American Journal of Hematology 76:143–147 (2004)

Thromboembolic Complications After Splenectomy

for Hematologic Diseases
Martin Mohren,1* Ilka Markmann,1 Ulrike Dworschak,1 Astrid Franke,1 Christian Maas,1
Sabine Mewes,1 Günter Weiss,2 and Kathleen Jentsch-Ullrich1

Thromboembolic complications following splenectomy for hematologic diseases occur in

up to 10% of patients and may range from portal vein thrombosis (PVT) to pulmonary
embolism (PE) and deep vein thrombosis (DVT). Up to now there exist no recommendations
for the duration and intensity of prophylactic anticoagulation, which usually follows local
institutional protocols. We report on three consecutive patients with severe portal vein
thrombosis and/or pulmonary embolism—one with fatal outcome—7 to 35 days after splen-
ectomy for autoimmune hemolytic anemia, immunothrombocytopenia, and indolent lym-
phoma, respectively. Incidence and pathophysiology of thromboembolic events (TE) in this
patient group as well as prophylactic anticoagulation will be discussed, including a review
of the current literature on this topic. Am. J. Hematol. 76:143–147, 2004. ª 2004 Wiley-Liss, Inc.
Key words: splenectomy; hematologic diseases; thromboembolism; prophylactic anti-
coagulation

INTRODUCTION heparin until complete mobilization, had no prior

Splenectomy is performed for diagnosis or treatment
of a variety of hematologic diseases such as immune
cytopenia, indolent non-Hodgkin’s lymphoma (NHL),
and, rarely, for myeloproliferative diseases (MPD) [1].
Within the recent years, laparoscopic splenectomy has
been increasingly employed when possible and is asso-
ciated with a decreased rate of intraoperative organ
damage and quicker postoperative recuperation of the
patient [2].
Postoperative complications most often comprise
infections, particularly with encapsulated organisms
such as pneumococci or haemophilus influenzae type
B following a sometimes fulminant course, thus pre-
operative vaccination has become standard practice [2].
Additionally, patients with immunethrombocytopenia
and low platelet counts may have an increased peri-
operative bleeding risk. In contrast to the former,
thromboembolic risk has so far attracted considerably
less attention in this patient group, and published
reports of thromboembolism (TE) so far have come
from surgeons rather than hematologists.
Within a period of 6 months we saw severe TE in
3 patients several days to weeks after splenectomy
for various hematologic disorders. All 3 patients had
received anticoagulation with low molecular weight
ª 2004 Wiley-Liss, Inc.
history of thrombophilia, and exhibited no further
obvious clinical risk factors for TE, such as prolonged
immobilization, infection, or heart failure except for
smoking in patient I.
PATIENTS
Patient I
A 24-year-old female patient with a 2-year history
of autoimmune hemolytic anemia of the warm anti-
body type had been successfully treated with pred-
nisone and relapsed under a maintenance dose of
5 mg daily. The hemoglobin normalized when steroids
were increased to 100 mg/day, and the patient was
referred for splenectomy after the dose was tapered to
<40 mg/day. Laparoscopic splenectomy (spleen weight
395 g) was performed, and after quick recovery she was
*Correspondence to: Dr. Martin Mohren, Klinik für Hämatologie/
Onkologie, Universität Magdeburg, Leipziger Str. 44, 39120
Magdeburg, Germany.
E-mail: Martin.Mohren@medizin.uni-magdeburg.de
Received for publication 26 May 2003; Accepted 17 October 2003
Published online in Wiley InterScience (www.interscience.wiley.com).
DOI: 10.1002/ajh.20018
144 Case Report: Mohren et al.

Fig. 2. Inferior cava vein thrombosis.

Fig. 1. Portal vein thrombosis with Perfusion defects in
the liver.

discharged on day 7 after the operation with a platelet

count of 519 G/l. Anticoagulation with low molecular
weight heparin was carried out until discharge. The
next day, the patient collapsed at home and was
found to have extensive pulmonary embolism as well
as portal vein thrombosis (Fig. 1) on CT scan. Platelets
were at 499 G/l. Although anticoagulation with full-
dose heparin was started immediately, the further
course was complicated by mesenteric vein thrombosis
with severe bowel ischemia, subsequently leading to
death from multiorgan failure.
Patient II
A 21-year-old male patient with chronic immune-
thrombocytopenia first diagnosed 8 years prior to
admission had received a short course of therapy with
immunoglobulins and steroids, whereafter the platelet
count had remained stable at 70 G/l with no further
medication. At relapse, high steroid doses were neces-
sary to maintain platelets above 40 G/l, and splenec-
tomy was recommended. Laparoscopic splenectomy
(spleen weight 350 g) was carried out at a platelet
count of 60 G/l with no bleeding complications.
He received anticoagulation with low molecular
weight heparin until discharge on day 5 postsplenect-
omy. By that time, platelets were at 360 G/l. Two
days later the patient complained of acute-onset
shortness of breath and left-sided upper abdominal
pain. The platelet count was at 1,013 G/l. Embolism
of the right pulmonary artery and portal and inferior
cava vein thrombosis (Fig. 2 and 3) were diagnosed,
and full-dose heparin was begun immediately fol-
lowed by oral anticoagulation for 1 year. He is cur-
rently well with no further anti-coagulation and/or
immunosuppressive medication.
Fig. 3. Right pulmonary embolism.
Patient III
A 63-year-old male patient presented to our clinic
because of left-sided upper abdominal pain secondary
to splenomegaly. Bone marrow histology revealed infil-
tration by indolent lymphoma, and open splenectomy
was performed, yielding a massively enlarged organ, of
size 24.5  14  6.5 cm and weight 1,660 g. Histologic
findings were consistent with marginal zone lymphoma.
There was also involvement of abdominal lymph nodes.
The postoperative course was uneventful, and he
received prophylactic anticoagulation with low mole-
cular weight heparin for 30 days due to our previous
experience. However, on day 35 postsplenectomy, this
patient experienced chest pain and dyspnoea of
sudden onset; ventilation and perfusion scintigraphy
revealed pulmonary embolism in segment 2 in both
lungs. The platelet count was at 1,158 G/l. Full-dose
heparin was started and was later switched to a vita-
min K antagonist for a period of 6 months.
DISCUSSION
We saw 3 patients with severe thromboembolic
events (TE) after splenectomy for hematologic
 Case Report: Thromboembolic Complications After Splenectomy
diseases within a short period of time. TE is a rare
complication of splenectomy that usually does not
receive as much attention as secondary infections or
hemorrhage. However, an increased incidence of TE,
portal vein thrombosis in particular, following splen-
ectomy has been reported in patients with hematolo-
gic diseases.
A Dutch study of 563 splenectomies published in
2000 found postoperative portal vein thrombosis in
9 patients (2%). The latter was more likely to occur
in patients with hematologic diseases, such as auto-
immune hemolytic anemia and myeloproliferative
syndrome (incidence 10%) [3].
In a recently published report investigating portal
vein thrombosis after splenectomy, an incidence of
8% was found among 101 patients, 74% of whom
had hematologic disease [4]. Among 8 patients who
developed PVT, 4 had a myeloproliferative disorder.
Interestingly PVT occurred as late as 3 years follow-
ing splenectomy [4].
Another report of 223 splenectomies in patients with
myeloid metaplasia, however, detected thrombosis
(location not further specified) in 7.2% only. Postsplen-
ectomy thrombocytosis in these patients was associated
with postoperative thrombosis. Unfortunately, platelet
numbers are not provided by the authors [5].
An additional analysis of 26 splenectomized patients
with myelofibrosis revealed venous thrombosis in 12%
[6]. In an American investigation carried out with 50
splenectomized patients whose underlying disease was
not further specified, 5 of 50 patients (10%) were found
to develop PVT that was then successfully treated with
anticoagulation [7]. In an Italian report on patients
undergoing splenectomy for treatment of thalassemia,
an incidence of TE as high as 29% was seen [8].
One of 12 patients with immunethrombocytopenia
developed PVT without any further complications in
another Italian report [9]. A French group performed
repeat Doppler ultrasound studies in 60 consecutive
splenectomized patients on days 7 and 30 postopera-
tively and found 1 symptomatic and 3 asymptomatic
PVT (8%) in this patient cohort [10]. Thus asympto-
matic PVT may occur more frequent than clinically
apparent TE. A review of imaging findings in postsplen-
ectomy patients revealed PVT in 12 of 123 patients
(9,8%), all of whom had hematologic disease [11]. A
small number of TE (1%) was reported by a French
group that performed laparoscopic splenectomy in 275
patients with hematologic disease [12], and a lower
rate of TE was also found in another report on laparo-
scopic splenectomy [5]. In contrast to these findings,
2 of our patients with severe TE had undergone laparo-
scopic splenectomy.
There is little data on the incidence of TE other
than PVT, such as DVT and/or PE in splenectomized
145
patients [13]. In one report, 4 of 8 patients with post-
splenectomy PVT developed TE later in the course
[5], whereas 2 of our patients had both, PVT and PE
at the same time.
So far the pathogenesis of TE after splenectomy is
poorly understood.
Portal vein thrombosis may result from local fac-
tors, such as intraoperative manipulation of visceral
vessels, the splenic vein in particular [14], and may
account for the particular risk in patients with
massively enlarged organs. Five of 31 patients (16%)
with mesenteric venous thrombosis (MVT)—a rare
complication of PVT—had previously undergone
splenectomy, whereas 13 (42%) of these patients had
a hypercoagulable state (protein C, protein S, and
ATIII deficiency) without local risk factors [15]. Mes-
enteric thrombosis, as was also seen in patient I of
our report, is associated with a poor prognosis due
to subsequent bowel ischemia and multiorgan failure
[4,15,16].
Thrombocytosis that accompanies splenectomy in
most cases may, although poorly defined, play an
etiologic role.
In several reports, large spleen size, thrombocytosis,
and a myeloproliferative disease (MPD) as the under-
lying disease were described as risk factors for TE [4,7].
Patients with MPD and a large spleen (>3,000 g) had
a 75% incidence of PVT [4]. Splenic weight >1,000 g
was found to be associated with significant morbidity
after laparoscopic splenectomy [17], yet none of our
patients with severe TE had MPD and only one patient
had a massively enlarged spleen.
The association of postsplenectomy thrombocyto-
sis and PVT is unclear, particularly in patients with-
out MPD, since not all patients with thrombocytosis
develop PVT and PVT also occurs in patients with
normal platelets [4]. Patient I in our report, for exam-
ple, had slightly elevated platelets only (499 G/l), and
not all of the reported patients with MVT following
splenectomy had thrombocytosis [15].
In a study of 129 patients with extreme thrombo-
cytosis (>1,000 G/l), 72 in MPD and 57 with reactive
thrombocytosis, thrombosis was found in 3–4% [18],
but a prospective observational study of patients with
essential thrombocythemia (ET) and a platelet count
<1,500 G/l as compared to an age- and sex-matched
control group failed to show an increased risk for
thrombosis in ET patients [19]. A recent study investi-
gating etiology and clinical significance of thrombocy-
tosis found a higher incidence of venous and arterial
thrombosis in patients with primary thrombocytosis
as compared to patients with secondary thrombocyto-
sis (12.4% vs. 1.6%). The authors conclude that post-
splenectomy thrombocytosis is not associated with an
increased risk for hemostatic complications [20]. Thus
146 Case Report: Mohren et al.
thrombocytosis by itself does not seem an indication
to initiate anti-coagulation or antiplatelet or platelet-
lowering therapy unless it is being used as secondary
prophylaxis.
There exists little data on activation of plasmatic
coagulation and/or lack of coagulation inhibitors,
such as protein C or S deficiency after splenectomy or
the presence of the factor V Leiden mutation or anti-
phospholipid antibodies. Few reports demonstrated a
systemic hypercoaguable state in patients with PVT
[14,15], but these reports focused on the etiology of
PVT and MVT without special attention to the post-
splenectomy state.
Disseminated intravascular coagulation was not
found in splenectomized patients in a systemic evalua-
tion [5].
How can TE be prevented in splenectomized
patients?
Preventive measures still rely preferentially on per-
sonal experience, because accumulated data is incon-
clusive and general recommendations for duration
and intensity of anticoagulation after splenectomy
do not exist [2,21].
So far there is little evidence indicating that the
method of splenectomy has an impact on subsequent
thromboembolic complications. General prophylactic
measures such as early mobilization and refraining
from smoking may be of some help. Until now, pro-
phylactic low-dose anticoagulation has failed to show
sufficient effect [4] to prevent PVT but may reduce the
risk for TE in other locations, such as deep venous
thrombosis and pulmonary embolism. However, 3 of
our patients had pulmonary embolism—2 of whom
also had portal vein and inferior cava vein and/or
mesenteric vein thrombosis—although they had re-
ceived prophylactic anticoagulation with low molecular
weight heparin until complete mobilization and dis-
charge from the hospital. Antiplatelet agents such as
aspirin in postsplenectomy thrombocytosis have not
been thoroughly looked at—most experience of as-
pirin use in thrombocytosis stemming from patients
with ET, a disease with a different impact on platelet
function—and may increase bleeding complications as
has been shown in patients with ET [22]. However,
antiplatelet agents may be of benefit in patients with
extreme thrombocytosis (>1,500 G/l) particularly if
there are additive cardiovascular risk factors [22].
Although extension of prophylactic anticoagula-
tion with low molecular weight heparin until day 30
postsplenectomy was started after we saw TE in the
first 2 patients, patient III developed pulmonary
embolism on day 35, which was 5 days after cessation
of anticoagulation. Thus even our newly established
approach may not be sufficient in preventing TE in
splenectomized patients with hematologic disease.
In concordance with our findings, the majority of
patients (approximately 70%) developed PVT while
receiving prophylactic anticoagulation with low-dose
heparin [4,8]. Although a combination of heparin and
antiplatelet agents or even vitamin K antagonists has
been suggested in high-risk splenectomized patients
[4], its use is problematic due to an increased risk of
bleeding in the postoperative period and the abnor-
mal platelet function seen in some patients with
hematologic disease [5]. However, routine postopera-
tive surveillance ultrasound imaging may be of bene-
fit in patients with myeloproliferative disease and a
large spleen [11] for prompt diagnosis of PVT.
CONCLUSIONS
Although it does not receive as much attention as
infectious complications, TE occurs in approximately
10% of splenectomized patients, and prophylactic
anticoagulation is warranted in patients undergoing
splenectomy for hematologic diseases. Duration and
intensity of mandatory anticoagulation has yet to
be established. At our institution, patients currently
receive low molecular weight heparin for a period of
30 days after splenectomy. However, as has been
shown in one of our patients and in further previous
reports, TE may occur ever after a prolonged time
interval after surgery, thus physicians as well as
patients should be alert to symptoms of TE in splen-
ectomized patients, even years after the operation to
ensure rapid diagnosis and proper treatment.
REFERENCES
1. Xiros N, Economopoulos T, Christodoulidis C, et al. Splenectomy
in patients with malignant non-Hodgkin’s lymphoma. Eur J Hae-
matol 2000;64(3):145–150.
2. Marcaccio MJ. Laparoscopic splenectomy in chronic idiopathic
thrombocytopenic purpura. Semin Hematol 2000;37(3):267–274.
3. van’t Riet M, Burger JW, van Muiswinkel JM, Kazemier G,
Schipperus MR, Bonjer HJ. Diagnosis and treatment of portal
vein thrombosis following splenectomy. Br J Surg 2000;87(9):
1229–1233.
4. Winslow ER, Brunt LM, Drebin JA, Soper NJ, Klingensmith ME.
Portal vein thrombosis after splenectomy. Am J Surg 2002;184(6):
631–635.
5. Tefferi A, Mesa RA, Nagorney DM, Schroeder G, Silverstein MN.
Splenectomy in myelofibrosis with myeloid metaplasia: a single-
institution experience with 223 patients Blood 2000;95(7):
2226–2233.
6. Akpek G, McAneny D, Weintraub L. Risk and benefits of splen-
ectomy in myelofibrosis with myeloid metaplasia: a retrospective
analysis of 26 cases. J Surg Oncol 2001;77(1):42–48.
7. Hassn AM, Al-Fallouji MA, Ouf TI, Saad R. Portal vein throm-
bosis following splenectomy. Br J Surg 2000;87(3):362–373.
 Case Report: Thromboembolic Complications After Splenectomy
8. Cappellini MD, Robbiolo L, Bottasso BM, Coppola R, Fiorelli G,
Manucci AP. Venous thromboembolism and hypercoagulabiliy
in splenectomized patients with halassaemia intermedia. Br J
Hematol 2000;111(2):467–473.
9. Valeri A, Venneri F, Presenti L, Grossi A, Borelli D. Portal throm-
bosis. A rare complication of laparoscopic splenectomy. Surg
Endosc 1998;12(9):1173–1176.
10. Chaffanjon PC, Brichon P-Y. Portal vein thrombosis following
splemectomy for hematologic disease: prospective study with
Doppler color flow imaging. World J Surg 1998;22:1082–1086.
11. Loring LA, Panicek DM, Karpeh MS. Portal system thrombosis
after splenectomy for neoplasm or chronic hematologic disorder:
is routine surveillance imaging necessary? J Comput Assist Tomogr
1998;22(6):856–860.
12. Delaitre B, Champault G, Barrat C, et al. Laparoscopic splenec-
tomy for hematologic diseases. Study of 275 cases Ann Chir 2000;
125(6):522–529.
13. Stephens BJ, Justice JL, Sloan DA, Yoder JA. Elective laparo-
scopic splenectomy for hematologic disorders. 1997;63(8):700–703.
14. Valla DC, Condat B. Portal vein thrombosis in adults: patho-
phyiology, pathogenesis and management. J Hepatol 2000;32:
865–871.
147
15. Morasch M, Ebaugh MD, Chiou AC, Matsamura J, Pearce W,
Yao J. Masenteric venous thrombosis: a changing clinical entity.
Vasc Surg 2001;34(4):680–684.
16. Balz J, Minton JP. Mesenteric thrombosis following splenectomy.
Ann Surg 1975;181(1):126–128.
17. Patel AG, Parker JE, Wallwork B, et al. Massive splenomegaly is
associated with significant morbidity after laparoscopic splenec-
tomy. Ann Surg 2003;238(2):235–240.
18. Buss DH, Stuart JJ, Lipscomb GE. The incidence of thrombotic and
hemorrhagic disorders in association with extreme thrombocytosis:
an analysis of 129 cases. Am J Hematol 1985;20(4):365–372.
19. Ruggeri M, Finazzi G, Tosetto A, Riva S, Rodeghiero F, Barbui T.
No treatment for low-risk thrombocythaemia: results from a pro-
spective study Br J Haematol 1998;103(3):772–777.
20. Griesshammer M, Bangerter M, Sauer T, Wennauer R. Bergmann L,
Heimpel H. Aetiology and clinical significance of thrombocytosis:
analysis of 732 patients with an elevated platelet count. J Intern Med
1999;245:295–300.
21. Benoist S. Median and long-term complications of splenectomy
Ann Chir 2000;125(4):317–324.
22. Schafer A. Thrombocytosis and thrombocythemia. Blood Rev
2001;15:159–166.