HAR occurs within a few minutes to a few hours after the reperfusion of the graft posttransplant. Preformed antibodies directed against antigens presented by the graft mediate activation of complement, activation of endothelial cells, and formation of microvascular thrombi, leading to graft thrombosis and loss [5]. The process is irreversible; currently, no treatment is available. Because HAR is mediated by circulating preformed antibodies normally directed against ABO system (comprising the four main blood types, i.e., A, B, AB, and O) antigens or against major HLA antigens, thorough screening of potential transplant recipients should prevent nearly all HAR.

The panel-reactive antibody (PRA) assay is a screening test that examines the ability of serum from potential transplant recipients to lyse lymphocytes from a panel of HLA-typed donors. A numerical value, expressed as a percentage, indicates the likelihood of a positive cross-match to the donor population. Therefore, patients lacking preformed antibodies to random donor lymphocytes are defined as having a PRA of 0% and have a very low probability of eliciting a positive lymphocyte cross-match to any donor. The finding of a higher PRA identifies patients at high risk for a positive cross-match and thus for HAR and for subsequent graft loss. Most often, such patients were previously sensitized by childbirth, blood transfusions, or a prior transplant.

Pretransplant, cross-match testing is performed to identify preformed antibodies against class I HLAs (T-lymphocyte cross-match testing) and class II HLAs (B-lymphocyte cross-match testing). A strong positive class I-HLA cross-match immediately pretransplant is ordinarily an absolute contraindication to renal and pancreas transplants. At most centers, heart and liver transplants are performed without a cross-match, unless the recipient is highly sensitized or has previously received a graft possessing major antigens in common with the current donor (i.e., donor-specific antibody [DSA]). A positive B-lymphocyte crossmatch indicates preformed antibodies directed against class II HLAs and is a relative, but not absolute, contraindication to a transplant. Recent studies confirmed the
efficacy of plasmapheresis followed by administration of immune globulin to reduce PRA levels and to convert strongly positive crossmatch results to weakly positive or negative results, thereby allowing organs to be transplanted across what were previously considered as strong immunologic barriers [6].

Crossmatch testing is a vital tool to identify the presence of antibodies against potential donor antigens and to assess the risks of posttransplant rejection and subsequent graft loss. Ironically, cross-match testing methods are not standardized. Since the mid-1960s, cross-match testing was based on the complement-dependent cytotoxicity (CDC) assay. The CDC assay was further refined by adding a wash step and an antihuman globulin (AHG) step, to increase its sensitivity and specificity. Then, with the introduction of technology based on flow cytometry (FC), the presence of recipient antibody on the surface of donor lymphocytes could be detected independent of complement binding. The FC method further enhances the sensitivity of crossmatch and, since the late 1980s, has been adopted by an increasing number of transplant centers [7].

The latest development in anti-HLA antibody screening was the introduction of Luminex® technology, using HLA-coated fluorescent microbeads and FC. This method in theory pinpoints the DSAs in sera of recipients with high PRA levels. Since all transplant donors are HLA typed nowadays, a negative cross-match for recipients with high PRA levels can be ensured by avoiding the selection of donors carrying unacceptable antigens (virtual cross-match) [8].

The main concerns with these new developments in antibody typing and crossmatch testing are between-center test variability and the thresholds of defining false-negative results (results that could deny recipients with high PRA levels a chance for a potential lifesaving transplant). Currently, it is up to an individual transplant center to implement its own HLA typing and cross-match policy, depending on the center’s experience and clinical outcomes.

Although screening has all but eliminated HAR as a clinical problem, active investigation is nonetheless directed at dissecting the underlying pathophysiologic mechanisms of HAR. Another research focus is on the similar rapid rejection of xenoreactive antigens that serve as a barrier to the development of xenotransplantation.

AR is the most common form of graft rejection in modern clinical transplantation. It may develop at any time, but is most frequent during the first several months posttransplant. Rarely, it occurs within the first several days posttransplant, a process termed accelerated acute rejection (AAR), most likely a combination of amnestic immune response driven by sensitized memory B lymphocytes and activation of the direct allorecognition pathway. Under such circumstances, the donor antigen exposure often occurred in the distant past, so the level of circulating DSAs would have been too low to be detected by conventional crossmatch techniques. Once challenged by the same donor antigens introduced by the organ transplant, dormant memory lymphocytes reactivate, replicate, and differentiate. Within several days, large numbers of antibodies are directed against the donor tissue and result in graft rejection.

Cellular rejection and antibody-mediated rejection (AMR) are not mutually exclusive in AR. Histologically, AR generates an infiltration of activated T lymphocytes into the graft, resulting in gradually progressive endothelial damage, microvascular thrombosis, and parenchymal necrosis. Pathologic grading schemes have been developed regarding the extent to which AR involves vascular damage, cellular infiltration, or a combination of both. Vascular AR is thought to be mediated by the presence of DSAs, albeit not in sufficient numbers to cause HAR. C4 d, a complement split product detected immunohistochemically in the capillaries of biopsied graft specimens, is highly correlated with AMR [9]. Without intervention, AR inevitably progresses to graft loss. The clinical presentation of AR varies markedly, depending on the specific organ, on the level of immunosuppression, and on the attendant reduction of inflammation in the affected tissues.

Unless the host immune system is suppressed pharmacologically, a transplant inevitably leads to AR. A combination of immunosuppressive agents is typically used chronically to prevent AR, including a lymphocyte antagonist (usually a calcineurin inhibitor [CNI] such as cyclosporine or tacrolimus) and an antiproliferative agent (such as azathioprine or mycophenolate mofetil), with or without corticosteroids. Antilymphocyte antibody therapy is often added during induction of immunosuppression or for treatment of “steroid-resistant” AR.

In the last decade, immunosuppression for transplant recipients has been undergoing a paradigm shift. Since the mid-1990s, the use of antibody induction in solid-organ transplant recipients has increased from 25% to more than 60% [10]. In particular, monoclonal antibodies such as basiliximab and daclizumab (both anti-CD25 [IL-2 receptor]) as well as alemtuzumab (Campath-1 H, anti-CD52) were proven to be effective induction agents in preventing AR in renal or pancreas transplantation [11,12,13]. Furthermore, strategies such as corticosteroid avoidance and CNI-reduced or CNI-free maintenance immunosuppression were shown to be equivalent to traditional triple-drug maintenance [14,15,16]. Nonetheless, all immunosuppressive agents carry some risk of toxicity and adverse reactions that may complicate therapy (Table 186.2).

CR remains a common yet poorly understood clinical problem, with slightly different manifestations in each type of graft. Over time, the accumulation of microvascular injury in a graft degrades graft function, with eventual graft loss. This process appears to be mediated by multiple mechanisms, likely including both immune and nonimmune factors. Evidence for the contribution to CR of immune factors includes the observation that AR episodes significantly increase the likelihood of CR as well as the correlation, observed in renal transplant recipients, between a poor response to AR treatment and the subsequent development of CR [17]. A similar association between a poor response to AR treatment and the subsequent development of CR has been observed in liver transplant recipients, although reversible AR has little impact. Nonimmune factors likely also contribute to the development and progression of CR, including the toxic effects of immunosuppressive medication and cumulative injury from infection such as that caused by cytomegalovirus (CMV) [18] and polyomavirus [19]. CR nearly always eventuates in graft loss, although the rapidity of the process varies considerably.

Current reports indicate that about 10% to 25% of renal transplant recipients experience an episode of AR. Because most episodes are clinically silent, the diagnosis of
AR must be considered in recipients whose serum creatinine, blood urea nitrogen, and urinary output values have normalized and whose graft function has been stable in the outpatient setting, but whose serum creatinine and blood urea nitrogen values subsequently rise while their urinary output decreases. The presence of hypovolemia, drug nephrotoxicity (e.g., high calcineurin levels), ureteral obstruction or leak, lymphocele, or vascular anastomotic complications should be excluded, and the diagnosis of AR should be established via histologic examination of a percutaneous graft biopsy specimen. Rarely, tenderness and swelling in the area of the graft occur, and occasionally fever or other signs of systemic inflammation, although such findings used to be common.

As discussed earlier, most AR episodes occur in the early posttransplant period. Among the subset of recipients who experience delayed graft function, up to 30% exhibit evidence of AR on biopsy [20]; 20% of recipients who require dialysis posttransplant have AR [21]. Intriguingly, up to 30% of recipients with well-functioning grafts also have AR, per early posttransplant protocol biopsies, but whether such findings are clinically important and whether mild episodes should invariably be treated remain controversial [22]. Recent studies have provided data that may allow prediction of individual risk of AR, with the potential for individualizing immunomodulatory therapy. For example, donor IL-6 genetic polymorphism is strongly associated with an increased incidence of AR posttransplant [23], and recipients with elevated levels of serum C-reactive protein (CRP), presumably indicative of systemic inflammation, have a higher rate of AR and a shorter time to AR than those with lower CRP levels [24]. Other biomarkers (such as soluble CD30, gene expression assays on peripheral blood samples, urinary proteomics, and T-lymphocyte subset analysis) were shown to be predictive for rejection or transplant tolerance, and are currently undergoing various clinical investigations [25].

The diagnostic workup for AR includes studies that may identify alternative causes of recipient graft dysfunction (Table 186.3). It is vital to consider alternative diagnoses, particularly in the early postoperative period, including vascular problems with the arterial or venous anastomoses, ureteral obstruction, or urinary leak. Other common causes of apparent graft dysfunction include the acute tubular necrosis associated with delayed graft function, hypovolemia and attendant prerenal azotemia, and the nephrotoxic effects of cyclosporine and tacrolimus. To rule out the vascular and ureteral problems discussed previously, a duplex ultrasound study of the renal graft is commonly obtained. Several ultrasound findings may suggest the diagnosis of AR: increased size of the graft, increased cortical thickness, enlargement of the renal pyramids, and decreased
graft renal artery blood flow [26]. The diagnosis of AR is clearly established by percutaneous allograft biopsy and histologic examination. Biopsy is generally safe when performed by experienced practitioners; however, complications include bleeding, hematoma and arteriovenous fistula formation, and ureteral or major vascular injury.

Rejection is graded according to a standardized histologic classification scheme, the modified Banff Criteria, which may be used to guide therapy [27]. Fine-needle aspiration biopsy has been used by some centers to establish the diagnosis of AR; however, some consider the loss of microstructural data, as compared with traditional core biopsy, to be a weakness of the technique. In particular, the diagnoses of acute vascular rejection and CR are difficult to make using fine-needle aspiration biopsy.

The treatment of AR in renal transplant recipients varies between centers. High-dose methylprednisolone (500 to 1,000 mg per day or every other day [2 to 3 doses] is common) is often the initial approach. Corticosteroid-resistant AR, or AR that is histologically graded as severe or vascular, is often treated with potent depleting antilymphocyte antibodies such as murine monoclonal IgG2a antibody (OKT3) or polyclonal antithymocyte globulin (antithymocyte gamma globulin, Thymoglobulin). Alemtuzumab was selectively used to treat AR in some centers [28]. Since some AR episodes occurred while the recipients were on stable immunosuppression, their maintenance therapy was switched from cyclosporine to tacrolimus or from azathioprine to mycophenolate mofetil. Most AR episodes are reversible with current therapies; however, as noted previously, the long-term outlook for preservation of graft function is lessened with each episode, especially when the posttreatment serum creatinine level does not return to the pre-AR baseline.

CR in renal transplant recipients is a persistent clinical problem and appears to be multifactorial, with immunologic and nonimmunologic factors driving the gradual loss of graft function. As described earlier, minimizing the frequency and severity of AR episodes is important in decreasing the likelihood of eventual CR. Nonimmunologic factors thought to contribute to CR include (a) episodes of infection, particularly due to CMV and BK virus (vide infra); (b) the nephrotoxicity of CNI therapy; (c) ischemia-reperfusion injury and delayed graft function in the peritransplant period; and (d) innate cell senescence within the graft [29]. Attention is being directed toward identifying inflammatory activity within the graft, in response to both immune and nonimmune insults that may contribute to the development of CR. One of the leading causes of kidney retransplants is CR. It remains a formidable problem that is still poorly understood.

The hepatic graft is considered to be immunologically “privileged” in that evidence of some degree of immune tolerance occurs in a substantial number of liver transplant recipients over time. Despite that observation, all forms of rejection can occur posttransplant. At one time, it was thought that HAR did not occur in the hepatic graft; this idea is now known to be incorrect, as anti-HLA antibody-mediated HAR has been described in liver transplant recipients [30,31]. Unlike the renal graft, the hepatic graft undergoes HAR over a number of days, not minutes to hours, probably secondary to its ability to absorb a large amount of antibody before the onset of the significant microthrombosis and vascular damage seen in HAR. A more delayed form of antibody-mediated rejection is seen in up to 33% of patients who undergo liver transplants across ABO-incompatible blood groups [32], but even this barrier appears surmountable with the use of plasmapheresis along with aggressive immunosuppression [33].

AR remains an important clinical problem in liver transplantation; even with the use of standard multiagent immunosuppression, the incidence of AR ranges from 30% to 80%. In two large, multicenter trials, double therapy with a CNI and steroids resulted in a 60% to 80% incidence of AR [34,35]. Triple therapy with Neoral® or Sandimmune®, along with azathioprine and prednisone, resulted in a 30% to 45% incidence of AR [36]. Substitution of mycophenolate mofetil for azathioprine further reduced the incidence of AR to 26% [37]. The latest liver transplant regimen, consisting of two doses of a monoclonal anti-IL2 receptor (basiliximab) as induction therapy and dual maintenance therapy with a CNI and mycophenolate mofetil, was shown to lessen the severity of rejection without increase the infection rate [38,39].

The diagnosis of AR in liver transplant recipients is normally suggested by elevated levels of transaminases, bilirubin, or alkaline phosphatase. Among patients with T-tube drainage (which is increasingly uncommon), the biliary drainage may be seen to thicken, darken, and decrease in amount. The suspicion of AR mandates graft biopsy and studies to eliminate other possible causes of early hepatic graft failure. Duplex ultrasonography and, in some cases, cholangiography are increasingly being replaced by magnetic resonance imaging. Biopsy findings are classified, according to a standardized set of criteria, as mild, moderate, and severe, with clear implications for prognosis [40]. AR is normally treated with high-dose corticosteroids, but 5% to 10% of cases are steroid-resistant; such recipients are then treated with an antilymphocyte antibody. Interestingly, in large population studies, the incidence of AR is associated with improved long-term patient survival rates [41], albeit thought to be due to the higher incidence of AR in younger, healthier recipients. Even adjusting for recipient characteristics, AR has not been clearly associated with either decreased graft or patient survival rates; however, frequent AR episodes are a risk factor for subsequent CR, so continued pursuit of immunosuppressive strategies that reduce the risk of AR is imperative.

CR in liver transplant recipients is characterized by vascular obliteration and bile duct loss (“the vanishing duct syndrome”). Seen in 5% to 10% of recipients, it is more common in those with vasculitic findings during AR episodes; if larger vessels are not seen on biopsy, the diagnosis of CR may be misread as AR. The incidence of CR appears to be decreasing, perhaps as a result of changes in immunosuppressive regimens [42]. In addition to multiple AR episodes, other factors associated with an increased risk of CR include CMV infection, chronic hepatitis, increased donor-recipient histocompatibility differences, and increased ischemia time. CR does not always herald graft loss; long-term patient survival and even regeneration of bile ducts have been described. Tacrolimus has been used to salvage grafts in recipients with CR on cyclosporine-based immunosuppression, with a 73% success rate [43].

At most centers, patients undergoing a pancreas transplant alone or a simultaneous pancreas–kidney transplant receive more potent immunosuppression than do renal transplant recipients, thanks to initial studies demonstrating a higher rate of AR after those two types of pancreas transplant [44]. Overall success rates continue to improve: the risk of AR has been reduced by standardized induction therapy with antilymphocyte antibody preparations, and it may be further reduced with mammalian target of rapamycin (mTOR) inhibitors and/or with IL-2 receptor monoclonal antibodies [45].

Establishing the diagnosis of AR in pancreas transplant recipients may be difficult. Hyperglycemia is a late finding that only occurs with substantial loss of functional islet-cell mass.
By the time hyperglycemia is seen, it may be too late to retain a functional graft. Clinical findings may include fever and graft tenderness; however, pancreas graft rejection is often clinically silent.

For pancreas grafts transplanted along with a renal graft, a rising creatinine level is often used as a surrogate marker of rejection, with antirejection therapy aimed at both the pancreas graft and the renal graft. However, isolated pancreas graft rejection is observed in up to 20% of simultaneous pancreas–kidney transplant recipients who have AR [46].

In pancreas transplant recipients with exocrine bladder drainage, a decreasing urinary amylase level may be used as a marker of graft rejection [47]. Other possible markers of rejection (serum anodal trypsinogen, serum amylase, soluble HLA, and analysis of glucose-disappearance kinetics during a brief glucose tolerance test) have been examined but have failed to gain wide acceptance.

The diagnosis of pancreas graft rejection is confirmed by biopsy, which may be performed percutaneously or, in bladder-drained recipients, through a cystoscopic, transduodenal approach. Complications (bleeding, arteriovenous fistula formation, graft pancreatitis) have been described, but most biopsies do not lead to complications. Pancreas transplant recipients with early evidence of graft dysfunction should undergo Doppler ultrasonography to rule out graft thrombosis, which occurs in up to 10% to 20% of them [48].

Treatment of AR for pancreas transplant recipients is similar to that for renal or liver transplant recipients. High-dose corticosteroids are given initially, but a low threshold is maintained for possibly switching to antibody-based therapy, given the relatively common steroid resistance. Most AR episodes are reversed with treatment.

Rejection in heart transplant recipients is a major obstacle to long-term success and accounts for up to a third of the deaths. All forms of rejection are seen in heart transplant recipients. Albeit rare, HAR due to preformed antigraft antibodies occurs within minutes to days; it manifests with rapid deterioration of cardiac function, with prolonged need for inotropic support. In recipients whose grafts fail to recover rapidly, an attempt to reverse HAR by plasmapheresis may be made, but success is uncommon, and an immediate retransplant is usually required.

AR in heart transplant recipients is common and usually occurs in the first 3 to 4 months posttransplant. At one time, the diagnosis was made on the basis of the development of congestive heart failure or the elaboration of electrocardiographic abnormalities. However, the present-day use of protocol endomyocardial biopsies has eliminated such late findings of AR, except in noncompliant recipients. Most centers use frequent percutaneous transjugular right ventricular endomyocardial biopsies as part of a standardized surveillance protocol. Biopsies are evaluated histologically, according to an international grading system [49], and therapy is directed accordingly.

Several investigators have developed noninvasive approaches to establishing the diagnosis of AR, including electrocardiographic frequency analysis, nuclear scintigraphic techniques, and echocardiography; however, no approach has attained sufficient sensitivity to eliminate the need for protocol biopsies. The need for continued endomyocardial biopsies later than 1 year posttransplant is controversial, and many centers discontinue performance of biopsies at 1 year unless indicated on clinical grounds.

The treatment of AR is based on histologic findings. Bolus steroid therapy is used in lower-grade rejection without hemodynamic compromise; oral prednisone therapy for mild AR also has been used with success [50]. Salvage therapy with an antilymphocyte antibody agent is most common in recipients with histologic findings of more severe rejection, in recipients with steroid-resistant rejection, and in recipients with signs of hemodynamic compromise. In a series of 100 of such high-risk recipients, AR was reversed in 90% of those treated with 10 to 14 days of OKT3 [51]. However, other investigators have had markedly lower rates of success with OKT3 in the treatment of steroid-resistant rejection [52]. Methotrexate also has been used to reverse AR that fails to respond to steroids or that is refractory to OKT3.

Other approaches include switching from cyclosporine-based to tacrolimus-based immunosuppression as rescue therapy in recipients with refractory AR, a strategy that was proved to be safe and efficacious [50]. Photopheresis has been used in the treatment of recipients with T-cell lymphoma and autoimmune disease. Studies of photopheresis and triple-drug immunosuppression have provided evidence of a decrease in the total number of AR episodes, as compared with triple-drug immunosuppression alone [50]. Of note, photopheresis has reversed refractory high-grade rejection in small numbers of heart transplant recipients [53].

CR manifests in heart transplant recipients as cardiac allograft vasculopathy (CAV), an entity that is the major cause of late-term morbidity and mortality. The pathologic findings of CAV include progressive intimal thickening in a concentric manner, which begins distally within the cardiac vasculature. It is associated with the loss of response to endogenous (and pharmacologic) vasodilators [50]. CAV is thought to be immunologically mediated, because HLA donor-related matching is clearly associated with reduced rates of CAV [54]. Nonimmunologic mechanisms are also thought to be involved; identifiable risk factors for CAV include hyperlipidemia, donor age older than 25 years, recipient weight gain, CMV disease, preexisting donor or recipient coronary artery disease, and increasing time posttransplant [50]. Another nonimmunologic risk factor for CAV is ischemic time during the peritransplant period. As in other solid-organ transplant recipients, the use of mycophenolate mofetil is associated with a reduction in the incidence of CR in heart transplant recipients [55].

The lung graft is highly prone to rejection—nearly all lung transplant recipients experience at least 1 AR episode. The clinical difficulty posed by rejection is in distinguishing it from other causes of decreased graft function, most commonly infection.

HAR of the lung graft [56] is mediated by recipient preformed antibodies to the donor graft, in a fashion similar to other organs. The clinical manifestation is similar to the more common ischemia-reperfusion injury, which, unlike HAR, usually resolves. HAR of the lung graft is rare and only described in case reports. To date, we know of no lung transplant recipients who have survived HAR. It must be prevented via initial cross-match testing and exclusion of immunologically unsuitable donor organs.

Most AR episodes occur during the first 3 to 6 months posttransplant. Some recipients experience symptoms, including fever, cough, and dyspnea, but many are asymptomatic. Early diagnosis of AR in lung transplant recipients is essential: untreated AR can lead to respiratory insufficiency or failure, and repeated AR episodes are associated with an increased risk of bronchiolitis obliterans and eventual graft failure [57].

The diagnosis of AR is made by transbronchial biopsy, although less invasive techniques continue to be assessed [58]. Bronchoalveolar lavage (BAL) is also performed to rule out
infection before increasing immunosuppression; infection and rejection may occur simultaneously in up to 25% of lung transplant recipients with AR [59]. Early diagnosis of AR may be aided by spirometry; decreases in timed forced expiratory volume, in pulmonary capillary blood volume, and in the diffusing capacity of the lungs for carbon monoxide are associated with AR and should prompt investigation. Radiography is not ordinarily helpful. The histologic findings of AR include lymphocytic infiltrates into the perivascular and interstitial spaces; AR is graded according to histologic findings [60].

The initial treatment of AR in lung transplant recipients typically entails high-dose corticosteroids; if they are not successful, anti–T-cell antibody therapy is tried next. Many recipients initially respond to the steroid pulse therapy, yet it may not completely clear their AR, and secondary episodes are common, so additional therapy may be required. For that reason, surveillance bronchoscopy with transbronchial biopsies and BAL are common after initial treatment [61].

CR in lung transplant recipients is extremely common, affecting up to 40% of recipients at 2 years posttransplant and up to 70% of recipients after 5 years [62]. The mean time to diagnosis of graft dysfunction posttransplant is 16 to 20 months. A definitive histologic diagnosis of early bronchiolitis obliterans may be difficult to obtain, so it must be established largely on clinical grounds. Radiography, again, is not specific. Typical presenting symptoms are cough, progressive dyspnea, and loss of exercise tolerance. The use of home spirometry can point to the diagnosis based on a 20% reduction in timed forced expiratory volume on successive measurements [63]. Factors associated with accelerated bronchiolitis obliterans include multiple episodes of AR, CMV pneumonitis/infection, Pneumocystis jiroveci pneumonia (PCP), and episodes of airway ischemia [62,64].

Many different therapies have been tried for recipients with bronchiolitis obliterans, but with little success. Increases in immunosuppression, antilymphocyte antibody therapy, and inhaled cyclosporine have all been tried. Ultimately, the progress of bronchiolitis obliterans is inexorable, with continued loss of graft function and subsequent death. A lung retransplant is the only viable option [65].