Prognostic Features of Early Renal Transplant Rejections

Prognostic Features of Early Renal Transplant Rejections William J. C. Amend, Jr., M.D., Flavia Vincenti, M.D., Robert Duca, B.A., Nicholas J. Feduska, M.D., Kent Cochrum, D.V.M., Donald Potter, M.D., and Oscar Salvatierra, Jr., M.D. In an analysis of 632 cadaver transplants , the early renal transplant course gave important prognostic information depending on temporal and/or renofunctional characteristics of rejection episodes . Two transplant rejections occurring within the first 2 mo posttransplant were associated with either 37%, 27%, or 6% l-yr graft survivals depending on whether these episodes were separate, temporally back-to-back, or were without interrejection renofunctional recovery, respectively . This compares to l-yr graft survivals of 89% or 73%

I

N THE early investigations of human renal transplantation, much emphasis was directed towards defining clinical and immunologic parameters of early acute rejection. 1 These pioneering efforts to discriminate and recognize acute renal transplant rejection had as their basis the underlying assumption that: "Prompt institution of therapy may prevent some irreversible damage to the transplanted organ. "1 The proper, as well as prompt, use of the currently available, nonspecific posttransplant immunosuppressive therapy is crucial to optimal graft survival, while simultaneously allowing for optimal patient welfare. It has long been recognized that in the early cadaver-donated posttransplant course, rejection episodes are the rule rather than the exception. Within this group of patients who experience a transplant rejection, there are three potential courses: (1) a single rejection with immediate allograft loss; (2) a single rejection with complete or partial recovery without subsequent rejection; and (3) temporary allograft recovery with a second and/or a third rejection episode shortly thereafter. There are many unresolved questions with this third course - multiple early rejections. Williams and associates 2 pointed out that most patients in this third category had impairment of renal function . These patients usually went on to complete and early allograft failure despite repeated immunosuppressive adjustments. Pathologic changes of obliterative vascular disease were usually present in these patients . These vascular alterations are not felt to be modified by further increases in immunosuppression . 3 Also, Opelz

in those patients who had no rejection or one with recovery early posttransplant. Patient survival in groups with multiple early rejections was also associated with a poor prognosis. Ninety to 93% l-yr patient survival was noted when there was no or one rejection. There was only a 74%-83% l-yr patient survival with two early treated rejections. Transplant rejection therapy must be individualized or even withheld in order to ensure optimum graft and patient survival.

et al. 4 found highly significant correlations between the early transplant course and ultimate graft behavior: early rejection episodes were associated with poor clinical outcomes. Hamburger, 5 on the other hand, in a study of rejection episC!des in the first 2 wk posttransplant, concluded that there was no correlation between the number of early rejection crises and any long-term result. Many of these transplants, however , were from related donors. In an attempt to improve patient mortality and morbidity following cadaver renal transplantation, we have been interested in delineating any clinically recognizable features of a rejection episode that might give helpful prognostic information. MATERIALS AND METHODS Six-hundred and thirty-two cadaver transplants at UCSF, from November 1970 through April 1978, were evaluated for the occurrence of multiple early transplant rejections. Six patients with an early technical loss were excluded. Onehundred and ninety-two patients had at least two rejection episodes diagnosed in the first 60 days after renal transplant. This included 178 recipients of primary and 14 recipients of secondary cadaver transplants. Procurement, storage, tissue typing. and donor-recipient selection at UCSF have been previously described. 6 All consecutive cadaver transplants were uniformly treated, with the sole modification of a lowered glucocorticoid regimen

From the Kidney Transplant Service, University of Cali· fornia, San Francisco, Calif. Presented in part at the American Society of Nephrology, Boston , Mass. , November 1979. Reprilll requests should be addressed to William J. C. Amend, Jr .. M.D., Kidney Transplant Service , 884 Moffitt Hospital , San Francisco. Calif. 94143. © 1982 by The National Kidney Foundation , Inc. 0272-6386/82/050281 -07$01 .00/0

American Journal of Kidney Diseases, Vol. I, No.5 (March), 1982

281

282

AMEND JR. ET AL.

begun in August 1972. 7 Transfusion data had been previously obtained as part of a larger study. 6 These two early posttransplant rejection episodes were further subdivided into three groups: (1) two separate rejections, (2) camelback (time), and (3) camelback (creatinine). Two separate rejections involved two rejections that had an interrejection interval of more than 2 wk and a return of renal function to the prerejection level during the interrejection period. The latter two groups were empirically defined by either a temporal relationship - two rejection episodes occurring back-to-back less than 14 days apart - or by a functional relationship - the interrejection serum creatinine before the second episode being significantly greater (>25%) than the prerejection creatinine before the first rejection episode. Patient and graft survivals from the time of these second rejection episodes were determined. Patient death from any cause was included as a graft loss. Percentage survivals were calculated by actuarial methods and are expressed as percentage survival plus or minus the standard error (±SE).8 When actuarial comparisons were made between groups, the chi-square test was used to evaluate statistical significance. 9 Rejection episodes were diagnosed on the basis of oliguria, fever, and allograft enlargement coupled with an increased serum creatinine (1-day rises of ;. 20%). 1311_ hippurate scans 10 and negative urine cultures corroborated the diagnosis. Renal scan deterioration during initial acute tubular necrosis in the absence of prerenal factors or infection was diagnostic of rejection if coupled with physical findings. Both early rejection episodes were treated with increases of intravenous prednisolone (1970- 1972) or oral (August 1972-1978) prednisone, as previously described. 6 An irreversible graft rejection was diagnosed by the absence of 131I-hippurate uptake and DTPA-uptake on renal scans during or following this supplemental prednisone administration. It was confirmed by gross appearance and surgical findings at the time of open biopsy and/or transplant nephrectomy.

RESULTS

Following a renal transplant, the profiles of serum creatinine versus time curves could be characterized as "humps" of serum creatinine elevations during rejection(s). Examples of the two camelback varieties of early multiple renal transplant rejections are depicted in Figs. 1 and 2. These

lit - PRE-CREATININE=l.l mg% 2nd· PRE-CREATININE=1.3 mg%

!:l!

Tx·NTx-.

180~

DAY 47

Sl120

Z ~ if

60 0

o

20

10

30

DAYS

Fig. 1. An example of a camelback-time early transplant rejection.

10

180~

120

2

60

o

o

10

20

30

DAYS

Fig. 2. An example of a camelback-creatinine early transplant rejection.

are contrasted with one early rejection cases with immediate loss or cases with one rejection with continued recovery. They are also contrasted to the two early but separate rejection types in which the time interval between the two rejection episodes exceeds 14 days and the prerejection serum creatinine levels are equal to each other. In both cases (Fig. 1 and Fig. 2), the rejection episodes resulted in transplant nephrectomy (Tx NTx). In Fig. 1, the prerejection creatinine levels are similar, but the second rejection occurred within 14 days of the first. This type of second early posttransplant rejection is termed camelbacktime. There appears to be high-dose steroid dependency in this case. In Fig. 2, after an initial period of acute tubular necrosis, the patient's renal function appeared to temporarily improve by the eighth day of postrejection therapy. The serum creatinine, however, never returned to the level held prior to the first rejection. After an interval of 15 days, the clinicians diagnosed and treated a second allograft rejection with a poor result, as noted. This type of second early posttransplant rejection is termed camelback-creatinine. There appears to be high-dose steroid resistance in this case. We observed a high frequency (30%) of two early transplant rejections in these 632 cadaver recipients. As seen in Table 1, the frequency of this finding varied slightly with each transplant year but did not vary by the UCSF clinical team (1970-1973 = 27%; 1974-present = 32%). Also, the frequency of two early rejections was approximately the same with the two UCSF steroid-pulse protocols 12 (1970- August 1972 = 28%; August 1972-present = 31%). Combined data of cadaver renal transplants from these 8 yr, involving two clinical teams and two immunosuppressive steroid-pulse regimens, are presented

283

PROGNOSIS OF EARLY TRANSPLANT REJECTION

Table 1. Year

Total Cadaver Tx

1970* 1971 1972 1973 1974 1975 1976

10* 54 78 82 92 98 112

19n

90

1978*

Two Early Transplant Rejections - by Transplant Year

16*

Totals

Percent Having 2 Episodes

(n) 2 Episodes

632

(n) 2-Separate

(n) 2-Camelback

2 19 20 21 19 29 38 36 8

20% 35% 23% 26% 21% 30% 34% 40% 50%

11 13 15 12 21 26 25 6

1 8 7 6 7 8 12 11 2

192

30%

130

62

*Partial years.

in the O'verall results . AlsO', it appears frO'm Table 1 that the camelback variety O'f secO'nd rejectiO'n was equally diagnO'sed and treated in each O'f the years . This type O'f secO'nd rejectiO'n accO'unted fO'r roughly O'ne-third O'f the O'verall cadaver transplants with twO' early rejectiO'ns (62/ 192). The mean time (Table 2) fO'llO'wing the cadaver transplant O'f the first rejectiO'n dO'es nO't delineate which patients will immediately 100se O'r tempO'rarily recO'ver their allO'graft functiO'n befO're having a secO'nd rejectiO'n . The mean O'nset O'f rejectiO'n in thO'se patients whO' had O'ne rejectiO'n with recO'very, hO'wever, was significantly later pO'sttransplant than in the O'ther rejectiO'n categO'ries (19.7 days versus 10-12 days). There was a wide range fO'r the day O'f the first rejectiO'n O'nset in all groups. The mean time interval to' the secO'nd rejectiO'n was 10 days in the camelbacktime variety O'f secO'nd rejectiO'n . The time interval between the twO' early rejectiO'n episO'des was the same in bO'th the separate and camelbackcreatinine varieties (26 and 27 days). The serum creatinine in the camelback-creatinine cases, to' Table 2. Rejection Type

0 1 With loss 1 With recovery 2 Separate 2 Camelback (A) Time (8) Creatinine Total

*p < 0.Q1 .

reemphasize, had nO't returned to' the level befO're the first rejectiO'n. The actuarial survival O'f cadaver transplants after these early events is depicted in Fig. 3. Time from event is used rather than time from transplant, since prO'babilities O'f O'utcO'me were analyzed accO'rding to' the pO'sttransplant cO'urse after the significant rejectiO'n event. Graft survival was significantly 100wer at 5 (p < 0.01) and 9 mO' (p < 0 .001) after the secO'nd rejectiO'n if the secO'nd rejectiO'n was O'f the camelback type cO'mpared to' a secO'nd , but separate, early rejectiO'n type. This reduced transplant survival was primarily the result O'f pO'O'r functiO'nal O'utcO'mes in the camelback-creatinine group. One-year graft survival in the camelback-time patients was better , but still significantly reduced when cO'mpared to' twO' separate rejectiO'ns. Patient survival frO'm the time O'f these early events is depicted in Fig. 4. The 1-yr patient mO'rtality in the group(s) with twO' treated rejectiO'ns is significantly greater (p < 0 .01) than in thO'se with nO' or O'ne rejectiO'n. This was seen with

Time After Transplant to "Event" First Rejection (Days)

Second Rejection (Days)

21 .7% 17.7% 30.2% 20.6%

0 10.4 ± O.?' 19.7± 1.0* 12.2 ± 0.6 (3-27)

0

28

4.4%

34

5.4%

11 .1 ± 0 .8 (6-20) 12.1 ± 0.8 (4-36)

Patients (n)

Percent of Total

137 112 191 130

632

38.6 ± 2 .4

21.0 ± 1.9 39.3 ± 3.1

284

AMEND JR. ET AL.

100

Table 3.

•'

Type of Posttransplant Event

...

, ,______ ~

.--.--------------I ~-o--~

20

__

~

............

o

3

9 12 15 18 21 6 TIME AFTER EVENT (months) KEY

.6.

-• -• ~

0

24

TYPE OF EARLY EVENT

.6.

'"

0

NO REJECTION ONE REJECTION·RECOVERY TWO SEPARATE CAMELBACK·TlME CAMELBACK-CREATININE

Fig. 3. Actuarial cadaver renal transplant survival depending on the type of early transplant rejection.

both two separate and the camelback rejection types. Although the numbers are too small to allow for a significant difference in mortality between the two rejection groups, there is a tendency for a lower patient survival in the arbitrarily assigned camelback-time patient group. This is particularly noteworthy, since much of thi s group's reduced allograft survival (in Fig. 3) may have had part of this reduction on the basis of a statistically insignificant yet important reduction in patient survival. One-year cadaver graft and

.....

""-------

80

.:(

~ 60 ~

::>

VI

-

40

Graft Survival (% ) (± SEM)

None 1 With recovery 2 Separate 2 camelback-time 2 Camelback- creatinine

88 ,9 (±2.7)· 72,5 ( ± 3,3)·t 36,7 (±4,3)H 21.4 (± 7,6):I: 5,9 ( ± 4,0):I:

Patient Survival 92,5 (±2,2) 89,9 ( ± 2,3) 82.4 (± 3.4) 73.4 (:!:8,6) 83,0 (:!: 5,6)

] § ] §

om , om , om ,

'p < differences significant. tp < differences sign~icant , :j:p < 0,001 . differences significant. §o < differences sign~icant.

patient survivals from the event are summarized in Table 3. Additionally , 19 cadaver recipients had a third diagnosed and treated rejection in the first 60 days posttransplant. The type and outcome of the second and third rejections are listed in Table 4. Only two individuals had a camelback-creatinine type rejection for their third episode. The majority, 12/19 patients , had three separate rejection episodes . The first two rejections had been fully reversed with steroid-pulse therapy. Only one of these 12, the only one of the 19 overall, regained and retained normal transplant renal function . Pretransplant blood transfusions are assessed in Table 5. Patients in the no or one rejection with recovery categories (with high 2-yr graft survivals), had a greater likelihood of having been transfused pretransplant (91 % and 75 %) . Patients with the worst transplant outcomes - one rejection with loss and two early rejections - had significantly greater likelihoods of not being transfused (50% and 59%). DISCUSSION

Despite improvements in patient survival following cadaveric renal transplantation, 7.11

~

20

Table 4.

o

One-Ye.. Survivals After Cadaver Transplantation

3

9 6 TIME AFTER EVENT (months)

KEY

TYPE OF EARLY REJECTION

.6.--.6. t:r----6 0---0

NO-REJECTION ONE· REJECTION-RECOVERY TWO-SEPARATE CAMELBACK·TIME CAMELBACK-CREATININE

.--. ----

Fig. 4. Actuarial patient survival following cadaver renal transplantation depending on the type of early transplant rejection.

Number 01 Patients

Outcomes With Three Early Rejections Type of Early Rejection

1 (1) Separate 2 (2) Separate 13 (6) Separate (5) Third, camelback- time (2) Third, camelback- creatinine 3 (3) Separate Total

19

Result Normal (6 yr) Reduced function (both failed < 1 yr) Immediate Graft failure «

3 mol

Early patient deaths « 3 mol 1 Long·term success

285

PROGNOSIS OF EARLY TRANSPLANT REJECTION Table 5.

Group

o Rejection

Pretransplant Patient Characteristics Mean Blood Units Transfused

Monthly Cytotoxic Antibody Screens (n)

Pre-Tx Dialysis Time (mo)

Age

Average Match

Percent of Patients Transfused

34.1

1.6

91%*t

8.2

7.5

16.6

37.1

1.5

50%*

5.4

6.8

15.8

34.8

1.2

75%+

6.6

6.6

15.5

35.0

1.5

59"/0t+

8.2

5.6

14.2

36.1 37.4 35.1

1.5 1.4 1.6

59"/0 75% 47%

6.6 8.3 4.4

5.7 6.5 5.1

14.9 17.2 13.0

1 Rejection immediate loss

1 Rejection recover

2 Separate rejections

2 Camelback rejections (A) Time (8) Creatinine

*p < 0.001 (chi-square). tp < 0.001 (chi-square). tp < 0.01 (chi-square).

much expense and patient morbidity still exist. In a recent analysis,12 both l-yr costs and morbidity were highest in transplant recipients having two rejections. Once the transplant is in situ, sequential clinical decisions must be made in order to optimalize allograft survival and reduce patient morbidity and mortality. Each case must be treated (or not) depending on a particular patient's course. There are little existing data 2,4,5 that give guidance concerning eventual likely outcomes after the first 2 mo posttransplant. Probability analysis, based on the posttransplant course, might provide a helpful rationale to assist therapeutic decision making. Some data, derived from a multicenter study, relate to the question of the probability of longterm transplant outcomes. 4,13 Opelz and coworkers have determined the probabilities of eventual graft and patient outcomes given the patient's initial 3-mo course 4 following a cadaver renal transplant. Early rejection episodes coupled with a semiquantitated severity grade of rejections were associated with a markedly diminished chance of l-yr graft success. Characteristics of the rejection episodes (particularly of a temporal nature, such as time of onset of rejection or intervals between rejections) were not detailed in this review. Treatments of the rejection episodes were varied owing to center-to-center immunosuppressive differences. We concluded, from our single-center results, that meaningful differences in subsequent transplant function can be predicted

by the assessment of the early posttransplant clinical course. In our study we also noted, not unexpectedly, that the groups with the best prognoses had the highest number of transfused patients. This is not the case with the multiple early rejections and one rejection with loss groups in which significantly fewer patients were transfused. This reconfirmed our 14 and other'sl5 recent studies showing the benefits of pretransplant blood transfusions. The camelback-time group, interestingly, had a high frequency of being transfused, yet as previously mentioned, this group was arbitrarily assigned. Since these patients were given higher immunosuppression over a short period of time, much of the graft loss was related to early patient mortality (see Fig. 4). Immunologic testing, through demonstrations of alterations in serially obtained in vitro assays, might provide better opportunities to prognosticate likelihoods of eventual graft function. Such tests, if highly reliable, would be welcome in making posttransplant management decisions. The presence of pre- and posttransplant antibodydependent cell-mediated cytotoxicity (ADCC) , 16,17 donor-specific lymphocyte-mediated cytotoxicity (LMc),18 and donor-specific anti-B lymphocytotoxins 19 have each been reported to be associated with poor graft results. The above-mentioned studies were expensive, time-consuming, of a conflicting nature, and not 100% predictive. These inherent problems limit their practical

286

AMEND JR. ET AL.

utilization for clinical decision making at this time. We did not correlate our retrospective analysis of these clinical courses with posttransplant immunologic testing , but it is quite possible that these cases of multiple early rejections had measurable degrees of donor-specific immune responsiveness. Whether the patients with a camelback rejection have a less steroid-responsive type of an immune rejection mechanism is not known. It is apparent, however, that the shortterm outlook of transplant function is poor in these cases despite readministration of high-dose steroid therapy. It might be recommended that multiple transplant rejections either not be treated at all or that nonsteroidal alternative therapies be offered. The latter might include the use of antithymocyte globulin,21 plasmapharesis, lymphopharesis, or allograft irradiation. We have used local allograft irradiation in just such second rejection circumstances in the past 2 yr without evident beneficial effects. Additional studies of alternative posttransplant therapies are needed. It must be cautioned that these results represent

the experience at one center. These patients did not receive antithymocyte globulin, which might alter the time of onset, severity, or frequency of subsequent cadaver allograft rejections. 22.23 The reported delay in the timing of a first rejection episode suggests that there might be differences in graft survival in such treated subjects. All rejection episodes in our series were treated with increments of intravenous prednisolone or oral prednisone. A patient with multiple early rejections can be given a progressively lower probability of long-term transplant function depending on the temporal and/or renal function features of second or subsequent rejections. There appear to be certain types, characterized here as camelback posttransplant rejections, which have a uniformly poor prognosis. This suggests to us that either no further steroid increments be prescribed in these situations or that alternative methods of immunosuppression be used. ACKNOWLEDGMENTS The authors wish to acknowledge the secretarial and technical assistance of Robin Sterin. Peggy Doeskin. and Pat Lovelock .

REFERENCES l. Carpenter CB. Austen FK: The early diagnosis of renal

allograft rejection in. Rapaport FT, Dausett J (eds): Human Transplantation. New York, Grune & Stratton, 1968, pp 151-169 2 . Williams GM , White HJO, Hume DM: Factors inHuencing the long tenn functional rate success of human renal allografts. Transplantation 5:837-843, 1967 3. Busch GJ, Reynolds ES, Galvanek EG, et al: Human renal allografts: The role of vascular injury in early renal failure. Medicine 50:29-83 , 1971 4. Opelz G, Sasaki N, Terasaki PI: Prediction of longterm kidney transplant survival rates of monitoring early graft function and clinical grades. Transplantation 25:212-,?15, 1978 5. Hamburger J: A reappraisal of the concept of organ "rejection," based on the study of homotransplanted kidneys . Transplantation 5:870-884, 1967 6. Vincenti F, Duca RM , Amend W, et al: Immunologic factors determining survival of cadaver-kidney transplants: The effect of HLA serotyping, cytotoxic antibodies and blood transfusions on graft survival. N Engl J Med 299:793-798, 1978 7. Salvatierra 0, Potter D, Cochrum KC, et al: Improved patient survival in renal transplantation . Surgery 79: 166171, 1976 8. Merrell M, Shulman LE: Determination of prognosis in chronic disease, illustrated by systemic lupus erythematosus. J Chron Dis 1:12-32, 1955 9. Mantel N: Evaluation of survival data and two new rank

order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966 10. Salvatierra 0, Powell MR , Price D, et al: The advantages of 13 1I-orthoiodohippurate scintiphotography in the management of patients after renal transplantation. Ann Surg 180:336- 342, 1974 11. Tilney NL, Strom TB, Vinehard GC, et al: Factors contributing to the declining mortality rate in renal transplantation. N Engl J Med 299:1321-1325, 19~8 12. Salvatierra 0, Feduska NJ , Vincenti F, et al: Analysis of costs and outcomes of renal transplantation at one center: Its implications . JAMA 241:1469-1473,1979 13. Opelz G, Mickey MR, Terasaki PI: Calculations on long-tenn graft and patient survival in human kidney transplantation. Transplant Proc 9:27-30, 1977 14. Feduska NJ , Vincenti F, Amend WJ, et al: Do blood transfusions enhance the possibility of a compatible transplant? Transplantation 27:35-38 , 1979 15. Opelz G, Terasaki PI: Improvement of kidney-graft survival with increased numbers of blood transfusions. N Engl J Med 299:799-803, 1978 16. Gailiunas P. Person A, Suthanthiran M, et al: Pretransplant assays of alloimmunity in predicting renal graft outcome. Proc Clin Dial Transplant Forum 7:1-3, 1977 17. Gailiunas P, Suthanthiran M, Person A, et al: Posttransplant immunologic monitoring of the renal allograft recipient. Transplant Proc 10:609-611, 1978 18. Stiller CR, Sinclair NR, Abrahams S, et al: Anti-donor

PROGNOSIS OF EARLY TRANSPLANT REJECTION immune responses in prediction of transplant rejection. N Engl J Med 294:978-982, 1976 19. Ettenger RB, Terasaki PI, Ting A, et al: Anti-B Iymphocytotoxins in renal-allograft rejection. N Engl J Med 295:305-309, 1976 20. Opelz G, Iwaki Y, Terasaki PI , et al : B-cell antibodies and kidney transplants: confused? Dial Transplant 8:479484, 1979

287 21. Shields CF, Cosimi AB, Tolkoff-Rubin N, et al: Use of antithymocyte globulin for reversal of acute allograft rejection. Transplantation 28:461-464, 1979 22. Najarian JS, Simmons RL: The clinical use of antilymphocyte globulin. N Engl J Med 285:158-166, 1971 23. Thomas F, Mendez-Picon G, Thomas J, et al: Effect of antilymphocyte globulin potency on survival of cadaver renal transplants. Lancet 2:671 - 674, 1977