From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
Evidence That Sustained Growth Suppression of Intestinal Anaerobic Bacteria
Reduces the Risk of Acute Graft-Versus-Host Disease After Sibling
Marrow Transplantation
By Dietrich W. Beelen, Elsa Haralambie, Heinrich Brandt, Gotz Linzenmeier, Karl-D. Muller, Klaus Quabeck,
Herbert G. Sayer, Ullrich Graeven, Hossam K. Mahmoud, and Ulrich W. Schaefer
The influence of intestinal bacterial decontamination on the
occurrence of grades I1 t o IV acute graft-versus-host disease
(GVHD) was retrospectively analyzed in 194 predominantly
adult patients treated by genotypically identical sibling marrow transplantation under conditions of strict protective
isolation and intestinal antimicrobial decontamination. Fortyfive patients (23%) developed acute GVHD and univariate
analysis identified four features that significantly increased
the risk for this reaction: chronic myeloid leukemia as the
underlying disease, as compared with all other disease
categories (P < .0001); female marrow donors for male recipients, as compared w i t h other gender combinations
(P .005); ineffective, as compared with sustained growth
suppression of intestinal anaerobic bacteria (P < .006); and
methotrexate as the sole immunprophylactic compound, as
compared with cyclosporine containing regimens (P < .05).
Using the duration of anaerobic growth suppression as a
time-dependent explanatory variable, proportional hazards
regression analysis confirmed these features as independent
predictors for acute GVHD with relative risk estimates of 1.9
(95% confidence interval [CI], 1.3 t o 2.7) for the immunprophylactic regimen (P < .0004), of 1.8 (95% CI, 1.3 t o 2.5) for
the underlying disease (P .0005), of 1.7 (95% CI, 1.2 to 2.5)
for anaerobic decontamination (P .002), and of 1.3 (95% CI,
1.1 t o 1.6) for the donor/recipient gender combination
(P < .008), respectively. Best subset selection modeling also
identified the quality of anaerobic decontamination as the
third most important predictor for acute GVHD, when all four
significant features were included. Estimates of acute GVHD
stratified by the quality of anaerobic bacterial growth suppression showed a strong influence of anaerobic decontamination in patients burdened by at least one of the other
unfavorable factors (P < .009). In conclusion, this study provides strong evidence that sustained growth suppression of
intestinal anaerobic bacteria after clinical sibling marrow
transplantation can independently modulate the occurrence
of grades II t o IV acute GVHD, which is in concordance with
previous results from animal transplantation models. Antimicrobial chemotherapy specifically targeted t o the intestinal
anaerobic bacterial microflora may be complementarily useful in preventing acute GVHD and should be investigated in a
prospective trial.
0 1992by The American Society of Hematology.
A
Studies in germ-free or completely decontaminated rodents showed that the absence or complete elimination of
the intestinal microflora prevents the development of
delayed-type acute GVHD in recipient animals of MHCmismatched transplant^.^^-^^ In nonhuman primates, complete decontamination was found to be similarly effective in
preventing acute GVHD after MHC-mismatched, one-log
T-lymphocyte-depleted tran~p1ants.l~
The significance of the intestinal microflora in modulating the occurrence of acute GVHD after human marrow
transplantation currently remains controversial. Clinical
trials on the role of intestinal decontamination in human
marrow transplant recipients led to inconsistent results as a
consequence of nonuniform definitions of decontamination
efficacy, varying microbiologic techniques for isolation,
quantification and differentiation of the intestinal flora,
small patient numbers, and differing patient populat i o n ~ .Furthermore,
~~-~~
most of these analyses did not take
other potential factors influencing the pathogenesis of
acute GVHD into account.
The present report focuses on the association between
the suppression of the intestinal bacterial microflora and
the risk of acute GVHD after non-T-lymphocyte-depleted
HLA-genotypically identical sibling marrow transplantation in a larger single-center patient population.
CUTE GRAFT-VERSUS-HOST disease (GVHD) remains a leading cause of morbidity and mortality after
allogeneic bone marrow transplantation (BMT). It is commonly thought to result from differences between the
marrow donor’s and recipient’s gene products encoded by
the human major histocompatibility complex (MHC) and
other currently not well-defined polymorphic systems outside the MHC. Besides the outstanding significance of the
degree of histoincompatibility, the occurrence of acute
GVHD can be modulated by a variety of other factors that
have been partially identified by means of retrospective
analyses and comparative clinical trials. While the influence
of the type of immunpharmacologic prophylaxis, of patient
or donor age, and of the state of alloimmunization (by
transfusions or pregnancies) of females donating marrow
for male recipients on acute GVHD is generally regarded
as well-defined, other potential hazards await further confirmati0n.1-1~
From the Departments of Bone Marrow Transplantation, Medical
Microbiology, and Biomathematics, University Hospital of Essen,
Essen, Germany.
Submitted October IS, 1991; accepted July 24, 1992.
Supported by grants of Deutsche Forschungsgemeinschaft, Sonderforschungsbereich 102, TP E3, and C40.
Address reprint requests to Dietrich W Beelen, MD, Department of
Bone Marrow Transplantation, University Hospital of Essen, Hufelandstr. 55, W-4300Essen I , Germany.
The publication costs of this article were defrayed in pari by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C.section 1734 solely to
indicate thB fact.
0 I992 by The American Society of Hematology.
0006-4971/92/8010-0027$3.00/0
2668
PATIENTS AND METHODS
Two-hundred eleven patients received marrow transplants from
HLA-genotypically identical and mixed lymphocyte culture nonreactive siblings between the initiation of the marrow transplant
program at the University Hospital of Essen in December 1975 and
January 1989. Twelve of these patients were excluded from the
analysis because they refused to be treated under gnotobiotic
Blood, Vol80, No 10 (November 15), 1992: pp 2668-2676
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
2669
INTESTINAL ANAEROBIC BACTERIA AND ACUTE GVHD
conditions or because they were unable to tolerate the discomfort
inherent in the oral decontamination medication. Another five
patients died before hematopoietic engraftment and were therefore considered not evaluable for the development of acute
GVHD. The demographic characteristics, underlying diseases, and
details of the preparative regimens before marrow transplantation
of the remaining fully eligible 194 patients are summarized in
Table 1. Treatment of the patients enrolled in the transplant
program followed guidelines approved by the Committee on
Treatment of Human Subjects in Research at the University
Hospital Essen.
Table 1. Demographic Parameters and Treatment Modalities
Median age (yr) (range)
Median donor age (yr) (range)
Sex (F/M)
Donor sex (F/M)
Underlying disease (no. of patients)
Malignant disease
AML first CR
AML advanced disease
ALL first CR
ALL advanced disease
CML first CP
CML advanced disease
Others'
Nonmalignant disease
SAAt
Preparative regimen (no. of patients)
Malignant disease
Single-dose TBI Cy
FractionatedTBI Cy
BU Cy
Nonmalignant disease
28 (7-49)
27 (4-61)
97/97
97/97
CY
TNI Cy
lmmunoprophylaxis (no. of patients)
MTX
CSP
sMTX CSP
T-cell MoAbS CSP
Protective environment (no. of patients)
Laminar air flow system
Reverse isolation room
13
6
+
+
+
+
+
+
52
31
13
13
49
9
8
19
60
98
17
110
5
72
7
51
143
Abbreviations: AML, acute myeloid leukemia; ALL, acute lymphoblastic leukemia; SAA, severe aplastic anemia; CR, complete remission; CP,
chronic phase: TBI, total body irradiation (single, 1 x 8.6 Gy; fractionated, 4 x 2.5 Gy); Cy, 2 x 60 mg/kg of body weight cyclophosphamide
IV (malignant disease), 4 x 50 mglkg of body weight cyclophosphamide IV (nonmalignant disease); BU, 1 mg/kg of body weight busulfan
every 6 hours over 4 days orally; TNI, total nodal irradiation (4 x 3.2
Gy); MTX, methotrexate 15 mglm2 of body surface area IV on day 1.10
mg/mz MTX IV on days 3, 6, and 11, and once weekly until day 102;
sMTX, short course of methotrexate (15 mg/m2 of body surface area
MTX IV on day 1.10 mg/m2MTX IV on days 3,6, and 11; CSP, 3 mg/kg
cyclosporine as continuous infusion day -1 until day42.
*Four patients with myelodysplastic syndromes in leukemic transformation, two patients with acute osteomyelofibrosis, one patient with
medulloblastoma, and one patient with multiple myeloma.
tlncluding one patient with SAA associated with paroxysmal nocturnal hemoglobinuria.
SFive to 10 mg of murine MoAb directed to the constant part of the
cr/p T-lymphocyte receptor IV on days 10 through 19.
Gnotobioticmeasures and supportive care. Patients were treated
under conditions of strict protective isolation using either downstream laminar air flow systems with air filtration (n = 51) or
reverse isolation rooms with conventional air supply (n = 143) as
previously described." Assignment of patients to the type of
protective environment was based on its availability. Nursing staff,
physicians, and other attendants used aseptic techniques with
complete surgical dressing during patient contacts throughout the
isolation period, which was initiated at least 2 weeks before
transplantation and was generally maintained until discharge
(usually on day 50 after transplantation [day 0 designates the day of
marrow infusion]). With the intention to prevent infections and to
attain decontamination of the intestinal microflora, nonabsorbable
antibiotic and antimycotic compounds (320 mg gentamycinl
tobramycin or 600 mg netilmycin in combination with 2,800 mg
amphotericin-B and 5.2 x 1 8U nystatin) were administered orally
in four divided daily doses. The first 22 patients received in
addition four daily oral doses of 500 mg cephazolin solution. If
analyses of fecal samples indicated colonization of the gastrointestinal tract with bacteria resistent to the decontamination medications, additional treatment with oral or parenteral antibiotics was
initiated according to results of sensitivity testings. Isolation and
oral decontamination measures were supplemented by daily skin
cleansing in a sterile water bath containing 1-propanol, 2-propanol,
and 2-biphenylol, disinfection of the oral mucous membranes using
polyvinyl-pyrrolidon-iodinesolution three times daily, restriction
to autoclaved nutrients or sterile beverages, and sterilization of all
items introduced into the protective environment.
To prevent pmumocysfis carinii infections, oral trimethoprimsulfamethoxazole was administered over a 2-week period until the
day before transplantation. For treatment of suspected or documented bacterial infections, all granulocytopenic patients developing fever greater than 38.5"C received a systemic broad-spectrum
antibiotic chemotherapy consisting first-line of an ureido-penicillin
in combination with an aminoglycoside and a penicillinaseresistent penicillin. If signs of infection persisted despite adequate
antibacterial treatment and a deep or systemic mycotic infection
was suspected or documented, intravenous (IV) amphotericin B
was additionally instituted. Herpes virus infections were treated
with IV acyclovir, which was generally not used for the prevention
of endogenous reactivation in seropositive patients. A variety of
mostly unsuccessful approaches (eg, high doses of IV acyclovir
alone or in combination with hyperimmunglobulines) were used
for the treatment of infections with cytomegalovirus (CMV) hefore
ganciclovir became available.
During the early posttransplant period, all patients received a
regular substitution with platelet concentrates prepared by blood
cell separators until self-sustaining platelet counts in excess of 20 x
109/Lwere reached. If the peripheral blood hemoglobin content
declined below 8 g/dL, packed red blood cells were substituted.
Since the mid 1980s, blood preparations from CMV antibodynegative blood donors were used for all patients irrespective of
their CMV serostatus. All blood products had been y-irradiated
immediately before transfusion with 15 to 20 Gy.
GVHD. Pharmacologic prophylaxis of acute GVHD consisted
of IV methotrexate (MTX) according to the standard Seattle
protocol in the first 110 patients.25 Seventy-two patients received a
regimen with a short course of MTX in combination with continuous intravenous cyclosporine (CSP), as bas been described previously.%CSP alone was used in five patients due to contraindications to
MTX. The remaining seven patients received a monoclonal antibody directed to the a/p T-lymphocyte receptor between days 10
and 19 after transplantation in combination with CSP (Table
l).27,28
The assessment and grading of acute and chronic GVHD
was performed on the basis of clinical findings together with
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
2670
BEELEN ET AL
histologic evaluation of biopsies from skin, liver, or gastrointestinal
on these three time intervals, an explanatory variable was entered
tract and followed the commonly accepted criteria proposed by
with different values in the logistic regression models. In addition,
Glucksberg et aIz9and Thomas et aL30
a three-level time variable corresponding to these time intervals
was analyzed as an explanatory variable in the model. To account
Microbiologic surveillance. Starting at admission, samples of
oral washings, urine, feces, and swabs from multiple skin sites were
for the interactions of acute GVHD and intestinal bacterial
obtained at least once weekly for bacterial and mycotic examinadecontamination, the duration of bacterial growth suppression was
defined by a time-dependent covariate function. Bacterial growth
tions. All fecal samples were regularly cultured for aerobic,
suppression took different values at a given time point after
microaerophilic, and anaerobic bacteria. Details of the culture
techniques used for the present analysis have been p u b l i ~ h e d . ~ ' . ~ ~transplant depending on whether or not the duration of a patients'
decontaminated status exceeded that time interval. Thus, the value
Aerobic cultures were monitored for Escherichia coli and other
of bacterial growth suppression changed for patients in each risk
potential pathogens (including klebsiella, proteus, enterobacter,
set. Because multiple comparisons were performed, only those
pseudomonas) as well as for enterococcus, lactobacillus, and
explanatory variables with a two-tailed significance level of less
staphylococcus species. Samples for anaerobic cultures were immethan 1% after adjustment for all variables in the model were
diately processed after collection. These cultures allowed for the
considered significant. Maximum partial-likelihood estimates and
detection of different species of the genera bifidobacterium,
estimates of conditional risk ratios and their 95% confidence
bacteroides, fusobacterium, clostridium, eubacterium, peptococintervals (CI) were derived from regression analyses after adjustcus, peptostreptococcus, and veillonella?* Quantitative growth of
ments for all significant variables in the model^?^,^^
bacteria was expressed as the common logarithm of colony-forming
Best subset selection analysis was performed for all significant
units (CFU) with a threshold of detection below lo3CFU per gram
variables of the proportional hazards regression analysis using the
of sample. While we were aware of some degree of intersection, we
leap and bound algorithm of Furnival and Wilson.38The criterion
categorized bacteria according to culture growth conditions as
used to determine the best model was based on the global score x2
either aerobic or anaerobic. Culture proof of any bacterial growth
statistic. All analyses were performed on data prospectively colwould have classified the respective fecal samples as contaminated.
lected in the BMT program data base of the University Hospital of
In patients who later developed acute GVHD, only samples taken
Essen.
before diagnosis of this immunologic reaction were considered
eligible for the analysis. The observation period of microbiologic
RESULTS
surveillance covered 1 week before and 5 consecutive weeks after
transplantation.
Decontamination efficacy. The overall intestinal bacteThe pretransplant serologic status for herpes simplex virus,
rial growth suppression achieved in the present study is
varicella zoster virus, and CMV of recipients and donors was tested
given in Table 2. The median bacterial content of fecal
in the vast majority of individuals by enzyme-linked immunosorsamples showed a steep decline 2 weeks after marrow
bent assay (ELSA). In the first 20 donor/recipient combinations,
transplantation
and remained on a low level until the end of
complement fixation assay was exclusively used.
the
surveillance
period. Complete suppression of bacterial
Srutisticul analysis. All analyses involving the clinical grades of
growth, as expressed by the percentage of germ-free fecal
severity of acute GVHD as an end point were performed on a
samples, was seldom attained during the week before
two-level classification (patients with grades 0-1 versus those with
transplantation, but reached up to 65% posttransplant
grades 11-IV acute GVHD). To evaluate the influence of intestinal
bacterial decontamination on the development of acute GVHD,
(Table 3). The percentage of completely germ-free samples
patients were classified as being sustained decontaminated if
of patients treated in laminar air flow systems was significontinuous bacterial growth suppression could be shown in the
cantly increased over that of patients in reverse isolation
time interval between the day of marrow transplantation and the
rooms (Table 3). This also applied to anaerobic bacterial
end points of the analysis. In patients surviving beyond the
growth suppression, but reached only marginal significance
observation period of microbiologic surveillance, the end point of
for aerobic bacteria (P= .06) (Table 3). The numbers of
analysis was day 35. If patients contracted acute GVHD or
patients classified as being sustained decontaminated for
deceased without acute GVHD before day 35, the respective event
the aerobic, the anaerobic, and the entire bacterial microtimes were taken as end points of the analysis. These definitions
flora were 47 (24%), 41 (21%), and 15 @%), respectively.
were applied to the aerobic, the anaerobic, as well as to the entire
intestinal bacterial microflora.
Differences between frequencies were compared by the twoTable 2. Growth Suppression of Intestinal Bacteria (CFU per gram of
tailed Fisher's exact test. Estimates of failure times were deterfecal sample) in the Microbiologic Observation Period
mined by the product limit method with right-censoring of subjects
Bacterial
at the last time point at which they were at risk for a given event.33
Growth (%)
Time'
The Mantel-Haenszel test was used for testing the homogeneity of
Prestudy
1.2 x 1010 1 x 109
survival functions across ~ t r a t a . 3To
~ further define their influence
-1
3.1 x lo9 2 3.4 x lo8 (26)
on failure times, explanatory variables with a significance level of
+1
1 x lo9 2 8.9 x lo7 (8)
less than 10% in univariate analysis were included in proportional
t2
1 x lo4 2 1 x lo3 (0.00008)
hazards logistic regression models using forward and backward
+3
i x 105 2 9.7 x 103 (o.ooo8)
stepwise selection processes.3s To account for the retrospective
+4
1
x 104 2 1 x 103 (o.oooo8)
nature of the analysis, all explanatory variables were associated
+5
1 x 107 1 . 1 x 106 (0.08)
with discrete time intervals of the study period before model
building. These time intervals were chosen according to changes in
Median absolute number of CFU per gram of fecal sample in each
major treatment strategies that might have influenced the occurtime interval ? standard error of the median. Numbers in brackets
rence of acute GVHD. Changes in major treatment strategies
represent percentages of CFU in each time interval as compared with
during the study period were the introduction of reverse isolation
prestudy values.
rooms in 1982 and of CSP-containing regimens in 1986. Depending
*Weeks before and after transplantation.
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
267 1
INTESTINAL ANAEROBIC BACTERIA AND ACUTE GVHD
Table 3. Decontrmlnotlon Efficacy (percemtages of germ-free fecal samples) by Isolotlon S p t m and Acute GVHD
lsolstoon System
-
LAF
Time'
-1
+1
+2
+3
+4
+5
Total
Bs*uist
Aerobic
Anaerobic
All
Aerobic
Anaerobic
All
Aerobic
Anaerobic
All
Aerobic
Anaerobic
All
Aerobic
Anaerobic
All
Aerobic
Anaerobic
All
Aerobic
Anaerobic
All
In
51):
-
RIR
In
l431t
46
NS
NS
NS
NS
NS
< .01
< .006
NS
< .005
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
< .03
33
<.om
34
34
7
5
63
40
36
79
66
11
5
49
32
20
57
55
41
63
62
64
58
65
43
65
71
48
56
65
45
60
53
41
P Value5
46
65
63
48
56
59
41
54
-
-
-
1l.N
In
451:
PVslue5
21
10
5
36
21
9
61
43
35
59
43
35
63
45
37
64
61
46
51
36
27
37
10
5
57
37
28
63
63
51
62
69
48
66
72
52
56
61
42
57
51
37
Abbreviations: LAF, laminar air flow system: RIR. reverse isolation room; NS, not significant.
*Weeks before and aher transplantation.
tNumben of bacterial cultures included in the analysis: LAF, n .c 670; RIR. n = 1,912; grades 0-1, n
were considered eligible until the end points of the analysis.
*Numbers of patients included in subsets of the analysis.
ISignificance by two-tailed Fisher's exact test.
Acute G W D . Of the 194 patients. 45 (23%) developed
acute GVHD of grades 11-IV at a median of 33 days (range,
15 to 52 days). Significantly less samples from these patients
were decontaminated from all bacteria as well as from
anaerobic bacteria when compared with the frequency of
decontaminated samples from patients with grades 0-1
acute GVHD (Table 3). sustained growth suppression of
anaerobic bacteria was associated with a striking and
significant reduction of the cumulative incidence of acute
GVHD (P < .OM) (Fig 1) and this resulted from a significant reduction in each target organ involvement with acute
GVHD (skin: 8% t 8% v 30% f 8% [ P < .OM]; liver:
3% t 3% v 16% t 6% [P < .MI;intestinal tract: 0% v
19% t 6% [P < .W]).
In contrast, no significant association between growth suppression of aerobic bactcria and
acute GVHD could be dctcctcd (Fig 1). The same applicd
to patients in whom sustaincd growth suppression of the
entire bacterial microflora could be demonstrated (Fig 1).
and this was most probably connected to the small number
of patients who achieved complete decontamination.
To assess other potentially relevant factors that might
have contributed to the development of acute GVHD,
categorized features (Table 4) wcrc first tested in univariate
analysis. Among these, the underlying discasc had the
strongest influcncc on acute GVHD, with a more than
2.5-fold higher cumulative incidcnce for patients with
chronic myeloid lcukcmia (CML) as opposcd to othcr
disease categories (P < .MMl). Recipients of marrow from
a donor of an opposite gender had a higher probability of
Acute GVHO of Grades
0.1
(n 1491t
NS
NS
NS
< .005
< .02
< .002
NS
< .02
< .05
NS
< .002
NS
NS
< .002
NS
NS
NS
NS
NS
< .0005
< .002
2,020: grades Il-IV, n = 562. Bacterialcultures
acute GVHD as compared with sex-matched combinations
(P < .03). In male recipients of female donor transplants,
the risk of acute GVHD was increased nearly twicc compared with other donor/rccipicnt gender combinations
(P < .OM). This was further enhanced by an alloimmu-
40
nr.
ru.
30
20
l4f9
10
0
8f8
d
47
147
rmMc
buruia
.
41
153
M..mbk
b.CM1.
129
15
sll
M.h
Fig 1. Product-lhnlt dmmtes (i953CCI) of gf.d.r I1 t o N u u t e
GVHD stmtlfled by the quality of intestinal bacterial growth suppression. Patients were classified as being sustained decontaminated if
complete growth suppression of the respective intestinal bacteria
was attained in the time period between the day of marrow transplantation and the end point of the analysis. Numbers of patients are
indicated at the bottom of bars. Significances were obtained by the
Mantel-Haenuel test.
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
BEELEN ET AL
2672
Table 4. UnivariateAnalysis of Patient, Donor, and Treatment
Characteristicson the Risk of Acute GVHD
No. of
Patients
Age* (vr)
5 28
> 28
Age category (yr)
5 20
21-30
31-40
241
Donor age* (yr)
s 27
> 27
Donor age category (yr)
s 20
21-30
31-40
241
Cumulative Incidence
of Grades IIto IV
Acute GVHD*
PValuet
99
95
19f8
30f 10
NS
31
77
56
30
2 2 2 16
19 c 9
32k 13
33f 18
NS
102
92
21 k 9
302 10
NS
28
2 0 2 16
84
54
28
21 k 9
272 12
40 k 21
NS
97
97
21 f 9
30f 10
NS
97
97
27 9
23 2 9
49
49
48
48
98
96
146
48
170
24
142 10
2 0 k 12
26.- 13
40 2 14
17 2 8
33f 10
21 f 7
402 14
22 2 7
52 f 21
83
26
58
27
58
136
15 f 8
18 f 16
45f 13
18.- 14
45% 13
1727
< .0004
110
84
32 f 8
18f9
< .05
133
61
26 f 8
18f 10
NS
51
143
15 f 10
29 f 8
< .07
116
78
106
88
74
120
30 f 9
18* 10
25 f 8
2 6 2 10
302 12
22 2 8
44
Sex
F
M
Donor sex
F
M
Donorlrecipient sex combination
F+F
M-M
M-F
F-M
Sex-match
Sex-mismatch
All except F + M
F-M
All except sF M
sF + M
Disease category
AML
ALL
CML
SAA and others
-
CML
all others
lmmunprophylaxis
MTX
CSP-containing regimens
Disease stage
Standard risk
High risk
Protectiveenvironment
Laminar air flow system
Reverse isolation room
Pretransplant herpes virus serology
.-
NS
< .03
< .03
< ,005
<.001
< ,0001
CMV
Recipient -ve
Recipient +ve
Donor -ve
Donor +ve
Recipient and donor -ve
Recipient andlor donor +ve
NS
nized female donor state (P < .OOl). Pharmacologic immunprophylaxis with CSP-containing regimens decreased
the cumulative incidence of acute GVHD as opposed to
MTX alone (P < .05). Furthermore, patients nursed in
laminar air flow systems tended to have a reduced risk of
developing acute GVHD as compared with those treated in
reverse isolation rooms (P < .07).
Features with no detectable influence on the risk of acute
GVHD in univariate analysis were donor and recipient age
(by decades), donor and recipient pretransplant herpes
serology, and the stage of the underlying disease (Table 4).
The significant features in univariate analysis were then
evaluated in a stepwise proportional hazards regression
analysis. As indicated in Table 5, all four explanatory
variables found to be significant in univariate analysis
model. Using subset selection analysis, the best sequence of
the four significant variables selected by the global score
criterion was as follows: (1) immunprophylactic regimen;
(2) underlying disease; (3) growth suppression of intestinal
anaerobic bacteria; and (4) gender matching (global score
value, 36.6).
Estimates of acute GVHD stratified by the quality of
anaerobic bacterial decontamination were then calculated
depending on whether other unfavorable prognostic factors
were present. This analysis showed that in patients burdened by at least one of the unfavorable factors, the risk of
acute GVHD was reduced by 50% if sustained anaerobic
bacterial growth suppression was attained (P < .009) (Fig
2).
Chronic G W D . One hundred fifty of the 194 patients
(77%) survived more than 70 days after transplant and were
thus eligible for chronic GVHD. With a median onset at
day 91 (range, 45 to 555 days), 58 of these patients (39%)
contracted clinical symptoms of chronic GVHD, corresponding to a cumulative incidence of 43%. The clinical grades of
severity of chronic GVHD were scored as extensive in 35
patients (23%) and as limited in 23 patients (15%), respectively.
Preceding acute GVHD most strongly influenced the
development of chronic GVHD (P < .0001). Other signifi-
NS
Table 5. Multivariate Analysis of Risk Factorsfor Acute GVHD
NS
HSV
Recipient -ve
Recipient +ve
Donor -ve
Donor +ve
Recipient and donor -ve
Recipient andlor donor +ve
150
55
139
24
170
2 2 + 14
26 k 8
30k 12
23 8
272 18
25 f 8
Recipient -ve
Recipient +ve
Donor -ve
Donor +ve
Recipient and donor -ve
Recipient andlor donor +ve
Recipient -ve for 1 to 3 viruses
Recipient +ve for all 3 viruses
a9
105
78
116
43
151
153
41
25k 10
25 f 8
18 f 10
31 f 10
182 12
28 k 8
28 2 8
15 f 12
vzv
Factor
NS
95% C l t
PValueS
1.9
1.3-2.7
< .0004
1.8
1.3-2.5
< .0005
1.7
1.2-2.5
< ,002
1.3
1.1-1.6
<.008
lmmunprophylaxis (MTX v
NS
CSP regimens)
NS
NS
Relative
Risk"
Disease category (CML v
other diseases)
Anaerobic decontamination
(not sustained v sustained)
NS
Gender combinations (F + M
NS
v other combinations)
NS
" M a x i m u m partial-likelihood estimates in stepwise proportional hazAbbreviations: sF, sensitized female marrow donor (by transfusions or pregnancies); standard risk, first remission of acute leukemia, first chronic phase chronic
myeloid leukemia, severe aplastic anemia; high risk, advanced malignancies; HSV,
herpes simplex virus; VZV, varicella zoster virus; -ve, negative pretransplant
herpes virus serology; +ve, positive pretransplantherpes virus serology.
*Product-limit estimates (+95% confidence limits).
tsignificance derived from comparisons between strata using the MantelHaenszel test.
*Categorized by age median.
ards regression analysis after adjustment f o r all covariates in t h e model.
Time period of transplant as w e l l as all other factors f o u n d t o b e
significant by univariate analysis d i d n o t reach t h e level of significance
(P
< .01) t o enter m o d e l building.
t U p p e r a n d lower b o u n d s of 95% confidence interval derived f r o m
the regression model.
*Significance i s based o n an asymptomatic
xz distribution.
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
2673
INTESTINAL ANAEROBIC BACTERIA AND ACUTE GVHD
504
x
0
1
I
0
0
- 3 nskkmrs - dgatamnaMm (n=lZl) D
I I
I
M)fW!aciu++(n-8)A
.
1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 1 0 0
Days post transplantation
Fig 2. Product-limit estimates of grades I1 to IV acute GVHD
stratified by the quality of intestinal anaerobic decontamination in
patients with no other risk factor (curves A and B) or with any of the
other risk factors (curves C and D), as identified by proportional
hazards general linear regression analysis. Risk factors were the
disease category (CML v other diseases), the gender combination
(female donors for male recipients v other combinations), and the
type of immunprophylaxis (sole MTX- v CSP-containing regimens).
The difference between curves C and D is significant (Pe .OM).
cant factors identified by univariate analysis were the
underlying disease ( P < .mol), the immunprophylactic
regimen (P < .0002), the donor/recipient gender combination (P < .003), and the type of protective environment
(P < .04). Multivariate analysis, however, confirmed only
preceding acute GVHD as an independent predictor for
chronic GVHD, with a relative risk estimate of 3.3 (95% CI,
2.3 to 4.6) compared with patients without acute GVHD
(P < .0001).
DISCUSSION
This retrospective study was aimed at defining the
contribution of growth suppression of the intestinal bacterial microflora on the risk of acute GVHD after clinical
sibling marrow transplantation. Nearly identical techniques
for protective isolation and intestinal bacterial decontamination, as well as for microbiologic surveillance were used
in a predominantly adult patient population over a time
period of 13 years, enabling us to accumulate comprehensive data on this topic. Previous clinical trials dealing with
the association between the intestinal microflora and acute
GVHD generally involved considerably smaller patient
numbers as well as differing (eg, pediatric patients, HLAidentical and nonidentical transplant recipients) populat i o n ~ . ~Most
~ , importantly,
~ ~ ~ ~ ~ these
, ~ ~trials were generally
focused on this association without adequate statistical
evaluation of other contributing factors for this immunpathologic condition.
Sustained complete intestinal bacterial decontamination
could only be shown in a small minority of patients in the
present analysis, which is in contrast to the higher proportions of successfully decontaminated patients previously
reported by other^.^^,^^ However, there are several impor-
tant distinctions in evaluating the effects of intestinal
decontamination between the cited trials, which may explain in part these differences. Among others, these include
different regimens of antimicrobial medications, differing
observation periods of microbiologic surveillance, and nonuniform definitions of complete decontamination. Microbiologic monitoring did not include culturing of strictly
anaerobic bacteria in some trials, while other studies,
including this study, analyzed the complete spectrum of
intestinal bacteria. Further difficulties in comparing decontamination efficacy result from the generally sparse descriptions of the threshold for detection of bacterial growth in
most trials and undefined end points of evaluation in
patients contracting acute GVHD. The latter point is of
critical importance, because acute GVHD by itself may
influence decontamination efficacy due to a reduced ability
of affected patients to tolerate the oral antimicrobial
medications or by shortened intestinal contact times and
lower concentrations of these compounds as a consequence
of diarrhea. To allow meaningful comparisons, we therefore considered patients only evaluable for decontamination efficacy until the diagnosis of acute GVHD was
established.
The lowest frequencies of decontaminated samples were
observed in the immediate pretransplant period, which on
the one hand may point to incomplete patient compliance
during high-dose radiochemotherapy preceding transplantation or, on the other hand, may indicate that longer time
periods before transplantation are required to achieve a
decontaminated state. In contrast, 30% to 70% of samples
during the posttransplant observation period did not contain detectable numbers of either anaerobic or aerobic
bacteria. Whether the increased posttransplant decontamination efficacy reflects better tolerance of the oral antimicrobial medication or other factors adding to the suppression
of intestinal bacterial growth, eg, the indispensible administration of systemic antibiotics in severely granulocytopenic
patients, was not distinguishable. Using otherwise identical
procedures, the decontamination efficacy of patients in
laminar air flow systems was higher compared with patients
in reverse isolation units. The reasons for this phenomenon
remain essentially unknown and deserve further investigation. However, the protective environment alone had no
influence on acute GVHD after adjustment for significant
factors contributing to this condition.
The key finding in the present analysis was the markedly
reduced risk of acute GVHD of grades I1 to IV in patients
in whom sustained growth suppression of the anaerobic
bacterial microflora was attained. A comparable low incidence of acute GVHD was also observed in the small
fraction of completely decontaminated patients. However,
this small patient number was insufficient to show a
statistically significant advantage with regard to the modulation of acute GVHD opposed to the group of patients that
failed to achieve sustained growth suppression of the entire
intestinal bacterial microflora.
Studies in germ-free or completely decontaminated experimental animals that showed that elimination or complete
suppression of the intestinal microflora can abrogate acute
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
2674
GVHD even in recipient animals of MHC-mismatched
transplants, have stimulated the clinical application of
gnotobiologic techniques for the prevention of acute
GVHD.15-19However, the biologic basis for this immunmodulatory effect is currently poorly understood. It has been
hypothesized that increased lymphokine secretion of phagocytes stimulated by endotoxines or other bacterial products
penetrating the intestinal mucosa may lead to a nonspecific
activation and polyclonal expansion of graft-derived T
lymphocyte^.^^ This, in turn, may promote a specific alloimmune response of donor lymphocytes against nonshared
histocompatibility antigens of the recipient tissue. Experimental evidence from studies in rodents also suggests that a
specific allorecognition process of donor T lymphocytes
may be induced by cross-reactions between intestinal bacterial antigens and antigens of the intestinal mucosal epithelium.40
In contrast to defined settings in animal studies, the
clinical pathogenesis of acute GVHD is superimposed by a
variety of predisposing and partially undefined factors, and
further depends on treatment strategies such as the type of
immunpharmacologic pr~phylaxis.~-l~
To evaluate the effect
of intestinal decontamination on a clinical basis, theinultifactorial pathogenesis of this immunpathologic reaction
has, therefore, inevitably to be taken into consideration.
Using the duration of sustained intestinal bacterial growth
suppression as a time-dependent variable, we applied
proportional hazards regression analysis techniques with
adjustments for other risk factors of acute GVHD and
confirmed consistently that decontamination of intestinal
anaerobic bacteria independently reduced the probability
of acute GVHD. Recent experimental data from a murine
transplantation model also support that the anaerobic
intestinal microflora has a dominant influence on the
occurrence of acute GVHD.39Whether this is connected to
specific properties of anaerobic bacteria currently remains
to be defined. However, it can be speculated that complete
suppression of the preponderant resident anaerobic microflora in the intestinal lumen, which amounts to greater than
99% of all bacteria, results in a some log higher quantitative
reduction of the bacterial load compared with successful
decontamination of intestinal aerobic bacteria. These quantitative relationships, in turn, might explain why the proposed interactions between intestinal bacteria and grafted
immune cells could be influenced more strongly by anaerobic than by aerobic microorganisms.
With regard to other significant hazards for acute GVHD
in the present study, the influence of the underlying disease
must be regarded currently as controversial in light of large
retrospective analyses from the European and International Bone Marrow Transplant Registries that did not
demonstrate an independent association between the underlying disease and acute GVHD.@ The probability of acute
GVHD for CML patients in the present study is in good
concordance with the majority of reports from single
institutions and registry analyse~.l93.~,~J~
In contrast, patients with other underlying diseases showed a comparatively low probability of acute GVHD. This difference was
BEELEN ET AL
clearly not connected to the age of recipients or donors,
because both had no influence on the risk of acute GVHD.
Splenectomy did not have to be considered as another
potential factor for an increased risk of GVHD in CML
patients in the present analysis, because only 3 of the 194
patients underwent splenectomy before transplant.1°J3
Therefore, one has to take other factors into consideration
that might explain the effect of the underlying disease on
acute GVHD. In this context, a report of the Basel
Transplant Group in which a strong association between an
increased culture growth of blood-derived macrophages
from CML patients and the development of acute GVHD
has been described is of interest.4l The investigators postulated that excess macrophages may trigger GVHD in
patients with CML by enhancing presentation of recipient
antigens to donor T lymphocytes. Further evidence that
CML patients might be particularly prone to acute GVHD
comes from recent investigations, in which an increased
early release of tumor necrosis factor-a (TNF-a) was
preferably found in marrow graft recipients with CML (E.
Holler, personal communication, and Holler et a142).TNF-a,
produced primarily by macrophages, is increasingly recognized as an important mediator of different steps in the
pathogenesis of acute GVHD.424 It might be speculated
therefore that increased TNF-a levels produced by macrophages of recipients with CML additionally promote the
development of acute GVHD and that this pathway is less
sensitive to gnotobiotic measures.
The improved prophylactic efficacy of combined CSPcontaining regimens (especially a combination with a short
course of MTX) compared with MTX alone has been
shown in a prospective randomized trial for patients with
severe aplastic anemia and is further supported by retrospective analyses.2JI The effect of female marrow donors for
male recipients is in concordance with most reports from
single centers and the registrie~.5-~,~,l~,l~
Although we observed the strongest influence in male recipients of transplants from alloimmunized female donors in univariate
analysis, multivariate analysis confirmed an independent
hazard of female donors for male recipients irrespective of
donor alloimmunization.
Patient or donor age were not independent hazards for
acute GVHD in the present study. Most other studies
examining age as a risk factor failed to show an incremental
risk of acute GVHD with increasing age in patients older
than 20 y e a r ~ . ~ - ~It .thus
~ - ~appears
J~
justified to assume that
the lack of an age-dependent influence on acute GVHD
resulted from the small proportion of pediatric and adolescent patients (10%) in the present study. Furthermore,
pretransplant herpes virus serology of recipients or marrow
donors did not contribute to the risk of GVHD. This is in
contrast to some, but not all, retrospective analyses dealing
with pretransplant herpes virus serology as a risk factor for
GVHD and, therefore, this issue cannot be regarded as
precisely defined at
The only independent predictor for chronic GVHD was
preceding grades I1 to IV acute GVHD. This is in concordance with a recent report from the International Bone
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
2675
INTESTINAL ANAEROBIC BACTERIA AND ACUTE GVHD
Marrow Transplant Registry by Atkinson et a!l5 However,
it might be argued that diminishing the risk of acute GVHD
by anaerobic decontamination at the same time prevents
the development of (secondary) chronic GVHD.
In conclusion, the present analysis provides for the first
time strong evidence that the intestinal anaerobic microflora independently influences the clinical pathogenesis of
acute GVHD after allogeneic sibling marrow transplantation and confirms similar observations from experimental
animal transplantation models. Antimicrobial chemotherapy specifically targeted to intestinal anaerobic bacteria
should therefore be investigated in a prospective trial as an
adjunct for the prevention of acute GVHD.
ACKNOWLEDGMENT
We are indebted to all supporting institutions of the University
Hospital of Essen for their cooperation in this study; to the nursing
staff of the Department of Bone Marrow Transplantation for their
excellent care of the patients; and to the technicians of the
Department of Medical Microbiology, whose outstanding efforts
made this work possible.
REFERENCES
1. Storb R, Deeg HJ, Whitehead J, Appelbaum F, Beatty P,
Bensinger W, Buckner CD, Clift R, Doney K, Farewell V, Hansen
J, Hill R, Lum L, Martin P, McGuffin R, Sanders J, Stewart P,
Sullivan K, Witherspoon R, Yee G, Thomas ED: Methotrexate and
cyclosporine compared with cyclosporine alone for prophylaxis of
acute graft-versus-host disease after marrow transplantation for
leukemia. N Engl J Med 314729,1986
2. Storb R, Deeg HJ, Farewell V, Doney K, Appelbaum F,
Beatty P, Bensinger W, Buckner CD, Clift R, Hansen J, Hill R,
Longton G, Lum L, Martin P, McGuffin R, Sanders J, Singer J,
Stewart P, Sullivan K, Witherspoon R, Thomas ED: Marrow
transplantation for severe aplastic anemia: Methotrexate alone
compared with a combination of methotrexate and cyclosporine for
prevention of acute graft-versus-host disease. Blood 68:119,1986
3. Storb R, Deeg HJ, Pepe M, Appelbaum F, Anasetti C, Beatty
P, Bensinger W, Berenson R, Buckner CD, Clift R, Doney K,
Longton G, Hansen J, Hill R, Loughran T, Martin P, Singer J,
Sanders J, Stewart P, Sullivan K, Witherspoon R, Thomas ED:
Methotrexate and cyclosporine versus cyclosporine alone for prophylaxis of acute graft-versus-host disease in patients given HLAidentical marrow grafts for leukemia: Long-term follow-up of a
controlled trial. Blood 73:1729,1989
4. Bross DS, Tutschka PJ, Farmer ER, Beschorner WE, Barine
HG, Mellits ED, Bias WB, Santos G W Predictive factors for acute
graft-versus-host disease in patients transplanted with HLAidentical bone marrow. Blood 63:1265,1984
5. Atkinson K, Farrell C, Chapman G, Downs K, Penny R, B i g s
J: Female marrow donors increase the risk of acute graft-versushost disease: Effect of donor age and parity and analysis of cell
subpopulations in the donor inoculum. Br J Haematol63:231,1986
6. Gale RP, Bortin MM, van Bekkum DW, Biggs JC, Dicke KA,
Gluckman E, Good RA, Hoffmann RG, Kay HEM, Kersey JH,
Marmont A, Masaoka T, Rimm A, van Rood JJ, Zwaan FE: Risk
factors for acute graft-versus-host disease. Br J Haematol 67:397,
1987
7. Gratama JW, Zwaan FE, Stijnen T, Weijers TF, Weiland HT,
D’Amaro J, Hekker AC, The TH, De Cast GC, Vossen JMJJ:
Herpes-virus immunity and acute graft-versus-host disease. Lancet
1:471,1987
8. Gratwohl A, Hermans J, Barrett AJ, Ernst P, Frassoni F,
Gahrton G, Granena A, Kolb HJ, Marmont A, Prentice HG, Speck
B, Vemant JP, Zwaan FJ: Allogeneic bone marrow transplantation
for leukaemia in Europe. Report from the working party on
leukaemia, European group for bone marrow transplantation.
Lancet 1:1379,1988
9. Bostrom L, Ringdtn 0, Gratama JW, Jacobsen N, Prentice
HG, Zwaan FE, Nilsson B for the Leukaemia Working Party of the
European Group for Bone Marrow Transplantation: A role of
herpes virus serology for the development of acute graft-versushost disease. Bone Marrow Transplant 5321,1990
10. Bostrom L, Ringdtn 0, Jacobsen N, Zwaan FE, Nilsson B
for the Leukaemia Working Party of the European Group for Bone
Marrow Transplantation: A European multicenter study of chronic
graft-versus-host disease. The role of cytomegalovirus serology in
recipients and donors-acute graft-versus-host disease, and splenectomy. Transplantation 49:1100,1990
11. Gratwohl A, Hermans J, Barrett AJ, Ernst P, Frassoni F,
Gahrton G, Granena A, Kolb HJ, Prentice HG, Vernant JP,
Zwaan FE: Allogeneic bone marrow transplantation for leukemia
in Europe: Regional differences. Report from the leukemia working party of the European group for bone marrow transplantation.
Bone Marrow Transplant 5:159,1990
12. Flowers MED, Sullivan Pepe M, Longton G, Doney KC,
Monroe D, Witherspoon RP, Sullivan KM, Storb R: Previous
donor pregnancy as a risk factor for acute graft-versus-host disease
in patients with aplastic anaemia treated by allogeneic marrow
transplantation. Br J Haematol74:492,1990
13. Michallet M, Corront B, Bosson JL, Molina L, Peissel B,
Maraninchi D, Reiffers J, Chabannon C, Gaspard MH, Stoppa
AM: Role of splenectomy in incidence and severity of acute
graft-versus-host disease: A multicenter study of 157 patients.
Bone Marrow Transplant 8:13,1991
14. Weisdorf D, Hakke R, Blazar B, Miller W, McGlave P,
Ramsay N, Kersey J, Filipovich A Risk factors for acute graftversus-host disease in histocompatible donor bone marrow transplantation. Transplantation 51:1197, 1991
15. Jones JM, Wilson R, Bealmear PM: Mortality and gross
pathology of secondary disease in germfree mouse radiation
chimeras. Radiat Res 45577,1971
16. Van Bekkum DW, Roodenburg J, Heidt PJ, van der Waaij
D: Mitigation of secondary disease of allogeneic mouse radiation
chimeras by modification of the intestinal microflora. J Natl Cancer
Inst 52:401,1974
17. Heit H, Wilson R, Fliedner TM, Kohne E: Mortality of
secondary disease in antibiotic treated mouse radiation chimeras,
in Heneghan JB (ed): Germfree Research: Biological Effects of
Gnotobiotic Environments. New York, NY,Academic, 1973, p 477
18. Truit RL, Winter M, Winter S: Application of germfree
techniques to the treatment of leukemia in AKR mice by allogeneic
bone marrow transplantation, in Waters H (ed): The Handbook of
Cancer Immunology. Volume 5: Immunotherapy. New York, NY,
Garland STPM, 1978, p 431
19. Wagemaker G, Heidt PJ, Merchav S, van Bekkum DW:
Abrogation of histocompatibility barriers to bone marrow transplantation in rhesus monkeys, in Baum SD, Ledney GD, Thierfelder S
(eds): Experimental Hematology Today. Basel, Switzerland, Karger,
1982, p 111
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
2676
20. Buckner CD, Clift RA, Sanders JE, Meyers JD, Counts GW,
Farewell VT, Thomas ED, and the Seattle Marrow Transplant
Team: Protective environment for marrow transplant recipients. A
prospective study. Ann Intern Med 89:893,1978
21. Skinh~jP, Jacobsen N, H0iby N, Faber V, and the Copenhagen Bone Marrow Transplant Group: Strict protective isolation in
allogeneic bone marrow transplantation: Effects on infectious
complications, fever and graft versus host disease. Scand J Infect
Dis 19:91,1987
22. Storb R, Prentice RL, Buckner CD, Clift RA, Appelbaum F,
Deeg J, Doney K, Hansen JA, Mason M, Sanders J, Singer J,
Sullivan KM, Witherspoon RP, Thomas ED: Graft-versus-host
disease and survival in patients with aplastic anemia treated by
marrow grafts from HLA-identical siblings. Beneficial effect of a
protective environment. N Engl J Med 308:302, 1983
23. Vossen JM, Heidt PJ, van den Berg H, Gerritsen EJA,
Hermans J, Dooren LJ: Prevention of infection and graft-versushost disease by suppression of intestinal microflora in children
treated with allogeneic bone marrow transplantation. Eur J Clin
Microbiol Infect Dis 9:14, 1990
24. Mahmoud HK, Schaefer UW, Schiining F, Schmidt CG,
Bamberg M, Haralambie E, Linzenmeier G, Hantschke D, GrosseWilde H, Luboldt W, Richter HJ: Laminar air flow versus barrier
nursing in marrow transplant recipients. Blut 49:375,1984
25. Storb R, Epstein RB, Graham TC, Thomas ED: Methotrexate regimens for control of graft-versus-host disease in dogs with
allogeneic marrow grafts. Transplantation 9:240,1970
26. Beelen DW, Quabeck K, Kaiser B, Wiefelsputz J, Scheulen
ME, Graeven U, Grosse-Wilde H, Sayer HG, Schaefer UW:Six
weeks of continuous intravenous cyclosporine and short-course
methotrexate as prophylaxis for acute graft-versus-host disease
after allogeneic bone marrow transplantation. Transplantation
50421,1990
27. Beelen DW, Graeven U, Ryschka U: Primaly treatment and
prophylaxis of acute graft versus host disease with a combination of
a monoclonal antibody against the human a@-T-cell-receptor and
ciclosporin. Blut 59:299,1989 (abstr 220)
28. Beelen DW, Grosse-Wilde H, Ryschka U, Quabeck K, Sayer
HG, Graeven U, Schaefer UW: Initial treatment of acute graftversus-host disease with a murine monoclonal antibody directed to
the human a/P-T-cell-receptor. Cancer Immunol Immunother
34:97,1991
29. Glucksberg H, Storb R, Fefer A, Buckner CD, Neiman PE,
Clift RA, Lerner KG, Thomas ED: Clinical manifestations of
graft-versus-host disease in human recipients of marrow from
HLA-matched sibling donors. Transplantation 18:295,1974
30. Thomas ED, Storb R, Clift RA, Fefer A, Johnson FL,
Neiman PE, Lerner KG, Glucksberg H, Buckner CD: Bone
marrow transplantation. N Engl J Med 292895,1975
BEELEN ET AL
31. Haralambie E, Linzenmeier G, Nowrousian M, Schaefer
UW,Schmidt CG: Gnotobiotische Untersuchungen-Ein mikrobiologischer Beitrag zur Dekontamination von knochenmarktransplantierten Leukamiepatienten. Immun Infekt 8:89,1980
32. Haralambie E: Clinically applied gnotobiotics with special
regards to anaerobic bacteria, in Ducluzeau R (ed): Gnotobiology
and its Applications. Proceedings of the IX International Symposium on Gnotobiology. Lyon, France, Editon Foundation Marcel
MeriCw, 1980, p 175
33. Kaplan EL, Meier P Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457,1958
34. Mantel N: Evaluation of survival data and two new rank
order statistics arising in its consideration. Cancer Chemother Rep
50163,1966
35. Cox DR: Regression models and life-tables. J R Stat SOC
34:187, 1972
36. Cox DR: Partial likelihood. Biometrika 62:269,1975
37. Kalbfleisch JD, Prentice R L The Statistical Analysis of
Failure Time Data. New York, NY, Wiley, 1980
38. Furnival GM, Wilson R W Regressions by leaps and bounds.
Technometrics 16:499,1974
39. Heidt PJ: Gnotobiotics and Bone Marrow Transplantation:
Experimental and Clinical Studies. Rijswijk, The Netherlands,
Radiobiological Institute of the Division for Health Research
TNO,1989, p 21
40. Van Bekkum DW, Knaan S: Role of bacterial flora in
development of intestinal lesions from graft-versus-host reaction. J
Natl Cancer Inst 58:787,1977
41. Nissen C, Gratwohl A, Tichelli A, Speck B: Abundant
macrophage growth in culture from patients with chronic myelogenous leukemia: A risk factor for graft-versus-host disease after
bone marrow transplantation. Experientia 44:167, 1988
42. Holler E, Kolb HJ, Moller A, Kempeni J, Liesenfeld S,
Pechumer H, Lehmacher W, Ruckdeschel G, Gleixner B, Riedner
C, Ledderose G, Brehm G, Mittermiiller J, Wilmanns W: Increased serum levels of tumor necrosis factor a precede major
complications of bone marrow transplantation. Blood 751011,
1990
43. Piguet PF, Grau GE, Allet B, Vassalli P: Tumor necrosis
factor/cachectin is an effector of skin and gut lesions of the acute
phase of graft-versus-host disease. J Exp Med 166:1280,1987
44. Symington FW, Sullivan Pepe M, Chen AB, Deliganis A:
Serum tumor necrosis factor alpha associated with acute graftversus-host disease in humans. Transplantation 50:518, 1990
45. Atkinson K, Horowitz MM, Gale RP, van Bekkum DW,
Gluckman E, Good RA, Jacobsen N, Kolb HJ, Rimm AA, RingdCn
0,Rozman C, Sobocinski KA, Zwaan FE, Bortin MM: Risk factors
for chronic graft-versus-host disease after HLA-identical sibling
bone marrow transplantation. Blood 75:2459, 1990
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
1992 80: 2668-2676
Evidence that sustained growth suppression of intestinal anaerobic
bacteria reduces the risk of acute graft-versus-host disease after
sibling marrow transplantation
DW Beelen, E Haralambie, H Brandt, G Linzenmeier, KD Muller, K Quabeck, HG Sayer, U
Graeven, HK Mahmoud and UW Schaefer
Updated information and services can be found at:
http://www.bloodjournal.org/content/80/10/2668.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.