Management of the Postoperative Cardiac Surgical Patient

Published on 22/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1616 times

194 Management of the Postoperative Cardiac Surgical Patient

Sajid Shahul, Daniel Talmor, Alan Lisbon

The first days of care for the cardiac surgery patient present multiple challenges for the intensivist. The intensive care unit (ICU) stay for most of these patients lasts for only 24 to 48 hours, but during this period, life-threatening problems such as low cardiac output (CO), arrhythmias, and coagulopathy may become apparent. After 48 hours in the ICU, the problems encountered by postoperative cardiac surgery patients tend to become more like those experienced by other groups of critically ill patients.

The Cardiac Surgery Patient in the Intensive Care Unit

History of Cardiac Surgery Linked to the History of Intensive Care

The development of modern cardiac surgery has been intimately related to the development of the ICU. This relationship has worked in both directions. Until the 1950s, cardiac surgery was limited to control of traumatic injuries and the closed repair of valves. Development of the extracorporeal pump oxygenator in 1953 by Gibbon ushered in the era of open-heart surgery.¹ Heart valve replacement then became possible. Subsequently, in the 1960s, coronary artery bypass grafting (CABG) for ischemic heart disease was developed and rapidly popularized.²

Several studies have demonstrated that risk-adjusted mortality rates after CABG vary significantly among surgeons and hospitals and that mortality is related both to the number of surgeries performed by each surgeon and the total volume of procedures performed at the hospital.³^–⁵ For high-risk surgical patients, survival is also related to the characteristics of the ICU care.⁶

The Changing Epidemiology of Cardiac Surgery

Over the last decade, the population of patients treated with cardiac surgery has changed dramatically. Advances in cardiology including reperfusion therapy, angioplasty, stenting, and drug-eluting stents, have obviated the need for surgical approaches to treatment except for particularly complex problems or after failure of other less invasive modalities. In the year 2000, 561,000 patients in the United States underwent percutaneous transluminal coronary angioplasty (PTCA), an increase of 262% relative to 1987. In the same year, 314,000 patients underwent CABG. Multiyear trends, represented in Figure 194-1, show a leveling off and subsequent decrease in the overall number of patients undergoing CABG.⁷ The recently developed sirolimus-coated coronary stent has been associated with even better results.⁸ Studies comparing the use of stents versus CABG for left main disease have found no significant difference in rates of death or of the composite endpoint of death, Q-wave infarction, or stroke between patients receiving stents and those undergoing CABG. However, stenting, even with drug-eluting stents, was associated with higher rates of target-vessel revascularization than was CABG.⁹

Figure 194-1 Trends in cardiovascular operations and procedures in the United States, 1979-2000. PTCA, percutaneous transluminal coronary angioplasty.

(From American Heart Association. Heart disease and stroke statistics—2003 update. Dallas, TX: AHA; 2003.)

Even as younger patients are being treated with interventional techniques, the elderly are increasingly referred for operation. Although these operations are successful even in most octogenarians, they are associated with increased hospital mortality and longer ICU and hospital stays. It is clear, however, that good results in terms of long-term survival and quality of life are achievable.

Alternative Techniques for Cardiac Surgery

The increasing age of patients undergoing cardiac surgery and the relatively high incidence of adverse effects related to cardiopulmonary bypass (CPB) in these patients have led to the development of less invasive cardiac surgical techniques. These techniques are intended to decrease postoperative morbidity, reduce hospital length of stay, reduce costs, and hasten recovery of lifestyle (Table 194-1). Three major techniques have been proposed.

TABLE 194-1 Comparison of Minimally Invasive Cardiac Surgery Techniques

Minimally invasive direct coronary artery bypass (MIDCAB) differs from conventional CABG mainly in the type of incision used for access. In place of the conventional median sternotomy, access is obtained via a left or right thoracotomy, a parasternal incision, or a partial sternotomy. The proposed benefit of such an approach is the reduction in morbidity related to median sternotomy. This proposed advantage has not been demonstrated. MIDCAB grafting is a challenging technique and should be performed only in selected patients with favorable coronary anatomy. Both bare metal and drug-eluting stenting have been shown to be inferior to MIDCAB for proximal left anterior descending (LAD) coronary artery lesions, owing to higher reintervention rates with similar results in mortality and morbidity.^4,¹⁰

Off-pump coronary artery bypass (OPCAB) is performed on a beating heart without benefit of CPB. The proposed benefit of this procedure is reduction of morbidity related to hypothermia and CPB. The procedure is undertaken using partial to full heparinization. Extubation may be achieved earlier in these patients because they do not require rewarming and are less coagulopathic. A subset of patients cannot tolerate the extent of retraction of the heart required for the surgery and need to be urgently placed on CPB. These patients may suffer ischemic myocardial injury and require support with inotropes or intraaortic balloon pumping (IABP) during the postoperative period. A retrospective study of 1398 patients showed that use of the OPCAB technique for multivessel myocardial revascularization in high-risk patients significantly reduced the incidence of perioperative myocardial infarction (MI) and other major complications, length of stay in the ICU, and mortality.¹¹ In a single-center non-randomized registry, the incidence of major cardiac events were similar in OPCAB versus sirolimus-eluting stents in diabetic patients with multivessel disease.¹²

A third method of minimally invasive cardiac surgery is the port-access technique. This operation entails obtaining access for CPB with the use of endovascular catheters. This allows surgery to be performed using CPB via either a left or right thoracotomy. The technique is particularly useful for mitral valve replacement through a right thoracotomy and for redo CABG (avoiding the complications associated with repeat sternotomy). The port-access technique has been shown to be safe and is associated with shorter lengths of stay, reduced transfusion requirements, fewer infections, decreased incidence of renal failure, and less atrial fibrillation when compared with conventional techniques.¹³ In outcome data using propensity score analysis for mitral valve repair, minimally invasive repair had similar results to open repair. There was an increase in cross-clamp and bypass times, but early outcome was similar.¹⁴ Widespread adoption of this technique has been limited by the technical complexity of placing the required catheters, which requires both extra time and a specially trained and skilled operative team.

The techniques of minimally invasive cardiac surgery are still evolving. The intensivist caring for cardiac surgical patients must continue to keep abreast of these new methods.

Organization of the Postoperative Cardiac Surgery Unit

Optimal results from cardiac surgery require a skilled, dedicated, and multidisciplinary ICU team. Patients undergoing cardiac surgery are usually admitted to the hospital on the day of surgery. They arrive in the ICU directly from the operating room (OR). The typical patient is transferred to a step-down unit on the morning after surgery. This unit allows continued monitoring with telemetry for an additional 24 to 48 hours. Patients remaining in the ICU beyond 48 hours tend to become similar to a standard ICU population, as they develop secondary complications such as sepsis, pneumonia, and acute respiratory distress syndrome (ARDS).

Guidelines developed by the American Heart Association and the American College of Cardiology outline the requirements for cardiac surgical ICUs.¹⁵ These include the development of protocol-driven care, a minimum number of cardiac surgical ICU beds that is half the number of surgeries performed per week, and one-to-one nursing care during the first night in the unit. ICU coverage by a dedicated intensivist has been shown to improve outcomes in other types of major surgery and should be recommended after cardiac surgery as well.⁶

Separation from Cardiopulmonary Bypass and the End of Surgery

Successful management of the postoperative cardiac surgery patient begins by understanding what occurs in the OR. Problems encountered in the OR often persist after transfer to the ICU. An understanding of the technical and pathophysiologic aspects of CPB can help the intensivist better manage cardiac surgical patients in the ICU.

Cardiopulmonary Bypass

The goal of CPB is to separate the heart and lungs from the systemic circulation so that the heart can be arrested while the surgical repair is constructed. Blood is drained from the right side of the heart, either by gravity or with vacuum assistance, via a cannula in the right atrium directly or via a cannula in the femoral vein that is advanced into the right atrium. The blood is collected in a reservoir and then pumped through an oxygenator that contains a membrane where the blood is oxygenated and carbon dioxide is removed (Figure 194-2). The perfusionist controls both the fraction of inspired oxygen and the rate of oxygen flow through the circuit, thereby controlling the patient’s arterial oxygen and carbon dioxide levels, respectively. The treated blood then passes through an air filter and is returned to the patient via an arterial cannula placed in either the ascending aorta or the femoral artery. The perfusionist controls the amount of flow provided to the patient (i.e., CO). Mild to moderate systemic hypothermia (28°C-34°C) is used during bypass to minimize oxygen consumption by both the body and the brain. After adequate CPB is established, an aortic cross-clamp is applied to the ascending aorta, between the aortic cannula and the heart. The interval when the cross-clamp is applied is referred to as “ischemic” time, because no blood is circulated through the heart during this period. The heart is arrested by infusion of a high-concentration potassium solution into the native coronary arteries (antegrade cardioplegia) via a cannula placed between the aortic cross-clamp and the heart. Cardioplegia may also be given “backwards,” through the venous system of the myocardium (retrograde cardioplegia) via a catheter placed in the coronary sinus. Potassium is used as the arresting agent because it stops the heart from beating and minimizes myocardial oxygen consumption.

Figure 194-2 Cardiopulmonary bypass circuit.

(Adapted with permission from Gravlee GP, Davis RF, Kurusz M, Utley JR, editors. Cardiopulmonary bypass: principles and practice. 2nd ed. Baltimore: Lippincott Williams & Wilkins; 2000, p. 70.)

Myocardial Protection

Several measures are taken to protect the heart during ischemic time, because irreversible myocardial damage may otherwise occur. Electromechanical arrest is the most important protective measure, because the beating action of the heart accounts for about 85% of the heart’s total oxygen consumption. The heart is usually cooled to about 10°C with a cold cardioplegia solution (4°C) supplemented with topical ice slush. Additionally, the left ventricle is “vented” to prevent distention, which could lead to subendocardial ischemia. Finally, various additives are included in the cardioplegia solution to minimize myocardial edema, maintain normal intramyocardial pH, and provide substrates for anaerobic metabolism. The adequacy of intraoperative myocardial protection is critical for determining the subsequent course and final outcome of the patient

Separation from Cardiopulmonary Bypass

Weaning from CPB is the process whereby cardiopulmonary function is transferred from the bypass system back to the patient’s own heart and lungs. Successful separation from CPB requires that the metabolic, cardiac, and respiratory parameters are as close to normal as possible. Separation from CPB implies that the native circulation will be required to support the body’s metabolic demands. The surgical team manipulates the heart rate and rhythm, preload, afterload, and myocardial contractility to achieve this goal.

In most cases, normal sinus rhythm is restored after discontinuation of cardioplegia and rewarming of the heart. Occasionally, discontinuation of cardioplegia and rewarming leads to the onset of ventricular fibrillation; in such cases, electrical defibrillation is required. Other dysrhythmias commonly encountered are atrioventricular disassociation and atrial fibrillation. An attempt should be made to convert these to sinus rhythm by pharmacologic means. Bradyarrhythmias are treated by pacing, using temporary epicardial wires placed by the surgeon after completion of the repair. A heart rate of 70 to 90 beats/min usually is optimal. Pharmacologic support of the circulation may be needed to provide appropriate afterload or systemic vascular resistance (SVR) during separation from CPB. Most patients are vasodilated to some extent, possibly as a result of a systemic inflammatory response to CPB or the effects of rewarming, or both. As a consequence, infusion of a vasoconstrictor is often required. Care must be taken to strike a proper balance so that increased SVR maintains adequate arterial blood pressure without excessively increasing left ventricular afterload and compromising CO.

Most often, myocardial function is adequate, and infusion of an inotrope is not necessary. However, inotropic support often is needed for patients with a poor preoperative ventricular function or inadequate myocardial protection or revascularization during CPB. The optimal inotrope in this situation is a matter of considerable debate, and data are lacking to support a strong recommendation for a specific agent. Epinephrine, norepinephrine, dopamine, dobutamine, amrinone, and milrinone have all been used successfully. Intraoperative monitoring using transesophageal echocardiography (TEE) is particularly useful for titration of inotropic therapy.

Once all preparations for separation have been made, the perfusionist begins to wean the patient from bypass. This is done by slowly decreasing the amount of blood drained from the right atrium while simultaneously reducing flow into the aorta. Once the patient is off bypass (i.e., no blood is being drained from the right atrium into the CPB circuit), the perfusionist, at the direction of the anesthesiologist or surgeon, may continue to infuse through the aortic cannula. This maneuver allows optimization of ventricular filling or preload. Care must be taken, however, not to overdistend the heart; again, during this period, TEE is extremely useful.

Reversal of Anticoagulation

After weaning from CPB, protamine is given to neutralize any residual heparin. Dosing can be based on the patient’s weight, the total amount of heparin given, or an assay of residual heparin activity. Institutional preference governs the technique employed, and all have been proven effective. Several adverse responses to protamine administration are possible, including histamine-induced systemic hypotension, immunoglobulin E–mediated allergic reactions, and complement-mediated catastrophic pulmonary hypertension.

Transport and Admission to the Intensive Care Unit

After chest closure, confirmation of hemodynamic stability, and adequate medical and surgical hemostasis, the patient may be transferred to the ICU. Transport of a critically ill patient is a potentially dangerous process and requires extreme vigilance. Transport between the operating room and the ICU should be done with the same degree of monitoring as would be available at either end. This usually includes continuous monitoring of arterial blood pressure, pulmonary artery pressure and/or central venous pressure (CVP), electrocardiogram (ECG), and pulse oximetry. The transport bed should be equipped with a full oxygen tank, Ambu bag and mask, intubation equipment, resuscitation drugs, and a defibrillator. Care must be taken to ensure that infusions of vasoactive drugs are not interrupted.

On arrival in the ICU, the ICU team assumes care of the patient. A detailed sign-out from the operative team ensures continuity of care. The sign-out should include a detailed history including an assessment of preoperative cardiac functional status, a list of preoperative medications, and a detailed description of the surgery. Key facts are the type of repair performed, target vessels (if the patient has undergone CABG), duration of CPB and cross-clamping, difficulties encountered in separation from CPB, presence of abnormal bleeding, and postoperative assessment of cardiac function. All treatments administered in the OR should be detailed—in particular, fluids, blood products, and vasoactive drugs.

Once care has been handed over to the ICU team, a thorough examination of the patient should immediately follow. This examination should include verification of endotracheal tube placement, type and position of arterial or central venous lines, chest tube position and patency, and the presence and location of any epicardial pacing wires.

Monitoring the Postoperative Cardiac Surgery Patient

Hemodynamic Monitoring

All patients admitted to the ICU after cardiac surgery will have their blood pressure continuously monitored using an intraarterial line. This is usually placed in either a radial or femoral artery. Accuracy of the measurements depends on strict attention to calibration, leveling, and removal of air from the tubing. After CPB, femoral arterial pressure may more accurately reflect central aortic pressures,¹⁶ but this problem has usually resolved by the time the patient arrives in the ICU. If the radial artery is cannulated, the hand should be examined for signs of ischemia.¹⁷ Vascular complications of femoral arterial lines are extremely rare, but femoral catheters may be associated with an increased incidence of infection.¹⁸

Central venous access is required in all patients for drug administration and hemodynamic monitoring. In the low-risk patient, a CVP catheter may be all that is needed, particularly if echocardiography is available as a backup. Pulmonary artery catheters have the advantage of allowing measurement of pulmonary artery occlusion pressure (PAOP), thermodilution, and CO, as well as sampling of the mixed venous blood saturation (SvO₂). Use of the pulmonary artery catheter remains controversial. Improved outcome due to use of a pulmonary artery catheter for monitoring of cardiac surgical patients has not been demonstrated.¹⁹ Some studies showed an increased risk of death or adverse outcome when treatment was guided by the use of a pulmonary artery catheter.^20,²¹ However, many of these studies have been criticized on methodological grounds, and use of the catheter in cardiac surgery remains widespread.²² Current guidelines recommend use of the pulmonary artery catheter in high-risk patients undergoing surgery in an appropriate practice setting.²³ Such a setting is one in which the physician and nursing staff are familiar with the catheter and trained to properly interpret the information obtained. If echocardiography is readily available, it is possible to manage even high-risk patients using a CVP catheter.

Electrocardiography

On admission to the ICU, the patient is connected to a continuous ECG monitor, and a formal 12-lead ECG is obtained. The cardiogram is examined for rate, rhythm, QRS complex morphology, and signs of myocardial ischemia. For patients who are being paced postoperatively, the type of pacing and the degree of capture should be assessed.

Continuous ECG monitoring allows detection of arrhythmias. If an arrhythmia is detected, a 12-lead ECG should be obtained, and serum electrolyte concentrations should be measured. Treatment of arrhythmias should be carried out using established protocols.²⁴ If a malignant arrhythmia occurs, myocardial ischemia should be considered as a possible precipitating cause.

Monitoring of trends in ST-segment elevation or depression allows early detection of postoperative myocardial ischemia. Although transient ST-segment changes are relatively common and of unclear significance, persistent changes should be investigated by obtaining a 12-lead ECG and measuring circulating levels of creatine kinase myocardial band (CK-MB), troponin-T, or troponin-I.^25,²⁶ If ischemia is strongly suspected, then echocardiography followed by coronary angiography should be considered. Findings from these studies may indicate the need for further coronary revascularization.

Chest Radiography

The postoperative chest radiograph should be systematically evaluated. Proper placement of the endotracheal tube and any central lines inserted should be confirmed. If a pulmonary artery catheter is place, the location of its tip should be noted and adjusted as needed. The lung fields should be examined for the presence of pneumothorax or collapse. Additional air may be noted as subcutaneous emphysema or as pneumopericardium, although these findings are of little clinical significance. Further examination of the lung fields commonly shows small areas of atelectasis and pleural effusion. The cardiac silhouette is often enlarged after surgery as a result of myocardial edema and accumulation of fluid in the open pericardial sac. Increasing size of the cardiac silhouette or pleural effusions on serial chest radiographs may be evidence of ongoing mediastinal bleeding.

Echocardiography in the Intensive Care Unit

Echocardiography is an excellent tool for evaluating chamber size and function and the adequacy of valve repair or replacement. Indications include postoperative assessment of left ventricular function, assessment of unexplained sudden hemodynamic deterioration, evaluation to rule out pericardial tamponade, and workup of new cardiac ischemia. Limitations to transthoracic echocardiography (TTE) include inadequate windows early after operation due to air and edema in the soft tissues and wound dressings.

TEE is being used as a tool to facilitate decision making in the management of critically ill patients, including cardiac surgical patients. In the cardiac surgical ICU, this modality may have a particularly high yield when it is used to establish the cause of postoperative hypotension.²⁷ In one large series, a new diagnosis was established or an important pathology was excluded in 45% of TEE examinations performed in the ICU. Pericardial tamponade was diagnosed in 34 cases (11%) and excluded in 36 cases (12%). Other diagnoses included severe left ventricular failure and presence of large pleural effusions. The results of TEE had an impact on therapy in 220 cases (73%) by leading to a change of pharmacologic treatment and/or fluid administration, reoperation, or a decision that reoperation was unnecessary.²⁸

Clinical Manifestations of the Postbypass Period

The Normal Course

Patients are typically admitted to the ICU intubated and ventilated. Sedation with a short-acting agent, typically propofol, is continued until the patient is ready for extubation. Once hemodynamic stability is ascertained and chest tube drainage is judged to be under control, the patient is allowed to awaken. There is no need for prolonged weaning from mechanical ventilation. A short trial of spontaneous ventilation is sufficient to determine whether respiration will be adequate without mechanical support. The rapid shallow breathing index (RSBI) has been shown to be a sensitive way to assess the likelihood of successful extubation.²⁹ The RSBI is calculated by dividing the respiratory rate (in breaths per minute) by the tidal volume (in liters). A value of lower than 105 predicts successful extubation. Chest tubes are commonly removed on the first postoperative day. The pulmonary artery catheter, if present, is discontinued, and the patient may be transferred to a step-down unit.

Fast-tracking of cardiac surgical patients refers to a comprehensive program designed to reduce both length of stay and hospital costs.^30,³¹ As a part of this program, multiple anesthetic techniques designed to allow earlier postoperative extubation have been proposed, studied, and shown to be safe. These techniques may allow extubation in the OR.³² The key to proper use of this technique is patient selection. Although the criteria are expanding, patients with unstable angina or a high degree of congestive heart failure are generally not appropriate candidates for fast-tracking. In a retrospective review comparing 4020 patients undergoing cardiac surgery with a conventional anesthetic versus 3969 patients with a fast-track anesthetic, the fast-track group had shorter extubation times, shorter ICU or PACU stays, and a lower incidence of low cardiac output syndrome.

Low Cardiac Output

Low CO is the most common problem encountered in the postoperative cardiac surgical patient. A hallmark of low CO is low blood pressure. However, a patient may have a low CO with tissue hypoperfusion and still maintain what appears to be an adequate blood pressure. In the postoperative state, the physician must continuously examine and monitor the patient for signs of hypoperfusion. Physical signs of inadequate tissue perfusion include altered mental status; cool, pale, or even cyanotic extremities; diaphoresis; and low urine output. Global measures of hypoperfusion include increasing base deficit, elevated blood lactate concentration, and decreased SvO₂. Although the clinician must consider CO in terms of adequacy of perfusion, blood pressure per se is still important. Both the brain and kidneys depend on adequate blood pressure to maintain tissue perfusion. Additionally, coronary artery blood flow is dependent on a diastolic blood pressure, a key determinant of coronary artery perfusion pressure.

When assessing a patient with hypotension or signs of hypoperfusion, it is useful to consider the problem in relation to the components of CO; namely, preload, contractility, afterload, and rate and rhythm.

Preload

Preload refers to the stretch of the left ventricle at the end of diastole and is determined by the extent of ventricular filling during diastole. Adequate filling is required to ensure ejection in the subsequent systole. The most common cause of inadequate preload in postoperative patients is hypovolemia. Intravascular volume status should be continually monitored by assessing changes over time with respect to physical examination, chest tube output, and filling pressures (CVP, PAOP, or pulmonary artery diastolic pressure). Because none of the clinically measured filling pressures correlates perfectly with actual ventricular preload (i.e., end-diastolic volume), and correlation is particularly poor when the heart is diseased, it is often useful to obtain a “snapshot” of ventricular filling using echocardiography. By this means, it is possible to assess the relationship between measured filling pressures and actual preload in a specific patient. Preoperative catheterization data also can be helpful for determining this relationship. Hypovolemia should be treated with fluid replacement. Crystalloids are generally used. Surprisingly, there is no generally accepted hemoglobin concentration or hematocrit that should be used as a trigger for ordering transfusion of packed red blood cells. Red cell transfusion has been associated with early morbidity as well as long-term adverse sequelae.

In some cases, low preload is not caused by absolute hypovolemia but by relative or distributional hypovolemia. CPB and subsequent rewarming may lead to vasodilatation and a subsequent hypotension. Intravascular volume expansion may be required to maintain perfusion. An acceptable alternative is administration of a low dose of vasopressor such as phenylephrine or norepinephrine to maintain an adequate perfusion pressure. Recently, vasopressin in doses between 0.01 and 0.1 units/min has been demonstrated to be effective in this situation.^33,³⁴ Vasodilatation is usually a transient problem that resolves during the first several hours after separation from CPB. Continued vasodilatation after this period should prompt a search for another cause, particularly infection.

Pump Failure

Either or both ventricles may fail postoperatively. Decreased myocardial contractility may be caused by impaired preoperative function, inadequate revascularization at surgery, post-CPB reperfusion injury, or perioperative myocardial ischemia or MI. The incidence of infarction is approximately 5% in large series.³⁵ Preoperative myocardial function and the adequacy of revascularization at surgery should be clear from the history. Determination of circulating levels of CK-MB or troponin postoperatively can provide evidence of perioperative ischemia or infarction.^25,²⁶ Often, diminished contractility after operation is caused by inadequate myocardial protection during surgery. Decreased myocardial contractility secondary to inadequate myocardial protection usually resolves within the first 24 hours postoperatively. ECG changes are nonspecific.

Persistent new myocardial dysfunction associated with ECG changes and echocardiographic evidence of new wall-motion abnormalities should raise suspicion that the problem is an occluded graft and MI. Measurements of CK-MB in serum are of limited usefulness because levels of this enzyme are commonly elevated after surgery due to manipulation of the heart and incision of the atria, structures that are rich in the enzyme. If CK-MB levels are very high, greater than 80 mg/dL, then perioperative MI is likely.³⁶ Cardiac troponins are more specific for the diagnosis of perioperative infarction. A comparison of CK-MB, troponin-T, and troponin-I showed that a troponin-I level of greater than 5 µg/L was the most accurate indicator of MI, being superior to either troponin-T or CK-MB.³⁷ Elevated serum concentrations of troponin-I are associated with a cardiac cause of death and with major postoperative complications.³⁸ In addition, troponin-T concentrations measured after surgery are an independent predictor of in-hospital death after cardiac surgery.²⁶ If ischemia or MI is diagnosed, the patient may be taken for angiography or re-exploration and revascularization.

Postoperative valvular insufficiency can occur not only in patients with preexisting valvular lesions but also as a result of injury during surgery. The mitral valve is most commonly affected. Ischemia of the papillary muscles due to inadequate myocardial protection or perioperative MI can lead to acute mitral regurgitation in the postoperative period. Diagnosis is often made by TEE in the OR, but inadequate CO and a new systolic murmur should prompt echocardiographic evaluation.

Rate and Rhythm

CO is the product of heart rate (HR) times stroke volume (SV). Many dysrhythmias can adversely affect CO. If HR is too low, CO can be compromised. If HR is too fast, ventricular filling during diastole can be impaired, decreasing CO. Rhythm disturbances are common after cardiac surgery and may be divided into bradyarrhythmias and tachyarrhythmias; these categories are further divided into atrial and ventricular arrhythmias.

Bradycardia can lead to ventricular distention, increasing wall tension, and decreasing coronary perfusion pressure, factors that can promote development of ischemia and failure. HR of 80 to 90 appears to be optimal, allowing adequate filling and preventing overdistention but not causing rate-related ischemia. Bradycardia can be corrected by pacing. In general, epicardial pacing wires are left in place after chest closure and are attached to an external pacemaker in the immediate postoperative period. If the dysrhythmia is sinus bradycardia, atrial pacing is usually optimal. The second most common cause of bradyarrhythmia after cardiac surgery is atrioventricular dissociation. The combination of atrial and ventricular leads allows atrioventricular pacing for management of disassociation. Synchronization of the atrioventricular interval between 0.1 and 0.225 second optimizes CO.³⁹

Atrial fibrillation is the most common tachyarrhythmia. It occurs in 10% to 35% of patients after cardiac surgery, usually on the second or third postoperative day. Postoperative atrial fibrillation is associated with increased morbidity and mortality and with longer, more expensive hospital stays.⁴⁰ The Multicenter Study of Perioperative Ischemia (McSPI) group examined 2417 patients undergoing CABG with or without concurrent valvular surgery.⁴¹ The overall incidence of postoperative atrial fibrillation was 27%. Independent predictors of postoperative atrial fibrillation included advanced age, male sex, a past history of atrial fibrillation, a past history of congestive heart failure, and a pre-CPB heart rate greater than 100 beats/min. Surgical practices such as pulmonary vein venting, bicaval venous cannulation, postoperative atrial pacing, and longer cross-clamp times also were identified as independent predictors of postoperative atrial fibrillation. Patients who developed postoperative atrial fibrillation had longer lengths of stay, both in the ICU and in the ward, compared with patients who did not develop the complication.

Although premature ventricular contractions (PVCs) are common, sustained ventricular arrhythmias are far less frequent. Severe ventricular arrhythmias occurring after cardiac surgery are related to ischemia, hypoxemia, hypovolemia, electrolyte abnormalities, the effects of vasoactive drugs, or an underlying preexisting cardiomyopathy.⁴² In a series of 2100 cardiac operations, only 16 patients (0.8%) developed ventricular fibrillation or a sustained ventricular tachycardia during the interval from 3 days to 3 weeks after surgery. Ten of these patients had undergone valve surgery.⁴³ Prognosis in these patients is dependent on the preoperative ventricular prognosis; it is excellent in those with good function. In those with a left ventricular ejection fraction of less than 40%, the mortality rate may be as high as 75%.⁴⁴

Afterload

Ventricular afterload is the impedance to ventricular ejection during systole. Hypertension develops in as many as 60% of patients after surgery. Increased arterial blood pressure occurs even among patients without a preoperative history of hypertension. Predisposing factors include hypoxemia, hypercapnia, inadequate rewarming, pain, fluid overload, and increased sympathetic tone. Perioperative discontinuation of β-adrenergic blockers also may contribute to the development of postoperative hypertension.

Hypertension and increased afterload can lead to myocardial ischemia by augmenting ventricular stroke work. Additionally, hypertension may lead to bleeding from surgical sites, aortic dissection, and increased risk of stroke.

Tamponade

Tamponade refers to the hemodynamic consequences of a collection of blood or other fluid in the pericardial sac. In postsurgical patients, the presentation of tamponade may be subtle and differ significantly from classic descriptions. Equilibration of filling pressures typically is not seen. More commonly, patients present with isolated elevation of right atrial pressure due to compression of the right atrium and superior vena cava. After cardiac surgery, as many as 66% of pericardial fluid collections are loculated posterior effusions.⁴⁵

Bleeding from the atrial cannulation site is a common cause of tamponade. As the pressure on the right atrium increases, ventricular filling is impaired, and CO decreases. Diagnosis of tamponade is made difficult by the high overall frequency of pericardial effusions after surgery. Echocardiographic studies have shown that moderate effusions are present in 30% of patients on the eighth postoperative day, with 2% of patients having large effusions.⁴⁶

Diagnosis of tamponade in the postoperative patient requires a high index of suspicion and prompt intervention. Any hemodynamic instability should be assessed for tamponade. Low CO, hypotension, and tachycardia accompanied by an elevation of the left, the right, or both atrial pressures should lead to a prompt echocardiogram. Other signs that may be present include a widened mediastinum on chest radiography, dysrhythmias, and decreased ECG voltage. Because of the influence of positive pressure ventilation, the classic sign of pulsus paradoxus may not be present.

If time permits, the diagnosis of tamponade can be confirmed with the use of echocardiography. Although effusions are common, signs of compression or collapse of either atrium or of the right ventricle are diagnostic.⁴⁷^–⁴⁹ It is important to remember that the diagnosis may be made on clinical suspicion alone, and that treatment should not be withheld to await confirmation. Once tamponade is diagnosed, volume transfusion may temporize the situation. Pericardiocentesis is not effective in this situation, and prompt re-exploration for hemostasis and evacuation of clot is indicated.

Respiratory Complications

Patients undergoing cardiac surgery are at risk for multiple pulmonary complications. These include pneumothorax and pleural effusion in the immediate postoperative period. After the first 24 hours, patients sometimes develop acute lung injury (ALI), ARDS, or pneumonia. Diaphragmatic dysfunction secondary to phrenic nerve injury can occur.

Residual pneumothorax is often seen on the initial postoperative chest radiograph. The pneumothorax is commonly on the left side, and it is a result of opening the left parietal pleura during dissection of the left internal mammary artery. The pneumothorax usually resolves spontaneously as the chest tubes are placed on suction. Occasionally, a pneumothorax is seen on the right side as a result of accidental incision of the right parietal pleura. Right pneumothorax can progress to tension pneumothorax and significant hemodynamic deterioration. This diagnosis should be considered in any unstable patient. Treatment consists of insertion of an additional chest tube.

Pleural effusion in the first 24 hours after cardiac surgery should raise the suspicion of hemothorax. Effusions should be watched carefully for expansion and correlated with other signs and symptoms of continued bleeding. Massive, expanding hemothorax is an indication for re-exploration and hemostasis. Pleural effusion after the first 24 hours is generally a benign process. Most pleural effusions resolve spontaneously. Thoracocentesis should be performed only if the effusion occupies more than 50% of the lung field on radiography or if the patient has significant impairment of respiratory function.

ALI and ARDS are rare complications after cardiac surgery, CPB, and blood transfusion. In one retrospective study of 3278 cardiac surgical patients, only 13 (0.4%) developed ARDS during the postoperative period. The mortality rate associated with this complication was 15%. Another study reported a much higher mortality rate (70%).⁵⁰ The patients who developed ARDS were more likely than their matched controls to have had previous cardiac surgery. During the postoperative period, patients with ARDS received more blood products and developed shock more frequently than patients without ARDS.⁵¹

Nosocomial pneumonia can complicate any ICU stay. Patients who require mechanical ventilation for longer than 48 hours are at particular risk. These pneumonias are usually caused by aspiration of oral or gastric secretions into the lungs. The incidence of nosocomial pneumonia can be reduced by diligent mouth care to prevent pooling of secretions and elevation of the head of the bed to greater than 30 degrees. Nosocomial pneumonia carries a mortality rate of 24% to 50% and warrants appropriate broad-spectrum antimicrobial chemotherapy.⁵² The antibiotic prescription can be tailored once the results of sputum cultures are available.

Diaphragmatic dysfunction is usually caused by cold-induced injury of the phrenic nerve due to application of ice slush to the heart as part of the cardioplegia regimen. This complication occurs in up to 2% of patients undergoing cardiac surgery with topical hypothermia; more rarely, it can occur even if topical cooling was not applied.^53,⁵⁴ While the patient is being ventilated with positive pressure, this injury will not be apparent. If preoperative pulmonary function was normal, unilateral diaphragmatic paralysis usually is well tolerated. Pulmonary function can be severely compromised, however, if pulmonary problems were present preoperatively or, in rare instances, if bilateral diaphragmatic injury occurs.⁵⁵ These patients are at increased risk for development of nosocomial pneumonia, failure to wean from the ventilator, and death. Diaphragmatic dysfunction usually resolves spontaneously within 3 to 4 months.

Continued Bleeding

Continued bleeding is a common problem and requires immediate and aggressive management before the onset of further complications. The reasons for continued bleeding are often multifactorial and include inadequate surgical hemostasis, platelet dysfunction, coagulopathy, and inadequate heparin reversal. Often these factors occur in combination. Patients undergoing valve replacement are at increased risk.⁵⁶

Multiple clotting abnormalities are possible, most of which result either directly or indirectly from the use of CPB.⁵⁷ The tubing, blood reservoir, and oxygenator membrane are all foreign surfaces that can activate the clotting cascade. Because the pump must be primed with either normal saline or lactated Ringer’s solution, the priming process leads to substantial dilution of all blood components including red cells, platelets, and clotting factors. After CPB, the platelet count is decreased, and the remaining platelets are functionally deranged.^58,⁵⁹ There is sequestration of platelets in the liver, spleen, and in the CPB circuit itself. Systemic fibrinolysis due to activation of this system by the CPB circuit occurs.

Inadequate reversal of heparin should be diagnosed at the bedside by the activated coagulation test (ACT) or by measurement of the activated partial thromboplastin time (APTT). Because the half-life of heparin is longer than that of protamine, heparin-induced anticoagulation can rebound in the immediate postoperative period. The treatment is administration of additional protamine.

Renal Dysfunction

Mild renal dysfunction is a common postoperative event. One multicenter study demonstrated significant worsening of renal function in 7% of patients undergoing myocardial revascularization.⁶⁰ Approximately 1% of patients with postoperative acute renal failure (ARF) require renal replacement therapy. These patients have increased morbidity and mortality. Development of ARF can prolong ICU length of stay as much as fivefold.⁶⁰

A multicenter study of 2222 patients undergoing CABG identified five independent preoperative predictors of renal dysfunction: age 70 to 79 years or age 80 to 95 years, congestive heart failure, previous myocardial revascularization, type 1 diabetes mellitus, or preoperative serum glucose levels exceeding 300 mg/dL and preoperative serum creatinine levels of 1.4 to 2.0 mg/dL. Independent perioperative factors that exacerbated risk were CPB lasting 3 hours or longer and various measures of ventricular dysfunction.⁶⁰ The predominant predisposing factor appears to be low CO. This factor may be exacerbated by concurrent use of vasopressors such as phenylephrine.⁶¹

Renal dysfunction tends to follow one of three main patterns.⁶² Abbreviated ARF is a transient event, most probably related to intraoperative renal ischemia. The serum creatinine concentration can be expected to peak on day 4 after surgery. Overt ARF occurs when the duration of the predisposing insult, usually low CO, is longer. The serum creatinine concentration peaks at a higher level than with abbreviated ARF and then decreases over a period of several weeks. Protracted ARF occurs when a second insult, commonly sepsis or hypotension, is superimposed on the resolving renal function. This event triggers a further, often irreversible, decrease in renal function.

Neurologic Complications

Neurologic sequelae of CPB range from subtle neurocognitive deficits (appearing in up to 80% of patients) to stroke. In order to estimate the relative risks of neurologic sequelae associated with various clinical factors, a logistic regression model was applied to prospectively collected data from 273 patients enrolled at 24 American medical centers.⁶³ Adverse cerebral outcomes occurred in 16% of patients and were almost equally divided between type I outcomes (8.4%; 5 cerebral deaths, 16 nonfatal strokes, and 2 new transient ischemic attacks) and type II outcomes (7.3%; 17 new cases of intellectual deterioration persisting at hospital discharge and 3 cases of newly diagnosed seizure disorder). Resource utilization for these patients was significantly increased; median ICU stay was prolonged from 3 days to 6 to 8 days. Total duration of hospitalization was increased by 50% (type II, P = .04) to 100% (type I, P < .001). After discharge from the acute care setting, specialized care was required for 69% of the patients with adverse neurologic sequelae. Risk factors for type I outcomes related primarily to embolic phenomena including proximal aortic atherosclerosis, intracardiac thrombus, and intermittent clamping of the aorta during surgery. Risk factors for type II outcomes included, in addition to these factors, a preoperative history of endocarditis, alcohol abuse, perioperative dysrhythmia, poorly controlled hypertension, and low CO after CPB.

Gastrointestinal Complications

Acute abdominal complications are relatively rare after cardiac surgery. If they do occur, they are associated with extremely high rates of morbidity and mortality. One prospective study of 1116 patients undergoing CPB found that abdominal complications occurred in 23 (2.1%). Ten of these patients underwent subsequent abdominal surgery, and 20 died. Early complications occurred on postoperative days 6 and 7 and consisted of bowel ischemia or hepatic failure. These complications are probably related to perioperative hypotension and low CO.⁶⁴ Late complications consisted of pseudomembranous colitis, cholecystitis, pancreatitis, and rupture of a septic spleen.⁶⁵

Mild transient increases in circulating levels of hepatocellular enzymes are common after surgery. These changes are generally of no consequence; however, increased serum transaminase levels, if sustained or very high (e.g., serum alanine aminotransferase concentration greater than 500 IU/L), may represent evidence of severe ischemic injury of the liver. Severe ischemic liver injury after cardiac surgery carries a high mortality and is strongly associated with low CO and increased filling pressures, suggesting that liver ischemia is induced by a combination of decreased perfusion and congestion.⁶⁶

Management of Common Postoperative Problems

Optimization of Cardiac Output

Treatment of hypotension and low CO must be tailored to the cause. Again, it is useful to consider treatment in terms of preload, contractility, afterload, and rate and rhythm. Inadequate filling pressures are treated with volume infusion. The intravascular volume expander may be a crystalloid solution, a colloid solution, or packed red blood cells if hematocrit is low or there is evidence of ongoing bleeding. It is important to remember that inotropic therapy is ineffective and possibly detrimental if adequate blood volume is not restored.

If CO or blood pressure remains low despite intravascular volume resuscitation, then it is necessary to institute inotropic or vasopressor support. No single agent is optimal in all cases. Rather, selection of the agent should be based on the suspected cause of low CO or hypotension and knowledge of the pharmacologic effects of the various inotropic and vasopressor drugs that are available (Table 194-2). If the primary cause of hypotension appears to be vasodilatation, administration of a vasoconstrictor (e.g., phenylephrine, norepinephrine, vasopressin) is indicated. If hypotension is related to inadequate ventricular ejection, then inotropic therapy with a β-adrenergic agent should be instituted. Epinephrine, norepinephrine, dopamine, and dobutamine are all reasonable choices. In patients with chronic systolic dysfunction, response to these agents may be impaired. Chronically elevated levels of circulating catecholamines deplete myocardial norepinephrine stores and down-regulate expression of myocardial β-adrenergic receptors. In these patients, tachyphylaxis to β-adrenergic agonists can develop rapidly. Addition of a phosphodiesterase inhibitor such as amrinone or milrinone is often effective in these patients.^67,⁶⁸ In all cases, agents should be titrated to achieve adequate perfusion.

TABLE 194-2 Comparison of Relative Activity of Available Vasoactive Agents

Mechanical Support of the Circulation

Failure to respond to appropriate inotropic therapy may necessitate mechanical support of the circulation. IABP is the most commonly used method. The balloon is positioned in the aorta just distal to the take-off of the left common carotid artery. Inflation of the balloon during diastole increases diastolic pressure, thereby increasing coronary perfusion pressure. Deflation during systole decreases left ventricular afterload. This combination of hemodynamic effects ameliorates myocardial ischemia and improves CO.

Ventricular assist devices (VADs) are more effective than IABP for maintaining CO. Either the left ventricle, the right ventricle, or both can be supported with VADs. Currently, VADs may be used either as a bridge to transplantation or as a bridge to recovery. Either situation assumes that the VAD is a time-limited intervention. There are some data to support the view that resting the heart through the use of a VAD can allow some recovery of acutely injured myocytes, permitting eventual withdrawal of mechanical support. One case series showed that when VAD was used as a bridge to recovery, 66% of patients were eventually able to wean from support and be discharged home.⁶⁹ If the heart is chronically diseased, there is little hope of recovery, and the VAD serves to support the patient until transplantation becomes possible.^69,⁷⁰

Ongoing clinical trials are investigating the use of VADs as definitive therapy rather than as a bridge to transplantation. Implantation of these devices may increase the long-term survival of patients with end-stage heart failure.⁷¹

Correction of Arrhythmias

Atrial fibrillation is the most commonly encountered arrhythmia after cardiac surgery. Prophylactic use of β-adrenergic blockers reduces the incidence of postoperative atrial fibrillation, and they should be administered after cardiac surgery to all patients unless specific contraindications are present.⁷² Prophylactic treatment with amiodarone and atrial overdrive pacing should be considered for patients who are at high risk for postoperative atrial fibrillation (e.g., those with a history of previous atrial fibrillation or mitral valve surgery).^40,⁷³

If atrial fibrillation develops after cardiac surgery, the intensivist needs to determine whether the primary strategy should be to control the ventricular rate or to restore normal sinus rhythm. If atrial fibrillation is associated with hemodynamic instability or anticoagulation is contraindicated, rhythm management using electrical cardioversion or amiodarone is preferred.^74,⁷⁵ Overdrive pacing using atrial pacing wires also can be effective. The appropriate strategy for most stable patients may be control of ventricular rate, because most will spontaneously revert to sinus rhythm within 8 weeks after discharge.^76,⁷⁷ Appropriate agents to achieve ventricular rate control include intravenous or oral β-adrenergic blockers or calcium channel blockers. All patients with atrial fibrillation persisting for longer than 24 to 48 hours should be anticoagulated unless there is a specific contraindication. Long-term outcomes are similar regardless of whether the rate-control strategy or the rhythm-control strategy is selected.^78,⁷⁹

Postoperative ventricular arrhythmias should be treated immediately according to current Advanced Cardiac Life Support (ACLS) protocols.²⁴ Any postoperative ventricular arrhythmia should prompt a search for an underlying cause. Importantly, ischemia should be ruled out. Patients with sustained ventricular arrhythmias should undergo electrophysiologic testing before long-term antiarrhythmic therapy is instituted. The implantable cardioverter-defibrillator (ICD) device has been shown to be superior to drug therapy for patients with hemodynamically significant arrhythmias.⁸⁰

Hypertension in the Postoperative Period

Hypertension leading to an increase in ventricular afterload is a common cause of decreased CO. Hypertension can be controlled by an intravenous infusion of sodium nitroprusside, nitroglycerin, β-adrenergic antagonists, or calcium channel blockers. These agents should augment CO by reducing blood pressure and afterload in the hypertensive patient. Frequently, acute hypertension resolves within 24 to 48 hours postoperatively. If hypertension persists beyond this initial period of recovery, intravenous agents should be weaned and oral therapy initiated. Both β-adrenergic blockers and angiotensin-converting enzyme (ACE) inhibitors have been shown to confer a long-term mortality benefit and should be started. If hypertension was not a problem preoperatively, prolonged antihypertensive therapy postoperatively usually will not be necessary.

Correction of Coagulopathy

Postoperative coagulopathy can promote bleeding and accumulation of blood in the chest or pericardial cavity. Aggressive measures must be used to correct the coagulopathy. A systemic approach to the evaluation and treatment of continued bleeding is needed; one such approach is outlined in Table 194-3. Hypothermia can contribute to coagulopathy. Therefore, profoundly hypothermic ICU patients must be actively rewarmed with the use of a warm air device. Laboratory evaluation of suspected coagulopathy should include measurements of platelet count, prothrombin time (PT), APTT, ACT, and bleeding time.

TABLE 194-3 Evaluation and Treatment of Postoperative Coagulopathy

Coagulation Test	Normal Range	Suggested Treatment
Body temperature	—	If less than 35.5°C, the patient should be actively rewarmed.
Prothrombin time (PT)	11-13.3 sec	Administer fresh-frozen plasma.
Partial thromboplastin time (PTT)	21-32 sec	Consider additional protamine.*
Platelets	140,000-440,000/µL	If <100,000, transfuse platelets.
Fibrinogen	150-360 mg/dL	If <100, transfuse cryoprecipitate.
Bleeding time	2.5-9.5 min	If prolonged and platelet count is normal, consider platelet dysfunction, and treat with desmopressin acetate (DDAVP) and/or cryoprecipitate.
Activated coagulation test (ACT)	90-120 sec	Consider additional protamine.*

* Excessive protamine may itself cause bleeding.¹⁰²

Postoperative Bleeding

Bleeding that continues after correction of coagulopathy needs to be aggressively treated. Venous bleeding in the chest can be partially controlled by application of positive end-expiratory pressure (PEEP).^81,⁸²

Continuing mediastinal hemorrhage, or the suspicion of cardiac tamponade, is an indication for immediate re-exploration. Exsanguinating hemorrhage or impending arrest from tamponade may require that re-exploration be carried out at the bedside in the ICU. Bleeding that is unresponsive to medical therapy and requires re-exploration is usually associated with a surgical source. Accepted guidelines for re-operation include bleeding rates of 400 mL/h for 1 hour, 300 mL/h for 2 hours, or 200 mL/h for 3 hours. A sudden decrease or total cessation of drainage from mediastinal tubes may be equally ominous. Cessation of drainage from a mediastinal or chest tube can be caused by clotted blood occluding the tube. If bleeding persists but drainage ceases, the result can be tamponade.

Re-exploration is associated with increased morbidity and mortality. However, this increased mortality and morbidity may be partially explained by delays in the decision to re-explore that lead to avoidable open-chest resuscitations in the ICU.^56,^83,⁸⁴

Postoperative Renal Failure

The cornerstone of prevention and treatment of renal failure in the cardiac surgical patient is the maintenance of adequate renal perfusion. This goal is best achieved by optimizing circulating blood volume and CO. Multiple pharmacologic regimens for renal protection have been described. Dopamine at low “renal” doses (1-3 µg/kg/min) has been used. The rationale for this strategy is that dopamine activates type 1 dopaminergic (DA1) receptors, leading to renal artery dilation, natriuresis, and diuresis. However, numerous human studies have failed to show that low-dose dopamine prevents renal failure or improves survival.⁸⁵ Even low doses of dopamine increase CO, and this may be the basis for any increase in urine output observed.⁸⁶ Fenoldapam⁸⁷ and dopexamine⁸⁸ are DA1 receptor antagonists that also have been proposed as renal protective agents and used with mixed success.⁸⁹

Loop diuretics such as furosemide have been proposed as renal protective agents, not only because of their ability to produce diuresis and natriuresis, but also because these drugs may reduce medullary tubular oxygen consumption. Mannitol, an osmotic diuretic, been used to prevent development of ARF. Neither mannitol nor furosemide has been shown to improve outcome for patients with ARF.⁶⁰ Indeed, these drugs may be deleterious because of their ability to promote diuresis and thus exacerbate hypovolemia and inadequate renal perfusion. Some success has been reported with the combination of mannitol, furosemide, and dopamine.⁹⁰ Infusion of a solution containing these three agents promoted diuresis in patients with acute postoperative ARF and adequate CO and significantly decreased the need for dialysis in the majority of patients.⁸⁸ Early administration of this solution in ARF caused early restoration of renal function to normal or baseline status.⁹⁰

The failure of pharmacologic means of preventing and treating renal failure has led to interest in other methods. Early and intensive use of continuous venovenous hemofiltration achieved a better than predicted outcome in a series of 65 consecutive patients with severe ARF who underwent cardiac operations.⁹¹

Glucose Control

Recent studies have shown that tight control of blood glucose level in the ICU is associated with an increase in morbidity and mortality (Table 194-4). Hyperglycemia and insulin resistance are common in critically ill patients, even those who have not previously had diabetes. Results of a prospective randomized controlled study⁹² in which 6104 critically ill adult patients were randomly assigned to receive either intensive insulin therapy (maintenance of blood glucose concentration between 80 and 108 mg/dL) or conventional treatment (infusion of insulin to keep blood glucose level 180 mg/dL or less) showed that at 3 months, the intensive insulin therapy group had an increase in ICU mortality, with an increase in hypoglycemic episodes in the treatment group.

TABLE 194-4 Protocol for Blood Sugar Control in the Postoperative Period

Decision to initiate IV insulin
↓
If BG <200 mg/dL, begin D₅

NS at 60-100 mL/h
If BG >300 mg/dL, give stat dose of IV insulin, 0.1 U/kg body weight
↓
Initiate an hourly rate (total daily dose of insulin divided by 24)
For patients who have never taken insulin, give 0.02 U/kg body weight per hour*
↓
Check BG hourly and adjust according to table below
Recheck BG hourly
↓
If in desirable range (101-150 mg/dL), continue to check BG every 2 h and adjust as necessary

Current BG (mg/dL)	Previous BG (mg/dL)
<60 60-80 81-100 101-150 151-200 201-250 251-300 301-400 >400
<60	Withhold drip and give 1 ampule of 50% glucose; check BG every 30 min until >100 mg/dL, then reinitiate drip at 50% of previous rate
60-80	Withhold drip; check BG every 30 min until >100 mg/dL, then reinitiate drip at 50% of previous rate
81-100	↓ Rate by 1 U/h No change ↓ Rate by 25% or 0.5 U/h^† ↓ Rate by 25% or 1 U/h^† ↓ Rate by 50% or 2 U/h^†
101-150	No change ↓ Rate by 25% or 1 U/h^†
151-200	↑ Rate by 1 U/h ↑ Rate by 0.5 U/h ↑ Rate by 25% or 1 U/h^† No change ↓ Rate by 25% or 1 U/h^†
201-250	↑ Rate by 25% or 2 U/h^† ↑ Rate by 25% or 1 U/h^† ↑ Rate by 1 U/h No change
251-300	↑ Rate by 33% ↑ Rate by 25% ↑ Rate by 25% ↑ Rate by ↑ Rate by ↑ Rate by No or 2.5 U/h^† or 1.5 U/h^† or 1 U/h^† 1 U/h^† 1.5 U/h^† 25% or 2 U/h^† change
301-400	↑ Rate by 40% or 3 U/h^†
>400	↑ Rate by 50% or 4 U/h^†
Before discontinuing insulin infusion:
Ensure that patient is able to tolerate oral intake
Write orders for alternative glycemic management
Precede discontinuation by 1-2 h with subcutaneous dose of very rapid or rapid insulin. If patient has never taken insulin, use a dose equal to twice the hourly rate of IV insulin. Otherwise, use the dose of insulin or oral agent given before surgery/admission.

BG, blood glucose concentration; D₅ NS, 5% dextrose in half-normal saline; IV, intravenous; U, units.

* For patients undergoing major surgery (e.g., cardiothoracic surgery, transplantation), higher doses may be necessary.

^† Whichever is greater.

Copyright © 2003 by Joslin Diabetes Center. All rights reserved. These Guidelines are the property of Joslin Diabetes Center and are protected by copyright. Any reproduction of this document which omits Joslin’s name or copyright notice is prohibited. This document may be reproduced for personal use only. It may not be distributed or sold. It may not be published in any other format (e.g., book, article, Web site) without the prior, written permission of Joslin Diabetes Center, Communications Department, 617-732-2695.