Evaluation and Treatment of Graft Dysfunction
John R. Crew
David J. Cohen
Division of Nephrology, Columbia University Medical Center, New York, New York 10032
INTRODUCTION
The importance of monitoring and responding in a timely manner to changes in allograft function cannot be overestimated. As in virtually every other area in medicine, early diagnosis and appropriate intervention produces improved outcomes. The renal allograft recipient, even with excellent allograft function, is by definition a patient with chronic renal insufficiency. A renal transplant faces a hostile environment, subject to injury at the time of organ donation, preservation and implantation, nephrotoxic medications, immune attack, hypertension, hyperlipidemia, hyperfiltration, and possibly recurrent disease. Nephron mass and glomerular filtration require constant attention and protection. Patients can ill afford reductions in allograft function. The absolute level of allograft function (as measured by serum creatinine) in the first year following transplantation as well as any irreversible deterioration in renal function over this period are highly significant determinants of long-term success (1). In addition, any permanent elevation in serum creatinine over baseline by as little as 0.3 mg/dL during the first posttransplant year greatly increases the risk of allograft failure (deceases the graft half-life) over the period of follow-up. Although conservation of allograft function during the first posttransplant year (and likely the first 3 to 6 months posttransplantation) may be the most important for long-term success, any irreversible loss of function at any time following transplantation from any cause is equally likely to accelerate the development of further structural damage and shorten the duration of overall graft survival given the relentless stresses to which the allograft is exposed. A 30% chronic decline in inverse
serum creatinine (1/creatinine) occurring at any time after transplantation has been found to be a strong and independent predictor of late allograft failure (2,3).
serum creatinine (1/creatinine) occurring at any time after transplantation has been found to be a strong and independent predictor of late allograft failure (2,3).
Therefore, accurate diagnosis of the etiology and effective treatment of allograft dysfunction to minimize any irreversible injury is a critical factor in maximizing long-term allograft success rates. Although immune-mediated processes such as acute rejection are more likely to occur early after transplantation, they remain a constant threat throughout the life of the allograft, and the differential diagnosis of allograft dysfunction is broad at all time periods.
MEASUREMENT OF ALLOGRAFT FUNCTION
Monitoring allograft function includes direct and indirect measurements of the glomerular filtration rate (GFR) as well as quantitation of urinary protein excretion. The frequency of evaluation is determined by the risk of allograft dysfunction. More frequent monitoring is required early (first 6 months) after transplantation because the risk of acute rejection is highest during this time period. Virtually all causes of dysfunction are more easily reversed when detected early. Later after transplantation, as the risk of acute rejection decreases, testing can be performed at greater intervals. The American Society of Transplantation has made recommendations on how often to evaluate renal transplant function (4). Assessment of allograft function two to three times per week for the first month, every week for the second month, bi-weekly for months three and four, monthly until the end of the first year, every 2 months during the second year, then every 3 to 4 months thereafter is recommended. The authors admit that these recommendations are not based on any scientific data but do fit to current practice in many transplant centers.
The best way to assess allograft function remains debatable. The ideal diagnostic test would be cheap and accurate, correlate with patient and graft outcomes, be widely available, and be sensitive to small changes in allograft function. Currently used tests that assess allograft GFR are discussed below. Another important, easily detected indicator of allograft dysfunction is proteinuria, which demonstrates the presence of either recurrent glomerular disease or transplant glomerulopathy, and is a measure of cardiovascular risk. In addition, several serologic and urinary tests correlating with antiallograft immunologic activity and suggestive of impending allograft injury are available and will be discussed as well.
Serum Creatinine
The standard test used to monitor graft function is measurement of the serum creatinine. It is widely available, cheap, and the results are available quickly. Studies have correlated elevation in serum creatinine and increases in serum creatinine from baseline with the risk of graft loss (5). Graft dysfunction detected by elevated serum creatinine also identifies patients at risk for cardiovascular death. The serum creatinine concentration, however, is an indirect measurement of GFR, and by itself is a relatively insensitive indicator of changes in graft function. At low serum creatinine levels, small changes in creatinine can indicate large changes in function. For example, an increase in creatinine from 1.0 to 1.2 represents a 20% decline in kidney function, potentially an absolute decrease in GFR of 18 to 20 mL/min. A similar change in creatinine from 2.0 to 2.2 represents only a 10% decline in kidney function and a smaller absolute change in GFR. Additionally, patients with small muscle mass, such as chronically ill, debilitated patients, or elderly patients, generate less creatinine on a daily basis; “normal” levels of creatinine in this population may actually mask significant levels of allograft dysfunction. Several formulas have been developed that take into account age and patient weight to adjust for these situations (see below). Significant damage, as well as antigraft immune activity can develop in the transplanted kidney despite a stable serum creatinine concentration. Studies using protocol biopsies on patients with stable serum creatinine levels have shown that despite apparent clinical stability, patients may be experiencing medication toxicity, fibrosis, and subclinical tubulitis/acute rejection (6, 7, 8, 9).
24-Hour Creatinine Clearance
Twenty-four-hour urine collection and measurement of creatinine clearance improve the accuracy of assessing GFR, particularly in patients with significantly reduced muscle mass (creatinine production) (10,11). However, the completeness of urine collection and tubular secretion of creatinine reduce the accuracy of the test. Collection of urine over a 24-hour period is frequently complicated by under collection or over collection. Additionally, storage at room temperature or low pH for 24 hours can increase the conversion of urinary creatine to creatinine, falsely elevating the clearance rate (12). Tubular secretion contributes to the creatinine clearance beyond that filtered by the glomerulus; therefore, 24-hour creatinine clearance tests systematically overestimate the GFR. Patients with renal transplants appear to have a similar rate of tubular creatinine secretion as other patients who have a solitary kidney-transplant donors or patients undergoing nephrectomy for malignancy (13). Unfortunately, the inter-individual and intra-individual variability in tubular secretion of creatinine is high, and the relative proportion of tubular creatinine secretion rate to GFR increases as the GFR declines (12). All of these factors limit the accuracy of estimating the true GFR from the measurement of creatinine clearance.
Formulas That Estimate Glomerular Filtration Rate
Several formulas have been developed that overcome the limitations of simple measurement of the serum creatinine. These formulas frequently include other anthropometric data (height, weight, sex, and race) as well as additional laboratory
information (BUN, albumin, and serum creatinine) in an attempt to more accurately estimate GFR.
information (BUN, albumin, and serum creatinine) in an attempt to more accurately estimate GFR.
Most formulas were developed for chronic kidney disease patients in general and may not be accurate in the setting of renal transplantation. To address this, Nankivell et al (18,19) published 3 formulas based on an evaluation of 146 renal transplant recipients who had laboratory evaluations, anthropometric data, and GFR determinations by 99mTc-DTPA. When compared to an independent, random sample drawn separately from this population, the Nankivell formulas were more accurate than the Cockroft-Gault, Mawer, and Siersback-Neilson formulas.
Other commonly used formulas for determining GFR in clinical nephrology were derived by Levey et al using data gathered during the Modification of Diet in Renal Disease (MDRD) study, a study involving 1,628 patients with 125I-Iothalamate clearances (20). One formula has proved to be more accurate than serum creatinine and creatinine clearance in estimating GFR (4,10). The MDRD formula has been validated in patients with renal transplants (10,21, 22, 23). Using inulin clearance as the gold standard, the MDRD formula was a more accurate predictor of GFR than the 24-hour creatinine clearance, Cockroft-Gault formula, and Nankivell formulas.
Serum Cystatin C
Serum cystatin C is a small, nonglycosylated protein (122 amino acids, 13 kDa) belonging to a family of cysteine protease inhibitors that is made by all nucleated cells at a relatively constant rate (24). It is freely filtered by the glomerulus, reabsorbed and degraded in the proximal tubule. Filtered cystatin C does not return to the circulation, nor is it secreted by the renal tubular cells. Its production is not dependent on sex, age, or muscle mass (25). It thus has many attractive features as an ideal molecule to assess the GFR. It has been validated as a useful marker of GFR in different kidney diseases, including transplantation, and may be more sensitive than serum creatinine at detecting early stages of renal dysfunction (12,24,26, 27, 28). In transplantation, serum cystatin C may underestimate GFR (29,30). This may be due to the effect of corticosteroids elevating the serum levels, but not all studies have shown this effect (31,32). Serum cystatin C appears to be superior to serum creatinine alone, even among patients using steroids. Similar to serum creatinine, elevated levels of serum cystatin C in transplant patients may predict graft failure (25). The limitations of cystatin C are its lack of availability and lack of widespread use. It may be used in the future to monitor transplant patients with good allograft function, since elevations of cystatin C may occur earlier than creatinine (33). Further studies correlating serum levels and clinical outcomes are needed to validate its utility.
Radionuclide and Radiocontrast Determinations of Glomerular Filtration Rate
Radionuclides and radiocontrast agents can provide extremely accurate measurements of graft function. Rather than relying on compounds generated by the body that may be affected by medications or changes in metabolism, these techniques allow for direct determination of the GFR. If a compound is cleared only by the kidneys, the GFR can be calculated by its rate of disappearance from the circulation, which can be measured by serial blood tests, or even by direct measurement with a gamma camera over the allograft (12). Blood measurements are more frequently used because of concerns regarding cost. Several different compounds have been used for this purpose. The most commonly used agents today are 51Cr-labeled ethylene diaminetetra-acetic acid (51Cr-EDTA), 125mTechnetiumlabeled diethylenetriamine penta-acetic acid (125mTc-DTPA), and 125I- Iothalamate (34). With earlier techniques, patients required intravenous injection followed by serial blood draws and urine collections to follow the clearance over time. The time and frequency of blood draws required were inconvenient, and ensuring complete bladder emptying in patients with bladder dysfunction posed another problem. With current techniques, a single subcutaneous injection followed by a single blood draw 4 to 5 hours later accurately predicts GFR (35,36). These techniques remain accurate in renal transplant patients (37). Use of radioactive materials has fallen out of favor with some physicians and patients. The amount of radiation used is less than in many radiologic studies, but the agents are concentrated in the kidney and urologic tract, increasing exposure in these tissues (12). To minimize radiation exposure of the genitourinary tract, patients are recommended to empty their bladders frequently, and increase their water intake to increase urine volume.
To avoid exposure to radiation, the nonradioactive contrast agents iohexol and iothalamate have been used. Serum concentrations can be measured with x-ray fluorescence, and GFR can be calculated with a single blood draw (38). The dose of contrast, frequently less than 30 mL, is usually free of nephrotoxicity (12). A potential benefit of using iothalamate to directly determine the GFR is to detect changes in renal function that have not affected serum creatinine levels (39). It remains to be seen whether the increased sensitivity to detect small changes will translate into clinical benefit for transplant patients.
Ultrasonography
Ultrasonography is critical in the evaluation of transplant dysfunction. The transplanted kidney’s location in the iliac fossa, close to the skin with little intervening tissue, improves the ability of ultrasound to detect structural changes and makes Doppler studies of the vasculature more reliable (40,41). It is most important in excluding causes of dysfunction other than acute transplant rejection. With ultrasound it is possible to diagnose obstruction from ureteral disorders or lymphoceles, other perinephric collections including abscesses, hematomas, urinomas, arteriovenous fistulas, and rarely intraparenchymal posttransplant lymphoproliferative disorder (42). Doppler ultrasound evaluates disorders of the larger vessels, particularly renal artery stenosis, renal artery and vein thrombosis, and rarely renal
vein stenosis. In kidneys with poor function immediately following implantation, Doppler ultrasound is essential to detect abnormalities of blood flow. Lack of blood flow immediately after transplantation suggests either hyperacute rejection or a vascular occlusion.
vein stenosis. In kidneys with poor function immediately following implantation, Doppler ultrasound is essential to detect abnormalities of blood flow. Lack of blood flow immediately after transplantation suggests either hyperacute rejection or a vascular occlusion.
One commonly measured parameter derived from Doppler interrogation of the intragraft circulation is the resistance index (also reported as the resistive index). The resistance index is calculated as:
This measurement attempts to describe the intrinsic resistance to flow in the small blood vessels and glomeruli. In theory, the more abnormal the small arteries and glomeruli are, the worse the flow in diastole, and the higher the resistance index will be. Stenosis of a large artery can lead to poststenotic vasodilatation, increased diastolic blood flow, and a decreased resistance index. However, many other factors affect the resistance index, including the compliance of the arterial circulation, the pulse pressure, and the patient’s heart rate, limiting its diagnostic use (43).
An increased resistance index is often associated with acute rejection (44,45), but may also be seen with acute tubular necrosis (ATN), cyclosporine toxicity, and obstruction; it is not specific enough to be diagnostic in the setting of acute renal dysfunction (41,46). The utility of the resistance index in patients with chronic allograft dysfunction is still being evaluated (see discussion on chronic allograft nephropathy below) (47).
Radionuclide Imaging
Nuclear imaging complements ultrasound evaluation of the renal transplant. A radiolabeled tracer is injected into the bloodstream, filtered by the glomerulus, and excreted in the urine. A gamma camera placed over the allograft can follow the blood flow and excretion. 99mTechnetium mercaptoacetyltriglycine (99mTc-MAG-3), 99mTc-DTPA, 123I-Hippuran have all been used. 99mTc-MAG-3 is the agent of choice at most centers for patients with poor renal function because it is also extracted from the blood and secreted into the urine by the renal tubules (48). In patients with delayed graft function (DGF) after transplantation, renal scintigraphy can be used to be certain that blood flow is present. It is also useful in the setting of urinary tract obstruction. The sensitivity of renal transplant ultrasound is very high for obstruction, but may not show dilatation early in the course (41). Since mild degrees of hydronephrosis are common after transplantation and do not always indicate that functional obstruction is present, nuclear imaging can provide useful diagnostic information in this setting. Finally, renal scintigraphy can diagnose urine leaks and urinoma formation, as the radiotracer leaves the ureter to enter a collection or the abdominal cavity (42). Radionuclear imaging does not appear to be sensitive or specific enough to reliably diagnosis acute rejection (49).
Noninvasive Diagnosis of Acute Rejection
Currently, renal biopsy is the gold standard for diagnosing acute rejection but is not without potential complications (50,51). A report on complications of 2,127 protocol renal biopsies at four centers in Europe identified only one patient with graft loss, two with hemorrhage requiring surgical intervention, one hemorrhage requiring interventional radiology intervention, three hemorrhages requiring blood transfusion, and two episodes of peritonitis from a perforated bowel (52). Additionally, inadequate tissue sampling has been reported to occur in 5% to 20% of biopsies (53). To obviate the need for renal or to complement information obtained by biopsy, there has been a long search for reliable serum and urine tests indicating the cause of allograft dysfunction, specifically whether acute rejection is occurring. Ideally, any surrogate for renal biopsy should also provide information regarding the severity and type of rejection. Despite much effort there is only beginning to be success at solving these issues.
Cytokines
Given that acute rejection is an inflammatory state, many attempts at noninvasive diagnosis of acute rejection have focused on the expression of cytokines. Elevated levels of the cytokines IL-1, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, INF-γ, and TNF-α have been described in the serum and urine (54, 55, 56, 57, 58, 59, 60). Although promising, cytokine markers lack specificity for acute rejection. Many of these inflammatory markers are elevated in the serum and urine in the setting of infections, and most show considerable overlap between quiescence and rejection. Serum and urine levels of IL-6 are elevated in the presence of acute rejection and DGF (57,61,62). Decreasing urinary concentration of IL-6 correlates with resolution of the intraparenchymal inflammation; patients with steroid-resistant rejection have been found to have persistently elevated urine levels until the rejection resolves. Unfortunately, the sensitivity and specificity of urine IL-6 levels were only 80% and 75% (61,63). IL-6 levels are higher in patients with chronic allograft nephropathy than stable renal transplant patients or controls (54,64). Altered serum and/or urine levels of any single cytokine do not appear to be sufficient to diagnose acute rejection: combinations of cytokine levels, cytokine mRNA expression profiles in circulating lymphocytes, and cytokine genetic polymorphisms may be useful in diagnosing or predicting which patients are at a highest risk for rejection (65, 66, 67, 68).
Urine Enzymes
Rather than using nonspecific markers of inflammation, Li et al (69) used levels of mRNA expression in urinary cells for enzymes produced by cytotoxic cells, specifically perforin and granzyme B. They found increased mRNA expression from urinary cells for both enzymes in the setting of biopsy-proven acute rejection but not in cases of chronic
allograft nephropathy (69). A follow-up study compared transplant patients with acute rejection to three other groups: transplant patients with urinary tract infections (UTIs), nontransplant patients with UTIs, and transplant patients without acute rejection or UTI. They found elevated levels of granzyme B mRNA expression only in urinary cells of patients with acute rejection (70); granzyme B mRNA levels were not different between the control groups. One important limitation of urinary granzyme B mRNA testing is that it does not deliver specific information about the type of rejection or the severity to help guide treatment.
allograft nephropathy (69). A follow-up study compared transplant patients with acute rejection to three other groups: transplant patients with urinary tract infections (UTIs), nontransplant patients with UTIs, and transplant patients without acute rejection or UTI. They found elevated levels of granzyme B mRNA expression only in urinary cells of patients with acute rejection (70); granzyme B mRNA levels were not different between the control groups. One important limitation of urinary granzyme B mRNA testing is that it does not deliver specific information about the type of rejection or the severity to help guide treatment.
Proteomics
Proteomics is another method of urine evaluation that may show promise in the noninvasive diagnosis of acute rejection in the future. Proteomics allows the separation and analysis of large numbers of proteins at one time (71). Investigators do not need to know the identity, sequence, or function of the proteins to do the initial investigations. Clarke et al (72) subjected urine samples from 17 patients with acute rejection compared to 15 stable renal transplant patients, and identified several “peaks” that identified acute rejection with 100% specificity and 83% sensitivity. These data need to be replicated on a larger sample and need to be compared to other subsets of transplant patients, including those with chronic rejection and infections.
Donor-Specific Antibodies
The presence of donor-specific anti-human leukocyte antigens (HLA) IgG antibodies prior to transplantation is associated with poor graft outcomes, and until recently was considered an absolute contraindication to transplantation. Testing the kidney transplant recipient after transplantation for the presence of antidonor HLA antibody has assumed greater importance since the discovery that C4d staining in the peritubular capillaries is specific for an anti-body-mediated response against the allograft, particularly in acute humoral rejection. C4d represents the remainder of the C4bBb complement complex that becomes covalently bound to the endothelial cell membrane following activation of the complement cascade. The presence of donor-specific anti-HLA antibodies is now required for the diagnosis of acute humoral (antibody-mediated) rejection; all patients with pathologic features of acute humoral rejection and peritubular capillary C4d staining in renal biopsies should be tested for donor-specific antibodies (73). This is particularly important since treatment with plasmapheresis, intravenous immunoglobulin, and modification of the immunosuppressive regimen has resulted in markedly improved outcomes in patients with antibody-mediated rejection (74).
The value of routine monitoring for the presence of donor-specific anti-HLA antibodies in the absence of acute humoral rejection is still debated. Twelve percent to 60% of patients with a negative crossmatch at the time of transplantation will develop donor-specific anti-HLA antibodies (75). Several studies have demonstrated that the development of anti-HLA donor-specific antibodies after transplantation adversely affects graft outcome (76, 77, 78, 79). In one study, the early development of donor-specific anti-HLA antibodies was associated with a 1-year graft survival rate of 37% compared to 86% in patients who did not develop an antibody response (79). The development of anti-HLA class I antibodies is more associated with the development of acute rejection, while anti-HLA class II antibodies are more associated with chronic rejection (75,80). In the study by Worthington et al, development of anti-HLA class I antibodies was associated with graft loss after a mean 2.7 years later, compared to 3.9 years for anti-HLA class II antibodies (80,81). These antibodies can develop immediately after transplantation, before and after acute rejection, and after graft loss/return to dialysis (81,82). The type of antibody formed may also be important with rejection being more associated with IgG antibodies, while IgM antibodies had little or no effect on kidney transplant patients (75). The pathology in these series has not been well delineated, but the development of anti-HLA antibodies has been associated with acute humoral, acute cellular, and a five- to six-fold increased risk of chronic rejection/chronic allograft nephropathy. Recent histologic studies of chronic allograft nephropathy show peritubular capillary C4d staining occurs in approximately 30% to 83% of patients (83, 84, 85). Anti-HLA class II antibodies have separately been associated with positive C4d staining (85,86). These findings together suggest that the humoral immune system and anti-HLA antibodies frequently contribute to chronic allograft nephropathy, and that these patients could indeed be properly described as having “chronic rejection.” As the contribution of anti-donor HLA antibodies to graft outcome is better defined, detection of these antibodies may become an important adjunct in routine monitoring in all recipients, particularly those with allograft dysfunction.
There are several ways of measuring donor-specific antibodies, including standard cytotoxicity (CDC), antihuman globulin-enhanced cytotoxicity (AHG-CDC), flow cytometric crossmatching (FCXM), ELISA assays, and flow cytometry using beads coated with known HLA molecules. Cytotoxicity and AHG-CDC require donor lymphocytes which may not be available months or years after cadaveric transplantation. Both flow cytometry and ELISA assays are more sensitive at detecting donor-specific antibodies than the cytoxicity assays. Patients with positive flow cytometry testing but negative CDC assay for donor-specific antibodies are at a higher risk of acute rejection and graft loss (87,88).
DYSFUNCTION IMMEDIATELY AFTER TRANSPLANTATION
The approach to diagnosis and management of allograft dysfunction is best understood by the time period after transplantation. Although many of the causes of allograft dysfunction
may occur at virtually any time point after implantation (such as acute rejection, recurrent glomerular disease, and ureteral obstruction), there is enough difference in the distribution of diseases, and unique conditions occurring only at specific time points (hyperacute rejection, chronic allograft nephropathy), to make this the most informative way to understand the diagnosis and management of allograft dysfunction.
may occur at virtually any time point after implantation (such as acute rejection, recurrent glomerular disease, and ureteral obstruction), there is enough difference in the distribution of diseases, and unique conditions occurring only at specific time points (hyperacute rejection, chronic allograft nephropathy), to make this the most informative way to understand the diagnosis and management of allograft dysfunction.
Following implantation of a living donor kidney, in the absence of technical difficulties with the donor nephrectomy or with implantation or hemodynamic compromise in the recipient, immediate allograft function is expected. Urine output should be brisk (>100 mL/hour) and the fall in serum creatinine rapid (>20% each day). For cadaveric kidneys, poor to no function immediately postimplantation is a relatively common event, occurring in 5% to 35% of patients. Most commonly, as will be discussed below, this is due to DGF (a form of ATN), which can be expected to recover without any specific therapy. However, DGF cannot be assumed to be the cause. The differential diagnosis of nonfunction in the immediate posttransplant period is broad— encompassing disastrous events such as hyperacute rejection, arterial or venous thrombosis, or generally reversible causes such as ureteral obstruction and urine extravasation, as well as DGF—and requires immediate diagnosis, and therapy when indicated (Table 12.1).
Poor allograft function may be manifest as anuria or oliguria with little fall in serum creatinine. A modest diuresis (<1,000 cc/24 hours) with a slow fall in serum creatinine (10%-15%/day) without the need for dialysis-termed slow graft function—is a clinical scenario which should be evaluated with the same concern as oliguria with no fall in serum creatinine, as many of the same problems may manifest as this clinically milder form of posttransplant graft dysfunction (89). In the absence of a brisk diuresis and a sustained, progressive fall in serum creatinine following implantation, immediate diagnostic efforts should be undertaken to investigate the cause. It is important to remember that patients who have maintained a relatively normal urine volume from their native kidneys just prior to transplantation can be expected to maintain a normal, or even elevated, urine volume following implantation. In this setting, urine volume or even a modest decline in serum creatinine may not reflect allograft function. Thus, the patient’s residual urine volume is an important element of the medical history to be considered when evaluating allograft function in the immediate postimplantation period.
TABLE 12.1. Causes of allograft nonfunction immediately posttransplantation | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||
Evaluation
Doppler ultrasound imaging of the transplant is a simple noninvasive test that can easily be done immediately postoperative in the recovery room or intensive care unit, and is the first step in evaluating function immediately after transplantation. Ultrasound can evaluate transplant blood flow characteristics, assess patency of the transplant renal artery and vein, and determine if there is ureteral dilatation, or bladder distension suggesting obstruction to urine flow. In addition, the presence or absence, size, location, and characteristics of any perinephric collections can be determined. Similar although less detailed information may be obtained by radionuclide examination. Further diagnostic and/or therapeutic interventions will depend upon allograft function and the results of initial imaging studies. Analysis of the electrolyte content of wound drainage, if present, or of any perinephric collection may also be of value, particularly in diagnosing urinary extravasation.
Hyperacute Rejection
Hyperacute rejection (HAR) is caused by the presence of undetected antidonor HLA antibodies present in the recipient at the time of transplantation. Once circulation is reestablished to the newly implanted allograft, these antibodies immediately bind to HLA antigens expressed on donor endothelium resulting in endothelial damage and immediate graft thrombosis. This may be noticed by the transplant surgeon following revascularization of the graft by softening or flaccidity of the kidney, or cyanosis or mottling of the graft surface. It may also occur within several hours after implantation and not be evident in the operating room. Pathologically, this is evident as polymorphonuclear leukocyte accumulation in the glomerular and peritubular capillaries, endothelial damage, and diffuse microcirculatory thrombosis (90). Anuria with complete absence of allograft function and absence of blood flow to the graft is invariably the result. No therapy has been found to be effective for hyperacute rejection, and immediate graft nephrectomy is generally indicated. Donor-specific antibody may be detected in a repeat crossmatch at that time or later. The inability to detect donor-specific antibodies should not deter the diagnosis of HAR, as at the time of HAR, donor-specific antibodies may be adsorbed onto the rejected allograft, and may not reappear in the circulation until weeks later. Additionally, non-HLA antigens, such as the endothelial monocyte antigen, may be the target of antidonor antibodies and lead to HAR (91, 92, 93). These are highly unlikely to be detected by standard crossmatch techniques and need to be specifically looked for. There has been an entity called “delayed HAR,” coined to describe sudden severe graft dysfunction or thrombosis
24 to 48 hours after implantation. This poorly characterized syndrome, on occasion reversible, may result from the same mechanism with perhaps a very low titer or low affinity antidonor antibodies, often combined with a component of cellular rejection (94).
24 to 48 hours after implantation. This poorly characterized syndrome, on occasion reversible, may result from the same mechanism with perhaps a very low titer or low affinity antidonor antibodies, often combined with a component of cellular rejection (94).
Vascular Occlusion
Other causes of immediate nonfunction with absence of renal blood flow by ultrasonography or other imaging include renal vein or artery thrombosis. This is reported to occur in up to 2% to 6% of all renal transplants, with over 90% presenting in the first 7 days after transplantation (95,96). A higher incidence was noted with donors at both extremes of age, female donors, and with prolonged total ischemic time. Renal vein thrombosis is almost twice as frequent as renal artery thrombosis, and often presents with sudden pain and swelling of the allograft. Graft rupture may occur and represents a surgical emergency. Angiography, if immediately available, may be of use in distinguishing HAR from arterial thrombosis, since in the former condition, renal arterial patency is generally maintained, although secondary thrombosis of the renal artery or vein can occasionally be seen secondary to stasis of blood flow. The demonstration of thrombosis of the renal artery or vein should lead to immediate surgical reexploration of the patient, as the allograft may be salvaged if arterial or venous thrombosis is recognized shortly after its occurrence and renal circulation is successfully reestablished. However, the window of opportunity for recovery of function is small, as the nonperfused allograft at body temperature (warm ischemia) will suffer irreversible injury within a brief period of time. A delay of several hours to obtain further imaging of a kidney that has no flow by ultrasonography is rarely justified if salvage of renal function is anticipated. Transplant renal artery thrombosis may be due to one of several causes: endothelial damage, dissection of the artery, kinking due to a shift in position of the allograft, or, less likely, extrinsic compression due to hematoma or lymphocele. Transplant renal vein thrombosis may similarly be due to a variety of causes including damage to the vein during preservation or implantation, external compression, or kinking due to a shift in allograft position. A thrombophilic state in the recipient may also be a contributing factor in some cases (97).
Pathologically, the characteristic microcirculatory thrombosis of HAR may extend into larger-sized arteries and veins, including the main renal vein and artery if there is renal infarction. The histology of primary renal vein thrombosis is characterized by hemorrhagic infarction of the allograft with thrombosis of the venous microcirculation, and may be distinguished from HAR by the absence of microthrombosis on the arteriolar side of the circulation. Primary thrombosis of the transplant renal artery usually results in bland renal infarction.
Another clinical entity that may present with poor allograft perfusion and function in the immediate postimplantation period is severe preservation injury. This uncommon entity occurs following complications of organ recovery, preservation, or implantation, and may result from excessive warm or cold ischemic times. Pathologically, there are findings of severe ATN, on occasion accompanied by evidence of endothelial injury and thrombotic changes in the glomerular microcirculation. There is a high primary nonfunction rate in these allografts, particularly if there has been vascular injury, and those that do recover function are most often significantly compromised. There is no defined treatment to prevent or minimize irreversible injury or to promote more rapid recovery.
Delayed Graft Function
DGF is defined as the need for dialysis during the first week after transplantation, while slow graft function describes patients who do not have an adequate decline in creatinine or urine output immediately after transplantation (i.e., <30% decline in serum creatinine within 48 hours or <1,000 mL urine output/day), but do not require dialysis. The pathogenesis of both these situations is thought to be similar, different only in degree, caused by ischemia/reperfusion injury resulting in tubular damage (89). Pathologically, ATN is seen, postulated to be a result of tubular cell damage due to ischemia, free radical formation during reperfusion, activation of endothelial cells which increases expression of MHC class II and adhesion molecules, leukocyte diapedesis, increased cytokine expression, activation of mitochondrial apoptotic pathways, and damage due to cold storage (98, 99, 100, 101, 102, 103). The reported incidence of DGF is quite variable, ranging from 5% to 40% for cadaveric transplants compared to 2% to 5% for living related donors (104). Risk factors for DGF include increased cold ischemia time, use of an expanded criteria donor (ECD) kidney, elevated donor creatinine at time of organ harvesting, donor hypertension, kidney from a non-heartbeating donor, recipient panel of reactive antibodies (PRA) >50%, recipient race, perioperative blood pressure, degree of HLA mismatch, recipient of prior transplantation, and method of organ preservation (104, 105, 106, 107, 108). The higher incidence of DGF in patients with high PRA suggests that immunologic factors also play an etiologic role in some cases (105,109,110). Any evidence of clear-cut immune attack contributing to or masquerading as DGF should be treated aggressively (see below). The mean duration of DGF varies between studies, averaging 7 to 10 days (111,112). DGF is associated with worse long-term graft outcome, an effect that appears to be stronger the longer the duration of graft dysfunction (104,113, 114, 115). Results between studies have varied, but this effect is likely independent of acute rejection, leading to a graft half-life of 9.7 years compared to 14.2 years in those patients without DGF (114,116).
DGF blinds the clinician to what may be occurring immunologically in the kidney. Since there is oliguria and an elevated serum creatinine whose level in many patients is
determined by dialysis, clinically detecting a superimposed acute rejection becomes difficult without a biopsy. The incidence of acute rejection in patients with DGF that persists for 7 to 10 days is significant, varying from 18% to 48% (117, 118, 119). The presence of acute rejection in biopsies of patients with DGF was associated with increased primary nonfunction and worse long-term graft survival (117,119). A transplant biopsy should be performed in patients with persistent DGF (>7days), since early detection and treatment of superimposed rejection is likely to improve graft function and survival, although this has not been formally studied. Repeat imaging of the allograft should also be performed at least once during the first week of DGF, and weekly during prolonged DGF to assure that graft thrombosis has not occurred, and that there is no obstruction to urine flow in the kidney recovering from DGF.
determined by dialysis, clinically detecting a superimposed acute rejection becomes difficult without a biopsy. The incidence of acute rejection in patients with DGF that persists for 7 to 10 days is significant, varying from 18% to 48% (117, 118, 119). The presence of acute rejection in biopsies of patients with DGF was associated with increased primary nonfunction and worse long-term graft survival (117,119). A transplant biopsy should be performed in patients with persistent DGF (>7days), since early detection and treatment of superimposed rejection is likely to improve graft function and survival, although this has not been formally studied. Repeat imaging of the allograft should also be performed at least once during the first week of DGF, and weekly during prolonged DGF to assure that graft thrombosis has not occurred, and that there is no obstruction to urine flow in the kidney recovering from DGF.
Strategies to limit DGF include improved organ preservation, shorter cold and warm ischemia times, and improved care of brain-dead donors. Intraoperative administration of a polyclonal antilymphocyte globulin (Thymoglobulin) may decrease the incidence of DGF. In one study, 58 patients were randomized to receive Thymoglobulin intraoperative or 6 hours after reperfusion (120). The incidence of DGF was 14.8% in the intraoperative group compared to 35.5% in the postprocedure group (P <0.05). Further, the intraoperative group had a decreased length of stay and lower serum creatinine at discharge. The long-term clinical benefit is unclear since renal function at 1 year was similar between groups. Studies to prevent DGF or to improve recovery from DGF using anti-ICAM1 antibodies have not shown any benefit, and the growth factor IGF-1 is still being studied (121,122).
There is no specific treatment for DGF to accelerate recovery of renal function. Management of immunosuppressive medications during DGF has been the subject of numerous investigations. However, no widely agreed upon strategy has been established. The goal is to balance adequate immunosuppression to prevent acute rejection against limited exposure to medications with a potential adverse impact on renal recovery. Most strategies have attempted to minimize exposure to calcineurin inhibitors (CIs) during DGF due to their nephrotoxic effects—renal vasoconstriction and direct tubular toxicity—while preventing rejection with the use of antilymphocyte antibody therapy (antithymocyte globulin, OKT3, or anti-IL2 receptor antibodies). In these treatment strategies, calcineurin inhibitor administration is often delayed until there is adequate recovery of renal function (often defined as serum creatinine <3 mg/dL). Other strategies have included antibody therapy with low-dose cyclosporine/tacrolimus, or sirolimus or mycophenolate mofetil (MMF) (or both) until renal function recovers (123, 124, 125, 126, 127). Some advocate immediate administration of full-dose calcineurin inhibitors regardless of renal function (128).
Early studies comparing antithymocyte globulin or OKT3 “prophylaxis” to immediate cyclosporine 10 mg/kg/day administration in patients with DGF suggested that cyclosporine use during DGF prolonged recovery (127,129,130). However, low blood levels of cyclosporine immediately after transplantation are associated with an increased risk of acute rejection (131). More recent studies have compared the time to recovery of renal function from DGF using two different immunosuppressant strategies: (a) administration of antithymocyte globulin while withholding calcineurin inhibitors and (b) immediate administration of CIs without antithymocyte globulin. The duration of DGF in patients treated immediately with CIs (cyclosporine 5 to 8 mg/kg/day or tacrolimus 0.15 to 0.3 mg/kg/day) compared to those treated initially with antithymocyte globulin was longer in one study but not in others (125,132, 133, 134). Sequential therapy using antithymocyte globulin or OKT3 until renal recovery followed by initiation of CI is safe and effective but may be associated with an increased risk of infection (135). Anti-IL-2 receptor antibodies, in conjunction with cyclosporine and azathioprine or mycophenolate mofetil, have also been used successfully to decrease the incidence of acute rejection in patients with DGF (136,137).
Sirolimus is an attractive alternative to immediate CI use in patients with DGF: it is an effective immunosuppressive without vasoconstrictive effects or inherent nephrotoxicity. However, its potent antiproliferative activity may prolong recovery from DGF. In animal models, sirolimus use impairs resolution of ATN, and its use has been associated with prolongation of DGF (111,112,138,139). Prolonged recovery from DGF did not affect renal function at 1 year (112).
Mycophenolate mofetil has not been associated with significant nephrotoxicity and does not inhibit the proliferation of renal tubule epithelial cells in vitro. It appears safe to administer it during episodes of DGF.
No single immunosuppressive strategy is clearly superior to another during DGF. Careful monitoring of allograft status is essential, with those patients who have not recovered function within 7 to 10 days undergoing allograft biopsy.
Urologic Complications
Urologic complications—ureteral obstruction and urine extravasation—are another cause of poor allograft function in the immediate postoperative period.
Transplant ureteral obstruction in the immediate postoperative period may represent a technical problem such as ureteral necrosis due to inadequate vascular supply, or poor surgical technique in creating the uretero-vesicular anastomosis. External factors such as compression of the ureter by a hematoma or kinking of the ureter due to poor positioning of the graft are also potential causes of obstruction to urine flow. If there is good initial function but ureteral obstruction, hydronephrosis will develop quickly. However, if DGF is present, a urologic abnormality may not be evident until recovery of function, as urine volumes recover. Periodic imaging of the transplanted kidney and ureter is essential to ensure that the allograft has not begun to recover function, but oliguria continues due to obstruction.
Damage to the transplanted ureter or faulty ureteroneocystostomy can also result in extravasation of urine out of the bladder or ureter. This urine will most commonly either collect in the perinephric area or drain via the fresh transplant incision. Less often, it may enter the peritoneal cavity or track into the scrotum or labia. Diagnosis of a urine leak can be made by analysis of fluid collected from wound drainage or aspirated from a collection. Simultaneous measurement of the sodium, potassium, and creatinine concentration should easily distinguish between urine leaking into the tissue and normal tissue fluid that might collect in a lymphocele or drain from the wound. Normal extracellular tissue fluid should have a creatinine, sodium, and potassium concentration nearly identical to plasma, whereas urinary potassium and creatinine concentrations will be much higher than that in the serum, and the sodium concentration lower. Radionuclide imaging may also be of value in the diagnosis of urinary extravasation. Radiotracer exiting the urinary system into the soft tissues strongly suggests a leak. This type of test may, however, be of limited value if allograft function is poor and background counts are high. Confirmation of the presence of and the site of a leak can be obtained by cystoscopy or following placement of a nephrostomy tube. A small volume of urine output per urethra does not exclude the diagnosis of ureteral obstruction or urinary leak. Urinary retention from bladder outlet obstruction may also lead to oliguria with no fall in serum creatinine. This occurs most commonly in men at risk from prostatic hypertrophy, or in diabetics with neurogenic bladder dysfunction.
The management of ureteral obstruction generally requires the placement of a ureteral stent to allow for decompression of the urinary system. If a stent bypassing the obstruction cannot be placed, a nephrostomy tube may be necessary. Alternatively, reoperation and reimplantation of the ureter may be indicated if the obstruction is due to obstruction at the ureteroneocystostomy site.
The management of urinary extravasation is directed toward improved drainage of urine into the bladder via a ureteral stent, or external drainage via nephrostomy tube placement. Either will reduce the hydrostatic pressure within the urinary system and encourage healing of the leak. If the ureter has completely dehisced from the bladder, reimplantation may be indicated. Mild urinary extravasation with a fundamentally sound ureteroneocystostomy may be managed with bladder drainage. More significant leaks usually require reoperation, either immediate or after controlling the leak with a nephrostomy/stent combination.
Catheter obstruction should always be considered in evaluating anuria or oliguria following transplantation. This should be readily evident if imaging reveals bladder outlet obstruction. The Foley catheter may be kinked or obstructed by a blood clot. Flushing the catheter with sterile saline should either relieve the obstruction or indicate the need to replace the catheter. Measures to relieve bladder outlet obstruction should be undertaken immediately to prevent disruption of the uretero-vesicular anastomosis due to elevated intraluminal pressure within the bladder.
Hypovolemia
Hypovolemia is another cause of poor allograft function immediately following transplantation. This is most commonly due to over zealous dialysis immediately prior to surgery, coupled with inadequate replacement of fluid losses in the operating room. Third spacing of fluid and postoperative bleeding can also cause intravascular volume depletion, and hemoglobin/hematocrit should always be followed closely after surgery.
DETERIORATION OF ALLOGRAFT FUNCTION EARLY AFTER TRANSPLANTATION
Once baseline allograft function has been established, careful monitoring is essential for optimizing chances for long-term success. The determination of what is an acceptable baseline function for the allograft depends somewhat on donor-recipient characteristics. In general, a serum creatinine <1.5 mg/dL should be expected. Recipients of kidneys from older donors, cadaveric donors, extended criteria donor kidneys, and where there is a substantial size mismatch between donor and recipient (donor small, recipient large) or in cases where there are technical difficulties with implantation would be expected to have higher serum creatinine at baseline posttransplantation. The absolute nadir creatinine may not be the true “baseline.” The creatinine may reach a nadir value in the first week after transplantation, and then stabilize at a slightly higher level once therapeutic blood levels of calcineurin inhibitor are attained. The American Society of Transplantation guidelines provide a generally agreed upon schedule for follow-up monitoring in this critical early period posttransplantation (4).
Any deterioration of allograft function, particularly in the first 3 to 6 months after transplantation, warrants immediate investigation. As noted above, Hariharan et al (140) have demonstrated that absolute graft function and the degree of change in graft function in the first year after transplantation are strong and independent predictors of long-term success. The occurrence of any acute rejection has a negative impact on long-term success rates (140). However, the severity of the acute rejection episode and the degree of reversal in response to antirejection therapy are the main determinants of the long-term effects of acute rejection. Rejection episodes that are completely reversed in response to treatment and are mild are likely to have little impact on the subsequent development of chronic allograft nephropathy (CAN). Early diagnosis and treatment of acute rejection are essential for optimizing the chances to return allograft function to its previous baseline and to achieve long-term success. If the serum creatinine does not return to baseline in response to antirejection treatment, or if subclinical rejection goes unrecognized and untreated,
irreversible damage is very likely to occur, compromising long-term outcome (140, 141, 142, 143, 144, 145).
irreversible damage is very likely to occur, compromising long-term outcome (140, 141, 142, 143, 144, 145).
As noted above, serum creatinine values are an insensitive marker of changes in GFR. Diligence in the pursuit of accurate diagnosis and effective treatment of elevations in serum creatinine cannot be overemphasized.
Diagnosis
A rise in serum creatinine, or a fall in GFR by other measurements, should lead to a prompt review of the patient’s clinical status, including physical examination, review of medications, interval medical history, measurement of immunosuppressive drug levels, urinalysis, and complete blood count. Standard measures should be initiated for the diagnosis and treatment of such causes of acute renal dysfunction as volume depletion, urinary tract or other infection, congestive heart failure, or overtreatment of blood pressure. Measurement of calcineurin inhibitor blood level is essential, as acute nephrotoxicity due to cyclosporine or tacrolimus is one of the most common causes of acute allograft dysfunction. Review of all medications for other nephrotoxins—non-steroidal antiinflammatory drugs (NSAIDs), aminoglycosides, amphotericin—or other drugs potentially affecting renal function and/or serum creatinine level—trimethoprim/Sulfamethoxazole, Angiotensin converting enzyme unhibitors, Angiotensin 2 Receptor Blockers, and/or diuretics should be undertaken. Imaging of the allograft by either ultrasound or radionuclide technique should be a routine part of the evaluation to determine whether there is any obstruction to urine flow. Routine complete blood count should be done to evaluate the possibility of thrombotic microangiopathy—sudden anemia, thrombocytopenia, increased serum lactated enydrogenese (LDH), or decreased/absent haptoglobin are all suggestive findings (Table 12.2)
If a diagnosis of the cause of allograft dysfunction is not evident from these tests and/or a return to baseline function is not achieved with treatment of the presumed cause, allograft biopsy should be performed to determine whether acute rejection is present. Empiric treatment of presumed acute rejection without a tissue diagnosis is generally not recommended. Several groups have observed that allograft biopsy findings resulted in a change in management in 25% to 40% of patients in whom a presumptive diagnosis was made by experienced transplant physicians and surgeons based on the clinical scenario (146,147).
TABLE 12.2. Causes of acute allograft dysfunction early (first 3 to 6 months) posttransplantation | ||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||
Acute Rejection
Acute rejection remains a significant, albeit much reduced, problem with modern immunosuppressive regimens. Whereas 10 to 15 years ago, most centers reported a 30% to 50% incidence of acute rejection, the introduction of more potent immunosuppressants, and the use of antilymphocyte antibody induction have substantially reduced the reported rate of acute rejection. Registry data reviews indicate an overall average of 14.2% acute rejection rate within the first 3 to 6 months following primary renal transplantation for those receiving allografts in 1996 to 1997 (148). This rate is less than half of what was reported for 1988 to 1989 (31.4%). Mange et al (149) reported a 12% to 14% acute rejection rate in the initial 3 months in recipients of living donor kidneys transplanted between 1994 and 1997 (149). Decades of clinical experience have consistently demonstrated that approximately 75% of these acute rejection episodes occur in first 3 months after transplantation. Recent clinical trials support these findings, with clinical acute rejection rates of <20% for both cadaveric and living donor kidney recipients (150,151). At least half of these are histologically mild—Banff grade I or IIA.
In certain specific populations of patients, however, there may be a higher incidence of acute rejection. Patients who have received ABO incompatible transplants, or transplants from donors against whom there was a previously positive crossmatch, higher rates of acute antibody-mediated rejection can be expected (152,153). African Americans have also been considered to be at higher risk for acute rejection, although the careful use of more modern immunosuppressive agents appears to be able to reduce this risk (151,154,155). The timing of acute rejection is also of some importance. Rejections occurring more than 3 months after transplantation are generally associated with a worse long-term prognosis (156) (see below).
The clinical diagnosis of acute rejection requires confirmation with allograft histology. The clinical presentation of acute rejection is extremely variable. There is no clinical syndrome that automatically establishes a diagnosis of acute rejection or reliably indicates the type of rejection. As noted above, even in the most experienced centers, a renal biopsy is required to definitely establish the diagnosis of acute rejection. Furthermore, differentiating between acute cellular rejection and acute rejections that are mediated in part or entirely by antidonor antibodies is essential for directing appropriate therapy, and can only be accomplished by immunostaining of allograft tissue.
Most acute rejections are clinically mild, presenting with a modest rise in serum creatinine in an asymptomatic patient.
The deterioration in allograft function may not be progressive in the short term: the serum creatinine may increase above baseline and may remain relatively stable at a new, elevated level. The Efficacy Endpoints Conference on Acute Rejection proposed a definition of acute rejection that included an increase in serum creatinine 0.4 mg/dL. It is clear, however, that more modest elevations in serum creatinine may indicate the presence of acute rejection. In fact, recent studies involving “protocol” biopsies (allograft biopsies performed on a predetermined schedule, even when there is no indication of allograft dysfunction) have revealed evidence of apparent antigraft immune reactivity. The clinical significance of the lymphocytic infiltrates found in the biopsy specimens obtained from clinically stable grafts remains an area of some controversy (157). Alternatively, acute rejection may present as a rapid and dramatic deterioration in allograft function leading to oligo-anuria (94).
The deterioration in allograft function may not be progressive in the short term: the serum creatinine may increase above baseline and may remain relatively stable at a new, elevated level. The Efficacy Endpoints Conference on Acute Rejection proposed a definition of acute rejection that included an increase in serum creatinine 0.4 mg/dL. It is clear, however, that more modest elevations in serum creatinine may indicate the presence of acute rejection. In fact, recent studies involving “protocol” biopsies (allograft biopsies performed on a predetermined schedule, even when there is no indication of allograft dysfunction) have revealed evidence of apparent antigraft immune reactivity. The clinical significance of the lymphocytic infiltrates found in the biopsy specimens obtained from clinically stable grafts remains an area of some controversy (157). Alternatively, acute rejection may present as a rapid and dramatic deterioration in allograft function leading to oligo-anuria (94).
Criteria for the histologic diagnosis of acute rejection have been the subject of several international consensus conferences, with the most widely accepted definitions detailed in the Banff 1993 and more recently 1997 criteria (90) (Table 12.3). A second scoring system for acute rejection was developed in 1997 by the Cooperative Clinical Trials in Transplantation Group in association with clinical trials sponsored by the National Institutes of Health. This differs from the Banff criteria mainly in the inclusion of milder forms of inflammation (classified by Banff as “borderline” or “suspicious”) in the type I rejection category (158). The Banff classification of allograft pathology includes three grades of “acute/active rejection” and three grades of “chronic/sclerosing allograft nephropathy.” There is also a category for “borderline changes/suspicious for acute rejection,” and a recent modification to describe findings that establish a diagnosis of antibody-mediated rejection (detailed below).
In addition, the criteria for what is considered to be an adequate sample of allograft tissue was also articulated in the Banff 1997 classification. This is defined as one that contains at least 10 glomeruli and two arteries. Two separate cores should be obtained to minimize sampling error, with tissue analyzed by light microscopy and by immunofluorescence.
Acute Cellular Rejection
For many years, acute rejection was defined by T-lymphocyte attack on the allograft, as manifest most commonly by tubulitis, and less frequently by inflammatory changes in the blood vessels (“vascular rejection”—Banff class II and III). Cellular rejection is still by far the most common form of acute rejection. Most acute cellular rejections are clinically mild, without fever, graft tenderness or swelling, or change in urine volume. Common symptoms, if any, include fluid retention, elevated blood pressure, and/or a sense of malaise. These signs and symptoms, however, are nonspecific, and may result from many other causes of allograft dysfunction (Table 12.2).
The treatment of acute cellular rejection should be determined primarily by the histologic severity of the rejection process seen on biopsy, and to a lesser extent by the clinical setting in which the episode is occurring. The Banff classification appears to provide a useful framework for the initial treatment of acute rejection (159, 160, 161). The treatment of acute cellular rejection has been the subject of few recent multicenter trials. The data detailing the modality and efficacy of the treatment of acute rejection in transplant recipients receiving modern immunosuppressive regimens derives largely from clinical trials of MMF, tacrolimus, and sirolimus which focus on the prevention of rejection, rather than its treatment. Registry data and single center reports also provide helpful information. The treatment options include high-dose oral/intravenous corticosteroids, polyclonal antilymphocyte antibodies (rabbit, horse), the monoclonal antibody muromonab-OKT3. Alemtizumab (Campath-1H®), a humanized anti-CD52 monoclonal antibody, has shown efficacy in preliminary studies in a limited number of cases (162).
TABLE 12.3. Banff ’97 diagnostic categories for renal allograft biopsies | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Initial treatment for mild (Banff class IA, IB) acute rejection episodes is usually pulse methylprednisolone 250-100 mg/day for 3 to 5 days. There is no data to suggest that higher doses of bolus corticosteroids are superior to lower doses, or even that intravenous dosing is superior to oral dosing. The rate of successful reversal of mild acute rejection with steroid therapy—clinically judged—is reported to be 80% to 90%. With high-dose corticosteroid therapy, allograft function should begin to improve within 72 to 96 hours (163). Various investigators have noted that the time course of the change in serum creatinine concentration does not clearly differentiate steroid responsive rejections from steroid nonresponsive rejections until 3 to 5 days after the initiation of steroid therapy (164, 165, 166). If there has been no decline in serum creatinine by that time, in the absence of any other factors contributing to allograft dysfunction, steroid therapy should be judged ineffective, and an alternative therapy and/or repeat renal biopsy should strongly be considered. The determination that the treatment of acute rejection has been successful is generally based on the return of serum creatinine to its previous baseline value. Follow-up biopsies after clinically successful rejection therapy have revealed the continued presence of lymphocytic infiltrates (167,168). The significance of these is uncertain—they may represent residual/resolving rejection, the presence of regulatory T cells, or ongoing graft attack due to inadequate treatment. There are no reliable markers to distinguish among these possibilities.
An initial diagnosis of acute rejection Banff class IIA or higher (i.e., with “vascular” involvement—arteritis), or the failure of high-dose corticosteroid therapy should prompt the use of an antilymphocyte antibody treatment, either polyclonal or monoclonal. Data from the Efficacy Endpoints Conference indicated that only 42% of rejections grade 3 by the Banff 1993 criteria responded to steroid therapy. The currently available reagents are antithymocyte globulin (rabbit)—Thymoglobulin®; antithymocyte globulin (equine)—Atgam®; and muromonab/CD3—Orthoclone OKT3®. Alemtuzumab (Campath-1H®); a humanized anti-CD52 monoclonal antibody and currently a focus of tolerance induction studies, has also been used with some success in the treatment of acute rejection (162). Studies comparing one antilymphocyte antibody preparation to another in the treatment of acute rejection have generally been inconclusive. Gaber et al (169) demonstrated that Thymoglobulin had a higher rejection reversal rate and a lower incidence of recurrent rejection when compared to Atgam: 88% rejection reversal vs. 76%, and 17% vs. 36%, respectively (169).
OKT3 was the first monoclonal antibody approved for clinical use. Initial studies in 1985 established its superiority over high-dose corticosteroids in the treatment of acute rejection, with 94% rejection reversal rate versus 75% for corticosteroids. However, patients were not stratified for histologic grading of rejection in that study. Rejection reversal rates of 86% to 98% were observed for OKT3 in later reports that included patients with high-grade rejections (170, 171, 172). OKT3 has also proven effective for “rescue” therapy—treatment of steroid nonresponsive acute rejections, with a success rate ranging from 50% to 96% (163). Side effects commonly seen with the first dose (“cytokine release syndrome”) and the development of antimurine antibodies have limited its use as a first-line agent. Comparisons between monoclonal and polyclonal treatments of acute rejection have not consistently indicated superiority of either therapy (170,173, 174, 175).
Treatment of recurrent rejection episodes with repeated doses of either monoclonal or polyclonal antilymphocyte agents is problematic, as patients may make immune responses against the antilymphocyte antibody, producing antibodies that neutralize the therapy. This has been best documented with OKT3, but can also occur with polyclonal preparations (176,177). Multiple courses of antirejection treatment in close succession, or prolonged courses of antibody therapy greatly increase the risk of serious infection or lymphoproliferative disease, and should be undertaken with great caution.
Another important consideration in the overall approach to the treatment of acute rejection includes modification of the maintenance immunosuppressive regimen. If acute rejection has developed on a given maintenance immunosuppressive regimen, a modification of that regimen may be desirable to prevent recurrent rejection episodes. This may include any or all of the following measures: increased dosages of the same medications used at the time of rejection, addition of another immunosuppressant, substitution of a different immunosuppressant for one in the current regimen. Data suggest that the addition of MMF (CellCept) may be helpful if this is not already part of the maintenance regimen (178,179). Switching calcineurin inhibitor therapy from cyclosporine to tacrolimus has also been shown to be successful (180,181). Similarly, switching to a cyclosporine-based regimen may be helpful for tacrolimus-treated patients with acute rejection (151). Reinstitution of corticosteroids may be indicated for those patients experiencing acute rejection following steroid discontinuation.
Sporadic reports have described successful rescue therapy in patients failing more standard antirejection treatments using modalities not designed for use in patients experiencing acute cellular rejection, such as intravenous immunoglobulin G (IVIg) and anti-CD25 monoclonal antibodies (182, 183, 184).
Acute Antibody-mediated Rejection
It has long been recognized that not all acute rejection involves lymphocyte infiltration and tubulitis (“cellular rejection”). Hyperacute rejection (see above) is well known to be caused by antidonor HLA antibodies present in the serum of the recipient at the time of transplantation. Other forms of antidonor antibody-mediated rejection have been the focus of intense interest and there have been recent significant advances in diagnosis and treatment. It is now recognized that
antibody-mediated rejection can also present clinically in a less devastating fashion, as acute rejection or even as chronic rejection. Indeed, 37% of patient with steroid-resistant acute rejection were found to have evidence of antibody-mediated rejection (185). With the growing interest in new regimens to transplant ABO incompatible donor-recipient pairs, and protocols to desensitize transplant recipients who have a positive crossmatch against their prospective donor, and new diagnostic tools to recognize anti-body-mediated rejection, the diagnosis and treatment of anti-body-mediated acute rejection has assumed increased importance.
antibody-mediated rejection can also present clinically in a less devastating fashion, as acute rejection or even as chronic rejection. Indeed, 37% of patient with steroid-resistant acute rejection were found to have evidence of antibody-mediated rejection (185). With the growing interest in new regimens to transplant ABO incompatible donor-recipient pairs, and protocols to desensitize transplant recipients who have a positive crossmatch against their prospective donor, and new diagnostic tools to recognize anti-body-mediated rejection, the diagnosis and treatment of anti-body-mediated acute rejection has assumed increased importance.
The discovery that the deposition of a complement breakdown product C4d in peritubular capillaries correlates strongly with rejection mediated by antidonor antibodies has given new life and meaning to the long held, but poorly elucidated, distinction between “cellular” and “humoral” rejection. Antibody-mediated rejection criteria have now been adopted as a modification of the Banff 1997 Classification of Renal Allograft Rejection (Table 12.4) (73). The original Banff 1997 diagnostic categories for renal allograft biopsies list only “immediate” (hyperacute) and “delayed” (accelerated acute) as the forms of antibody-mediated rejection. Criteria for the diagnosis of acute antibody-mediated rejection (AMR, also called acute humoral rejection [AHR]) in renal allografts now include three features in addition to graft dysfunction: (a) histologic evidence of tissue injury; (b) immunopathologic evidence for antibody action—C4d staining in peritubular capillaries or immunoglobulin and complement in arterial fibrinoid necrosis; and (c) serologic evidence of circulating antibodies to donor HLA or endothelial antigens (73,90). The pathology of antibody-mediated rejection includes acute tubular injury, and/or neutrophil/monocyte margination in peritubular capillaries and/or glomeruli, and/or capillary thrombosis, and/or vascular injury (intimal arteritis or fibrinoid necrosis or intramural or transmural inflammation in arteries). Tubulitis is not a feature of AMR, but may be present if there is concomitant cellular rejection. Mauiyyedi (85) reported that widespread C4d staining was present in 30% of all acute rejection biopsies, with 30% of these demonstrating only AMR, 45% a combination of AMR and cellular rejection, 15% acute cellular rejection alone, and acute tubular injury in 11% (85). A minority of C4d-positive biopsies shows “pure” AMR (186).
TABLE 12.4. An addition to Banff ’97 classification of renal allograft rejection | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||
The clinical features of AMR are nonspecific: acute deterioration of allograft function occurs, varying from mild dysfunction to complete cessation of graft function with oligo-anuria. Antidonor-HLA antibody is detected in 56% to 90% of patients with C4d+ acute rejection. There are several techniques for measuring donor-specific antibody titers, each with differing sensitivity and specificity. The significance of clinical and histologic findings compatible with AMR, but no detectable donor-specific antibody, is unknown. There are preliminary reports suggesting that subclinical AMR may also occur (187). As is the case with cellular (C4d-negative) acute rejection episodes, most AMR occur within the first month after transplantation (188).
Risk factors for AMR include elevated PRA, prior transplantation, historically positive crossmatch, and female sex. No correlation with HLA match, ischemic time, or donor characteristics has been found (185,189, 190, 191).
Immunostaining biopsies for C4d in order to determine whether antibody-mediated rejection is present is of critical importance in the diagnosis of allograft dysfunction early after transplantation, as the treatments of cellular- and anti-body-mediated rejection are quite different (see below). In addition, AMR may have scant findings, possibly only tubular injury, easily mistaken for ATN, if C4d staining is not performed. Assessment of the presence of detectable antidonor antibody should also be done in cases where biopsy specimens suggest antibody-mediated rejection.
Treatment of Antibody-mediated Rejection
It is generally recognized that the treatments outlined above for cellular rejection are of limited efficacy as primary treatment of AMR. Antibody-mediated acute rejections are highly unlikely to respond to high-dose corticosteroid therapy. Graft loss at 1 year is greater than with cellular rejection: 30% overall vs. 4% (191, 192, 193, 194). The goal in attempting to reverse AMR is to eliminate the donor-specific antibody and inhibit its resynthesis. This has successfully been accomplished with a variety of different regimens, including combinations of high-dose tacrolimus and MMF, immunoadsorption, plasmapheresis and intravenous immunoglobulin administration, and antilymphocyte antibody treatment. Nickeleit at al (195) reported successful reversal of C4d+ acute rejection employing aggressive treatment with antilymphocyte antibodies (195). Crespo et al (185) reported success with plasmapheresis, and “tacrolimus/mycophenolate rescue.” They documented a fall in donor-specific antibody titer associated with resolution of acute rejection in 9 of 10 patients utilizing this regimen (185). Immunoadsorption with a protein A column accompanied by antithymocyte treatment has also been effective (196). More recently, good results have consistently been achieved
by several groups incorporating the use of intravenous immunoglobulin treatment along with plasmapheresis (152,153,197,198). A prolonged course may be necessary to reverse severe AMR. The titer of donor-specific antibody, if present, should be monitored during therapy. Rituximab (anti-CD20 monoclonal antibody) may also have a role, as its anti-B-cell effects appear to contribute to reducing antidonor antibody production (199,200). Delineation of the full spectrum of the histologic manifestations of antidonor antibody-mediated rejection, the optimal techniques for easy detection and characterization of these antibodies, as well as the determination of the most effective therapy are all rapidly developing areas which are only at the initial phases of understanding.
by several groups incorporating the use of intravenous immunoglobulin treatment along with plasmapheresis (152,153,197,198). A prolonged course may be necessary to reverse severe AMR. The titer of donor-specific antibody, if present, should be monitored during therapy. Rituximab (anti-CD20 monoclonal antibody) may also have a role, as its anti-B-cell effects appear to contribute to reducing antidonor antibody production (199,200). Delineation of the full spectrum of the histologic manifestations of antidonor antibody-mediated rejection, the optimal techniques for easy detection and characterization of these antibodies, as well as the determination of the most effective therapy are all rapidly developing areas which are only at the initial phases of understanding.
Related posts:
Detection of Recipient Pretransplant Alloreactivity
Surgical Issues in the Transplant Recipient
Pregnancy and Gender Issues in the Renal Transplant Recipient
Posttransplantation Liver Disease
Cytomegalovirus in Renal Transplantation
Medical Complications of the Eyes, Nasopharynx, Dentition, Oropharynx, and Hearing in the Kidney Transplant Recipient
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
