Recurrence of obliterative bronchiolitis and determinants of outcome in 139 pulmonary retransplant recipients

CARDIAC AND PULMONARY REPLACEMENT RECURRENCE OF OBLITERATIVE BRONCHIOLITIS AND DETERMINANTS OF OUTCOME IN 139 PULMONARY RETRANSPLANT RECIPIENTS

An international series of pulmonary r e t r a n s p l a n t a t i o n was u p d a t e d to identify the predictors of outcome and the prevalence and recurrence rate of obliterative bronchiolitis after operation. The study cohort included 139 patients who underwent r e t r a n s p l a n t a t i o n in 34 institutions in North America and Europe between 1985 a n d 1994. Eighty patients underwent r e t r a n s p l a n t a t i o n because of obliterative bronchiolitis, 34 because of acute graft failure, 13 because of intractable airway complications, 8 because of acute rejection, and 4 because of other indications. Survivors were followed up for a median of 630 days, with 48 patients alive at 1 year, 30 at 2 years, and 16 at 3 years after retransplantation. Actuarial survival was 65% -+ 4% at 1 month, 54% - 4% at 3 months, 45% + 4% at 1 year, 38% - 5% at 2 years, and 36% -+ 5% at 3 years; nonetheless, of 90-day postoperative survivors, 65% - 6% were alive 3 years after r e t r a n s p l a n t a t i o n . Life-table and univariate Cox analysis revealed that more recent year of r e t r a n s p l a n tation (p = 0.009), identical match of ABO blood group (p = 0.01), absence of a donor-recipient cytomegalovirus mismatch (p = 0.04), and being ambulatory immediately before r e t r a n s p l a n t a t i o n (p = 0.04) were associated with survival. By multivariate Cox analysis, being ambulatory before r e t r a n s p l a n t a t i o n was the most significant predictor of survival (p = 0.008), followed by reoperation in Europe (p = 0.044). Complete pulmonary function tests were done yearly in every survivor of r e t r a n s p l a n t a t i o n and bronchiolitis obliterans syndrome stages were assigned. Eleven percent of patients were in stage 3 at 1 year, 20% at 2 years, and 25% at 3 years after retransplantation. Values of forced expiratory volume in 1 second decreased from 1.89 - 0.13 L early after r e t r a n s p l a n t a t i o n to 1.80 - 0.15 L at 1 year and 1.54 +- 0.16 L at 2 years (p = 0.006, year 2 versus baseline postoperative value). Most of this decrease occurred in patients who underwent r e t r a n s p l a n t a t i o n because of obliterative bronchiolitis, whereas the pulmonary function of patients who underwent retransplantation because Richard J. Novick, MD, Hans-Joachim Schfifers, MD (by invitation), Larry Stitt, MSc (by invitation), Bernard Andr6assian, MD (by invitation), Jean-Pierre Duchatelle, MD (by invitation), Walter Klepetko, MD (by invitation), Robert L. Hardesty, MD, Adaani Frost, MD (by invitation), and G. Alexander Patterson, MD, London, Ontario, Canada;

Hannover, Germany; Clichy, France," Vienna, Austria; Pittsburgh, Pa.; Houston, Tex.; and St. Louis, Mo.

From the Departments of Surgery and Epidemiology and Biostatistics, University Hospital, the Robarts Research Institute and the University of Western Ontario, London, Ontario, Canada; Hannover, Germany; Clichy, France; Vienna, Austria; Pittsburgh, Pa.; Houston, Tex.; St. Louis, Mo.; and participating centers in the pulmonary retransplant registry. Address for reprints: Dr. Richard J. Novick, PO Box 5339, University Hospital, London, Ontario, Canada N6A 5A5. 1402

Supported by a grant from the Ontario Thoracic Society. Read at the Seventy-fifthAnnual Meeting of The American Association for Thoracic Surgery, Boston, Mass., April 23-26, 1995. J THORACCARDIOVASCSURG 1995;110:1402-14 Copyright © 1995 by Mosby-Year Book, Inc. 0022-5223/95 $5.00 + 0 12/6/67383

The Journal of Thoracic and Cardiovascular Surgery Volume 110, Number 5

Novick et at

1403

of other conditions did not significantly change. We conclude that survival after pulmonary retransplantation is improving. Optimal results can be obtained in patients who are ambulatory befbre retransplantation. Compared with recent data after primary lung transplantation, bronchiolitis obliterans syndrome does not appear to recur in an accelerated manner after retransplantation. As long as early mortality as a result of infection can be minimized, pulmonary retransplantation appears to offer a reasonable option in highly selected patients. (J THORACCARDmVASCSURG 1995;110:1402-14)

uring the p a s t d e c a d e lung t r a n s p l a n t a t i o n has evolved into a successful t h e r a p y for p a t i e n t s with e n d - s t a g e r e s p i r a t o r y failure. R e c e n t l y , d o n o r a n d r e c i p i e n t selection, o p e r a t i v e t e c h n i q u e , a n d p o s t o p e r a t i v e c a r e have b e e n refined, resulting in a significant i m p r o v e m e n t in s u r v i v a l ] D e s p i t e these advances, the p o s t o p e r a t i v e c o u r s e of a significant n u m b e r o:g p a t i e n t s is c o m p l i c a t e d by early graft dysfunction, 2-4 art airway h e a l i n g c o m p l i c a t i o n , 5' 6 or acute r e j e c t i o n ] ' s F u r t h e r m o r e , the d e v e l o p m e n t of o b l i t e r a t i v e bronchiolitis ( O B ) r e m a i n s a t h r e a t to i n t e r m e d i a t e - t e r m survivors of lung t r a n s p l a n t a tion. 9q2 O n occasion, t h e s e c o m p l i c a t i o n s m a y p r o v e so i n t r a c t a b l e that graft failure will e n s u e a n d the only possibility for c o n t i n u e d survival rests with r e t r a n s p l a n t a t i o n . Since 1988 an increasing n u m b e r of pulmonary, r e t r a n s p l a n t s have b e e n p e r f o r m e d b e c a u s e of acute o r chronic graft failure. ~3-21 T h e p r a c t i c e o f p u l m o n a r y r e t r a n s p l a n t a t i o n has raised c o n t r o v e r s y b e c a u s e of the s h o r t a g e o f lung grafts available for p r i m a r y t r a n s p l a n t a t i o n a n d t h e p o o r results in the early e x p e r i e n c e with t h e s e p r o c e dures. T h e p u l m o n a r y r e t r a n s p l a n t registry was est a b l i s h e d in 1991 to d o c u m e n t , in a large n u m b e r of patients, the o u t c o m e a n d p r e d i c t o r s o f survival after p u l m o n a r y retransplantation. 13 Recently, after additional :patient accrual and with increasing follow-up of retransplant recipients, the registry data were u p d a t e d to d e t e r m i n e the predictors of o u t c o m e and recurrence rate of OB after retransplantation.

D

Patients and methods Since mid-1991 patients undergoing pulmonary retransplantation have been recruited to the pulmonary retransplant registry by means of a standardized study questionnaire. These patients have been followed up prospectively with at leasl yearly updates of clinical status and pulmonary function. For the purposes of this report, the status of all study patients was determined in October 1994. Patients who underwent retransplantation for any indication were included in the study cohort. The 16 parameters listed in Tables I and II were analyzed in each patient. The chief outcome variables

Table I. Association of recipient variables with survival after pulmonary retransplantation

Univariate analysis Variable Age As a continuous variable <40 vs. >40 yr Female sex Original diagnosis Indication for retransplantation Ambulatory before retransplant Ventilator support before retransplant Recipient CMV positMty Waiting time As a continuous variable -<3 days vs. 4-14 days vs. 15-70 days vs. >70 days Interval between transplants As a continuous variable -<4 wk vs. >4 wk-1 yr vs. 1-2 yr vs. >2 yr -<1 yr vs. >1 yr

Multivariate analysis

Risk p Adjusted p ratio Value risk ratio Value * 1.08 1.05 * *

0.984 0.717 0.829 0.821 0.468

t t t t t

NS NS NS NS NS

0.59

0.045

0.46

0.008

1.32

0.208

t

NS

0.96

0.874

t

NS

* *

0.240 0.259

t t

NS NS

* *

0.146 0.144

t t

NS NS

1.14

0.553

t

NS

NS, Not significanton multivariateanalysis;CMV,cytomegalovirus. *Riskratio not calculated,becausevariablehas more than two categories. tAdjusted risk ratio not calculated,becausevariabledoes not enter into multivariatemodel.

included survival interval after retransplantation, cause of death, and pulmonary function test data. Study form results were incorporated into the pulmonary retransplant database with use of the FoxPro database management system (Microsoft Corporation, Redmond, Wash.) on a 486DX2/66 MHz computer. Statistical analysis was performed with the SAS statistical package, version 6.04 (SAS Institute Inc., Cary, N.C.). All data were expressed as mean plus or minus the standard error of the mean. Actuarial survival was calculated by the Kaplan-Meier method 22 and Cox proportional hazards methods 23 were used to determine which variables were associated with survival after pulmonary retransplantation. Variables exhibiting a p value less than 0.25 on univariate analysis were considered for entry into a multivariate model to deter-

14 0 4

The Journal of Thoracic and Cardiovascular Surgery November 1995

Novick et al.

Table II. Association of donor and operative variables with survival after pulmonary retransplantation Univariate analysis

Variable Year of reoperation As a continuous variable 1985-1991 vs. 1992-1994 Retransplant center (Europe vs. North America) Type of reoperation All 5 types Old graft in situ vs. not Individual center experience (<5 vs. ->5 retransplants) Identical match of donor-recipient A B O blood group Donor C M V positivity Donor-recipient C M V mismatch

BOS between patient groups at 1 and 2 years after retransplantation. A p value less than 0.05 was deemed significant.

Multivariate analysis

Risk ratio

p Value

Adjusted risk ratio

p Value

* 1.74 0.72

0.009 0.015 0.140

t t 0.61

NS NS 0.044

* 1.22 1.44

0.795 0.382 0.104

t t t

NS NS NS

0.55

0.015

t

NS

1.24 1.79

0.350 0.043

t t

NS NS

NS, Not significant on multivariate analysis; CMV, cytomegalovirus. *Risk ratio not calculated, because variable has more than two categories. tAdjusted risk ratio not calculated, because variable does not enter into multivariate model.

mine the independent predictors of survival after retransplantation. Furthermore, the risk ratio of each variable was expressed as a comparison of survival between groups, with a value of 1.0 indicating no survival difference. A major focus of this report has been graft function and the recurrence of OB in the intermediate-term after retransplantation. The detection rate of histologic OB after retransplantation would be expected to vary among contributing centers, because of the different frequency of the performance of postoperative transbronchial biopsies. We therefore chose a functional, rather than a histologic definition of OB, which thus permitted the standardization of postoperative results among institutions. 11 Complete pulmonary function test data were obtained from every survivor of retransplantation at yearly intervals. Bronchiolitis obliterans syndrome (BOS) stages were designated according to previously published criteria, based on values of forced expiratory volume in 1 second (FEV1).I, 11The changes in absolute FEV 1values at 1 and 2 years after retransplantation were calculated for the entire study cohort, for single versus bilateral lung retransplant recipients, and for patients who underwent retransplantation because of OB versus those who underwent retransplantation because of other indications. Paired, two-tailed t tests were used to compare changes over time within each patient group, whereas unpaired, two-tailed t tests were used to contrast results between patient groups. Furthermore, contingency table analyses were used to compare the prevalence of and freedom from stages 1, 2, and 3

Results

Thirty-four lung transplant centers participated in this study, including 19 from North America and 15 from Europe. Participating institutions and the number of patients contributed by each center are listed in the appendix. A total of 139 patients have undergone retransplantation, including 79 female and 60 male patients with a median age of 42 years (range 6 months to 62 years). Before the first transplantation procedure, 35% had a diagnosis of emphysema, 21% primary pulmonary hypertension or Eisenmenger's syndrome, 19% restrictive lung disease, 16% cystic fibrosis, and 9% miscellaneous conditions. The indications for retransplantation included OB in 80 cases, primary graft failure in 34 cases, intractable airway complications in 13 cases, histologically confirmed severe acute rejection in 8 cases, and other indications in 4 cases. Fifty-nine patients underwent repeat single-lung transplantation, 37 on the ipsilateral side and 22 on the contralateral side. Thirty patients underwent repeat doublelung transplantation, 24 double-lung transplantation after a previous single-lung transplant, and 26 singlelung transplantation after a previous double-lung or heart-lung transplant. Patient follow-up was 100% complete, with up-to-date survival and pulmonary function test data available in every patient. Survival. The actuarial survival of all study patients is shown in Fig. 1. Of the 139 retransplant recipients, 85 have died and 54 are still living. Actuarial survival was 65% ± 4% at 1 month, 54% _+ 4% at 3 months, 45% _+ 4% at 1 year, 38% _+ 5% at 2 years, and 36% _+ 5% at 3 years. Despite the significant patient attrition early after operation, 65% + 6% of 90-day postoperative survivors were alive 3 years after retransplantation. The median follow-up in current survivors is 630 days (mean 706 _+ 67 days, range 45 to 1781 days). Forty-eight patients have reached the first anniversary, 30 the second anniversary, 16 the third anniversary, and 4 the fourth anniversary of retransplantation. The association of the 16 variables that were analyzed in each patient with survival after retransplantation is depicted in Tables I (recipient variables) and II (donor and operative variables). Survival was not statistically different according to the age, sex, or original diagnosis of the recipients. Similarly, survival did not differ according to the predominant indication for retransplantation (Fig. 2). Univariate Cox analysis

The Journal of Thoracic and Cardiovascular Surgery Volume 110, Number 5

Novick et al.

1405

,oo[ 8O

_J > £E ~D CO l-Z LLI 0 r~ ILl Q-

40-

! 1

20-

C

a

I

6

12

I

I

i

I

I

I

I

I

18

24

30

36

42

48

54

60

INTERVAL

(months)

oo[

Fig. 1. Actuarial survival of 139 patients undergoing pulmonary retransplantation for all indications.

......

OBLITERATIVE BRONCHIOLITIS

.......... AIRWAY COMPLICATIONS GRAFT FAILURE

80 I'L _J

X

> or" 60~t\ "!"--. CO : L--I FZ ............. 't. ..... LIJ 4ot'--~ O Q/ LI.J Cl

tTi ,4 ............

,

20

0

i

I

r

6

12

r8

I

I

I

J

24 30 36 42 INTERVAL ( m o n t h s )

I

i

l

48

54

60

Fig. 2. Actuarial survival according to predominant indication for retransplantation; p = 0.468. has confirmed that survival after pulmonary retransplantation is improving (Table II). In particular, actuarial survival has been significantly higher in patients who underwent retransplantation between 1992 and 1994, as opposed to between 1985 and 1991 (Fig. 3).

Effect of clinical condition before retransplantation, waiting time, and interval between transplants on survival. Immediately before retransplantation, only 29% of patients were ambulatory (able to walk 50 m with or without assistance). By

The Journal of Thoracic and

1406

Novick et aL

Cardiovascular Surgery November 1995

I00 .......

1992-1994 1985-1991

80 _J

<> > cr

60

cO

]. ...........

z

LIJ £_3 O/ LIJ £1_

40

20

01

,

I

J

I

I

I

I

I

I

I

6

12

18

24

50

56

42

48

54

60

INTERVAL (months)

Fig. 3. Actuarial survival according to year of retransplantation; p = 0.015. I00 ......

AMBULATORY

- -

NON A M B U L A T O R Y

8C ._l

X

> cr

"I I. . . . .

1

60

oo }-z LIJ 0

rr LLI

40 |

|

Q_

[ 20

I

I

6

12

I

18

I

24

30

I

I

I

I

I

36

42

48

54

60

INTERVAL (months)

Fig. 4. Actuarial survival according to ambulatory status of recipient immediately before retransplantation;p = 0.045. univariate and multivariate Cox analyses, ambulatory patients had a significantly better survival than nonambulatory patients (Table I, Fig. 4). Before retransplantation, half of the patients

were ventilator d e p e n d e n t whereas half required supplemental oxygen but not mechanical ventilation. T h e r e were no significant differences in survival according to the preoperative ventilator

The Journal of Thoracic and

Novick et al.

Cardiovascular Surgery

1407

Volume 110, Number 5

I00 ......

ABO IDENTICAL

- -

NOT ABO IDENTICAL

8C

> cr"

60 b ~ q "L.

O3

-

'1.

.11

z u.J 4 0 o n.ill o_

¢. . . . . . . .

"t

I t. . . . . . . . . . . . . . .

20

01

f

I

I

I

I

I

I

I

I

I

6

12

18

24

30

56

42

48

54

60

INTERVAL (months)

Fig. 5. Actuarial survival according to presence of identical match of donor-recipient ABO blood group; p = 0.015.

status of the retransplant recipient (Table I). Neither the official waiting time for retransplantation nor the interval between transplant procedures had a significant impact on survival, regardless of whether these variables were analyzed continuously or after the grouping of patients into approximately equal quartiles (Table I). Impact of the type of retransplant procedure and the retransplant center on survival. Actuarial survival did not differ among patients undergoing the five different types of retransplant procedures (Table II). In 35% of cases, an old contralateral graft remained in situ after retransplantation, whereas in 65% o f cases the old graft or grafts had been excised. As shown in Table II, survival was not significantly different according to whether an old graft remained in situ after retransplantation. Sixty-one patients underwent retransplantation in Europe and 78 in North America. In contradistinction to our previous report on retransplantation for O B , 24 n o difference in survival was found between European and North American patients on univariate analysis (Table II). However, when adjusted for ambulatory status in the multivariate analysis, reoperation in Europe remained as a predictive variable (p = 0.044, adjusted risk ratio 0.61). Six of the 19 North American centers and 3

of the 15 European centers performed five or more pulmonary retransplant procedures. There was a trend toward improved survival in patients undergoing retransplantation in centers with experience in at least five pulmonary retransplant procedures (Table II). Impact of donor-recipient ABO blood group and cytomegalovirus serologic status on survival. Seventy-eight percent of patients undergoing retransplantation received an ABO-identical graft at reoperation, whereas 22% received a graft that was ABO compatible, but not identical. Survival was significantly better in patients who received an ABO-identical graft (Fig. 5). Neither donor nor recipient cytomegalovirus serologic status was predictive of survival after retransplantation; nonetheless, the 18 cases of cytomegalovirus mismatch (cytomegalovirus-positive donor, negative recipient) were associated with significantly decreased survival after retransplantation (Table II). Causes of death. As noted in previous reports from the registry, 13' 14, 24 infection was the predominant cause of death at all time intervals after retransplantation. In this series, 48 of the 85 deaths (56%) were caused by opportunistic infection by viral, bacterial, or fungal organisms. Twenty-two percent of deaths were caused by acute failure of the second graft early after operation, 14% by recurrent OB, 4% by an airway complication, and 4% by other

1408

The Journal of Thoracic and Cardiovascular Surgery November 1995

Novick et al.

[00 ......

1992-1994 1985-1991

80 _J

> rr

60

o9 }-Z LLJ L) rr LLJ EL

-]1. .................

L ............

40 [ m

-1

20

I

I

i

I

I

I

I

I

~

I

6

12

18

24

.30

:36

42

48

54

60

INTERVAL ( m o n t h s )

Fig. 6. Actuarial survival of 59 patients who underwent retransplantation because of indications other than OB, according to year of reoperation; p = 0.43. Table III. Prevalence of states O, 1, 2, and 3 BOS according to duration of follow-up after retransplantation Postop. No. of interval patients 1 yr 2 yr 3 yr

48 30 16

Stage0 (%)

Stage 1 (%)

Stage2 (%)

Stage3 (%)

79 63 69

4 10 6

6 7 0

11 20 25

causes. Interestingly, only 5 of the 80 patients who underwent retransplantation because of OB died of this complication after operation. Predictors of survival by multivariate analysis. As shown in Tables I and II, being ambulatory immediately before retransplantation was the most significant predictor of survival on multivariate analysis, followed by reoperation in Europe. Identical match of donor-recipient ABO blood group and the absence of a cytomegalovirus mismatch, which were associated with survival on univariate analysis, did not enter the multivariate model. Whereas more recent year of retransplantation was a significant predictor of survival on multivariate analysis in patients with O B , 24 the year of reoperation did not significantly influence survival in the 59 patients who underwent retransplantation because of other indi-

cations (Fig. 6). This resulted in the failure of year of reoperation to enter the current multivariate model as a factor predictive of survival after retransplantation. Pulmonary function and recurrence of OB in retransplant survivors. Complete, up-to-date FEV 1 data were available from every retransplant recipient who survived at least 1 year after operation. Table III shows the BOS stages of patients at yearly intervals after retransplantation. Of note is that the prevalence of stage 3 (severe) BOS was 11% at 1 year, 20% at 2 years, and 25% at 3 years after retransplantation. There were no significant differences in the BOS stage of single versus double lung retransplant recipients. Likewise, as shown in Fig. 7, there were no statistically significant differences in the prevalence of BOS between patients who underwent retransplantation because of OB versus those who underwent reoperation because of other indications (p = 0.49 and 0.45 at 1 and 2 years, respectively). Two years after retransplantation, the prevalence of stage 3 BOS was 33% in patients who underwent reoperation because of OB versus 7% in those who underwent retransplantation because of graft failure or an airway complication (p = 0.16). In the 30 recipients who survived beyond the second anniversary of retransplantation, absolute FEV 1 values decreased from 1.89 +_ 0.13 L 3 months

The Journal of Thoracic and Cardiovascular Surgery Volume 110, Number 5

N o v i c k et al.

I

-"":--.-7"---<:2°' --

O3 W (_9

1409

....

(n=15)

(n=28) " ' - , .

CO v

03 (D O3

60"O (n=15)

c)

O2 b_

40

© w w n~ LL

•

RETRANSPLANT FOR GRAFT FAILURE OR AIRWAY COMPLICATION O . . . . . O RE'TRANSPLANT FOR OBLITERATIVE BRONCHIOLITIS

20

•

o< I

I

12 INTERVAL ( months )

24

Fig. 7. Freedom from BOS stages 1, 2, and 3 in patients who underwent retransplantation because of graft failure or airway complication versus those who underwent retransplantation because of OB; p = 0.49 at 1 year and 0.45 at 2 years.

after reoperation to 1.80 _+ 0.15 L at 1 year (p = 0.32) and 1.54 + 0.16 L at 2 years (p = 0.006 versus baseline postoperative values). There were no significant differences in the rate of decrease of FEV1 values between patients who underwent single or double lung retransplantation. As shown in Fig. 8, FEV 1 values did not decrease significantly in the 15 2-year survivors who underwent retransplantation because of acute graft failure or an airway complication (p = 0.23, 2-year versus baseline value). However, in the 15 survivors Who underwent retransplantation because of OB, FEV 1 values decreased significantly by 2 years after operation (p = 0.02), although the absolute F E V 1 value was not statistically different from that of the non-OB group at that time (p = 0.29). Discussion

The results of this study confirm the determinants of outcome after pulmonary retransplantation that have been noted in previous reports from the registry.13, 14, 24 Increasing experience with these procedures in North America and Europe has resulted in improved survival after pulmonary retransplantation, consonant with the experience in renal aS' 26 and hepatic 27-29 retransplantation. Familiarity of individ-

ual centers with pulmonary retransplantation has played a role, inasmuch as the nine institutions that reported five or more retransplant procedures exhibited a trend toward improved patient survival compared with that of less experienced centers. Whereas reoperation in Europe appeared to confer a survival advantage in our previous analysis of retransplantation for O B , 24 the influence of continent of retransplantation on survival has begun to wane. This variable was no longer statistically significant on univariate analysis and had only a marginally significant p value on multivariate analysis in this study. Donor factors at reoperation continued to play an important role in determining intermediateterm survival after pulmonary retransplantation. The presence of a donor-recipient cytomegalovirus mismatch was associated with decreased survival on univariate analysis. The influence of donor-recipient cytomegalovirus matching on survival and on the prevalence of OB after primary lung transplantation is a subject of controversy, with some studies showing no discernible effect1 and other reports indicating that the presence of a cytomegalovirus mismatch is a highly significant predictor of early and late death after operation. 3° Although the immunosup-

1410

The Journal of Thoracic and Cardiovascular Surgery November 1995

Novick et al.

2.5

2.0

C.)

1.5-

CO

/ v

5 1.0tad LL

•

RETRANSPLANT FOR GRAFT FAILURE OR AIRWAY COMPLICATION O . . . . O RETRANSPLANT FOR OBLITERATIVE BRONCHIOLITIS

0.5-

O

I 3

•

I 12

24

I N T E R V A L (months)

Fig. 8. FEV 1 values in 15 2-year survivors who underwent retransplantation because of OB versus 15 2-year survivors who underwent retransplantation because of graft failure or airway complication; see text for p values.

pressive protocols after retransplantation differed among contributing centers, most retransplant recipients in the registry were administered augmented immunosuppression perioperatively, which perhaps accounted for the increased morbidity and mortality from primary cytomegalovirus infection in the 18 patients who received a cytomegalovirusmismatched graft. Another important finding in this study is that an identical match of donor-recipient ABO blood group is significantly associated with survival after retransplantation. A trend in this direction was noted in the first report from the retransplant registry in 1993.13 A similar analysis in a larger number of patients after primary lung transplantation has indicated only a trend toward improved survival (p = 0.11) in patients receiving an ABOidentical, as opposed to an ABO-compatible, graft (personal communication, Ms. Mary Pohl, St. Louis International Lung Transplant Registry). The importance of an identical match of donor-recipient ABO blood group in pulmonary retransplantation awaits confirmation by the long-term follow-up of a larger number of patients with secondary lung grafts. The multivariate analysis indicated that the most significant predictor of survival after retransplanta-

tion was the ambulatory status of the recipient immediately before reoperation. Patients who were able to walk more than 50 m before retransplantation were more than twice as likely to be alive at all time intervals after operation (adjusted risk ratio of death 0.46, p = 0.008). In 1994, almost half of the newly registered patients were ambulatory before retransplantation, as opposed to 26% in previous years. Similarly, very few patients underwent retransplantation in 1993 and 1994 in the setting of severe multiorgan failure, which has been shown to predict a greater than 90% perioperative mortality rate not only in pulmonary retransplantation TM z4 but also in hepatic retransplantation. 29 These data indicate that patient selection for pulmonary retransplantation may be improving, as additional experience with these procedures is gained in North America and Europe. Given the absolute necessity to make optimal use of scarce donor lung grafts, it is important to determine whether OB or BOS develops in an accelerated manner after retransplantation. We are aware of only three papers in which the prevalence of BOS after primary lung or heart-lung transplantation has been reported in an actuarial manner. A recent study from Stanford indicated that the actu-

The Journal of Thoracic and Cardiovascular Surgery Volume 110, Number 5

arial freedom from "clinical OB" was 71% at 1 year and approximately 55% to 60% at 2 and 3 years after heart-lung transplantation. 12 In a report from the University of North Carolina, the actuarial freedom from stages 1, 2, and 3 BOS was 85% to 90% at 1 year and 60% at 2 years after bilateral lung transplantation for cystic fibrosis. 31 In a recent study from Hannover, the prevalence of BOS in more than 100 patients who underwent primary lung transplantation was contrasted with that of 14 patients who underwent retransplantation. 21 At i year after operation, 88% of primary lung transplant recipients and 80% of patients who underwent pulmonary retransplantation were free from stage 3 (severe) BOS. At 2 years, however, 72% of primary lung graft recipients were free from stage 3 BOS, whereas :only 27% of retransplant recipients had avoided this complication.21 The Hannover group regards the development of severe BOS as a failure of immunosuppression and has recently switched to an FK506-based protocol in its pulmonary retransplant recipients in an attempt to slow the rate of development of BOS in this patient subset. 21 Other centers have had a favorable experience with FK506 in lung transplantation, 32 although further investigations are required to determine whether modified immunosuppression is capable of reducing the risk of graft loss in the intermediate-term after operation. The pulmonary function test analyses performed in this study have determined that the prevalence of stages 1, 2, and 3 BOS is almost identical at 1 and 2 years after retransplantation as in the previously noted series of primary lung transplantation. 12' 21, 31 Our 80% rate of freedom from stage 3 BOS 2 years after retransplantation compares favorably with the freedom from stage 3 BOS in recipients of primary lung grafts in the Hannover study.21 Furthermore, our 63% rate of freedom from stages 1, 2, and 3 BOS at 2 years is similar to the 60% freedom from the same BOS stages in the North Carolina experience. 31 Current evidence, therefore, indicates that in patents who undergo pulmonary retransplantation BOS deveiops with a similar prevalence and intensity as in recipients of primary lung grafts. Although in some patients rapidly progressive BOS develops after retransplantation, the majority of retranspl~nt survivors have acceptable pulmonary function in the intermediate-term after operation. Detailed anatysis of patient subgroups has confirmed that pulmonary function declines at a similar rate in single and bilateral lung retransplant recipiI

Novick et aL

1411

ents. Interestingly, FEV 1 values decreased significantly by 2 years after operation in patients who underwent retransplantation because of OB, but not in patients who underwent retransplantation because of acute graft failure or an airway complication. Although the differences in FEV 1 and the prevalence of stage 3 BOS were not statistically significant between these two patient groups at 2 years, it is possible that with further patient follow-up and additional patient accrual statistically significant differences will be realized. These results underline the importance of research on alternative immunosuppressive protocols in primary and secondary lung transplantation, so that the prevalence of BOS and its recurrence rate after retransplantation can be minimized. In summary, survival after pulmonary retransplantation is improving because of increasing experience and better patient selection. The goal of the retransplant registry has been to accelerate this trend by determining the factors associated with survival and, ultimately, with excellent graft function after retransplantation. Current evidence indicates that optimal results may be anticipated in ambulatory candidates who receive an ABO-identical graft that is matched for cytomegalovirus status. Retransplant recipients who survive the first 3 postoperative months have a medium-term prognosis similar to that of patients with first-time lung transplants. Efforts to minimize early mortality from infection should improve further the results of retransplantation and justify the continued performance of these procedures in highly selected patients. We thank the contributing thoracic surgeons, pulmonary medicine physicians, and recipient coordinators who have participated in the pulmonary retransplant registry. We also acknowledge the assistance of Heather Motloch in manuscript preparation and of Theresa Novick, MSc, in data collection and analysis. REFERENCES 1. Cooper JD, Patterson GA, Trulock EP, Washington University Lung Transplant Group. Results of single and bilateral lung transplantation in 131 consecutive recipients. J THORACCARDIOVASCSURG1994;107:46071. 2. Novick RJ, Menkis AH, McKenzie FN. New trends in lung preservation: a collective review. J Heart Lung Transplant 1992;11:377-92. 3. Burdine J, Hertz MI, Snover DC, Bolman RM. Heartlung and lung transplantation: perioperative pulmonary dysfunction.Transplant Proc 1991;23:1176-7. 4. Haydock DA, Trulock EP, Kaiser LR, Knight SR,

14 1 2

5.

6.

7.

8.

9.

10. 11.

12.

13.

14.

15.

16.

17.

18.

19.

The Journal of Thoracic and Cardiovascular Surgery November 1995

Novick et al.

Pasque MK, Cooper JD. Management of dysfunction in the transplanted lung: experience with 7 clinical cases. Ann Thorac Surg 1992;53:635-41. Schfifers H J, Haydock DA, Cooper JD. The prevalence and management of bronchial anastomotic complications in lung transplantation. J THOP,AC CARDIOVASe SURG 1991;101:1044-52. Schfifers H J, Schfifer CM, Zink C, Haverich A, Borst HG. Surgical treatment of airway complications after lung transplantation. J THORAC CARDIOVASCSURG 1994; 107:1476-80. Griffith BP, Hardesty RL, Armitage JM, et al. Acute rejection of lung allografts with various immunosuppressive protocols. Ann Thorac Surg 1992;54:846-51. de Hoyos A, Chamberlain D, Schwartzman R, et al. Prospective assessment of a standardized pathological rating system for acute rejection in lung transplantation. Chest 1993;103:1813-8. Burke CM, Glanville AR, Theodore J, Robin ED. Lung immunogenicity, rejection and obliterative bronchiolitis. Chest 1987;92:547-9. Theodore J, Starnes VA, Lewiston NJ. Obliterative bronchiolitis. Clin Chest Med 1990;11:309-21. Cooper JD, Billingham M, Egan T, et al. A working formulation for the standardization of nomenclature and for clinical staging of chronic dysfunction in lung allografts. J Heart Lung Transplant 1993;12:713-6. Sarris GE, Smith JA, Shumway NE, et al. Long-term results of combined heart-lung transplantation: the Stanford experience. J Heart Lung Transplant 1994; 13:940-9. Novick R J, Kaye MP, Patterson GA, et al. Redo lung transplantation: a North American-European experience. J Heart Lung Transplant 1993;12:5-16. Novick RJ, Andrdassian B, Schfifers HJ, et al. Pulmonary retransplantation for obliterative bronchiolitis: intermediate-term results of a North American-European series. J THOP,AC CARDIOVAS¢SURG 1994;107: 755-63. Shennib H, Novick R, Mulder D, et al. Is lung retransplantation indicated: report on four patients. Eur Respir J 1993;6:354-7. Miller JD, Patterson GA. Retransplantation following isolated lung transplantation. Semin Thorac Cardiovasc Surg 1992;4:122-5. Bjortuft O, Forester A, Boe J, Geiran O. Single lung transplantation as treatment for end-stage pulmonary sarcoidosis: recurrence of sarcoidosis in two different lung allografts in one patient. J Heart Lung Transplant 1994;13:24-9. Fournier M, Sleiman C, Mal H, et al. Single-lung retransplantation for late graft failure. Eur Respir J 1993;6:1202-6. Adams DH, Cochrane AD, Khaghani A, Smith JD, Yacoub MH. Retransplantation in heart-lung recipi-

20.

21.

22.

23. 24.

25.

26.

27. 28.

29.

30.

31.

32.

ents with obliterative bronchiolitis. J THORACCARD10VASe SURG 1994;107:450-9. Haverich A, Hirt SW, Wahlers T, Schfifers HJ, Zink C, Borst HG. Functional results after lung retransplantation. J Heart Lung Transplant 1994;13:48-55. Schfifers HJ, Hausen B, Wahlers T, Fieguth HG, Jurmann M, Borst HG. Retransplantation of the lung: a single center experience. Eur J Cardiothorac Surg (In press). Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1957;53: 457-81. Cox DR. Regression models and life-tables. J R Stat Soc Ser B 1972;34:187-220. Novick R J, Sch~fers H J, Stitt L, et al. Seventy-two pulmonary retransplantations for obliterative bronchiolitis: predictors of survival. Ann Thorac Surg 1995;60:111-6. Cecka JM, Terasaki PI. The UNOS Scientific Renal Retransplant Registry. In: Terasaki P, Cecka JM, eds. Clinical transplants 1992. Los Angeles: UCLA Tissue Typing Laboratory, 1992:1-16. Cecka JM, Terasaki PI. The UNOS Scientific Renal Transplant Registry. In: Terasaki P, Cecka J, eds. Clinical transplants 1993. Los Angeles: UCLA Tissue Typing Laboratory, 1993:1-18. Shaw BW, Wood RP. Improved results with retransplantation of the liver. Transplant Proc 1989;21:2407-8. Mora NP, Klintmalm GB, Cofer JB. Results after liver retransplantation in a group of 50 regrafted patients: two different concepts of elective versus emergency retransplantation. Transplant Int 1991;4:231-4. D'Alessandro AN, Ploeg RJ, Knechtle SJ, et al. Retransplantation of the liver: a seven-year experience. Transplantation 1993;55:1083-7. Bando K, Paradis IL, Komatsu K, et al. Analysis of time-dependent risks for infection, rejection and death after pulmonary transplantation. J THORAC CARDIOVASCSURG 1995;109:49-59. Egan TM, Detterbeck FC, Mill MR, et al. Improved results of lung transplantation for patients with cystic fibrosis. J TI4oP,ac CARDTOVASCSURG 1995;109:224-35. Griffith BP, Bando K, Hardesty RL, et al. A prospective randomized trial of FK506 versus cyclosporine after human pulmonary transplantation. Transplantation 1994;57:848-51.

Appendix Contributing centers, thoracic surgeons, pulmonologists, and coordinators are as follows: Allegemeine Krankenhaus, Vienna, Austria; Walter Klepetko, Bernhard Schlechta, Abelheid End (18 patients). Medizinische Hochschule Hannover, Hannover, Germany; HansJoachim Schgfers (14 patients). Presbyterian University Hospital, Pittsburgh, Pa.; Bartley Griffith, Robert Hardesty, Ann Lee, Debbie Opacic (12 patients). H6pital Beaujon, Clichy, France; Bernard Andr6assian, Jean-Pierre

The Journal of Thoracic and Cardiovascular Surgery Volume 110, Number 5

Duchatelle (10 patients). Baylor-Methodist Hospital, Houston, Tex.; Adaani Frost, George Noon, H. David Short, Janine O'Leary (10 patients). Toronto Hospital, Toronto, Ontario, Canada; Tim Winton, Janet Maurer, Massina Scavuzzo, Louise Won (9 patients). University of Minnesota Health Center, Minneapolis, Minn.; R. Morton Bolman, Marshall Hertz, Beth Dosland (6 patients). Loyola University Medical Center, Maywood, Ill.; James Houck, Edward Garrity (5 patients). St. Louis Children's Hospital, St. Louis, Mo.; Thomas Spray, George Mallory (5 patients). Washington University-Barnes Hospital, St. Louis, Mo.; Alec Patterson, Joel Cooper, Mary Pohl (4 patients). University of North Carolina Medical Center, Chapel Hill, N.C.; Tom Egan, Ellen Cairns (4 patients). Stanford University Medical Center, Stanford, Calif.; Bruce Reitz, Tom Bnrdon (3 patients). University of Michigan Medical Center, Ann Arbor, Mich.; Michael Deeb, Ros Florn (3 patients). University of Texas Medical Center, San Antonio, Tex.; John Calhoon, Kent Trinkle (3 patients). Montreal General Hospital, Montreal, Quebec, Canada; Hani Shennib (3 patients). Freeman Hospital, Newcastle-upon-Tyne, England; John Dark, Paul Corris (3 patients). Rikshospitalet, Oslo, Norway; Odd Gelran, Oystein Bjortuft (2 patients). University of Iowa Hospital, Iowa City, Iowa; Louis Lanza (2 patients). University of Pennsylvania Hospital, Philadelphia, Pa.; Larry Kaiser, Nancy Blumenthal, Angela Wurster (2 patients). Vanderbilt University Medical Center, Nashville, Tenn.; Jim Loyd, Bill Frist (2 patients). University of Wisconsin Medical Center, Madison, Wis.; Robert Love, Debbie Welter (2 patients). Papworth Hospital, Cambridge, England; John Wallwork (2 patients). H6pital d'Enfants de la Timone, Marseille, France; Dominique Metras (2 patients). H6pital XavierArnozan, Pessac, France; Louis Couraud, Claire Dromer (2 patients). Klinikum Grosshadern, Munich, Germany; Bruno Reichart, Florio Wagner (2 patients). Centre for Health Sciences, Winnipeg, Manitoba, Canada; Wayne Kepron, Helmut Unruh (1 patient). Brigham and Women's Hospital, Boston, Mass.; Stephen Mentzer, David Sugarbaker, Carol Coakley (1 patient). University of Virginia Health Sciences Center, Charlottesville, Va.; Curtis Tribble, John Truwit, Margaret Ball (1 patient). H6pital Erasme, Brussels, Belgium; Martine Antoine, Marc Estenne (1 patient). University Hospita], Rigshospitalet; Gosta Pettersson, Ulrik Gerner Svendsen (! patient). Helsinki University Central Hospital, Helsinki, Finland; Ari Harjula (1 patient). Centre MedicoChirurgical Foch, Suresnes, France; Alain Bisson (1 patient). Academisch Ziekenhuis Groningen, Groningen, The Netherlands; Wim de Boer (1 patient). Universitfitsspital Zfirich, Zfirich, Switzerland; Walter Weder (1 patient). Discussion Dr. C. G. A. McGregor (Rochester, Minn.). I would like to congratulate Dr. Novick and his colleagues for successfully forming the pulmonary retransplant registry. Clearly, no individual center develops enough experience with such relatively unusual techniques to be able to draw meaningful conclusions as to optimal management. For example, perhaps only 15 of the 77 lung transplant centers in the United States perform more than 10 transplants per year. I would encourage members of the Association to pursue similar efforts whether as registries of multicenter

Novick et al.

14 13

studies in pursuit of knowledge as to the best management of other rare clinical challenges in cardiothoracic transplantation. Examples that come to mind include cardiac transplantation for amyloid disease or cardiac tumors and pulmonary transplantation for systemic collagen diseases; there are many other examples. Pulmonary retransplant candidates comprise a group of patients with extremely challenging conditions in that they have received recurrent heavy immunosuppressive therapy for treatment of OB or recalcitrant rejection including steroids, cytolytic therapy, and sometimes total lymphoid irradiation. These patients are, therefore, in profoundly immunosuppressed conditions and not in a good state for retransplantation in terms of the likelihood of infection and perhaps the development of posttransplant lymphoproliferative disorders. It is no surprise, therefore, that 56% of the early deaths in this series were caused by infection. Only 29% of the retransplant recipients whose results are presented in this paper were ambulatory and half were ventilator dependent. Although the survival of 45% at 1 year is disappointing, it is important to note that those patients who survived for 90 days after operation had a survival thereafter comparable with that of primary lung transplant recipients. It is also interesting that the interval between transplantation and retransplantation and the reason for retransplantation did not have an impact on survival, unlike the situation in heart retransplantation in which early retransplantation, which is commonly necessitated by recalcitrant rejection, has a 1-year survival approximately half of that for late retransplantation, which is commonly done because of accelerated transplant atherosclerosis. This paper informs us that the preoperative status of the patient as reflected by ambulation immediately before retransplantation is the critical factor in a successful outcome. If, with this important information, we can reduce early mortality, then we know from this presentation that the rate of recurrence of OB will be no greater than that after a primary transplant and that satisfactory lung function will be achieved. I would like to ask Dr. Novick the following three questions. First, previous reports from the registry indicated that removal of the previously transplanted lung was important, but the current report suggests this is no longer true. Is this change simply a reflection of more numbers in the registry? Second, there seems to be a paradox in that being ambulatory is important in survival, and yet being ventilator dependent is not. Can you explain this? Third, can you comment on the contrast between the indication and time interval before retransplantation not being a factor in survival in pulmonary retransplantation compared with the situation in heart retransplantation? Dr. Noviek. In 1993 I presented a paper on 33 patients who underwent retransplantation because of OB. The survival of patients without an old, retained contralateral graft was significantly better than that of patients with an old contralateral graft. Two years have gone by, the number of patients in the registry has increased fourfold (to 139), and patient selection and postoperative treatment have improved. Likely, increased patient accrual to the pulmonary retransplant registry and these other factors have muted the importance of the presence of an old

14 14

N o v i c k et al.

contralateral graft as a predictor of poor outcome after retransplantation. There was no difference in survival between the 70 patients who were receiving ventilator support and the 69 patients who were not receiving ventilator support before retransplantation. Of the 70 patients who were on a ventilator, however, 33 were ambulatory; they were able to walk at least 50 m with assistance from the nursing and physiotherapy staff. Thus, if the patients are rehabilitated in every other way except for the ventilatory status before retransplantation, and if they survive the perioperative period, in all likelihood they will do well. Finally, concerning the interval between transplant operations, it is true that in cardiac retransplantation an interval of greater than 6 months correlates with a successful outcome. When analyzed as a continuous variable after pulmonary retransplantation, there is no relationship between the interval between transplants and survival. Dr. Alvaro Montoya (Chicago, Ill.). At Loyola University in Chicago we have done 135 lung transplants since 1991. Seven of these patients required retransplantation. We had four survivors. One of these survivors was a 21-year-old girl with cystic fibrosis in which OB developed. I elected to do a retransplant in her case. I retransplanted

The Journal of Thoracic and Cardiovascular Surgery November 1995

the heart and lung because of clinical feasibilities. OB developed in the new transplanted lungs 3 mouths after retransplantation and she died. We had been debating about doing a second retransplantation, but she died before we came to a conclusion. We now have six patients on the waiting list for retransplantation. One of these is a 32-year-old marathon runner who received double-lung transplantation because of alveolar cell carcinoma. Now he needs a retransplant and we are debating about whether to do it. We are committed to these patients. For their care we performed transplantation one time; maybe we should perform retransPlantation if it is required. My question is simple. Should we continue to do retransplants? We do not retransplant hearts. I have asked why and the answer was, we do not have donors. Should be continue to do retransplants on patients with lung disease? Dr. Noviek. In brief, I believe that as a minimum a pulmonary retransplant candidate should meet the same criteria as a first-time lung transplant candidate and perhaps be even more highly selected. Our findings, which show satisfactory results in the intermediate term of 90-day postoperative survivors of retransplantation, indicate that we should still b e retransplanting lungs, but that patient selection should be further refined.

B o u n d v o l u m e s a v a i l a b l e to s u b s c r i b e r s

Bound volumes of THE JOURNALOF THORACICANDCARDIOVASCULARSURGERYare available to subscribers (only) for the 1995 issues from the Publisher, at a cost of $91.00 for domestic, $118.77 for Canadian, and $111.00 for international subscribers for Vol. 109 (January-June) and Vol. 110 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNALname, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Subscription Serv!ees, 11830 Westline Industrial Drive, St. Louis, Missouri 63146-3318, USA; phone 1 (800) 453-4351 or (314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNALsubscription.