During orthotopic liver transplantation, the native organ is removed and replaced by a cadaveric liver. Extracorporeal venovenous bypass is often used during the procedure to decompress the systemic and splanchnic venous systems. Although this may help preserve hemodynamic stability during the operation and decrease intraoperative bleeding problems, potential complications include intraoperative air embolism73,74 and thromboembolism. The biliary tract will be reconstructed either by creating an end-to-end anastomosis of the donor and recipient common ducts (using a T-tube stent) or by connecting the donor's common duct to the recipient's jejunum. Anastomoses are created between the native and allograft cava (supra- and infrahepatic), portal veins, and hepatic arteries. Removal of the venous clamps leads to reperfusion of the organ, and this will often be associated with hemodynamic instability, coagulopathy, and electrolyte abnormalities (particularly hyperkalemia).
Recently, there have been a growing number of patients who undergo living-donor transplants. In Asia, almost all liver transplant procedures have involved living donors.75 In this procedure, the right hepatic lobe from a donor with a compatible blood type is implanted into the recipient following hepatectomy of the diseased organ. Along with an often substantially reduced waiting time for the procedure, living related transplants may also allow for better selection of healthy donors (and consequently donor organs) and a considerably decreased cold ischemia time. The elective nature of the procedure also enables potential recipients to be medically stabilized preoperatively. The major disadvantage is the small but significant risk of complications for the donor. Biliary complications may occur in up to 6% of donors,76 and other complications of abdominal surgery such as wound infection may develop; the reported mortality of donors following living-donor liver transplant is 0.28%.75 One study suggests that in the United States approximately 99% of living donors are genetically or emotionally related to the recipient,76 creating important ethical and psychosocial challenges.
Most patients will require mechanical ventilation for the first 24 to 48 hours following the liver transplant procedure. Extubation should not be considered until there is evidence that the allograft is functioning properly and the patient's level of arousal is sufficient to allow for adequate airway protection. This latter consideration is especially important given that anesthetic agents and sedating medications may be cleared more slowly from the circulation by the newly transplanted liver. The need for reintubation in liver transplant patients has been associated with poorer outcomes and even increased mortality.77 However, there is mounting evidence to support the notion that attempts at early extubation of liver transplant recipients may reduce ICU utilization. Selection of suitable candidates seems crucial.78–82
Monitoring of Liver Function
The function of the new allograft should be assessed in the same manner that liver function is assessed in any patient. Coagulopathy is often present in the immediate postoperative period, but should correct soon after the transplant. The prothrombin time (PT) or International Normalized Ratio (INR) should be followed closely. Administration of fresh frozen plasma will interfere with the ability to monitor these parameters. Consequently some programs avoid the transfusion of coagulation factors unless necessitated by bleeding complications.
Glucose should be monitored frequently during the ICU stay as liver failure will often result in refractory hypoglycemia. Conversely, the use of corticosteroids may lead to insulin resistance and hyperglycemia, which should be treated appropriately. Bilirubin often remains elevated for many days following the transplant, but should be followed closely as abrupt changes may herald complications involving the biliary tree or the vascular supply to the liver. Serum lactate levels are frequently elevated immediately following the operation, but if the graft is functioning properly they should return to normal within a few days.
Serum transaminase levels (AST/ALT) are usually measured postoperatively; these usually rise in the first 2 to 3 days following the procedure, but then can be expected to normalize. Similar to changes in bilirubin levels, abrupt changes or progressive elevations of these enzymes may also signify complications and mandate further evaluation.
Physiologic derangements following liver transplantation are common. Intraoperative blood and fluid loss may be considerable. The use of aprotinin during the operation has been advocated by some as a means to reduce blood loss.83–85 Interestingly, several studies also report decreased vasopressor requirements in patients who receive this therapy intraoperatively.86,87 Its use has also been associated with an improvement in renal function.88 However, this initial enthusiasm for aprotinin therapy needs to be tempered by concerns regarding a possible increased potential for thrombembolism.89,90
Following the operation, patients may lose significant amounts of fluid and protein via the abdominal drain, especially if massive ascites was present preoperatively owing to advanced cirrhosis and decompensated liver disease. These latter patients are at particularly high risk of developing intravascular volume depletion, and they must be monitored carefully and treated aggressively. Serial hematocrit determinations should be measured (usually every 4 to 6 hours) to ensure that significant bleeding is not present. It should be noted that the hyperdynamic circulatory state that characterizes liver dysfunction will often persist in the postoperative period, making interpretation of hemodynamic measurements difficult. An increased cardiac output with decreased systemic vascular resistance is often observed. Although aggressive fluid resuscitation is indicated to treat intravascular volume depletion, it should also be noted that patients are at particularly high risk of cardiopulmonary complications following liver transplantation.91 Electrolyte abnormalities, particularly hyponatremia, are often observed. Overly aggressive treatment of derangements of sodium can precipitate the osmotic demyelination syndrome (formerly called central pontine myelinolysis).
Particular care is required for the patient who has significant preoperative portopulmonary hypertension. These patients will often encounter pulmonary hemodynamic instability and may have increased cardiopulmonary mortality, especially if the preoperative mean pulmonary artery pressure is greater than 35 mm Hg.92 Invasive hemodynamic monitoring with a pulmonary artery catheter should be considered to guide therapy in these patients. Specific pulmonary vasodilators such as inhaled nitric oxide or nebulized prostaglandins may be required in the setting of right ventricular dysfunction in order to preserve cardiac output.
A system for classifying patients based on their anticipated need for fluid and electrolyte replacement has been proposed93 and advocated by several experts.94 Such an approach may be practically useful in that it creates a framework to understand the hemodynamic considerations for a given patient (Table 90-3). According to this classification, a patient with class I liver disease can be expected to have a normal postoperative response to intravenous fluid therapy. Patients with class II or III liver disease will have more ascites, leading to greater fluid and protein loss from the abdominal drain, and can be anticipated to need more fluids postoperatively and careful monitoring of electrolytes. Patients with class IV disease require the closest monitoring and may benefit from insertion of a pulmonary artery catheter to assist with the management of the portopulmonary hypertension and cardiac dysfunction that is often present.
Table 90–3. Classification of End-Stage Liver Disease Severity for the Purposes of Intensive Care Unit Management |Favorite Table|Download (.pdf)
Table 90–3. Classification of End-Stage Liver Disease Severity for the Purposes of Intensive Care Unit Management
|Class||Hyperdynamic Circulation||Hyponatremia||Malnutrition||Portopulmonary Hypertension||Cardiac Dysfunction|
Postoperative Renal Dysfunction
Following the liver transplant procedure renal dysfunction is frequently observed, and postoperative renal failure can be severe enough to require renal replacement therapy in up to one third of patients. However, kidney function generally improves, and only 5% of patients require chronic hemodialysis following liver transplant.95
The most common precipitant is intravascular volume depletion resulting in prerenal azotemia. However, nephrotoxicity due to the immunosuppressive drugs is also common. The development of hepatorenal syndrome occurs in 7% to 15%94,96 of patients with cirrhosis, and is commonly observed following orthotopic liver transplant. It is likely that strategies aimed at preventing hepatorenal syndrome are the most effective. Thus care should be taken to promptly treat bleeding insults and to avoid intravascular volume depletion. When CT scanning or other imaging techniques are being considered, the use of intravenous contrast dye should be avoided if possible. If intravenous contrast is required for proper imaging (and the benefits outweigh the risks of the test), consideration should be given to the use of acetylcysteine if renal function has not completely returned to normal, in order to protect the kidney from further insult and contrast nephropathy.97
Terlipressin (a long-acting vasopressin analogue) has been reported to be successful in improving urine output in cirrhotic patients with hepatorenal syndrome.98,99 The relative contribution of albumin coadministration to the success of this pharmacologic strategy has been evaluated and supports the notion that maintenance of intravascular volume is integral in the treatment of these patients.100 One study also advocates the addition of acetylcysteine to mitigate the potential ischemic nephropathy that may result from the vasoconstrictive effects of terlipressin on the kidney.99 The role of terlipressin in patients with established hepatorenal failure following transplantation has not been systematically evaluated. Indeed, in theory, liver transplantation should correct hepatorenal failure.101,102
Primary Graft Nonfunction
Primary graft nonfunction refers to a failure of the transplanted liver early in the postoperative period. The characteristics of this devastating complication include minimal bile output, refractory coagulopathy, progressive elevation of transaminases, acidosis, hypoglycemia, and cerebral edema. The incidence of this dreaded complication is likely only between 3% and 5%,103,104 but the associated mortality rate may be higher than 20%.105–107
Several considerations exist if the graft fails to work postoperatively. The possibility of vascular complications should be entertained and excluded with Doppler ultrasonography.108 Investigations to detect severe infection or acute rejection should be initiated. If all these tests fail to elucidate the cause of graft failure, primary graft nonfunction is the likely cause.
The incidence of vascular complications following liver transplant ranges from 8% to 14%.109,110 These develop most commonly on the first postoperative day, but may occur up to several weeks following the surgery.108 Thombosis is the most common early complication, and stenosis and pseudoaneurysm formation generally develop later in the patient's course.The presentation of hepatic artery thrombosis has been associated with hepatic artery reconstruction with an interposition graft to the supraceliac aorta.111 Despite reassurances, some concerns about an increased risk of hepatic artery thrombosis with the use of sirolimus remain.112 The clinical picture of this complication can vary from the asymptomatic rise in liver enzymes to fulminant hepatic failure. Urgent laparotomy and revision of the hepatic artery anastomosis is required if this complication develops, and unfortunately retransplantation is often necessary if hepatic necrosis has occurred.
Portal vein thrombosis develops less frequently and may present more insidiously. Ascites may be seen to develop or worsen, and variceal bleeding (usually from pre-existing varices) may ensue. Thrombectomy and anastomotic revision can be successful if this complication is diagnosed early.
Biliary complications occur in approximately 15% of patients following orthotopic liver transplantation.113 Of these, bile leak is the most common early complication. Symptoms are nonspecific, and can include fever, abdominal discomfort, and signs of peritoneal irritation. Though ultrasound may demonstrate an intra-abdominal fluid collection, cholangiography will provide a definitive diagnosis. Endoscopic insertion of biliary stents can sometimes provide satisfactory results, but surgical repair of the leak may be required.
Biliary strictures and stones typically appear later in the postoperative period. Obstructions usually can be managed endoscopically114 or through the use of interventional radiology techniques, but surgical correction is sometimes necessary. Strictures typically develop at the anastomotic site, and are likely the result of local ischemia. These may present as cholestasis or possibly as overt cholangitis. Balloon dilation of the stricture with or without stent placement usually is successful treatment, but surgical revision and even retransplantation may be required.
Intra-abdominal infections ranging from local abscess formation to overt peritonitis can occur following liver transplantation. Development of these complications should always lead the clinician to suspect that a leak has occurred from the biliary anastomosis or from the jejunojejunostomy; correction of these problems will often require laparotomy and surgical repair. However, abscesses can usually be treated with CT-guided drainage and subsequent serial CT scans to ensure that the collection has been adequately drained.
Rejection of the hepatic allograft is usually not seen until about 1 to 2 weeks following the procedure, and most often manifests as fever, right upper quadrant pain, and reduced bile pigment and volume. However, the most sensitive marker of early rejection is a rise in serum transaminase (AST/ALT) levels and bilirubin. A rise in total white blood count may also develop. The most important consideration when elevation of these serum enzymes occurs early in the postoperative course is the exclusion of one of the various mechanical complications (such as vascular compromise, biliary obstruction, and primary graft nonfunction). After the biliary tree and vascular structures have been imaged using ultrasonography, a liver biopsy will help confirm the diagnosis of acute rejection. As is the case in any solid-organ transplant recipient, treatment of acute rejection should be aggressive and must be instituted promptly. For liver transplant recipients, therapy usually involves antithymocyte globulin and increased or pulsed doses of methylprednisolone.
The management of chronic rejection of the transplanted liver usually does not involve the intensivist, but occasionally these patients may require ICU admission. Chronic rejection of the liver presents clinically as progressive cholestasis and histologically with mononuclear infiltration of the allograft, vascular abnormalities, and fibrosis. These findings are most commonly seen as part of the vanishing-bile-duct syndrome; treatment is often unsuccessful and may require retransplantation.
While combined heart-lung transplantation is still performed for patients with Eisenmenger syndrome and complex congenital cardiac defects that cannot be easily repaired at the time of the lung transplant, most patients will undergo either bilateral or single-lung transplantation. Bilateral lung transplantation is the preferred procedure for patients with septic pulmonary disorders such as cystic fibrosis and bronchiectasis. In contrast, single-lung transplantation can be performed in patients with chronic obstructive pulmonary disease (COPD) and idiopathic pulmonary fibrosis. The observation that right ventricular dysfunction generally improves in patients with pulmonary hypertension following single- or double-lung transplantation115 has obviated the need for performing combined heart-lung transplantation in this group of patients.
During single-lung transplantation, sequential anastomoses are performed to attach the donor and recipient mainstem bronchus and pulmonary artery. The donor left atrial cuff and pulmonary veins are then attached to the recipient's left atrium. In a similar manner, bilateral lung transplantation is performed as two sequential single-lung transplant procedures. Most patients can be supported intraoperatively through the ventilation of a single lung while surgery is performed on the contralateral side. Consequently, cardiopulmonary bypass is generally only required if profound hypoxia is encountered.
Since hyperacute rejection (due to ABO mismatch) has become extremely rare, the earliest serious complication of the lung transplant operation is ischemia-reperfusion (IR) injury (also called primary graft failure or pulmonary reimplantation response). IR injury is a syndrome of severe allograft dysfunction characterized histologically by marked alveolar damage which manifests clinically as severe pulmonary edema. The incidence of this complication ranges from 10% to 20%, depending on the reporting center and the definition used,116–118 and when severe, it is associated with a very poor prognosis; as many as 41% of affected patients may die.117 A recent study also suggests that IR injury may be associated with the development of bronchiolitis obliterans later in the patient's course.119 Most series of patients with IR injury demonstrate that recipients will develop pulmonary infiltrates on chest x-ray during the first 3 days following transplantation.120 When entertaining this diagnosis, it is essential that the clinician perform a careful evaluation to exclude other possible complications. Confirmation should be sought that a donor-recipient cross-match (hyperacute rejection) has not occurred, that bacterial pneumonia has not developed (negative cultures from bronchoalveolar lavage), and that pulmonary venous return is not compromised owing to technical problems with the left atrial anastomosis (transesophageal echocardiography should be performed to ensure adequate blood flow through pulmonary veins). Invasive monitoring with a pulmonary catheter will help exclude volume overload or cardiogenic pulmonary edema. Transbronchial biopsy may be required to rule out the possibility of acute rejection. Treatment of IR injury is similar to the management of ARDS. Extending the results of trials of mechanical ventilation in patients with ARDS we recommend that lung-protective ventilation strategies with low tidal volumes be employed, with the use of sufficient positive end-expiratory pressure (PEEP) to reduce the fraction of inspired oxygen to acceptable levels.121 Since IR is characterized by increased alveolar capillary permeability, care must be taken to provide judicious fluid management. Some success at attenuating the injury from IR has been observed with the administration of prostaglandin E1, though this may lead to hypotension owing to the effects of the drug on the systemic vasculature. Inhaled nitric oxide (NO) has been advocated as a means to both treat and prevent IR injury. Initial preclinical and uncontrolled clinical studies of inhaled NO suggest that administration prior to IR injury may improve graft function.122–126 However, a recent controlled study of NO administered 10 minutes after reperfusion failed to show any benefit in terms of graft function.127 NO has also been used to improve oxygenation and reduce pulmonary arterial pressures in patients with oxygenation and vasomotor dysfunction after lung transplantation.128–130 If inhaled NO is used for its pulmonary vasodilatory properties and to decrease ventilation/perfusion mismatch, it should be understood that these effects should be seen within minutes of the institution of this therapy. Daily measurements of methemoglobin levels are required. If one extrapolates from studies of NO use in acute lung injury, the effects of NO on lung function are likely short lived.131 Consequently the rationale for long-term use of NO is unclear. Furthermore, the benefit of therapy is often difficult to assess owing to the development of rebound during attempts to withdraw NO therapy.123 We often employ sildenafil to prevent such rebound associated with withdrawal of NO therapy in these patients.132
Following single-lung transplantation, IR injury will often compromise lung function in the allograft. To avoid exposing the native lung (which is often emphysematous and prone to hyperinflation) to potentially injurious high airway pressures and alveolar overdistention, independent lung ventilation is often used. If conventional ventilation is causing overexpansion of the native lung leading to an increase in its pulmonary vascular resistance, blood may be preferentially shunted to the compromised graft. Independent lung ventilation involves the use of a double-lumen endotracheal tube (avoiding intubation of the transplanted bronchus), and enables each lung to be ventilated with different airway pressures and tidal volumes. This strategy can dramatically improve ventilation-perfusion (V̇/Q̇) mismatch. In the extreme case, the native lung may become sufficiently overdistended to cause mediastinal shift and lead to hypotension as a result of compromised venous return or cardiac tamponade.133 Other measures that can ameliorate the V̇/Q̇ mismatch and other sequelae of native lung overexpansion include the positioning of the patient in a lateral decubitus position to direct the allograft superiorly (to preferentially perfuse the native lung) and the use of lower tidal volumes. The use of this latter strategy often comes at the expense of a mild respiratory acidosis. Rarely, lung reduction surgery performed on the native emphysematous lung may be required if the overdistension of this lung remains a problem.134,135
Airway and Mechanical Complications
Although the incidence of airway and other mechanical complications has decreased,136 it is important for the intensive care physician to have an understanding of these problems. Bronchial dehiscence is an infrequent yet potentially devastating complication. It may manifest within a few days of the operation and is usually heralded by persistent or large air leaks from thoracostomy tubes, but may present as pneumomediastinum with or without mediastinitis, pneumopericardium, or empyema. Usually the complication results from ischemia and necrosis at the anastomosis suture line. Although most cases can be managed conservatively by decompressing the large extrapulmonary air collections, bronchoscopic débridement and even bronchial stenting may be required to ensure airway patency. Complete dehiscence of the anastomosis mandates surgical correction or retransplantation.
Bronchial strictures leading to airway stenosis and compromise are sequelae of bronchial dehiscence and/or ischemia, but usually do not develop until several weeks following the transplant procedure. Strictures typically arise in the mainstem bronchus at or just distal to the anastomosis. Ischemic injury to the bronchi may also lead to bronchomalacia, which is characterized by weakened cartilage that is prone to (expiratory) dynamic collapse. Permanent stenting is the preferred approach to this problem. More commonly patients develop inspissation of secretions in the bronchi that cannot be cleared. This may manifest as areas of airspace disease on chest x-ray, atelectasis, or a worsening in oxygenation. In this setting, bronchoscopy is valuable to both establish the cause and clear the retained secretions.
Extracorporeal Membrane Oxygenation
Extracorporeal membrane oxygenation (ECMO) has been used with success to support patients with severe graft dysfunction resulting from ischemia-reperfusion injury. This therapy has generally been used only once conventional strategies to improve oxygenation have failed. Mechanical support is employed in some centers to treat patients with moderate to severe IR injury. ECMO employs a membrane with a large surface area through which oxygen and carbon dioxide may passively diffuse to equalize their respective concentration gradients in the patient's blood. This exploits the notion that resting the lung will allow time for the graft to overcome the initial IR insult. At present a systematic evaluation of this treatment is lacking and experience is restricted to case series and registry databases.137
Acute rejection may occur while the transplant recipient is being cared for in the ICU, and generally occurs within the first 100 days. The signs are nonspecific, and thus it is important to differentiate the bilateral airspace infiltrates that characterize this complication from those caused by infections. Other features are also unreliable discriminators, and include low-grade fever, impaired oxygenation, and leukocytosis. Findings on chest radiography other than alveolar infiltrates include nodular or interstitial opacities and pleural effusions; after the first month acute rejection is often undetectable radiologically.138 Routine bronchoscopy has become widely accepted as a means of differentiating infection from rejection. Bronchoalveolar lavage or protected-specimen brushing of the airways can exclude any important infection, whereas transbronchial lung biopsy (with at least five pieces of alveolated parenchyma, each containing bronchioles and more than 100 air sacs) can confirm rejection. Treatment of acute rejection entails the use of high-dose parenteral steroids (i.e., methylprednisolone 10 to 15 mg/kg daily), which usually provide a good response, though biopsy-proven rejection may persist in up to one third of patients.61
A detailed discussion of the problem of chronic rejection is beyond the scope of this chapter. However, intensivists may encounter these individuals who develop chronic airflow limitation (usually more than 6 months following the transplant) that is characterized histologically as bronchiolitis obliterans. Although aggressive immunosuppression is often employed, the prognosis for posttransplant bronchiolitis is still poor, with 2-year mortality rates approaching 40%.139,140 The definitive treatment remains retransplantation.
Although other variations have been described, the standard approach for heart transplantation involves the creation of four separate anastomoses between the recipient atria and great vessels and the atria and great vessels of the donor heart (the biatrial technique). There has been renewed interest in bicaval and pulmonary vein to pulmonary vein anastomoses, and the bicaval technique has now been recommended by many experts.141
The Immediate Postoperative Period
The care of the heart transplant recipient is similar in many respects to the care of patients who have undergone aortocoronary bypass and other cardiac operations involving cardiopulmonary bypass. The heart is denervated during the transplant procedure, and frequently a junctional rhythm develops, requiring atrial pacing and possibly isoproterenol infusion. Owing to this denervation, the transplanted heart will also have an altered physiologic response to stress. The heart becomes dependent on circulating catecholamines rather than direct sympathetic stimulation, and consequently increases in heart rate and contractility in response to stress will occur more slowly. Changes in cardiac output become much more dependent on changes in stroke volume, which are in turn mediated by changes in venous return. Sinus bradycardia is the most common dysrhythmia present following heart transplantation, and usually lasts less than a week. Isoproterenol is frequently used for both its chronotropic and inotropic properties, and is typically titrated to achieve a heart rate between 90 and 100 beats per minute. As already mentioned, temporary atrial pacing may be required. If the recipient's native atrial appendage and sulcus terminalis including the sinoatrial (SA) node are preserved during the surgical procedure, two P waves may be seen on the postoperative electrocardiogram (one from the donor heart and one from the recipient's SA node). Other atrial dysrhythmias are frequently seen and include atrial fibrillation and flutter in 15% to 20% of patients.142
Nearly all patients will have a pulmonary artery catheter in place for hemodynamic monitoring following the operation to help guide fluid and vasoactive management. Generally, the goal is to keep both the central venous pressure (CVP) and the pulmonary capillary wedge pressure (PCWP) <10 mm Hg if possible, though these goals should be individualized, and the management should be guided by the clinical picture and not purely based on hemodynamic measurements. Consequently intravenous fluids are used sparingly and aggressive diuresis is continued postoperatively. However, the denervated heart will not be able to respond acutely to hypovolemia with reflex tachycardia, and adequate preload must be present to maintain stroke volume and to preserve blood pressure. A decreased cardiac output can be supported with inotropes (dobutamine or milrinone), but transient support with intra-aortic balloon counterpulsation may be required. If the cardiac output acutely deteriorates, urgent echocardiography should be obtained to exclude the possibility of tamponade and to evaluate left and right ventricular dysfunction. Left ventricular function of the allograft may also be reduced and a restrictive physiology may be observed if there has been prolonged ischemic time and poor myocardial preservation. Other causes of ventricular dysfunction should be sought such as acidemia, hypovolemia, hypoxemia, and medications. Systemic vascular resistance can be decreased following the procedure, and should be supported with vasopressors.
If significant pulmonary hypertension was present before the transplant procedure, it may also persist in the postoperative period. This increase in right ventricular afterload can generate significant strain on the newly transplanted right ventricle (which is accustomed to pumping against a normal mean pulmonary artery pressure). Tricuspid regurgitation is often present following the transplant (especially if the biatrial technique was used), and this may be exacerbated by increased right ventricular strain. Through its effects as a pulmonary vasodilator, isoproterenol can often reduce the right ventricular afterload that is frequently present. However, when significant pulmonary hypertension persists postoperatively, intravenous nitroglycerine, nitroprusside, and prostaglandin E may be required. If vasodilators are used in this setting, systemic infusion of α-agonists such as norepinephrine or phenylephrine may be needed to support the systemic arterial pressure. Inhaled nitric oxide offers the advantage of reducing pulmonary vascular resistance while causing minimal systemic side effects and without increasing shunting.143–147 Routine measurement of methemoglobin level is required when nitrate-containing pulmonary vasodilators are employed. Recently, sildenafil, a cyclic guanosine monophosphate (cGMP)-specific phosphodiesterase type 5 inhibitor has been used with success for persistent pulmonary hypertension, and to prevent rebound after withdrawal of inhaled NO.132 Owing to the expense associated with the use of inhaled NO, other inhaled therapies have been evaluated (NO donors and prostaglandin analogues) and appear to be of equal benefit.148–151 However, the half-lives of these medications mandate repeated dosing.
Right and Left Ventricular Assist Devices
Patients with failing hearts are frequently bridged to transplantation using mechanical cardiac assist devices. Both left (LVAD) and right (RVAD) ventricular assist devices are used.152,153 Preliminary reports suggest that although mechanical devices may be associated with more serious adverse effects, they may be superior to medical therapy alone in supporting the patient to transplantation.154 The role of ventricular assist devices in right ventricular failure or primary graft failure following transplantation is evolving. A report from New York reviewed one institution's experience with assist devices in 462 transplants. Of the 20 patients diagnosed with primary cardiac graft dysfunction, 11 received RVAD, 4 required LVAD, 3 received biventricular support, and 2 were treated with intra-aortic balloon counterpulsation.155 This group with primary graft dysfunction tended to have longer ischemic times and developed early noncardiac organ dysfunction, which was the primary mode of death in the nonsurvivors. Forty-five percent were weaned from mechanical support; the mean duration of support was 114.7 ± 207 hours (median = 54 hours). The 2-year survival in this group with primary graft dysfunction was 67% compared to 79% in the entire cohort. The authors reported that ventricular recovery within 96 hours was associated with better early survival.
Though immunosuppressive regimens have improved the impact of episodes of acute rejection on graft function and overall survival among heart transplant recipients, this complication still is responsible for 7% of deaths within the first 30 days of the procedure.156 Most episodes of acute rejection are asymptomatic and are diagnosed on routine endomyocardial biopsy. When symptoms are present, they are often nonspecific and may include fever, malaise, hypotension, congestive heart failure, or reduced exercise tolerance and fatigue. Consequently, surveillance for rejection through the use of endomyocardial biopsy has become standard practice. When performed by an experienced operator, this procedure is associated with a morbidity of less than 1% and a procedure-related mortality of less than 0.2%.157,158 The most concerning complication of this procedure is cardiac perforation with acute pericardial tamponade or injury to the tricuspid valve. Typically biopsies (with multiple sampling) are performed 5 to 7 days following the transplantation and then twice weekly for the first 2 months.
Once rejection is diagnosed histologically, imaging to assess cardiac function should be obtained (two-dimensional echocardiography or multiple-gated acquisition).141 Mild rejection in the absence of cardiac dysfunction is usually treated conservatively with an increase in the dose of immunosuppressive agents. If moderate rejection or left ventricular dysfunction is present, the episode of rejection should be treated using high-dose pulsed steroids with or without cytolytic therapy (antithymocyte globulin or OKT3). Repeat endomyocardial biopsy should always be performed to assess the response to therapy.
Chronic rejection in the cardiac allograft typically manifests as aggressive and premature coronary artery disease. This complication usually develops months to years after the procedure, and usually will not involve the intensivist. In contrast to classic coronary artery disease, transplant-associated coronary artery disease is generally more diffuse, involving all the vessels of the heart including the arteries, veins, and great vessels. Because the allograft is denervated, classic angina only develops in a minority of patients, and coronary artery disease more typically presents with “angina-equivalent” symptoms such as dyspnea.
Kidney-pancreas transplantation is almost exclusively performed on patients with end-stage kidney disease secondary to diabetes mellitus.
Following kidney-pancreas transplantation many patients will not require ICU admission, and can be extubated before leaving the recovery room. However, ICU admission may be necessary for these patients because of pre-existing medical problems, due to difficulties encountered during the operation, or owing to complications that develop in the postoperative period.
The most common initial complication following the transplant of a kidney is oliguric or nonoliguric acute tubular necrosis (ATN) resulting from ischemic-preservation injury to the renal allograft. Care must be taken in this setting to avoid hypovolemia which can lead to prerenal azotemia and exacerbate the ATN. Urine output following the transplant procedure is usually high (800 to 1000 mL/h) and should be associated with a decline in creatinine; a lower urine output in the immediate postoperative period may herald graft dysfunction. Fluid management can often be facilitated through the use of a pulmonary artery catheter for hemodynamic monitoring. If oliguria develops, the urinary catheter should be flushed to ensure patency, especially if hematuria has been observed. Duplex sonography or radionuclide renal imaging should be considered to exclude the possibility of vascular thrombosis. If these interventions fail to elucidate the cause of the oliguria, the most likely remaining explanation is ATN and ischemic-preservation injury.
Following combined kidney-pancreas transplantation patients are at risk of developing metabolic derangements, and are particularly prone to developing hyper- or hypoglycemia. Usually a continuous intravenous infusion of insulin is started intraoperatively and continued through the postoperative period. A glucose-containing maintenance infusion should also be used to protect against the development of hypoglycemia. Care should be taken to avoid mixing medications in glucose-containing solutions, as these will provide boluses of glucose that can aggravate the control of blood sugar levels.
The diagnosis of rejection in recipients of pancreas transplants is often challenging. Overt hyperglycemia does not occur until approximately 80% to 90% of the allograft has been destroyed, and consequently by the time clinical evidence of rejection is present it is often too late for therapy to be effective. For this reason, diagnosis of pancreatic rejection is usually made by inference after diagnosing rejection in a kidney allograft. This likely explains the improved graft survival in recipients of combined kidney-pancreas transplants as compared to patients who only receive a pancreas transplant; in the former group rejection is likely diagnosed and treated at a much earlier stage. In contrast, rejection of a renal allograft is usually suspected when the serum creatinine rises by more than 10%. If another cause of renal dysfunction (such as prerenal azotemia, pyelonephritis, etc) is not readily apparent, kidney biopsy should be performed. This procedure is relatively safe and will enable treatment of rejection to be instituted early and promptly. Recent reports suggest that use of the renal arterial resistance index159 or DNA microarray profiling160 may become important methods for diagnosing rejection in the renal allograft in the future.161