderly, poor ventricular function, etc.) should be placed on special beds/mattresses in the OR.

     2. Patient movement from the OR table to the transport bed. Movement can cause hemodynamic instability, fluid shifts, and arrhythmias. Movement can also cause inadvertent loss of airway, vascular access, and interruption of intravenous infusions. Residual intracardiac air is a complication of many procedures (e.g., valve replacement) and this air may be easily dislodged when moving the patient. In addition, the position of a pulmonary artery catheter (PAC) can be altered during patient movement. Confirmation of the PAC position (i.e., pulmonary artery waveform rather than pulmonary artery occluded or RV waveform) before and after patient movement should be done. Sudden onset of dysrhythmia should trigger examination of the PAC. Ready access to a large-bore intravenous infusion port and to any ongoing or continuous infusions of medications is critical to managing this period safely and being able to respond promptly.

     3. Transport from the OR to the ICU

        a. Personnel. Generally, at least three members of the operative team should transport the patient from the OR to the ICU. This should include a member of the anesthesia care team, surgical team, and a nurse or technician. Additional team members (perfusionists, respiratory therapists, etc.) may be needed for patients on mechanical assist devices, inhaled pulmonary vasodilators, or those with acute lung injury (ALI) who require a transport ventilator.

        b. Equipment. ACCM guidelines recommend a minimum of a blood pressure monitor, pulse oximeter, and cardiac monitor/defibrillator for all transports of critically ill patients [1]. An additional monitor to consider is continuous end-tidal CO2 for intubated patients. Equipment and drugs for emergency airway management should be immediately available. An oxygen (O2) source with enough O2 for the duration of transport plus 30 min must be available. Basic emergency advanced cardiac life support (ACLS) drugs should be immediately available. All infusions should be checked and all pumps should be fully charged prior to transport. Supplemental O2 should be provided to all extubated patients. Bag mask ventilation (with or without PEEP valves) can be used for most patients. Transport ventilators may be needed for patients with ALI or acute respiratory distress syndrome (ARDS). Mechanical support device batteries should be immediately available.

        c. Monitoring. ACCM guidelines state that critically ill patients should “… receive the same level of basic physiologic monitoring and transport as they had in the ICU ….” The same concept applies to patients leaving the OR [1].

        d. IV Access. Every effort must be made to avoid a “tangle” of IV tubing. In general, it is best to have one large bore IV identified for rapid administration of fluids or emergency medications. This line should be easily identified and immediately accessible. Bolus medications should ideally be given via a central venous site for faster onset. Finally, all IV fluid bags should be full enough to give fluid boluses as needed.

        e. Sedation/Analgesia. In extubated patients, it is best not to give boluses of narcotics during transport. It is probably better and safer to give analgesics prior to transport, then give additional medications after arrival in the ICU. In intubated patients, it is best to start the postoperative sedation and analgesia plan prior to transport to minimize the need to give bolus medications during transport.

II. Transfer of care to the ICU team

   A. Importance of hand-offs. The hand-off of care from the OR team to the ICU team is a surprisingly hazardous and dangerous event. The Joint Commission identified that communication failure was the root cause of 65% of sentinel events in 2006 [2]. Numerous studies have shown that the best hand-offs occur when they are structured, standardized, and use checklists [36]. Recently, many centers are developing hand-off tools from the electronic medical record [7].


   B. Logistics. Ideally, each member of the OR and ICU teams should have specific tasks and the hand-off should occur in a standardized sequence [3,4]. One simple sequence would be transition from transport to ICU monitor, then initial ventilator settings, then formal structured hand-off.

   C. Transition to ICU monitors. The patient must be continuously monitored during this process. Ideally, each parameter (ECG, O2 saturation, etc.) should be transferred from the transport monitor to the ICU monitor in series, as opposed to unhooking all of them at once then hooking them up one at a time. Some systems allow for all the monitors to be almost instantly switched over by removing the entire “brick” at once. In any event, based on local monitors, there should be an orderly transition between both sets of monitors.


   D. Initial ventilator settings. Intubated patients must have their endotracheal tube evaluated for patency, security, and position. This can be accomplished with a chest x-ray or with a bedside bronchoscopy. Ventilator parameters including ventilator mode, rate, fraction of inspired oxygen (FiO2), PEEP, and pressure support must be selected. The patients who have no respiratory effort can be placed on assist-control (AC) or synchronized intermittent mandatory ventilation (SIMV) with an adequate rate, tidal volume, and PEEP. The patients who have regained spontaneous respiratory effort can be placed on SIMV or pressure support ventilation (PSV). PSV and SIMV modes can be combined. Excessive use of PEEP impedes venous return and may impair RV performance. The application of PEEP may decrease mediastinal bleeding, although the literature on this topic is inconsistent and this technique must be used with caution, as PEEP’s adverse effects on hemodynamics are well established.

   E. The actual hand-off. Once the monitors have been transferred to the ICU bedside monitor and oxygenation and ventilation have been confirmed, a structured hand-off should occur. This should include the patient’s name, age, allergies, medical history, all significant intraoperative events, and the immediate postoperation plan. One structured hand-off form generated from the electronic medical record (EMR) is shown (see Fig. 10.1). Time should be allowed for questions and answers from all members of OR and ICU teams.

Figure 10.1 Structured hand-off form generated from electronic medical records.

      1. The initial review of the patient upon his or her arrival to the recovery area includes the patient’s history, age, height, weight, pre-existing medical conditions, any allergies, a list of preoperative medications, and review of the most current laboratory findings (with special emphasis on potassium and hematocrit). The report should include a detailed review of the patient’s cardiac status, including ventricular dysfunction, valvular disease, coronary anatomy, and details of the surgical procedure.

      2. An anesthetic review should be presented, which includes types and location of intravenous catheters and invasive monitors, along with any complications that occurred during their placement. A brief description of the anesthetic technique should be discussed to help plan for a smooth emergence. Any difficulties with airway management should be emphasized, particularly when weaning and extubation protocols are utilized. The presence or absence of obstructive sleep apnea should be discussed and the need for continuing patients’ home continuous positive airway pressure (CPAP) or bi-level positive airway pressure (BiPAP) should be addressed. A post-cardiopulmonary bypass synopsis should be reported, including the use of vasoactive, inotropic, and antiarrhythmic drugs, as well as any untoward events such as arrhythmias and presumed drug reactions. This should also include an update on the presence or absence of bleeding prior to chest closure.

      3. Early upon arrival to the ICU, the patient’s heart rate, rhythm, and blood pressure should be determined. If the heart is being paced, the settings should be reviewed and all electrodes identified and secured, as the patient may be dependent on the device.

        a. If the patient has a permanent pacemaker or defibrillator, the settings should also be reviewed. The devices should be interrogated in the ICU and antitachycardia treatment should be activated. While waiting for the device to be activated, external defibrillator pads should be placed on the patient and a defibrillator should be immediately available [8].

        b. If the patient has a ventricular assist device (VAD), the monitor should be attached to a wall-based energy supply and the output of the device should be attached to the display module. The settings of the device and the position and location of the cannula should be reviewed.

        c. For patients on extracorporal membrane oxygenation (ECMO), the O2 and air supplies should be attached to the wall outlet supplies, and back-up tanks should be available.

   F. Laboratory Tests/electrocardiogram (ECG)/chest radiograph (CXR). After the hand-off is complete and questions are answered, baseline ECG, CXR, and labs should be obtained. An initial arterial blood gas (ABG) should be drawn to ensure the adequacy of oxygenation and ventilation, whether the patient is on a mechanical ventilator or breathing spontaneously. Potassium, blood glucose, and hematocrit levels should be obtained. Acid–base status should be reviewed from ABGs. Baseline coagulation parameters, including prothrombin time (PT), activated partial thromboplastin time (aPTT), and platelet count should be acquired if the patient is bleeding excessively.

III. Mechanical ventilation after cardiac surgery

   A. Hemodynamic response to positive-pressure ventilation (PPV). Heart–lung interactions of PPV are complex [9,10]. In patients with normal LV function, PPV increases intrathoracic pressure (ITP), which reduces venous return, afterload, and stroke volume (SV) and cardiac output (CO) (see Figure 10.2). On the other hand, in patients with LV dysfunction, decreased preload and afterload actually can improve LV performance and CO (see Fig. 10.3). PEEP further increases ITP and decreases venous return.

Figure 10.3 The effect of increasing and decreasing intrathoracic pressure (ITP) on the LV pressure–volume loop of the cardiac cycle in congestive heart failure when LV contractility is reduced and intravascular volume is expanded. The slope of the LV ESPVR is proportional to contractility. The slope of the diastolic LV pressure–volume relationship defines diastolic compliance [10].

Figure 10.2 The effect of increasing and decreasing intrathoracic pressure (ITP) on the pressure-volume loop of the cardiac cycle. The slope of the LV end-systolic pressure–volume relationship (ESPVR) is proportional to contractility. The slope of the diastolic LV pressure–volume relationship defines diastolic compliance [10].

   B. Pulmonary changes after sternotomy and thoracotomy. Cardiac surgery requires either a midline sternotomy or a thoracotomy. Both of these approaches temporarily compromise the function of the thoracic cage, which acts as a respiratory pump. One week after cardiac surgery, there is a significant reduction in total lung capacity, inspiratory vital capacity, forced expiratory volumes, and functional residual capacity compared to preoperative values [11]. Even at 6 wks postoperatively, total lung capacity, inspiratory vital capacity, and forced expiratory volume remained significantly below preoperative values. These findings suggest a marked tendency toward postoperative atelectasis and the possibility of hypoxemia from increased physiologic shunting. These changes in chest wall function can increase physiologic shunt to as much as 13% (compared to a baseline normal value of 5%).

       In addition to these changes in mechanics and volumes, there are also abnormalities in gas exchange, compliance, and work of breathing [12]. The cause of these abnormalities is multifactorial and may include inflammation, reperfusion, and other mechanisms.

   C. Choosing modes of ventilation

     1. Extubated patient. If the patient was extubated in the OR, supplemental oxygen may be all that is necessary postoperatively. Following a general anesthetic, patients will exhibit a mild increase in the PaCO2. Aggressive pulmonary toilet and frequent incentive spirometry must be performed to prevent the atelectasis and hypoxemia that may develop from changes in chest wall function.

     2. Noninvasive ventilation (NIV). NIV can be used to treat or prevent postoperative respiratory failure, and has been shown to prevent reintubation, decrease ventilator-associated pneumonia, and improve outcomes [13,14]. A sample protocol for the use of NIV is shown in Figure 10.4. Contraindications are shown in Table 10.1. Two types of NIV are commonly used:

Table 10.1 Contraindications to noninvasive positive-pressure ventilation

Figure 10.4 Protocol for initiation of curative postoperative NIV. PSV, pressure support ventilation; PEEP, positive end expiratory pressure. FiO2, fraction of inspired oxygen; Spo2, pulse oximetry saturation [13].

        a. CPAP, where constant airway pressure is applied during both inspiration and expiration.

        b. BiPAP, where PSV is applied during inspiration and PEEP is applied during expiration.

     3. Intubated patient

        a. If a patient returns from the OR with an endotracheal tube in place, an individualized plan of care should be developed for that patient. The choice of mechanical ventilation mode is based on the patient’s inherent respiratory effort. If a patient demonstrates an inspiratory effort, PSV or SIMV can be used.

            If a patient is not demonstrating spontaneous respiratory effort, AC or SIMV should be selected. In AC, a set respiratory rate is delivered regardless of the patient’s respiratory effort. If a spontaneous breath is initiated, the ventilator detects the trigger and delivers a set tidal volume (or pressure if on pressure control ventilation). In SIMV, a set respiratory rate is also delivered, but spontaneous breaths over the set rate are not fully supported (like they are in AC), but are dependent on the patient’s effort.

        b. Patients with severe hypoxemia, respiratory failure, ALI, or ARDS will need to be ventilated in a way that minimizes or avoids further “ventilator-induced lung injury” [15]. There have been several recent reviews on the ICU management of ARDS and ALI [1620].

            Initial ventilator settings in patients with ALI or ARDS would include (ardsnet.org):

           (1) Any ventilator mode.

           (2) Tidal volume (VT) 8 mL/kg predicted body weight.

           (3) Set respiratory rate (RR) so minute ventilation is adequate.

           (4) Adjust VT and RR to achieve a goal pH 7.30 to 7.40 and plateau pressure <30 cm H2O.

     4. Weaning from mechanical ventilation is multifactorial. In many postoperative environments, this can best be accomplished by using an algorithm so that weaning can proceed methodically and without interruption. Figure 10.5 shows an algorithm that could facilitate efficient weaning. Prior to attempts at weaning, the following parameters must be met:

Figure 10.5 Protocolized flow chart for ventilator discontinuation [22]. SBT, spontaneous breathing trial.

        a. Normothermia.

        b. Hemodynamically stable.

           (1) Stable vasoactive drug requirements.

           (2) Not requiring increasing doses or boluses of inotropes or vasopressors.

        c. Stable heart rate and rhythm.

        d. Normal acid–base and metabolic state.

        e. Not bleeding excessively (criteria vary, but generally <150 mL/hr chest tube drainage).

            If these criteria are met, the patient is ready to be liberated from mechanical ventilation.

   D. Liberation from mechanical ventilation

      1. Current recommendations are that patients should be liberated from mechanical ventilation as quickly as possible, and an attempt should be made at least daily [21,22].

      2. The first step is to assess the following to determine “readiness to wean”:


        a. PaO2/FiO2 > 200 mm Hg with PEEP  5 cm H2O.

        b. Hemodynamically “stable.”

        c. Awake, alert, and following commands.

        d. Able to cough effectively.

        e. Adequate reversal of neuromuscular blockade (negative inspiratory force [NIF] of 30 cm H2O or more, able to lift head off bed >5 s, no fade on train of 4, vital capacity >15 mL/kg, etc.).

        f. An RSBI (RR/VT in L) < 80 to 100 breaths/min/L after a 2- to 3-min spontaneous breathing trial.

      3. If the patient passes the readiness to wean screen, a trial of spontaneous ventilation via T-piece or with low levels of PSV (5 to 7 cm H2O) and PEEP (7 cm H2O) for 30 to 120 min is done. At the end of the of the trial, if the RSBI is <80 to 100, the patient should be considered for extubation, if they meet the following final criteria:

        a. Awake and alert.

        b. Able to cough and clear secretions.

           (1) The patients who require suctioning more often than every 2 hrs are at a higher risk for reintubation.

        c. No airway edema (as judged crudely by edema of tongue and presence of leak when endotracheal tube cuff is deflated).

        d. Hemodynamically stable.

           (1) Less than 10% to 20% change in HR, BP, pulmonary artery pressures, cardiac index, etc. during the trial.

        e. Normal oxygenation and ventilation.

      4. If the patient meets the criteria, extubation can be performed.

      5. If the patient does not meet the criteria, correctable causes need to be identified and optimized prior to another attempt.

      6. Patients who repeatedly fail spontaneous breathing trials may require more long-term weaning from mechanical ventilation [21].

   E. Incentive spirometry, deep breathing, and coughing maneuvers. Patients must be encouraged to use incentive spirometry and to do deep breathing and coughing maneuvers after extubation to reduce atelectasis. There are numerous physiologic causes of hypoxemia. Diffusion abnormality, low FiO2, hypoventilation, and V/Q mismatch along with shunt comprise the list of possibilities, with atelectasis (causing shunt) being the most common. If hypoxemia persists and atelectasis is the presumed cause, NIV can be used to improve oxygenation and decrease shunt.

IV. Principles of fast-tracking

   A. Goals of fast-tracking. Fast-track (FT) cardiac surgery was introduced to speed recovery and increase efficiency of limited resources (ICUs). Early extubation, ambulation, cardiac rehabilitation, and discharge are key goals of an FT program. Numerous randomized controlled trials have shown FT cardiac surgery is safe and less expensive than conventional cardiac anesthesia [23]. Initially, FT protocols were limited to young, low-risk patients; however, it can be used safely in older and higher risk patients as well [24].


   B. Methods of fast-tracking. A variety of anesthetic techniques can be used to facilitate fast-tracking. Shorter-acting intravenous narcotics can be combined with intrathecal opioids to enhance postoperative analgesia [25,26]. Propofol infusions are often used because of a predictable and rapid recovery profile that is almost independent of the duration of infusion. This property makes propofol a very good sedative agent in the early postoperative management of FT cardiac surgery patients, assuming that hemodynamic stability is not compromised by its use. Caution is needed when short-acting agents are used to set the stage for early extubation, as the incidence of intraoperative awareness may be as high as 0.3% [27]. Dexmedetomidine is an intravenous a-2 adrenergic agonist that may facilitate fast-tracking in cardiac surgical patients. Dexmedetomidine possesses both sedative and analgesic properties, and allows patients to follow commands despite adequate sedation, and it most often does not require weaning prior to extubation. Dexmedetomidine does not possess reliable amnestic properties.

       Dexmedetomidine has been evaluated in a number of trials in the ICU in both cardiac surgery and non-cardiac surgery patients. When compared to propofol, midazolam, and morphine in separate trials, dexmedetomidine has been shown to provide adjunctive analgesia, induce less delirium, and decrease the duration of mechanical ventilation [2831].

   C. Fast-tracking in the postanesthesia care unit. Many institutions prepare for the postoperative management of cardiac surgery patients by developing enhanced step-down or postanesthesia care units (PACUs), where postoperative management can occur safely and efficiently. These units require nurses who understand fast-tracking techniques, so that patients who have undergone cardiac surgery can move smoothly through early extubation in preparation for early transfer to a regular nursing unit. These specialized PACUs can be very effective in providing FT techniques because of their focused effort in caring for FT cardiac surgery patients [32]. Some investigators have even implemented ambulatory cardiac surgery programs [33]!

   D. Utilizing protocols. Developing and utilizing institution-specific FT protocols revolves around systematic plans for weaning patients from ventilators and managing routine postoperative issues to facilitate the progression toward early ICU and hospital discharge. Protocols ideally address most issues before they occur. Fast-tracking protocol development should involve all members of the perioperative care team before implementation.

V. Hemodynamic management in the postoperative period

   A. Monitoring for ischemia. Ischemia can be detected by utilizing a continuous ECG with ST segment analysis, although there is a slight delay in diagnosis of ischemia using this method. Many bedside ECG monitoring systems have ST-segment analysis built into their software algorithms, which is a cost-effective method of monitoring for ischemic events. It is important to ensure that the ECG is in diagnostic mode when evaluating potentially ischemic ECG changes. In monitor mode, the ECG filters out some electrical input (to decrease artifact) and may not accurately reflect ischemic changes. If continuous ST segment analysis is chosen, Leads II and V4 or V5 should be monitored, and sensitivity improves if three leads are used (Leads I, II, and V4 or V5, or Leads II, V4, and V5). Other indicators of myocardial ischemia include pulmonary artery pressures and CO, which tend to be less reliable and oftentimes late markers of myocardial ischemia. Transesophageal echocardiogram (TEE) segmental wall-motion abnormalities represent the most sensitive early detector of myocardial ischemia, but continuous monitoring usually is not done because the TEE probe (if used) typically is removed at the end of surgery. It is extremely important for the intensivist to recognize the changes in ECG and hemodynamics that can result from temporary epicardial ventricular pacing. Epicardial ventricular pacing can mimic septal wall dyskinesis that actually represents a pacemaker-induced change in the ventricular depolarization sequence.

       Intraoperative and ongoing postoperative ischemia can be detected as soon as 6 hrs after the event begins by examining some specific cardiac markers. The earliest and most useful marker is cardiac troponin I (cTnI). The ability to measure cTnI is particularly useful in cases where ECG monitoring is difficult to interpret, such as with left bundle branch block or LV hypertrophy. Elevated plasma levels of this biologic marker provide clear evidence of ischemia and may suggest a diagnosis of myocardial infarction.

       In postoperative cardiac surgical patients, all of the above methods have significant problems. Most often, myocardial ischemia is suspected by ECG changes or unexpected increases in vasoactive drug requirements. The diagnosis is best confirmed by TTE or TEE. Diagnosis may require cardiac catheterization. Treatment options include returning to the OR or medical management.

   B. Ventricular dysfunction after cardiac surgery. In addition to pre-existing ventricular dysfunction, postoperative causes of ventricular dysfunction include inadequate myocardial protection, myocardial stunning, incomplete revascularization, and reperfusion injury. Preoperative predictors of postoperative ventricular dysfunction include cardiac enlargement, advanced age, diabetes mellitus, female gender, high LV end-diastolic pressures at cardiac catheterization, small coronary arteries (for coronary revascularization procedures), and ejection fraction less than 0.40. Intraoperative predictors include longer cardiopulmonary bypass (CPB) and aortic cross-clamp times. These factors increase the likelihood of needing inotropic support in the postoperative period. The patients who have normal preoperative cardiac performance and short periods of CPB have a much lower likelihood of requiring postoperative inotropic support. The patients who fail to achieve adequate hemodynamics even with pharmacologic support will require mechanical cardiac assistance such as an intra-aortic balloon pump (IABP) or VAD. Recently, Hollenberg has written an excellent review article on vasoactive drugs in circulatory shock [34].

       The myocardium has both b1-adrenergric receptors and b2-adrenergric receptors, which contribute to inotropy and lusitropy (enhanced diastolic relaxation). The b-adrenergic agonists (b-agonists) are often the first-line agents used when there is a need to improve ventricular function after CPB. Depletion of endogenous catecholamines and the resulting b-receptor downregulation can blunt the response to b-agonists. Increased G-inhibitory proteins, reperfusion injury, tachycardia, incomplete revascularization, nonviable myocardium, preoperative use of b-agonists, and acute or chronic heart failure also may attenuate the response to b-agonists.

       The inotropic response to b1/b2-adrenergic receptor stimulation occurs via activation of the Gs protein and adenylyl cyclase leading to increased intracellular cyclic adenosine monophosphate (cAMP). It is important to recognize that lusitropy is an active, energy-consuming process; impaired ventricular relaxation (diastolic dysfunction) can cause heart failure alone or in combination with systolic dysfunction. Until recently, there have been no head-to-head clinical trials comparing the clinical outcomes of inotropes and vasopressors. In a trial of 30 patients with dopamine-resistant cardiogenic shock, the patients were randomized to receive either epinephrine alone or dobutamine in combination with norepinephrine. Both groups experienced an increase in cardiac index and a decrease in their creatinine. The group receiving epinephrine experienced more arrhythmias, transient lactic acidosis, and a decrease in splanchnic perfusion [35].

       Phosphodiesterase Type III inhibitors (amrinone and milrinone) augment b-adrenergic–mediated stimulation by inhibiting the breakdown of cAMP. Phosphodiesterase inhibitors (PDEIs) act either additively or synergistically with b-adrenergic agonists. PDEIs appear to have anti-ischemic effects and may favorably alter myocardial oxygen consumption [36]. PDEIs can be added to b-agonist therapy or employed as a first-line inotrope. Because PDEIs also induce systemic and pulmonary vasodilation (sometimes termed inodilators), clinical paradigms that favor their use include pulmonary hypertension, RV failure, aortic or mitral valvular regurgitation, and acute/chronic b1/b2-adrenergric receptor desensitization (long-standing CHF, use of b-agonist therapy preoperatively).

       Levosimendan is a novel inotrope that, at this time, is neither FDA-approved nor available in North America. Levosimendan is a myofilament “calcium-sensitizer,” which results in increased inotropy by improving the efficiency of the coupling of force-generating myocyte proteins in response to a given level of calcium [37]. Like PDEIs, levosimendan may augment inotropy without significantly increasing myocardial oxygen consumption, thus improving the myocardial oxygen supply/demand balance. In a recent meta-analysis analyzing the use of levosimendan in postoperative coronary artery bypass grafting (CABG), levosimendan was associated with improved mortality and morbidity [38].

       B-type natriuretic peptide (nesiritide) has been favored by some for the medical management of CHF [3942], although some work also associates the use of this agent with increased mortality in that setting. The role of nesiritide in the cardiac surgical population remains ill-defined.

       In patients with severely impaired cardiac performance, additional monitoring may be required to ascertain if the patient has optimal myocardial function. Oximetric PACs can provide real-time determinations of mixed venous oxygen saturation (SVO2). A normal SVO2 value of 75% corresponds to a PaO2 of approximately 40 mm Hg. Reductions in SVO2 result from either decreased oxygen delivery (decreased CO, decreased hemoglobin concentration, or decreased arterial oxygen saturation) or increased oxygen consumption. A sustained SVO2 below 40% is associated with increased morbidity and mortality. Similarly, some practitioners choose to use PACs with continuous cardiac output (CCO) determination in such patients, or with both SVO2 and CCO.

   C. Fluid management. Managing postoperative fluids after cardiac surgery can be challenging [43]. The effects of hypothermia (vasoconstriction) and hyperthermia (vasodilation) commonly complicate fluid management especially in the first few hours after cardiac surgery. Central venous pressure (CVP) and PAC are frequently utilized in the cardiac surgical population; however, it is imperative to recognize that these monitors measure pressure as a surrogate estimate of volume and/or cardiac performance. The use of filling pressures (CVP, PAOP, PADP) are poor predictors of assessing total blood volume and volume responsiveness [44]. Volume responsiveness is defined by an increase in cardiac index of >15% in response to a fluid challenge. For mechanically ventilated patients who have a regular respiratory pattern, the pulse pressure variation on an arterial pressure waveform can be a highly useful tool in predicting a hypotensive patient’s response to a fluid challenge [45,46].

       Cardiac surgical procedures, especially those involving CPB, typically result in fluid sequestration into the interstitial compartment. In addition to the changes in circulating blood volume from blood loss and other factors, fluid shifts into or out of the interstitial or the intracellular compartments can be anticipated in the hours following cardiac surgery. Most cardiac surgery patients reach the recovery area with excess total body fluids present that must eventually be mobilized. Healthy patients who have adequate cardiac and renal function typically diurese these fluids over the first two postoperative days without assistance. Other cardiac surgery patients, such as the elderly or those with renal or cardiac dysfunction, may require diuretic drugs (or possibly dialysis or hemofiltration) to remove excess body water.

       Management of blood components, particularly packed red blood cells (PRBC), in the cardiac surgical population is controversial [47,48]. Both transfusion of blood products and anemia are associated with increased perioperative morbidity and mortality. Even though cardiac surgical patients frequently require allogeneic blood products, establishing a transfusion trigger is difficult. Most patients probably require transfusion of PRBC at a hematocrit less than or equal to 21% (hemoglobin less than or equal to 7 g/dL) and few or none require transfusion when the hematocrit is greater than or equal to 30% (hemoglobin greater than or equal to 10 g/dL).


   D. Managing hypotension. Systematic evaluation of preload, afterload, contractility, and heart rate and rhythm should be performed in the hypotensive patient. If preload is adequate and an acceptable heart rate and a normal cardiac rhythm are present, hypotension represents either inadequate myocardial function or vasodilation. Inadequate cardiac function is managed with inotropes. Vasodilation is managed with vasoconstrictors.

     1. Vasodilatory Shock. Eight percent of cardiac surgery patients experience refractory vasodilatory shock after CPB. This refractory shock is associated with increased mortality (25% mortality in one case series). These patients do not respond to traditional treatments, vasopressors, or volume expansion. The causes of the problem are usually multifactorial: Long bypass run, preoperative use of angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers, calcium channel blocker agents, heart transplantation, VAD placement, and myocardial dysfunction. Small clinical trials have shown improved morbidity and mortality in this patient population with the use of an intravenous bolus of methylene blue followed by a continuous infusion [49]. Arginine vasopressin may also be useful in this patient population.

          Two other unique causes of hypotension that are difficult to diagnose without the aid of transesophageal echocardiography are systolic anterior motion (SAM) of the mitral valve and cardiac tamponade.

     2. SAM of the mitral valve should be assessed in the OR in patients undergoing mitral valve repair or septal myectomy. The c-sept distance (see Fig. 10.6) of 2.5 cm or less is associated with a high risk of SAM [50].

Figure 10.6 Schematic demonstrating the transesophageal echocardiographic measurements performed prior to and after mitral valve repair. The biplane image, obtained from the esophageal location at zero degrees, includes the left atrium (LA), left ventricle (LV), mitral valve, and the LV outflow tract. Lengths of the anterior and posterior leaflets were obtained using the middle scallops. AL, anterior leaflet length; CoaptAnn, distance from the mitral coaptation point to the annular plane; CSept, distance from the mitral coaptation point to the septum; LVIDs, LV internal diameter in systole; PL, posterior leaflet length [50].

     3. Tamponade. Although this topic is covered elsewhere, post-cardiac surgery tamponade should be addressed and discussed especially if the patient is demonstrating signs of a low CO shock. A large series of patients diagnosed with post-cardiac surgery tamponade demonstrated that the classical diagnostic signs such as equalizing filling pressures, increased jugular venous pressure, and pulsus paradoxus are frequently not present. Echocardiography can aid in the diagnosis; however, tamponade is a constellation of symptoms rather than a single echocardiographic finding [51].

   E. Dysrhythmia management. Managing postoperative dysrhythmias constitutes an important part of ICU care in cardiac surgery patients. A variety of atrial or ventricular dysrhythmias can occur. The patients with ongoing myocardial ischemia, possibly from incomplete revascularization or myocardial stunning, are predisposed to dysrhythmias. Atrial fibrillation is the most common dysrhythmia to occur after cardiac surgery, and it may occur in as many as half of the patients who undergo myocardial revascularization using CPB. Useful drugs for treating atrial fibrillation include magnesium, digoxin, diltiazem, esmolol, and amiodarone [52].

       It has been shown that various preoperative or postoperative pharmacologic prophylactic strategies may reduce the incidence of postoperative atrial fibrillation or other atrial dysrhythmias [5355]. It is important to identify the patients who are at increased risk for developing perioperative atrial fibrillation. These include the patients who have a previous history of atrial fibrillation, have undergone a combination valve and CABG procedure, are receiving inotropic support, or have pre-existing mitral valve disease, lung disease, or congenital heart disease. Prophylaxis against atrial fibrillation may decrease both the number of days spent in the ICU and the total length of stay in the hospital. Administration of a long-acting b-adrenergic receptor antagonist (i.e., atenolol or metoprolol) is frequently initiated on the first postoperative day following cardiac surgery.

       The prophylactic use of corticosteroids, in particular, hydrocortisone, has been shown in a recent meta-analysis of randomized controlled trials to be effective in decreasing the incidence of atrial fibrillation in a high-risk patient population [56,57]. It appears that a single dose of steroids during induction of anesthesia is sufficient to provide benefit to the patients.In the context of arrhythmia prevention, it is particularly important to maintain normal magnesium and potassium concentrations perioperatively [58,59].


   F. Perioperative hypertension. Perioperative hypertension can result from a number of causes:

      1. Etiologies of acute postoperative hypertension include emergence from anesthesia, hypothermia, hypercarbia, hypoxemia, hypoglycemia, intravascular volume excess, pain, and anxiety. One must consider iatrogenic causes, such as administration of the wrong medication or use of a vasoconstrictor when it is not necessary.

      2. Another cause of postoperative hypertension is withdrawal from preoperative antihypertensive medications. The b-blockers and centrally acting a2-agonists (clonidine) are known to elicit rebound hypertension upon withdrawal.

      3. Unusual causes include intracranial hypertension (from cerebral edema or massive stroke), bladder distention, hypoglycemia, and withdrawal syndromes (e.g., alcohol withdrawal syndrome, withdrawing from chronic opioid use).

      4. Rare causes to consider include endocrine or metabolic disorders such as hyperthyroidism, pheochromocytoma, renin–angiotensin disorders, and malignant hyperthermia.

   G. Pulmonary hypertension. Pulmonary hypertension may occur after cardiac surgery, the causes of which can be divided into new-onset acute pulmonary hypertension and continuation of a more chronic pulmonary hypertensive state. A primary consideration in the evaluation of pulmonary hypertension is the effect on RV performance. Echocardiography is critically important in diagnosing right heart failure. Pulmonary artery pressures may decrease and CVP may increase in the presence of worsening right heart failure. Because of the unique geometry of the right ventricle (RV), traditional echocardiography measurements of LV function cannot be applied to RV performance. Specific validated measurements such as tricuspid annular plane systolic excursion index [60] or Tei index [61,62] should be used to assess the RV function. Pulmonary hypertension and RV failure are particularly problematic following heart or lung transplantation and VAD placement [63].

     1. Chronic pulmonary hypertension is less responsive than systemic hypertension to traditional therapeutic interventions. Chronic elevation in the pulmonary vascular resistance (PVR) stresses the RV and can lead to RV dysfunction. In addition, the RV hypertrophy associated with chronic pulmonary hypertension enhances susceptibility to inadequate RV oxygen delivery. Chronic pulmonary hypertension is managed by continuing any ongoing medications that the patient has been taking, such as calcium channel blockers, along with utilizing therapeutic agents mentioned below for management of acute pulmonary hypertension.

     2. Acute postoperative pulmonary hypertension must be managed aggressively to avoid RV failure [6466]. Parameters that influence pulmonary hypertension (see Table 10.2) should be optimized. There are four major categories of focus in addressing right heart failure associated with acute pulmonary hypertension [67]. See Figure 10.7.

Table 10.2 Factors that contribute to pulmonary hypertension

Figure 10.7 Pathophysiology of RV failure in the setting of high PVR. CO, cardiac output; LV, left ventricle; MAP, mean arterial pressure; PVR, pulmonary vascular resistance; RV, right ventricle [67].

        a. Volume status of the RV (echocardiography). Echocardiography will provide a good understanding of the primary problem: Volume versus pressure overload. Chronic pressure overload causes RV hypertrophy, often with normal RV contractility and volume.

        b. RV function. Address the need for inotropic support (dobutamine, PDEIs, epinephrine). In addition, optimize the heart rate and rhythm.

        c. Offload the RV by correcting any existing acidosis, hypercarbia, or hypoxemia. Consider adding a pulmonary artery vasodilator (inhaled nitric oxide, inhaled prostacyclin, intravenous PDEIs, intravenous nitroglycerin), and minimize any harmful effects of PPV (high peak airway pressures, excessive tidal volumes, ventilator-induced lung injury).

        d. Maintain an adequate right coronary artery perfusion pressure by adding a vasopressor (norepinephrine, vasopressin, phenylephrine) or mechanical diastolic pressure support via an IABP.

VI. Postoperative pain and sedation management techniques

Managing postoperative pain and agitation are paramount in caring for the postoperative cardiac surgery patient. Pain represents a response to nociceptor stimulation from the surgical intervention. Patients may be agitated after cardiac surgery for a variety of reasons. Table 10.3 lists some possible causes of agitation that must be considered because they might be inappropriately “masked” by the administration of a sedative drug, or by residual neuromuscular blockade.


   A. Systemic opioids. A variety of techniques can be used to manage postoperative pain. It is very useful to initially discern the type, quality, and location of pain before administering an analgesic agent. Commonly used opioids include fentanyl, morphine, and hydromorphone.

Table 10.3 Causes of postoperative agitation

   B. Nonsteroidal anti-inflammatory drugs. Nonsteroidal anti-inflammatory drugs (NSAIDs) can be helpful when managing postoperative pain in cardiac surgery. A small amount of drug can provide analgesia without excessive sedation and other complications (e.g., respiratory depression) associated with opioid use. A concern with NSAIDs is their inhibition of platelet function and the potential for increased bleeding. NSAIDs also have been considered a poor choice after cardiac surgery because of their tendency to induce gastric ulcer formation and impair renal function. Renal insufficiency, active peptic ulcer disease, history of gastrointestinal bleeding, and bleeding diathesis should exclude the use of NSAIDs in the postoperative cardiac surgery patient [68].

   C. Intrathecal opioids. In the era of fast-tracking patients through the postoperative period, several regional analgesic techniques have been pursued to improve patient comfort. Systemic (intravenous, intramuscular, transcutaneous, or oral) opioids can cause respiratory depression and somnolence, making them potentially undesirable for fast-tracking. Intrathecal opioids (e.g., morphine 5 to 8 mg/kg up to 1 mg) constitute an alternative to systemic opioids for cardiac surgery [69]. This route has been explored in an attempt to improve patient comfort while minimizing respiratory depression and other side effects. Intrathecal or epidural opioids can facilitate early extubation and discharge from an ICU without compromising pain control or increasing the likelihood of myocardial ischemia [70]. Intrathecal morphine may be useful in attenuating the postsurgical stress response in coronary artery bypass graft (CABG) patients as measured by plasma cortisol and epinephrine concentrations [71]. This evidence suggests that intrathecal opioids may be an excellent pain management choice in preparing the cardiac surgical patient for early extubation and fast-tracking in the ICU. However, this approach has not gained widespread support perhaps because the principal proved advantage is decreased systemic opioid use.

   D. Nerve blocks. A variety of systemic and intrathecal analgesic techniques have been reviewed. Although these techniques are useful, each has inherent risks and complications. Nerve blocks constitute a potential alternative to these methods. Intercostal nerve blocks can be performed with ease during thoracic surgery procedures, as the intercostal nerves are easily accessible through the surgical field. These blocks can also be performed percutaneously by the anesthesia provider preoperatively or postoperatively. Intercostal nerve blocks do not provide satisfactory analgesia for a median sternotomy. Thoracic epidural analgesia (TEA) for cardiac surgical procedures [69] requiring CPB is considered acceptable by some practitioners and appears to be more accepted in Europe and Asia than in North America. Off-pump coronary artery bypass (OPCAB) procedures may be especially well suited to TEA. TEA decreases the risk of ischemia during cardiac surgery. However, many cardiac anesthesiologists consider the risk of epidural hematoma, however small, to be a deterrent in the face of the hypocoagulable state present during and after CPB. Paravertebral nerve blocks (PVBs) may be an alternative to TEA, which may be associated with less risk of epidural hematoma. PVB and bilateral PVB catheters have been utilized with success in cardiac surgical procedures [7274]. Recently parasternal nerve blockade has been described and may prove to be useful [75].

   E. Sedation. It is essential to understand the goals of sedation in a critically ill patient. In the ICU, one should titrate sedatives to the desired effect as precisely as one would titrate vasopressors, inotropes, and oxygen. Light sedation with aggressive analgesic techniques shortens ICU stay, decreases delirium, decreases post-traumatic stress disorder, and probably improves mortality. Sedation should be goal directed. The majority of patients in the ICU should be awake and interactive even if they require mechanical ventilation [76]. Not all postoperative cardiac surgery patients require sedation (see Table 10.4).

Table 10.4 ICU sedation

     1. Benzodiazepines, when used as sedatives, have been associated with prolonged mechanical ventilation, delirium, and possibly increased mortality when used as continuous infusions [77,78].

     2. Propofol is commonly administered as a continuous infusion in the ICU for sedation. It is easily titrated to effect but can produce significant vasodilation or myocardial depression with resultant hypotension. Propofol causes a burning pain during peripheral administration, so central venous administration is advisable if possible. In addition, propofol can be a source for sepsis since the lipid mixture can act as a medium for bacterial growth; strict aseptic technique must be used.

     3. Dexmedetomidine is a potent a2-adrenergic agonist that provides sedation, hemodynamic stability, decreased cardiac ischemia, improved pulmonary function, and analgesia, and can therefore be a good choice in cardiac surgical patients [7982]. Hypotension and bradycardia are occasionally attributed to its use. Dexmedetomidine is unique among sedatives in that patients remain cooperative yet calm and it does not require weaning to permit extubation. Dexmedetomidine activates endogenous sleep pathways, which may decrease agitation and confusion. Dexmedetomidine is administered as a continuous intravenous infusion (0.2 to 0.7 mg/kg/hr) and is approved for use up to 24 hrs. Clinical trials for extended infusion of dexmedetomidine at higher doses (up to1.5 mg/kg/hr) in a diverse ICU population have resulted in more ventilator-free days and more delirium-free days when compared to similar sedation end points with lorazepam or midazolam. Withdrawal syndromes (ethanol, chronic pain narcotic use, illicit drugs) can also be managed with the administration of a2-adrenergic agonists.

VII. Metabolic abnormalities

Many metabolic abnormalities can occur in the perioperative period. These irregularities result from the physiologic stress response or from the large fluid and electrolyte shifts that can derive from intravenous infusions or from CPB priming or myocardial protectant solutions.

   A. Electrolyte abnormalities

     1. Hyperkalemia can present from cardioplegia, overaggressive replacement, or secondary extracellular shifts associated with respiratory or metabolic acidosis. Hypokalemia increases the risk for dysrhythmia following cardiac surgery. Administration of mannitol in the CPB prime, improved renal perfusion, and aggressive treatment of blood glucose with insulin infusion all contribute to hypokalemia. Hypokalemia is more common than hyperkalemia. Potassium supplementation can be infused at a maximum rate of 20 mEq/hr via a central venous catheter. Rapid potassium infusion can induce lethal arrhythmias. The target serum potassium concentration should be 3.5 to 5 mEq/L.

     2. Hypomagnesemia is a common perioperative electrolyte abnormality and is associated with postoperative dysrhythmia, myocardial ischemia, and ventricular dysfunction [58,83]. Hypomagnesemia may result from dilution by large CPB priming volumes and from urinary excretion. Renal potassium retention requires adequate magnesium concentration and thus magnesium administration should always be considered in hypokalemic patients. If magnesium supplementation is required, it can be given in amounts of 2 to 4 g intravenously over 30 to 45 min. An infusion of 1 g/hr of magnesium sulfate can be used as well to assure a slow, steady infusion of this substance. If given too fast, it may cause hypotension or muscle weakness. In refractory dysrhythmias, particularly of the ventricular type, a normal serum magnesium concentration may not exclude the possibility of decreased total body stores of magnesium. The target serum magnesium concentration should be 2 to 2.5 mEq/L.

     3. Hypocalcemia may be present and may be related to rapid transfusion of large amounts of citrate-preserved bank blood. Hypocalcemia can be treated with 250 to 1,000 mg intravenous doses of calcium chloride or calcium gluconate, while paying careful attention to the potential for development of dysrhythmias. When following the calcium status, it is important to measure ionized calcium and not total calcium, because low albumin levels may decrease total calcium levels, whereas ionized calcium remains normal.

     4. Hypophosphatemia is a common problem encountered in the ICU. Hypophosphatemia can contribute to weakness and poor myocardial function, can change the ability of red blood cells to change shape, and can affect oxyhemoglobin dissociation [84,85].

   B. Shivering. The exact mechanism of shivering is difficult to discern, but it is thought to be associated with inadequate rewarming and the resulting hypothermic temperature “afterdrop.” Many patients are hypothermic when they arrive in the ICU and develop shivering as they emerge from anesthesia. Shivering can result in a 300% to 600% increase in oxygen demand, which potentially places unachievable oxygen delivery demand upon a compromised myocardium. The associated increase in CO2 production may cause respiratory acidosis. In patients with inadequate end-organ oxygen delivery, sustained shivering frequently will require mechanical ventilation and consideration should be given to administration of neuromuscular blockers to abolish the increased metabolic demand of shivering. Effective first-line treatments include active rewarming with forced air blankets and prevention of further temperature loss. Pharmacologic interventions that reduce shivering include intravenous meperidine (12.5 to 25 mg) or dexmedetomidine.

   C. Acidosis can be described as respiratory, metabolic, or mixed. Metabolic acidosis is divided into anion gap and non-anion gap acidosis. A complete discussion of acid–base disorders is beyond the scope of this chapter and can be found in standard critical care textbooks. We will briefly discuss the most common causes of acidosis in the early postoperative period.


     1. Respiratory acidosis results from hypoventilation or increased CO2 production. Residual anesthetics or an awakening patient with inadequate analgesia combined with impaired respiratory mechanics may lead to hypoventilation. Treatment consists of support of ventilation while treating the underlying cause.

     2. Metabolic acidosis, when present, is associated frequently with inadequate systemic perfusion because of compromised cardiac function. Treatment is directed at correcting the underlying cause of the acidosis. Metabolic acidosis in a cardiac surgery patient may require administration of sodium bicarbonate as a temporizing measure, especially in patients who are hemodynamically unstable.

     3. Lactic acidosis, a frequent finding in cardiac surgery patients, needs to be managed by assuring adequate CO and intravascular volume, and avoidance of shivering. There is some evidence that the use of sodium bicarbonate to treat lactic acidosis should be avoided; however, most of the critical care data assessing the effects of buffers in treating acidosis do not include patients with acute pulmonary hypertension or right heart failure. Epinephrine can produce a transient lactic acidosis that does not appear to reflect inadequate perfusion in the patient receiving the drug [35].

   D. Glucose management. Glycemic control in the critically ill is a highly controversial topic. Data from 2001 indicated that tight glycemic control in cardiac surgery patients (80 to 110 mg/dL) induced a remarkable improvement in mortality [86]; however, subsequent trials have failed to reproduce this dramatic effect. The most recent multicenter multinational interventional trial of tight glycemic control demonstrated an increase in mortality when hyperglycemia was aggressively managed to maintain a glucose range of 81 to 108 mg/dL [87]. Control of blood glucose levels at less than 150 mg/dL (or some would say less than 180 to 200 mg/dL, as there is no consensus standard) and avoiding hypoglycemia with frequent monitoring is a reasonable goal in cardiac surgical patients. Unrecognized hyperglycemia can result in excessive diuresis and the potential for a hyperosmolar or ketoacidotic state. Elevated serum glucose can be managed by using a continuous infusion of regular insulin, often starting at a dose of 0.1 units/kg/hr or less with titration to the desired serum blood glucose level.

VIII. Complications in the first 24 hrs postoperatively

A number of life-threatening complications can occur in the first 24 hrs after cardiac or thoracic surgery.

   A. Respiratory failure. Respiratory failure may be the most common postoperative complication of cardiac or thoracic surgery. Pulmonary dysfunction develops from the surgical incision and its attendant disruption of the thoracic cage. Postoperative pain exacerbates this effect. Respiratory failure can present as hypoxemia, hypercarbia, or both. Prompt identification and appropriate treatment is essential. Atelectasis is the most common pulmonary complication following cardiac surgery and can usually be managed with the application of PEEP, BiPAP, or CPAP.

   B. Bleeding. Postoperative hemorrhage from ongoing surgical bleeding or coagulopathy increases a patient’s length of stay and mortality [88]. Bleeding typically is monitored by the amount of blood that drains into the chest tubes after surgery. It is critical to differentiate a bleeding diathesis from a surgical bleeding situation requiring reoperation. Consequently, in bleeding patients it becomes essential to determine the status of the coagulation system, which is traditionally done by acquiring (at a minimum) PT, aPTT, fibrinogen concentration, and platelet count in addition to a chest x-ray. This panel of tests does not provide any indication of the functional status of the platelets. Thromboelastogram (TEG) or other platelet functional assays provide useful information about platelet functional status, and the TEG also provides information about plasma clotting function and fibrinolysis. Assessment of platelet function is particularly important in managing the cardiac patients who have received aspirin or other platelet inhibitors (e.g., clopidogrel, glycoprotein IIb/IIIa inhibitors) preoperatively. Transfusion of platelet concentrates may be appropriate when one suspects that the bleeding results from platelet dysfunction.


       Fresh frozen plasma (FFP) should most often be used to correct abnormalities in PT, INR, or aPTT, although modest elevations in these tests are often clinically insignificant. When associated with clinical bleeding, elevations of PT or aPTT in excess of 1.3 times the upper limit of normal or of INR in excess of 1.5 should probably be treated. Elevated aPTT may occur from deficiencies of plasma coagulation factors or from residual heparin. Fibrinogen deficiency (typically less than 75 mg/dL) is treated with cryoprecipitate, and fibrinogen concentrates may be available in some countries.

       The role of activated factor VII in the cardiac surgical population is controversial. A meta-analysis of 35 randomized clinical trials that included both cardiac surgery and non-cardiac surgery trials showed no statistically significant difference in venous thrombosis after administration of activated factor VII including doses up to 80 mcg/kg. However, there was a statistically significant increase in coronary arterial thromboembolic events when compared to placebo [89].

       Surgical bleeding is often considered once coagulopathy has been ruled out, and it may require a return to the OR for mediastinal or thoracic re-exploration to identify and control a bleeding site. Surgical re-exploration commonly fails to identify a specific source for the bleeding. However, irrigation of the pericardium and mediastinum will remove activated plasminogen (from dissolved blood clots), which may markedly decrease bleeding. In general, chest tube drainage greater than 500 mL/hr, sustained drainage exceeding 200 mL/hr, or increasing chest tube drainage justifies surgical re-exploration.

       Sudden hemorrhage from a suture line or cannulation site can cause profound hypotension from hypovolemia or tamponade. Rapid infusion of blood products, colloids, or crystalloids is necessary to maintain intravascular volume. Patients who can be quickly stabilized are transferred to the OR. In some instances, emergency sternotomy must be performed in the ICU to control life-threatening hemorrhage or tamponade.

   C. Cardiac tamponade. Excessive mediastinal bleeding with inadequate drainage or sudden massive bleeding can result in cardiac tamponade. Cardiac tamponade after cardiac surgery may seem impossible if the pericardium has been left open, suggesting to the inexperienced observer that tamponade cannot occur. However, this is not true because tamponade may occur in localized areas, affecting an area as circumscribed as the right atrium. As discussed above, it is important to recognize that traditionally described equalization of filling pressures is an unreliable sign of post-cardiac surgery tamponade. The differential diagnosis includes biventricular failure, and TEE or transthoracic echocardiography may be important to making the correct diagnosis. At times, cardiac tamponade is a clinical diagnosis that requires emergent surgical intervention.

   D. Pneumothorax. Pneumothorax can occur in patients who have undergone sternotomy, thoracotomy, or are undergoing PPV. Most cardiac surgical patients arrive in the recovery or ICU area with one or more chest (intrapleural) tubes in place, in addition to mediastinal tubes. These patients should have a baseline postoperative chest x-ray upon ICU arrival to confirm the adequacy of chest tube placement and the absence of a pneumothorax. Patients who have had a redo sternotomy or who have had an internal mammary artery used for CABG are at particular high risk for a pneumothorax and hemothorax. There is increasing evidence that bedside ultrasound is valuable in diagnosing pneumothorax, hemothorax, and lung consolidation. In some studies, it has higher accuracy than portable chest x-ray when compared to CT scan [90]. An intensivist who is skilled in performing the exam can obtain immediate data and perform repeat exams if needed. Pneumothorax can convert to tension pneumothorax when chest tubes function improperly. The resulting shift of mediastinal structures can obstruct the vena cava or distort the heart to cause a low CO state and hypotension.

   E. Hemothorax. Hemothorax can occur after coronary artery bypass surgery and must be considered in all patients who have undergone internal mammary artery dissection, which most often involves opening the left intrapleural space. These patients may need to be returned to the OR for surgical management.

   F. Acute graft closure. Acute coronary graft closure, while uncommon, can result in myocardial ischemia or infarction. If cardiac decompensation occurs and graft closure is the suspected cause, re-exploration should be performed to evaluate graft patency. However, this may be a diagnosis of exclusion, and re-exploration for this reason is uncommon. These patients may need to be taken to the cardiac catheterization laboratory where emergent cardiac catheterization can be performed to discern the presence of an occluded graft. Controversy continues about whether graft patency is similar or different in OPCAB and CPB CABG procedures either acutely or long-term [91,92].

   G. Prosthetic valve failure. Acute prosthetic valve failure should be suspected when sudden hemodynamic changes occur following open heart surgery, particularly if the rhythm is unchanged and intermittent loss of the arterial waveform is noted on the monitor screen. Immediate surgical correction is necessary. Valve dehiscence with a perivalvular leak usually does not present in the early postoperative period. TEE is the diagnostic modality of choice to evaluate prosthetic valve function in this setting. There is increasing experience and success with closing or reducing perivalvular leaks in the interventional cardiology laboratory using percutaneous atrial septal defect (ASD) and ventricular septal defect (VSD) closure devices.

   H. Postoperative neurologic dysfunction. Neurologic complications are frequently recognized in the postoperative period and remain one of the most important complications following cardiac surgery [93,94], so an early postoperative neurologic examination is very important. Neurologic complications can be divided into three groups: (1) Focal ischemic injury (stroke), (2) neurocognitive dysfunction (including diffuse encephalopathy), and (3) peripheral nervous system injury. Central nervous system dysfunction after cardiac surgery is discussed in detail in Chapter 22. Brachial plexus injury resulting from sternal retraction can occur, particularly on the left side during internal mammary arterial dissection. A thorough motor exam of the legs is crucial after descending thoracic aortic or thoracoabdominal aneurysm repairs. A delirium assessment, like the Confusion Assessment Method for ICU Patients (CAM-ICU), can identify patients with hypoactive delirium [95]. Hypoactive delirium in the ICU carries an increased morbidity and mortality. Vanderbilt University ICU physicians have developed an excellent resource about delirium for families and clinicians, which can be found at www.icudelirium.org.


IX. Discharge from the intensive care unit

Discharge from the ICU historically has occurred 1 to 3 days after cardiothoracic surgery. Reducing the amount of time spent in the ICU after cardiac surgery recently has become a priority. Many patients are now discharged from the ICU on the morning after routine CABG operations without compromise in patient care or safety. Complications such as those noted earlier often delay ICU discharge. Some centers place routine CABG patients in an ICU-level recovery area for several hours before discharging them to a “step-down” or intermediate care area, or even to a “monitored bed” postoperative nursing unit.

The criteria for ICU discharge vary depending upon the type of surgery. Predicting which patients can leave the ICU in an early FT style can be accomplished by reviewing a variety of preoperative risk factors. Reduced LV ejection fraction is a valid predictor of higher mortality, morbidity, and resource utilization [96]. Other preoperative predictors of prolonged ICU stays include cardiogenic shock, age greater than 80 years, dialysis-dependent renal failure, and surgery performed emergently [97]. These factors and others can be used to predict a patient’s length of stay and to plan for resource utilization.

X. The transplant patient.

The care of cardiac transplant patients is similar to other cardiac patients with several notable exceptions [98]. Pulmonary hypertension and RV failure are the primary challenges in the early postoperative period following heart or lung transplantation [63]. Heart transplant recipients may have varying degrees of end-organ dysfunction secondary to chronic low CO. Finally, these patients require meticulous attention to medications used to attenuate graft rejection.

XI. Patients with mechanical assist devices (see also Chapter 16)

Recent technologic advances have facilitated the development of mechanical assist devices for cardiac surgery patients with severely impaired RV or LV function. The number of assist devices available continues to grow, resulting in options for LV, RV, or biventricular mechanical assistance [99,100]. Postoperative management of patients with mechanical assist devices requires a thorough understanding of the technology underlying any device that may be chosen. The primary risks with mechanical assistance include thrombosis, bleeding from anticoagulation, infection from percutaneous catheters, and failure to wean from assist devices intended for temporary cardiac support.

   A. Intra-aortic balloon pump (IABP). An IABP is typically the first mechanical assist device utilized for cardiogenic shock in cardiac surgical patients [101] (see Chapter 22). The goals of IABP therapy are to acutely decrease the LV afterload, thereby improving forward blood flow, and to augment the LV coronary artery perfusion during diastole. Coronary blood flow is only augmented in patients with hypotension associated with their cardiogenic shock [102]. The IABP is most often placed percutaneously via the femoral artery into the descending thoracic aorta. There are two management strategies for weaning IABP support. Some clinicians wean pharmacologic support prior to IABP; this allows them to resume inotropic support should the patient decompensate after IABP removal. Other clinicians wean IABP support first due to concerns about lower extremity ischemia from arterial occlusion of the femoral and/or iliac arteries. Despite the clear risk of lower extremity ischemia from arterial occlusion, there is no decrease in embolic or thrombotic complications when the patient undergoes systemic anticoagulation [103]. In either strategy, the IABP is weaned from a 1:1 setting (each cardiac contraction triggers an IABP deflation/inflation cycle) to 1:2, then a brief trial at 1:3 precedes IABP removal.


   B. Ventricular assist device (VAD) (see Chapter 22). VAD can be utilized for support of the left (LVAD), right (RVAD), or both (BiVAD) ventricles [99,100]. There are three indications for VAD therapy:

      1. Temporary support (less than 14 days) in patients who fail to wean from CPB but are expected to recover sufficient cardiac function for removal of the VAD.

      2. Bridge to cardiac transplantation.

      3. Destination VAD therapy; patients who require long-term or permanent VAD support with the expectation that they will be discharged from the hospital with the VAD.

          In patients with LVADs, the possibility that LV decompression will result in geometric alterations and precipitation of right heart failure must be considered. Other considerations that must be addressed during the ICU period include adequate oxygenation and ventilation using a mechanical ventilator, along with maintenance of temperature, nutrition, acid–base balance, and electrolyte balance. The patients with mechanical assist devices can be weaned from the ventilator using standard weaning protocols depending upon their hemodynamic status, blood gas exchange, and neurologic stability. If a VAD has been implanted as short-term support with an anticipated return to the OR within a few days to remove the device, this tends to discourage endotracheal extubation.

XII. Family issues in the postoperative period

Interaction with families is important in communicating any patient’s status and in giving appropriate expectations about recovery.

   A. The preoperative discussion. In the preoperative discussion, it is important for the surgeon and anesthesiologist to give a detailed description of what to anticipate in the postoperative period. This information can be relayed either through preoperative visits or through video or website access describing typical postoperative events. Detailed discussions about planned extubation in the OR or in the ICU may prevent patients from misinterpreting early postoperative intubation as being awake during surgery. It is also important to discuss the goals of general anesthesia versus sedation in the ICU. Caution should be used when assuring the patient that they will have complete amnesia of the OR or being intubated. Family members should be told to expect that the cardiac surgical patient will have invasive monitors and may have significant edema, which alters their appearance. Reassurance that these are temporary cosmetic changes will provide comfort. Preoperative visits allow patients and their families the opportunity to ask questions and to understand the plan of movement through the postoperative course in a more relaxed setting than the preanesthesia holding area. Many times, anesthesiology preoperative visits may be compromised or precluded by admission day surgery patterns whereby cardiac surgery patients arrive at the hospital on the day of surgery. In these circumstances, opportunities for discussions with anesthesia care providers can be limited and should be anticipated at the preoperative visit with the surgeon, when information can be disseminated and questions can be answered. Some centers compensate for this practice pattern with anesthesia-specific pamphlets or web-based videos, often with a Frequently Asked Questions section.

   B. Family visitation. Family visitation occurs in the ICU or recovery room for many postoperative cardiac and thoracic surgery patients. This provides reassurance about the patient’s clinical course progression as well as encouragement toward subsequent postoperative care. Family members can be very important in encouraging adequate pulmonary toilet, coughing, deep breathing, and early ambulation to improve postoperative outcomes. Most cardiac surgery programs have designated personnel for the liaison between the professional staff caring for postoperative cardiac surgery patients and the family members who need education, encouragement, and the opportunity to assist in postoperative care.

   C. The role of family support. Family support is a vital link toward the early success of a fast-tracking program. Family members need adequate education by surgical and anesthesia staff, who can outline the expected early postoperative events. The role of family support is heightened when patients spend very short periods in postoperative areas such as the recovery room or ICU. Family members who are educated about the expected postoperative course can facilitate postoperative care and smooth the transition from the ICU to a regular nursing floor and finally to the patient’s home.


The authors would like to thank Mark Gerhardt, MD, the author of the previous edition of this chapter, and Jennifer Olin for her editorial assistance.


 1. Warren J, Fromm REJ, Orr RA, et al. Guidelines for the inter- and intrahospital transport of critically ill patients. Crit Care Med. 2004;32:256–262.

 2. Commission J. Improving America’s hospitals: The Joint Commission’s annual report on quality and safety. http//wwwjointcommissionreportorg/pdf/JC_2006_Annual_Reportpdf 2006.

 3. Dunn W, Murphy JG. The patient handoff. Chest. 2008;134:9–12.

 4. Saver C. Handoffs: what ORs can learn from Formula One race crews. OR Manager. 2011;27:1–13.

 5. Logio LS, Djuricich AM. Handoffs in teaching hospitals: situation, background, assessment, and recommendation. Am J Med. 2010;123:563–567.

 6. Cohen MD, Hilligoss PB. The published literature on handoffs in hospitals: deficiencies identified in an extensive review. Qual Safety Health Care. 2010;19:493–497.

 7. Raptis DA, Fernandes C, Weiliang Chua, et al. Electronic software significantly improves quality of handover in a London teaching hospital. Health Inform J. 2009;15:191–198.

 8. Apfelbaum JL, Belott R, Connis RT. Practice advisory for the perioperative management of patients with cardiac implantable electronic devices: pacemakers and implantable cardioverter-defibrillators: an updated report by the American Society of Anesthesiologists Task Force on Perioperative Management of Patients with Cardiac Implantable Electronic Devices. Anesthesiology. 2011;114:247–261.

 9. Frazier SK. Cardiovascular effects of mechanical ventilation and weaning. Nurs Clin North Am. 2008;43:1–15.

10. Singh I, Pinsky MR. Mechanical Ventilation. Philadelphia, PA: Saunders Elsevier; 2008.

11. van Belle AF, Wesseling GJ, Penn OCKM, et al. Postoperative pulmonary function abnormalities after coronary artery bypass surgery. Resp Med. 1992;86:195–199.

12. Cox CM, Ascione R, Cohen AM, et al. Effect of cardiopulmonary bypass on pulmonary gas exchange: a prospective randomized study. Ann Thor Surg. 2000;69:140–145.

13. Jaber SD, Chanques G, Jung B. Postoperative noninvasive ventilation. Anesthesiology. 2010;112:453–461.

14. Burns KEA, Adhikari NKJ, Keenan SP, et al. Use of non-invasive ventilation to wean critically ill adults off invasive ventilation: meta-analysis and systematic review. BMJ. 2009;338.

15. Gattinoni L, Protti A, Caironi P, et al. Ventilator-induced lung injury: the anatomical and physiological framework. Crit Care Med. 2010;38(10) Proceedings of a Round Table Conference in Brussels, Belgium, March 2010:S539–S548.

16. Diaz JV, Brower R, Calfee CS, et al. Therapeutic strategies for severe acute lung injury. Crit Care Med. 2010;38:1644–1650.

17. Esan A, Hess DR, Raoof S, et al. Severe hypoxemic respiratory failure. Chest. 2010;137:1203–1216.

18. Raoof S, Goulet K, Esan A, et al. Severe hypoxemic respiratory failure. Chest. 2010;137:1437–1448.

19. Liu LL, Aldrich JM, Shimabukuro DW, et al. Rescue therapies for acute hypoxemic respiratory failure. Anesth Analg. 2010;111:693–702.

20. Sud S, Sud M, Friedrich JO, et al. High frequency oscillation in patients with acute lung injury and acute respiratory distress syndrome (ARDS): systematic review and meta-analysis. BMJ. 2010;340.

21. Brochard L, Thille AW. What is the proper approach to liberating the weak from mechanical ventilation? Crit Care Med. 2009;37:S410–S415.

22. MacIntyre N. Discontinuing mechanical ventilatory support. Chest. 2007;132:1049–1056.

23. Constantinides VA, Tekkis PP, Fazil A, et al. Fast-track failure after cardiac surgery: development of a prediction model. Crit Care Med. 2006;34:2875–2882.

24. Kogan A, Ghosh P, Preisman S, et al. Risk factors for failed “fast-tracking” after cardiac surgery in patients older than 70 years. J Cardiothor Vasc Anesth. 2008;22:530–535.

25. Zarate E, Latham P, White PF, et al. Fast-track cardiac anesthesia: use of remifentanil combined with intrathecal morphine as an alternative to sufentanil during desflurane anesthesia. Anesth Analg.2000;91:283–287.

26. Latham P, Zarate E, White PF, et al. Fast-track cardiac anesthesia: a comparison of remifentanil plus intrathecal morphine with sufentanil in a desflurane-based anesthetic. J Cardiothor Vasc Anesth.2000;14:645–651.

27. Dowd NP, Cheng DCH, Karski JM, et al. Intraoperative awareness in fast-track cardiac anesthesia. Anesthesiology. 1998;89:1068–1073.

28. Maldonado JR, Wysong A, van der Starre PJA, et al. Dexmedetomidine and the reduction of postoperative delirium after cardiac surgery. Psychosomatics. 2009;50:206–217.

29. Shehabi Y, Grant P, Wolfenden H, et al. Prevalence of delirium with dexmedetomidine compared with morphine based therapy after cardiac surgery: a randomized controlled trial (dexmedetomidine compared to morphine-DEXCOM study). Anesthesiology. 2009;111:1075–1084.

30. Herr DL, Sum-Ping STJ, England M. ICU sedation after coronary artery bypass graft surgery: dexmedetomidine-based versus propofol-based sedation regimens. J Cardiothor Vasc Anesth.2003;17:576–584.

31. Riker RR, Shehabi Y, Bokesch PM, et al. dexmedetomidine vs midazolam for sedation of critically ill patients. JAMA: J Am Med Assoc. 2009;301:489–499.

32. Novick R, Fox S, Stitt L, et al. Impact of the opening of a specialized cardiac surgery recovery unit on postoperative outcomes in an academic health sciences centre. Can J Anesth. 2007;54:737–743.

33. Srivastava AR, Banerjee A, Tempe DK, et al. A comprehensive approach to fast tracking in cardiac surgery: ambulatory low-risk open-heart surgery. Eur J Cardio-Thor Surg. 2008;33:955–960.

34. Hollenberg SM. Vasoactive drugs in circulatory shock. Am J Respir Crit Care Med. 2011;183:847–855.

35. Levy B, Perez P, Perny J, et al. Comparison of norepinephrine-dobutamine to epinephrine for hemodynamics, lactate metabolism, and organ function variables in cardiogenic shock. A prospective, randomized pilot study. Crit Care Med. 2011;39:450–455.

36. Prielipp RC, MacGregor DA, Butterworth JF, et al. Pharmacodynamics and pharmacokinetics of milrinone administration to increase oxygen delivery in critically ill patients. Chest 1996;109:1291–1301.

37. Toller WG, Stranz C. Levosimendan, a new inotropic and vasodilator agent. Anesthesiology. 2006;104:556–569.

38. Maharaj R, Metaxa V. Levosimendan and mortality after coronary revascularisation: a meta-analysis of randomised controlled trials. Crit Care. 2011;15:R140–R150.

39. Arroll B, Doughty R, Andersen V. Investigation and management of congestive heart failure. Br J Anaesth. 2010; 341:C3657.

40. Ezekowitz JA, Hernandez AF, Starling RC, et al. Standardizing care for acute decompensated heart failure in a large megatrial: The approach for the Acute Studies of Clinical Effectiveness of Nesiritide in Subjects with Decompensated Heart Failure (ASCEND-HF). Am Heart J. 2009;157:219–228.

41. Krum H, Teerlink J. Medical therapy for chronic heart failure. Lancet. 2011;378:713–721.

42. Shah AM, Mann DL. In search of new therapeutic targets and strategies for heart failure: recent advances in basic science. Lancet. 2011;378:704–712.

43. Chappell D, Jacob M, Hofmann-Kiefer K, et al. A rational approach to perioperative fluid management. Anesthesiology. 2008;109:723–740.

44. Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid responsiveness?. Chest. 2008;134:172–178.

45. Michard F, Teboul J. Using heart-lung interactions to assess fluid responsiveness during mechanical ventilation. Crit Care. 2000;4:282–289.

46. Marik PE, Cavallazzi R, Vasu T, et al. Dynamic changes in arterial waveform derived variables and fluid responsiveness in mechanically ventilated patients: a systematic review of the literature. Crit Care Med.2009;37:2642–2647.

47. Ferraris VA, Brown JR, Despotis GJ, et al. 2011 update to The Society of Thoracic Surgeons and The Society of Cardiovascular Anesthesiologists blood conservation clinical practice guidelines. Ann Thoracic Surg.2011;91:944–982.

48. Varghese R, Myers ML. Blood conservation in cardiac surgery: let’s get restrictive. Semin Thor Cardiovasc Surg. 2010;22: 121–126.

49. Levin RL, Degrange MA, Bruno GF, et al. Methylene blue reduces mortality and morbidity in vasoplegic patients after cardiac surgery. Ann Thoracic Surg. 2004;77:496–499.

50. Maslow AD, Regan MM, Haering JM, et al. Echocardiographic predictors of left ventricular outflow tract obstruction and systolic anterior motion of the mitral valve after mitral valve reconstruction for myxomatous valve disease. J Am Coll Cardiol. 1999;34:2096–2104.

51. Russo AM, O’Connor WH, Waxman HL. Atypical presentations and echocardiographic findings in patients with cardiac tamponade occurring early and late after cardiac surgery. Chest. 1993;104:71–78.

52. Rho RW. The management of atrial fibrillation after cardiac surgery. Heart. 2009;95:422–429.

53. Bradley D, Creswell LL, Hogue CW, et al. Pharmacologic prophylaxis: American College of Chest Physicians guidelines for the prevention and management of postoperative atrial fibrillation after cardiac surgery. Chest.2005;128:39S–47S.

54. Halonen J, Loponen P, Järvinen O, et al. Metoprolol versus amiodarone in the prevention of atrial fibrillation after cardiac surgery. Ann Intern Med. 2010;153:703–709.

55. Chen WT, Krishnan GM, Sood N, et al. Effect of statins on atrial fibrillation after cardiac surgery: A duration- and dose-response meta-analysis. J Thor Cardiovasc Surg. 2010;140:364–372.

56. Marik PE, Fromm R. The efficacy and dosage effect of corticosteroids for the prevention of atrial fibrillation after cardiac surgery: a systematic review. J Crit Care. 2009;24:458–463.

57. Ho KM, Tan JA. Benefits and risks of corticosteroid prophylaxis in adult cardiac surgery. Circulation. 2009;119:1853–1866.

58. Booth JV, Phillips-Bute B, McCants CB, et al. Low serum magnesium level predicts major adverse cardiac events after coronary artery bypass graft surgery. Am Heart J. 2003;145:1108–1113.

59. Cook RC, Humphries KH, Gin K, et al. Prophylactic intravenous magnesium sulphate in addition to oral b-blockade does not prevent atrial arrhythmias after coronary artery or valvular heart surgery. Circulation.2009;120:S163–S169.

60. Forfia PR, Fisher MR, Mathai SC, et al. Tricuspid annular displacement predicts survival in pulmonary hypertension. Am J Respir Crit Care Med. 2006;174:1034–1041.

61. Tei C, Dujardin KS, Hodge DO, et al. Doppler echocardiographic index for assessment of global right ventricular function. J Am Society Echocardiogr. 1996;9:838–847.

62. Meluzín J, Špinarová L, Bakala J, et al. Pulsed Doppler tissue imaging of the velocity of tricuspid annular systolic motion;. a new, rapid, and non-invasive method of evaluating right ventricular systolic function. Eur Heart J. 2001;22:340–348.

63. Rosenberg AL, Rao M, Benedict PE. Anesthetic implications for lung transplantation. Anesthesiol Clin North Am. 2004;22:767–788.

64. Taylor MB, Laussen PC. Fundamentals of management of acute postoperative pulmonary hypertension. Pediatr Crit Care Med. 2010;11:S27–S29.

65. Gordon C, Collard CD, Pan W. Intraoperative management of pulmonary hypertension and associated right heart failure. Curr Opin Anesthesiol. 2010;23:49–56.

66. Lahm T, McCaslin CA, Wozniak TC, et al. Medical and surgical treatment of acute right ventricular failure. J Am Coll Cardiol. 2010;56:1435–1446.

67. Price LC, Wort SJ, Finney SJ, et al. Pulmonary vascular and right ventricular dysfunction in adult critical care: current and emerging options for management: a systematic literature review. Crit Care.2010;14:R169–R191.

68. Hynninen M, Cheng D, Hossain I, et al. Non-steroidal anti-inflammatory drugs in treatment of postoperative pain after cardiac surgery. Can J Anesth. 2000;47:1182–1187.

69. Chaney MA. Intrathecal and epidural anesthesia and analgesia for cardiac surgery. Anesth Analg. 2006;102:45–64.

70. Shroff A, Rooke GA, Bishop MJ. Effects of intrathecal opioid on extubation time, analgesia, and intensive care unit stay following coronary artery bypass grafting. J Clin Anesth. 1997;9:415–419.

71. Hall R, Adderley N, MacLaren C, et al. Does intrathecal morphine alter the stress response following coronary artery bypass grafting surgery? Can J Anesth. 2000;47:463–466.

72. Ganapathy S, Murkin JM, Boyd DW, et al. Continuous percutaneous paravertebral block for minimally invasive cardiac surgery. J Cardiothor Vasc Anesth. 1999;13:594–596.

73. Cantó M, Sánchez MJ, Casas MA, et al. Bilateral paravertebral blockade for conventional cardiac surgery. Anaesthesia. 2003;58:365–370.

74. Dhole S, Mehta Y, Saxena H, et al. Comparison of continuous thoracic epidural and paravertebral blocks for postoperative analgesia after minimally invasive direct coronary artery bypass surgery. J Cardiothor Vasc Anesth.2001;15:288–292.

75. McDonald SB, Jacobsohn E, Kopacz DJ, et al. Parasternal block and local anesthetic infiltration with levobupivacaine after cardiac surgery with desflurane: the effect on postoperative pain, pulmonary function, and tracheal extubation times. Anesth Analg. 2005;100:25–32.

76. Jackson DL PC, Cann KF, Walsh T. A systematic review of the impact of sedation practice in the ICU on resource use, costs and patient safety. Crit Care. 2010;14:R59.

77. Pandharipande P, Shintani A, Peterson J, et al. Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients. Anesthesiology. 2006;104:21–26.

78. Pandharipande P, Cotton BA, Shintani A, et al. Prevalence and risk factors for development of delirium in surgical and trauma intensive care unit patients. J Trauma. 2008;65:34–41.

79. Wijeysundera DN, Naik JS, Scott BW. Alpha-2 adrenergic agonists to prevent perioperative cardiovascular complications: a meta-analysis. Am J Med. 2003;114:742–752.

80. Dasta JF, Jacobi J, Sesti A-M, et al. Addition of dexmedetomidine to standard sedation regimens after cardiac surgery: an outcomes analysis. Pharmacotherapy. 2006;26:798–805.

81. Finfer S, Bellomo R, Boyce N. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004;350:2247–2256.

82. Aantaa R, Jalonen J. Periopertive use of alpha2-adrenoceptor agonists and the cardiac patient. Eur J Anaesthesiol. 2006;23: 361–372.

83. Chakraborti S, Chakraborti T, Mandal M, et al. Protective role of magnesium in cardiovascular diseases: a review. Mol Cell Biochem. 2002;238:163–179.

84. Brown G, Greenwood J. Drug- and nutrition-induced hypophosphatemia: mechanisms and relevance in the critically ill. Ann Pharmacother. 1994;28:626–632.

85. Davis SV, Olichwier KK, Chakko SC. Reversible depression of myocardial performance in hypophosphatemia. Am J Med Sci. 1988;295:183–187.

86. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001;345: 1359–1367.

87. Finfer S, Chittock DR, Su YS. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283–1297.

88. Hein OV, Birnbaum J, Wernecke KD, et al. Three-year survival after four major post-cardiac operative complications. Crit Care Med. 2006;34:2729–2737.

89. Levi M, Levy JH, Andersen HF, et al. Safety of recombinant activated factor VII in randomized clinical trials. N Engl J Med. 2010;363:1791–1800.

90. Xirouchaki N, Magkanas E, Vaporidi K, et al. Lung ultrasound in critically ill patients: comparison with bedside chest radiography. Intens Care Med. 2011;37:1488–1493.

91. Puskas JD, Williams WH, Mahoney EM, et al. Off-Pump vs conventional coronary artery bypass grafting: early and 1-year graft patency, cost, and quality-of-life outcomes. JAMA: J Am Med Assoc.2004;291:1841–1849.

92. Parolari A, Alamanni F, Polvani G, et al. Meta-analysis of randomized trials comparing off-pump with on-pump coronary artery bypass graft patency. Ann Thoracic Surg. 2005;80:2121–2125.

93. Stroobant N, Van Nooten G, Van Belleghem Y, et al. The effect of CABAG on neurocognitive functioning. Acta Cardiologica. 2010;65:557–564.

94. Deiner S, Silverstein JH. Postoperative delirium and cognitive dysfunction. Br J Anaesth. 2009;103:i41–i46.

95. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated patients. JAMA: J Am Med Assoc. 2001;286: 2703–2710.

96. Kay GL, Sun G-W, Aoki A, et al. Influence of ejection fraction on hospital mortality, morbidity, and costs for CABG patients. Ann Thoracic Surg. 1995;60:1640–1651.

97. Doering LV, Esmailian F, Laks H. Perioperative predictors of ICU and hospital costs in coronary artery bypass graft surgery. Chest. 2000;118:736–743.

98. Sista RR, Wall M. Postoperative care of the patient after heart or lung transplantation. Postoperative Cardiac Care. Richmond, VA: Society of Cardiovascular Anesthesiologists; 2011.

99. Thunberg CA, Gaitan BD, Arabia FA, et al. Ventricular assist devices today and tomorrow. J Cardiothor Vasc Anesth. 2010;24:656–680.

100. Naidu SS. Novel percutaneous cardiac assist devices. Circulation. 2011;123:533–543.

101. Trost JC, Hillis LD. Intra-aortic balloon counterpulsation. Am J Cardiol. 2006;97:1391–1398.

102. Williams D, Korr K, Gewirtz H, et al. The effect of intraaortic balloon counterpulsation on regional myocardial blood flow and oxygen consumption in the presence of coronary artery stenosis in patients with unstable angina. Circulation. 1982;66: 593–597.

103. Jiang CY, Zhao LL, Wang JA, et al. Anticoagulation therapy in intra-aortic balloon counterpulsation: does IABP really need anti. J Zhejiang Univ Sci. 2003;4:607–611.