British Iournal of Huernutology, 1989, 72, 439-444
Bone marrow histopathology of patients with severe
aplastic anaemia before treatment and at follow-up
MARIETM . DE P L A N Q U E , ~JOHAN H. J. M . V A N K R I E I c E N , ~HANNEKEC. K L U I N - N E L E M A N S , ~
LENY P . J. M. COLLA,l FRED VAN DER BURGH,l ANNEKEBRAND’ AND PHILIP M. K L U I N ~Departments of
’Haematology/lmmunohaernatologyand Bloodbank and ’Pathology, University Hospital, Leiden, The Netherlands
Received 20 September 1988; accepted 10 February 1989
Summary. Pretreatment bone marrow biopsies of 63 patients
with severe aplastic anaemia (SAA) who were not trans-
planted and of whom 5 5 received ATG, were evaluated
according to the amount and character of residual haemato-
poiesis (‘genuine’ aplasia/intermediate/hypoplastic myelo-
dysplasia (MD)), inflammatory infiltrate (Te Velde & Haak,
1977, grade I/II/III), and number of mast cells (normal or
slightly increased/increased). Of 6 1 evaluable biopsies, 4 7
were ‘genuine’aplastic, 11 intermediate and three hypoplas-
tic MD. Inflammatory infiltrates were graded as I11 in 36/60
evaluable biopsies, as I1 in 2 1 and I in three. A moderate to
marked increase of mast cells was seen in 19/61. Of grade I11
patients, 86% had a < 9 0 d interval between diagnosis and
administration of ATG, versus 50% of grade 1/11 patients
(P<O.OI). No other correlations with pretreatment charac-
teristics were found. No significant prognostic value for
survival or response to ATG of any of these three criteria has
been identified. More patients with grade 111 inflammation
tended to show adequate recovery at 4 and 6 months after
Acquired aplastic anaemia (AA) has been defined as a
pancytopenia due to unexplained bone marrow failure
(Gordon-Smith, 19 79). In most patients aetiologic events are
unclear, but in a minority a relationship with recent viral
infections or exposure to drugs or chemicals can be found.
Bone marrow histology confirms the diagnosis by excluding
disorders which might lead to pancytopenia.
The pathogenetic process which leads to this failure of
haematopoietic cells to proliferate is still unclear. Various
studies reveal ample evidence for a qualitative and/or
quantitative stem cell disorder, and for immunologic inhibi-
tion of stem cell proliferation due to suppressive cellular
activity, humoral inhibition, or lack of growth factors or of
Correspondence:Dr Mariet M. de Planque, Department of Immuno-
haematology and Blood Bank, Building 1, E 3-Q. University Hospital,
P.O. Box 9600. 2300 RC Leiden. The PIetherlands.
439
ATG. Stem cell damage, not identifiable morphologically,
and/or impairment of accessory cells might play a major role
in eventual outcome of SAA patients. Thirty-five patients are
currently alive, median 3.8 years (up to 12.4) after ATG.
Follow-up bone marrow aspirates and biopsies of 32 patients
were evaluable and none showed normal haematopoiesis.
One patient revealed persistent aplasia. Of the remaining 31,
haematopoiesis was decreased in 14 and increased in eight.
All had dyserythropoiesis, 2 8 dysplastic myelopoiesis and in
16/29 with evaluable megakaryocytes, dysmegakaryopoie-
sis was found. Sixteen patients had normo- to hypercellular
bone marrows with two dysplastic cell lines (consistent with
myelodysplastic syndrome (MDS) according to the FAB-
group). The prognostic impact of the dysplastic abnormalities
found in these patients needs longer follow-up. Close observa-
tion is indicated in view of the previously recognized, albeit
uncommon, evolution of SAA to MDS/acute non-lymphocy-
tic leukaemia.
response to these factors (Bacigalupoet al, 1980; Zoumbos et
al, 1985; Nissen et al. 1985a. b).
On the basis of both blood counts and bone marrow
cellularity. severe AA (SAA) has been defined and differen-
tiated from moderate or non-severe AA (Camitta et al, 1976).
Prior to the use of anti-thymocyte globulin (ATG). the
survival of SAA was less than 20% without bone marrow
transplantation (BMT) (Camitta et al, 1975). Following the
use of ATG, survival has improved (Young & Speck, 1984).
but studies with long-term follow-up identify late deaths due
to clonal evolution, suggesting that ATG does not cure the
disease (DePlanque et al, 1988a. b; Tichelli et al. 1988).
A relationship between AA and clonal disease such as
paroxysmal nocturnal haemoglobinuria. myelodysplastic
syndromes (MDS) and leukaemia has long been recognized,
but previously seen as a rare event (Dameshek, 1967: Milner
&?Geary, 19 79; Appelbaum & Fefer, 198 1). MDS as defined by
440 Mariet M . de Planque et a2
the French-American-British (FAB) group are characterized
by a normocellular or hypercellular bone marrow with at
least dysplastic features in two haematopoietic cell lines
(Bennett et al, 1982). Attempts have been made to separate
‘true AA’ from hypoplastic MDS on the basis of dysplastic
bone marrow characteristics, thus differentiating between
non-dysplastic AA which should not evolve and hypoplastic
MDS which might eventually evolve into clonal disease
(Fohlmeister et al, 1985b). Whether this differentiation is
indeed feasible and of eventual prognostic significance can be
questioned (De Planque et al. 1988a).
Both ATG and BMT are now considered effective treatment
for SAA, but short-term and long-term benefits differ from
patient to patient. Since therapy-related morbidity and early
mortality of BMT are higher, particularly in the older
patients, but also after multiple blood transfusions (Storb et al,
1980) a sound choice of treatment based on prognostic
factors is of major importance. Several studies have identified
prognostic factors for survival based on various patients’
characteristics (Williamset al, 19 78: Faille et al, 1979: Jansen
et al, 1982; Hunter et al. 1985; Torok-Storb et al, 1985). In
1977 Te Velde & Haak reported the prognostic value for
survival > 6 months of the intensity and distribution of
inflammatory infiltrates in methyl-methacrylate embedded
bone marrow biopsies of 15 AA patients treated with
conventional therapy. Since then ATG has gradually become
the treatment of first choice in our institution and has
improved SAA survival. Therefore, we were able to assess the
prognostic value of this grading of inflammatory infiltrates in
63 SAA patients, who were not transplanted and of whom
the majority were treated with ATG. In view of the above-
mentioned evolution in patients initially diagnosed as (S)AA
we subsequently evaluated bone marrow histopathology in
the SAA patients currently alive up to 12.4 years after ATG.
MATERIALS A N D METHODS
Patients. Between 1974 and July 1987 the diagnosis of
SAA (Camitta et al, 1976) has been confirmed in 82 patients,
42 males and 40 females. In 83% the disease was idiopathic,
in 11%a relationship with hepatitis and in 6% with drugs or
chemicals was suspected. Nineteen patients underwent BMT
either as initial treatment (11patients) or after failure of ATG
(eight patients). These 19 patients were excluded from this
analysis.
Of the remaining 63 patients (median age 30.0, range
13.8-84.7 years) 55 were treated with ATG. The interval
between diagnosis and ATG was < 30 d in 22 patients, 30-
60 in 13,60-90 in four and > 9 0 in 16. The majority of these
patients also received androgens prior to, or after ATG.
Twenty of the 5 5 ATG-treated patients have died. Eight
patients either died before treatment or received only hormo-
nal therapy (androgens and/or prednisone). Six of these eight
patients have died and two are lost to follow-up. Of the 35
patients currently alive following ATG therapy (median 3.8
years: range 0.3- 12.4) 3 2 gave permission to perform a bone
marrow biopsy and aspirate for follow-up evaluation.
Definition of response to ATG. Clinical responses were
defined as partial response: improvement in all three cell
lines, granulocyte count 3 0 . 5 x 1Oy/1, no transfusion re-
quirements: adequate recovery (‘complete response’, CR):
normal haemoglobin, granulocyte count 1.0 x 10y/land
platelet count 100 x 10y/l.
Bone marrow specimens. Bone marrow biopsy specimens
were obtained from the posterior superior iliac spine or iliac
crest using a bone drill or Y amshidi biopsy needle and after
fixation without decalcification embedded in methylmeth-
acrylate (Te Velde & Haak. 1977). Sections (3 pm thick) were
stained with Periodic Acid Schiff, Gallamine-Giemsa,
Gomori’s reticulin and Turnbull’s or Perl’s iron staining. For
initial diagnosis at least 30 mm2 of trabecular bone and
marrow surface per section was required. The cellularity was
determined semiquantitatively by analysis of intertrabecular
areas. Bone marrow aspirates obtained at the same time were
stained with May Griinwald-Giemsa and Prussian blue.
The most recent pretreatment bone marrow biopsies of all
63 patients were reviewed and categorized for evaluation of
prognostic value according to:
(i) residual haematopoiesis: (a) ‘genuine’ aplasia: severely
hypocellular bone marrow with inflammatory infiltrate. ‘Hot
spots’ of clustered erythropoiesis and limited remnants of
myelopoiesis could be present. No increase of reticulin fibres.
(b) Hypoplastic myelodysplasia (MD): bone marrow cellular-
ity <25% of normal with disturbed architecture of the
haematopoiesis. Dysplastic features of at least two cell lines.
Increase of fine reticulin fibres frequently present. (c) Inter-
mediate: biopsies which did not fulfil the criteria of the above
mentioned categories.
(ii) Amount and distribution of inflammatory infiltrate
acording to Te Velde & Haak (19 77): (a)grade I: sparse diffuse
inflammatory infiltrate: (b) grade II: moderate or extensive
inflammatory infiltrate in most but not all marrow fields; (c)
grade 111: moderate or strong inflammatory infiltrate in all
marrow fields.
(iii) Amount of mast cells, semiquantitatively scored as: (a)
low: normal or slight increase; (b) high: moderate to marked
increase.
Of pretreatment biopsies of three patients, one was not
available for review, one was not evaluable and one not
evaluable for grading of inflammatory infiltrate.
In the follow-up biopsies and aspirates, cellularity, overall
haematopoiesis, erythropoiesis, myelopoiesis and megakar-
yopoiesis were determined semiquantitatively. Dysplastic
features of the three cell lines were evaluated qualitatively as
was the presence or absence of an inflammatory infiltrate.
Tissue iron and the number of sideroblasts were scored as
decreased, normal or increased, and the presence or absence
of ringed sideroblasts was assessed. All bone marrow speci-
mens were examined independently by at least two of the
authors.
Statistical considerations. Overall survival was calculated
from the start of ATG therapy, or from diagnosis when no
ATG was administered, using the Kaplan-Meier method of
non-parametric estimation for complete observations. The
Fisher’s exact test was used for comparison of two groups of
patients.
 BM Histopathology and Severe Aplastic Anaemia 441

Fig 1.Methylmethacrylate-embeddedbone marrow biopsies, stained with periodic acid Schiff. (a) Biopsy of a patient with severe aplastic anaemia
showing only remnants of erythropoiesis with clustering of mature forms without atypia. (b) Area of dyserythropoiesis (abnormal nuclei and
dissociation of nucleus-cytoplasm ratio) in the follow-up biopsy of the same patient as in (a). (c) Abnormal myelopoiesis with many immature
forms and monocytoid cells. (d) Abnormal megakaryocytes in the follow-up biopsies of previously severe aplastic marrows with trilinear
dysplastic features.

Table I. Characteristics of pretreatment biopsy versus survival

Died Survived
C 6 months > 6 months Total
(n=13) (n = 48) (n=61)

Haematopoiesis
‘Genuine’aplasia 10 37 47
Intermediate/hypoplastic
myelodysplasia 3 11 14 P=0.626
Grade of inflammation.
Grade I and I1 5 21 26
Grade I11 8 26 34 P=0.470
No. of mast cells
Normal or slightly increased 7 35 42
Increased 6 13 19 P=0.163

* Only 60 biopsies could be graded for inflammatory infiltrate.

442 Mariet M . de Planque et a1
Table 11. Recovery after ATG of patients with inflammatory Table 111. Follow-up bone marrow abnormalities
infiltrate grade 1/11 versus grade I11

Total no. of patients 32

Response to ATC Grade 1/11 Grade I11 A1 1 P-value Persistent total aplasia 1
(n=20) (n=32) (ti=52) Evaluable bone marrows 31
Aspirates 30
At 4 months Biopsies 28
CR 1 5 6 P=0.242
No CR 19 27 46 Evaluable haematopoiesis 31
Haematopoiesis decreased 14
At 6 months Normal 9
CR 1 6 7 P = 0.1 61 Increased 8
No CR 19 26 45
Evaluable erythropoiesis 31
Minimal dysplasia 1
Moderate to severe dysplasia 30
RESULTS
Pretreatment biopsies Evaluable myelopoiesis 31
No dysplasia 3
‘Genuine’ aplasia (Fig l a ) was seen in 47/61 biopsies (77%), Minimal dysplasia 20
intermediate in 1 1 (18%)and hypoplastic MD in three (5%). Moderate dysplasia 8
The majority of the biopsies revealed a grade 111inflammatory
infiltrate (36/6O, 60%). Twenty-one biopsies ( 35%) were Evaluable megakaryopoiesis 29
No dysplasia 13
scored as grade 11, only three (5%) as grade I. A moderate to
Minimal dysplasia 12
marked increase in mast cells was seen in 19!61 ( 3 1 %) of the Moderate dysplasia 4
biopsies.
Comparison of overall survival and survival > 6 months of
patients with ‘genuine’ aplasia and intermediate/hypoplasic
MD. of patients with biopsies grade I and I1 versus grade 111 All bone marrow specimens revealed dyserythropoiesis
and of patients with a low number of mast cells in the marrow (Fig l b ) , which was minimal in one patient and moderate to
severe in 30/3 1. Biopsy specimens revealed clustering of
versus those with a high number of mast cells did not reveal
any difference (Table I). erythroid precursors in the same stage of maturation, lack of
normal erythron formation (25/31) and dissociation of the
Of 5 5 patients who received ATG 52 pretreatment biopsies
nuclear-cytoplasm ratio (2 1 patients). The smears also
were graded according to the inflammatory infiltrate. The
interval between diagnosis and ATG was < 9 0 d in 27/32 revealed clustering, megaloblastoid appearance (four
patients) and nuclear budding and multi-nuclearity (2 3
patients with a grade 111 infiltrate, but in only 10/2O with
patients). Tissue iron was increased in 22 patients, normal in
grade 1/11 (P=0.009). No difference was found between
patients with grade I and I1 versus grade I11 in overall two and decreased in eight. Sideroblasts were increased in 10
patients: only two revealed a few ringed sideroblasts.
response. However, more patients with grade 111 inflamma-
Dysplastic features of myelopoiesis (Fig l c ) could not be
tion showed adequate recovery (CR) at 4 and 6 months after
determined easily, in part due to decrease of the myelopoiesis.
treatment (respectively 5% versus 16% and 5% versus 19%.
In 22 biopsies myeloid precursors were localized centrally in
Table 11).
the marrow spaces instead of along the endosteal surface.
Follow-up bone marrows Pelger-Huet anomaly was seen in 18 smears and one patient
Of 32 patients currently alive after ATG. bone marrow revealed hypogranulation of myeloid precursors.
aspirates and biopsies were taken. One patient had remained In many patients megakaryopoiesis was diminished
totally aplastic (2.1 years after ATG). Of the remaining 3 1 revealing only one or very few megakaryocytes. Dysplastic
patients the blood values at time of biopsy were: haemoglobin megakaryopoiesis was found in 16/29 patients with evalu-
in females median 12.9 g/l (range 7.1-13.9), in males 14.2 able megakaryopoiesis. Clusters of small megakaryocytes
(10.6-16.3). granulocytes 2.0 x 109/1(0.7-6.1) and plate- were seen in aspirates and/or biopsies of 14 patients. Nuclear
lets 119 x 109/1(10-248). One aspirate and three biopsy abnormalities (non-lobulated or mononuclear cells) were
specimens were insufficient for evaluation. For these four found in smears and biopsies of 1 4 patients (Fig Id).
patients the data of the available specimen were used only. In All bone marrows still showed a t least some increase of
all other cases aspirate and biopsy data were combined when lymphoid cells. In seven marrows a slightly to moderately
applicable for an overall assessment, except for the cellularity increased number of mast cells was present, which did not
where histology prevailed. Results are summarized in Table correlate with the relative increase in the number of mast
111. cells in the initial biopsy. An increase of fine reticulin fibres
Cellularity was normal to increased in 171 3 1 patients and was seen in biopsies of 14 patients, Of 1 7 patients whose
decreased in 14. In almost all patients a (relative) increase of aspirates revealed a lower cellularity than their biopsies, 12
erythropoiesis was seen. The majority (25/31. 81%) of had a n increase in fine reticulin fibres.
patients had a decreased myelopoiesis and 16/31 (52%) a Various patient characteristics, such as age, quantitative
decreased megakaryopoiesis. and qualitative aspects of their initial biopsy, initial blood
 B M Histopathology and Severe Aplastic Anaemia
counts, rate and degree of recovery, blood counts at the time
of the follow-up biopsy were compared with cellularity and
degrees of dysplasia; however, no correlations were found.
Eight patients with a follow-up after ATG of at least 3 years
(up to 12.5)had one or more evaluable biopsies performed at
least 1year apart from ATG and the last biopsy. No consistent
pattern of cellularity and qualitative abnormalities of the
various cell lines was seen. Some showed a rather prolonged
period of hypocellular marrow, others an initial increase and
subsequent decrease of cellularity. In most patients dysplastic
features were present whenever there was evaluable haema-
topoiesis, but the severity of the dysplasia sometimes de-
creased or increased with time.
DISCUSSION
The putative immunologic pathogenesis of AA led to the
assumption of a prognostic value of the inflammatory
infiltrate usually seen in these biopsies. Te Velde & Haak
(1977) showed that 7/7 patients with a grade 1/11 infiltrate
survived > 6 months following conventional treatment and
only 1/8 with a grade I11 inflammation. These findings could
not be confirmed by Sale et al (198 7) and our data in patients
with SAA according to Camitta et a l ( l 976) who were treated
with ATG. The initial biopsies in the series of Te Velde & Haak
revealed an about equal percentage of patients with grade 1/11
and with grade 111 inflammatory infiltrate, but all grade 1/11
patients had non-severe and almost all long-standing AA. In
our series with only SAA patients the incidence of grade 1/11
and grade I11 was also about equal. However SO%, of grade I/
I1 patients, and a significantly higher percentage of grade 111
patients had a short interval between diagnosis and ATG
therapy. Although the patient population in the report of Sale
et a2 probably did not differ much from our series in severity
and duration of the disease, their incidence of grade 1/11 was
97% versus grade 111 3%. They found a significantly lower
median age in patients with low grade versus high grade
inflammation which was not seen in our primarily adult
patients. But in our study, patients with a grade 111infiltrate
who did respond to ATG therapy tended to reveal a higher
rate and degree of recovery which was also found by Te Velde
& Haak (1979).
All patients in this study showed severe hypoplasia of the
bone marrow. Usually only remnants of erythropoiesis were
found, sometimes with some myelopoiesis but almost always
without megakaryopoiesis. The pattern of residual haemato-
poiesis was not predictive for survival or response to ATG
therapy. Since severity of SAA, as measured by reticulocyte
and granulocyte counts, appears to correlate with long-term
survival (De Planque et al, 1988b) the lack of prognostic
value of residual bone marrow haematopoiesis could be due
to sampling errors. Another possibility is that the residual
haematopoiesis seen in the biopsy specimen does not reflect
the functional differentiating capacity of the haematopoietic
precursor cells, or the potential for proliferation of the stem
cells following therapy. It is also possible that there is a
difference in qualitative stem cell damage which cannot be
identified histologically.
Various case reports, and several studies, mention occasio-
nal patients who develop MDS and/or leukaemia following
443
AA (Delamore & Geary, 1971; Milner & Geary. 1979; Mir &
Geary. 1980; Streuli et al, 1980; Champlin et al, 1983; Doney
et al, 1987). The FAB cooperative group, defining the criteria
for the diagnosis of MDS, distinguished clearly between MDS
and AA by a normo- or hypercellular bone marrow of the
former (Bennett et al, 1982).However, patients with hypocel-
lular marrows with dysplastic haematopoiesis who subse-
quently developed leukaemia have been recognized (Fohl-
meister et al, 1985a). In a large retrospective study of 111
patients with AA, defined as pancytopenia and hypocellular
bone marrow, hypoplastic MDS, which might evolve to acute
non-lymphocytic leukaemia (ANLL), was differentiated from
non-dysplastic myelohypoplasia (Fohlmeister et al, 198 5b).
Five of the present 63 SAA patients of whom the initial bone
marrows showed severe hypoplasia or aplasia, have deve-
loped MDS and/or ANLL inspite of a minimal recovery or
even normalization of blood counts prior to evolution (De
Planque et al, 1988a).
All 31 patients in this study whose follow-up bone
marrows were evaluated, revealed dysplastic features in at
least one but often two or even three cell lines. Using the FAB
criteria (normo- or hypercellular bone marrow and at least
two dysplastic cell lines), 16 patients should be classified as
refractory anaemia (RA) on cytologic and also histologic
bone marrow examination. However, all 31 patients were
transfusion independent and the blood counts of most of
them differ from those of de novo diagnosed RA patients
(Kerkhofs et al, 1987). The remaining 15 patients had only
minimal dysplasia and/or hypocellular marrows and would
not be called non-dysplastic hypoplasia. For hypoplastic MDS
the risk of developing ANLL within 3 years after diagnosis has
been reported to vary between 9% and 82%depending on the
number of prognostic features present. These include marrow
fibrosis, dysplasia of the various cell lines and high number of
megakaryocytes (Fohlmeister et al, 1985b).
Clonal cytogenetic abnormalities discovered at the time of
SAA diagnosis will increase the risk of developing leukaemia
(Appelbaum et al, 1987). Three of the five patients in our
series who evolved to MDS/ANLL developed a clone of cells
with monosomy 7 after initial normal cytogenetic analysis
(De Planque et al, 1988a). None of the 31 patients in the
present study had any cytogenetic abnormality in unstimu-
lated marrow samples at the time of the follow-up biopsy
(data not shown). Therefore, the significance of the bone
marrow abnormalities found in these patients, who initially
presented with SAA needs further follow-up and evaluation.
Observation of SAA patients who survive following ATG
therapy is indicated and might eventually lead to a different
view on the relationship between SAA, MDS and ANLL.
ACKNOWLEDGMENTS
Thanks are due to the physicians of the Department of
Haematology who were responsible for the primary care of
these patients and kept records of their clinical course.
Furthermore we like to thank Dr H. L. Haak and Dr J. te Velde
for their suggestions and advice and Mrs M. L. de Vlieger-
Stokman for secretarial assistance.
This work was supported in part by grant 6.2.1 of the J. A.
Cohen Institute for Radiopathology and Radiation Protec-
tion, Leiden. The Netherlands.
444 Mariet M . de Planque et a1
REFERENCES
Appelbaum, F.R., Barrall. 1.. Storb. R.. Ramberg. R.. Doney. K.. Sale,
G.E. & Thomas E.D. (1987) Clonal cytogenetic abnormalities in
patients with otherwise typical aplastic anemia. Experimental
Hematology, 1 5 , 1134-1139.
Appelbaum, F.R. & Fefer. A. ( 1 981 ) The pathogenesis of aplastic
anemia. Seminars in Hematology. 1 8 , 241-257.
Bacigapulo, A.. Podesta, M., Mingari. M.C.. Moretta. L., Van Lint,
M.T. & Marmont. A.M. (1980) Immune suppression of hemato-
poiesis in aplastic anemia: activity of T-gamma lymphocytes.
Journal of Immunology. 1 2 5 , 1449-1453.
Bennett, J.M., Catovsky, D.. Daniel, M.T.. Flandrin, G.. Galton.
D.A.G., Gralnick. H.R. & Sultan, C. (1982) Proposals for the
classification of the myelodysplastic syndromes. British Journal of
Haematology. 5 1 , 189-199.
Camitta B.M.. Rappeport. J.M.. Parkman. R. &Nathan,D.G. (1975)
Selection of patients for bone marrow transplantation in severe
aplastic anemia. Blood. 45, 355-363.
Camitta. B.M.. Thomas, E.D., Nathan, D.G.. Santos. G.. Gordon-
Smith. E.C., Gale. R.P.. Rappeport. J.M. & Storb. R. (1976) Severe
aplastic anemia: a prospective study of the effect of early bone
marrow transplantation on acute mortality. Blood, 48, 6 3-70.
Champlin. R.. Ho. W. & Gale. R.P. (1983) Antithymocyte globulin
treatment in patients with aplastic anemia. New England Journal o/
Medicine. 308, 1 13-1 18.
Dameshek. W. ( 1 967) What do aplastic anemia. paroxysmal
nocturnal hemoglobinuria (PNH)and 'hypoplastic' leukemia have
in common? Blood. 30, 251-254.
Delamore, I.W. & Geary. C.G. (1971) Aplastic anemia, acute
myeloblastic leukaemia and oxymetholone. British Medical Jour-
nal, ii. 743-745.
De Planque. M.M., Kluin-Nelemans.J.C.. Van Krieken. H.J.M..Kluin.
Ph.M.. Brand, A,. Beverstock, G.C.. Willemze. R. & Van Rood, 1.1.
(1 988a) Evolution of acquired severe aplastic anemia to myelodys-
plasia and subsequent leukaemia in adults. British Journal oJ
Haematology. 70, 5 5-62.
De Planque M.M., Bacigalupo A,. Wiirsch. A.. Marsh, J.C.W..
Devergie. A,. Frickhofen. N.. Brand, A. & Nissen, C. (1988b)Long
term follow-up of severe aplastic anaemia patients treated with
antithymocyte globulin. Bone Marrow Transplantation. 3. (Suppl.
1). 239.
De Planque, M.M.. Brand, A.. Kluin-Nelemans. If.. Eernisse. j.G..
Van der Burgh. F., Natarajan, A.T.. Beverstock. G.C.. Zwaan. F.E..
Willemze, R. & Van Rood. j.J. (1989) Haematopoietic and
immunologic abnormalities in severe aplastic anaemia patients
treated with anti-thymocyte globulin. British Jounial of Haernato-
logy. in press.
Doney. K.. Storb. R.. Buckner, C.D.. McGuffin. R., Witherspoon. R..
Deeg. H.J..Appelbaum,F.R..Sullivan. K.M.&Thomas.E.D. (1987)
Treatment of aplastic anemia with antithymocyte globulin, high-
dose corticosteroids, and androgens. Experimental Hematology, 15,
239-242.
Faille, A,. Barrett, A.j., Balitrand. N.. Ketels, F.. Gluckman. E. &
Naiean. Y . (1979) Effect of antilymphocyte globulin on granulo-
cyte precursors in aplastic anaemia. British Iorrrnal of Haematology.
42, 371-380.
Fohlmeister. I.. Fischer. R. & Schaeffer. H.E. (19858) Preleukemic
myelodysplastic syndromes (MDS): Pathogenetical considerations
based on retrospective clinico-morphological sequential studies.
Anticancer Research. 5 , 179-188.
Fohlmeister. I., Fischer. R.. Modder. B.. Rister. M. & Schaefer. H.E.
( 198 5b) Aplastic anaemia and the hypocellular myelodysplasia:
histomorphological. diagnostic and prognostic features. Journal 01
Clinical Pathology. 39, 1218-1224.
Gordon-Smith, E.C. (1979) Clinical features of aplastic anaemia. In:
Aplastic Anemia-pathophysiology and approaches to therapy (ed. by
H. Heimpel. E. C. Gordon-Smith, W. Heit and B. Kubanek). pp. 9-
13. Haematology and Blood Transfusion, 24. Springer, Berlin.
Hunter. R.F.. Roth. P.A. & Huang, A.T. (1985) Predictive factors for
response to anti-thymocyte globulin in acquired aplastic anemia.
American Journal of Medicine. 79, 73-78.
Jansen. J., Zwaan. F.E.. Haak. H.L.. TeVelde, J., Guiot. H.F.L., Sabbe,
L.J.M.. Eernisse. J.G.. Tricot, G.J.K. & Van Rood, J.J. (1982)
Antithymocyte globulin treatment for aplastic anaemia. Scandina-
vian Journal of Haematology. 28, 341-351.
Kerkhofs, H., Hermans. J.. Haak. H.L. & Leeksma, C.H.W. (1987)
Utility of the FAB classifications for the myelodysplastic syn-
dromes: investigation of prognostic factors in 237 cases. British
lourrial of Haematology. 6 5 , 73-81.
Milner, G.R. & Geary. C.G. (1979) The aplasia-leukaemia syndrome.
Aplastic Anaemia (ed. by C. G. Geary). pp. 230-243. Bailliere
Tindall. London.
Mir. M.A. & Geary. C.G. (1980)Aplastic anaemia: an analysis of 174
patients. Postgraduate Medical Journal, 56. 322-329.
Nissen. C.. Moser. Y.. Weis. J. &Speck,B. (1985a) Stimulatory serum
factors in aplastic anaemia. I. Serum 'releaser' activity for
haemopoietic growth factors. a regulator? British Journal of
Haematology. 6 1 , 49 1-498.
Nissen. C.. Moser. Y.. Weis. J. & Speck, B. (1985b) Stimulatory serum
factors in aplastic anaemia. 11. Prognostic significance for patients
treated with high dose immunosuppression. British Journal of
Haematology, 61, 499-512.
Sale, G.E.. Rajantie. J., Doney, K.. Appelbaum, F.R.. Storb. R. &
Thomas. E.D. (1987) Does histologic grading of inflammation in
bone marrow predict the response of aplastic anaemia patients to
antithymocyte globulin therapy! British Journal of Haematology.
67, 261-266.
Storb. R.. Thomas, E.D.. Buckner. C.C., Clift, R.A.. Deeg, H.J.. Fefer,
A,. Goodell. B.W.. Sale, G.E.. Sanders, J.E.. Singer, J., Stewart, P. &
Weiden. P.L. (1 980) Marrow transplantation in thirty 'untrans-
fused' patients with severe aplastic anemia. Annals of lnternal
Medicine. 92. 30-36.
Streuli. R.A.. Testa, J.R.. Vardiman, J.W., Mintz, U., Golomb, H.M. &
Rowley. J.D. ( 1980) Dysmyelopoietic syndrome: sequential clini-
cal and cytogenetic studies. Blood, 55, 636-644.
Te Velde. J. & Haak. H.L. (1977) Aplastic anaemia: Histological
investigation of methacrylate embedded bone marrow biopsy
specimens: correlation with survival after conventional treatment
in 15 adult patients. British Journaf of Haematotogy, 35, 61-69.
Te Velde. J. & Haak. H.L. (1979) Histology of bone marrow failure, a
follow-up study in aplastic anaemia. In: Aplastic Anemia (ed. by H.
Heimpel. E. C. Gordon-Smith, W. Heit and B. Kubanek). Springer,
Berlin.
Tichelli. A.. Gratwohl. A., Wiirsch, A., Nissen, C. &Speck, B. (1988)
Late haematological complications in severe aplastic anaemia.
British Journal of Haematology, 69, 413-418.
Torok-Storb, B., Doney, K., Sale, G., Thomas, E.D. & Storb, R. (1985)
Subsets of patients with aplastic anemia identified by flow micro-
fluorometry. New England journal of Medicine, 312, 1015-1022.
Williams. D.M.. Lynch, R.E. & Cartwright. G.E. (1978) Prognostic
factors in aplastic anaemia. Clinics in Haematology. 7, 467-475.
Young, N. & Speck, B. (1984) Antithymocyte and antilymphocyte
globulins: clinical trials and mechanisms of action. Aplastic
Anemia: Stem cell biology and new clinical concepts (ed. by N. Young,
A. Levine and R. K. Humphries), pp. 221-226. Alan R. Liss. New
York.
Zoumbos. N.C., Gascaon, P.. Djeu. 1.Y.. Trost, S.R. & Young, N.S.
(198 5) Circulating activated suppressor T lymphocytes in aplastic
anemia. New England Journal of Medicine. 312. 257-265.