A 47 Year Old Man presenting with Upper Abdominal Pain

A 47 Year Old Man presenting with Upper Abdominal Pain
and Splenomegaly
Amulaya Mishra*, AK Agarwal**, RK Dudeja***, Ratnesh Singh Kanwar****,
Santanu Borgohain*, Ashok Kumar***, Sumeet Singla****
Presentation of the case
A 47 year old man, resident of Haryana, was
hospitalised for evaluation of left upper abdominal
pain which was present for about a fortnight.
Abdominal pain was dull in nature, poorly
localised, and accompanied with dragging
sensation and heaviness in left hypochondrium
but without any radiation to other parts of
abdomen. The heaviness used to increase
following intake of food, but no definite history of
exacerbation of pain after meals could be elicited.
He also gave history of decreased appetite and
occasional episodes of vomiting. Vomitus usually
contained food articles taken the previous night.
There was no history of haematemesis, malena,
or fever. He did not suffer from jaundice,
tuberculosis, hypertension, or diabetes in the past.
However, the patient did receive blood transfusion
about 11 years back when he was hospitalised
for head injury. He is a vegetarian, non-smoker,
and non-alcoholic.
On clinical examination he was fully conscious
and oriented. Patient did not look in agony, lying
comfortably in bed, and was of average built.
Pulse was 72/minute regular and blood
pressure was 130/90 mmHg. No pallor,
cyanosis, icterus, pedal oedema, or
lymphadenopathy was detected. JVP was
normal. Chest examination: bilateral vesicular
breath sounds with no accompaniments. CVS
and CNS examinations were unremarkable.
* Senior Resident
** Consultant in Medicine
*** Senior Physician
**** Postgraduate Student
Department of Medicine,
Dr. Ram Manohar Lohia Hospital,
New Delhi-110001 (India).
Abdominal examination revealed hepatomegaly
(2 cm, soft, and non-tender), splenomegaly (12
cm, firm in consistency, smooth surface, and mild
tenderness), but no evidence of free fluid in
peritoneal cavity. There were no visible/engorged
veins, or scars over abdominal wall. The bowel
sounds were neither decreased, nor increased.
There was no rub over the splenic area.
Primarily, our patient had significant splenomegaly
and vague gastrointestinal complaints. Five
general mechanisms may enlarge the spleen: (1)
reactive proliferation of lymphoid cells, (2)
infiltration by neoplastic cells or lipid laden
macrophages, (3) extramedullary haematopoiesis,
(4) proliferation of phagocytic cells, and (5)
vascular congestion. Diseases may cause
splenomegaly by one or by a combination of these
mechanisms. The causes of massive splenomegaly
(greater than 3,000 grams) are limited. The
myelodysplastic disorders and malignant lymphoid
disorders are the most common causes of chronic
massive splenomegaly in non-tropical countries.
Splenomegaly is common in tropical populations
and may be due to malaria, leishmaniasis (Kalaazar), and other disorders. Spleen may also be
markedly enlarged due to splenic vein
hypertension (vascular congestion) in a large
number of cases in our country. Chronic congestive
splenomegaly (Banti’s Syndrome) is characterised
by splenomegaly, pancytopenia, and
gastrointestinal bleeding secondary to portal
hypertension. The splenic vein hypertension is
either due to intrahepatic disease (e.g., cirrhosis)
or extrahepatic disease (such, as portal or splenic
vein thrombosis). Symptoms may range from
vague gastrointestinal complaints to catastrophic
bleeding from oesophageal or gastric varices.
Presence of ascites points towards the possibility
of cirrhosis of liver.
Splenomegaly can be present as an isolated
finding on physical examination, can exist in
association with a systemic disorder, or can be
discovered as a consequence of the secondary
haematologic effects of splenomegaly - the
hypersplenism syndrome. Symptoms arising from
splenomegaly may include pain from stretched
capsule of an enlarged spleen, particularly acute
enlargement, or shock from atraumatic rupture
of a tense capsule.
Evaluation of splenomegaly should include
examination of the peripheral blood and
frequently the bone marrow. In general, diagnostic
tests are not performed on the spleen itself; they
are oriented towards the diagnosis of disease
states producing splenomegaly. Chest
radiography, liver function tests, imaging
modalities (eg., ultrasonography, CT abdomen,
or MR imaging), liver biopsy, and a search for
varices may reveal the aetiology of splenic
enlargement. Evaluation for alcoholism and viral
hepatitis markers is often helpful. When systemic
symptoms accompany splenomegaly but no
lymphadenopathy is appreciated, even a
laparotomy with biopsies of liver, spleen, and
lymph nodes is sometimes needed. In a significant
number of cases, such a study may reveal presence
of lymphoma, congestive splenomegaly, or
inflammatory state.
Relevant investigations were carried out at this
stage, keeping in view the overall clinical picture.
The results are shown in table 1.
Portal vein thrombosis (PVT) is being
increasingly identified as a cause of abdominal
pain as a result of widespread use of ultrasound
for assessing the cause of abdominal symptoms.
With employment of such a practice, it has been
found that the proportion of recent portal vein
thromboses in cases of portal venous
obstruction has risen from less than 30% to more
than 60% in the last decade1,2. It is well known
that pulsed doppler is a valuable supplement
to real-time sonography for diagnosing this
condition 3,4.
Journal, Indian Academy of Clinical Medicine
Vol. 2, No. 3
PVT may develop following abdominal trauma or
intra-abdominal sepsis, or in association with
cirrhosis or hepatocellular carcinoma. In the
majority of cases, however, the cause is unknown.
Hypercoagulable states may also be responsible
for PVT. This is primarily a disease of children,
although adults may also develop PVT. The
diagnosis is suggested by the presence of portal
hypertension in a patient with normal liver biopsy.
The clinical features of recent PVT are generally
non-specific and failure to diagnose this condition
comprehensively and early may result in a delay
in institution of effective treatment5.
Extrahepatic portal venous obstruction (EHPVO)
is an important cause of portal hypertension in
our country. The spleno-portal axis obstruction,
results either from only portal vein thrombosis or
from the obliteration of the entire axis. In a study
from PGI Chandigarh, the PVT with patent splenic
vein was found in 66% of cases and total block in
29% of cases. Majority of Indian patients do not
have a demonstrable underlying cause6, whereas
an underlying cause is commonly found in patients
from the West.
In India, the most common clinical presentation
is with upper gastrointestinal variceal bleeding,
seen in about 90% of cases at presentation7.
Our patient never gave history suggestive of
upper gastrointestinal bleeding. The other
clinical presentation is the awareness of lump
in abdomen due to an enlarged spleen. The
splenomegaly is usually less than 7 cm below
the costal margin, in contrast to non-cirrhotic
portal fibrosis patients, who have bigger spleen.
In EHPVO only 18% of the patients were found
to have spleen size more than 11 cm6. In the
present case, spleen size was 12 cm below the
costal margin. Though 10% of patients of PVT
do not have haematemesis at presentation,
diagnosed cases eventually go onto bleed within
a mean period of 4 year8. Splenomegaly is seen
in 75-100% of patients and a substantial
number have mild hepatomegaly with
abdominal tenderness. Ascites is rare2.
July-September 2001
Table I : Investigations during initial work up.
Peripheral smear
12 gm%; 13.6 gm%; 13.3 gm%.
16,600/cmm; 20,000/cmm, 16,500/cmm
P83, L12 M2 E3; P89 L11 M0 E0; P84 L10 M0 E6
25 mm 1st hr (Westergren)
No malarial parasites seen, neutrophils show shift to the
left, platelets adequate.
5.43 x 106/cmm.
79 µm3
25.9 pg/cell
32.7 gm/dL
RBCs count
Platelets count
Reticulocytes count
Albumin: traces, Sugar: nil, 4-6 pus cells and 0-1 granular
cast per hpf.
Urine culture and sensitivity
Sterile culture
Biochemical laboratory investigations
Random blood sugar 73 mg%, blood urea 38 mg%, Serum
creatinine 1.0 mg%, serum Na+ 145 mEq/L, serum K+ 4.4
mEq/L, total serum bilirubin 0.3 mg/dL (direct 0.1, indirect
0.2 mg/dL), SGOT 51 U/L, SGPT 46 U/L, alkaline
phosphatase 267 µ/L.
Bone marrow aspiration
Mixed reaction, predominantly normoblastic, myeloid series
of cells show normal mutation, abundant functioning
megakaryocytes, no parasites seen.
Chest skiagram
Ultrasonography upper abdomen
Moderate splenomegaly with linear tubular mass in main
portal vein and its branches, liver echotexture normal.
Pulsed doppler scan
Portal vein dilated with echogenic material inside the lumen
with lack of flow inside. Evidence of peripheral collateral
flow within the epigastrium, features suggestive of portal
vein thrombosis with collaterals (Figure 1).
Our patient having evidence of portal vein
thrombosis with collaterals was subjected to further
investigations to know the extent of thrombosis
and its exact cause, if any. The results are shown
in table II.
Following above investigative evaluation it was
evident that the patient had symptomatic spleno-
mesenteric – portal venous thrombosis. Splenic
infarction and gastric varices could also be
demonstrated. No underlying cause for PVT
causing extrahepatic portal venous obstruction
could be demonstrated in our case. During
hospital stay his symptoms got aggravated.
Abdominal pain increased in intensity and became
troublesome. Patient complained of mild fever
Journal, Indian Academy of Clinical Medicine
Vol. 2, No. 3
July-September 2001
(37.6 to 37.8°C) and vomitings were more
anticoagulation therapy with low molecular
heparin alongwith oral anticoagulant (warfarin),
and symptomatic treatment was initiated. Low
molecular heparin was discontinued within one
week. There was definite improvement in the
overall clinical condition of the patient within 10
days of anticoagulation therapy. He was eager to
go home and resume his work. Therefore, further
evaluation to identify recanalisation could not be
possible. But, he was continued on warfarin even
after discharge from hospital. During follow up,
the dose of warfarin was adjusted according to
prothrombin time to achieve INR of about 3. He
remained asymptomatic upto 4 months following
discharge from hospital on oral anticoagulation.
Patient was also prescribed low dose propranolol
(20 mg three times a day). At this point of time he
was asked to attend gastroenterology department
of AIIMS, Delhi for further management. He has
not reported back since then.
Fig. 1 : Pulsed Doppler Scan
The diagnosis of spleno-portal and mesenteric
venous thrombosis is difficult to establish.
History and physical findings are non-specific,
consequently correct diagnosis and adequate
treatment may often be delayed with a major
risk of intestinal infarction.
The angiographic, CT, and sonographic diagnoses
Table II : Further investigations.
Prothrombin time
16 seconds, control 15 seconds
Bleeding time
1 minute 40 seconds
Clotting time
5 minutes
Protein C
78.40 percent (Ref. range 70.00-140.00 percent)
Protein S
68.99 percent (Ref. range 70.00-123.00 percent)
Antithrombin III
88.00 percent (Ref. range 80.00-120.00 percent)
Antiphospholipid antibodies
Hepatitis markers
Hepatitis B surface antigen, hepatitis B surface antibody, and
hepatitis C virus antibody were all negative.
ELISA for HIV I and II
Upper GI endoscopy
Normal oesophagus, stomach shows 2 small variceal tags
in fundus.
Contrast enhanced CT abdomen
Suggestive of hepatomegaly with thrombosis of main portal
vein, splenic vein, and extension into superior mesenteric
vein with splenic infarct and left sided pleural effusion (Figure
2 to 5).
Essentially normal histology.
Liver biopsy
Journal, Indian Academy of Clinical Medicine
Vol. 2, No. 3
July-September 2001
Fig. 2 : Section of CECT showing portal vein thrombosis with
collaterals at porta and splenomegly.
of portal venous thrombosis have been known. IV
contrast enhancement, particularly with rapid
sequence scanning, is essential for the CT
diagnosis of PVT9. MR is a valuable tool for
imaging PVT. The portal vein, its right and left
branches, and its splenic and superior mesenteric
tributaries are easier to see with MR than with either
CT or sonography. Furthermore, thrombosis of
these vessels and associated portal venous
collaterals can be detected with greater sensitivity
by MR than by CT or sonography10. But MR may
fail to detect calcification within a chronic, retracted
thrombus. Another limitation of MR in the
diagnosis of PT is the slow-flow-enhancement
artifact, which is more likely to occur in patients
with portal hypertension. Splenoportovenography
is seldom needed nowadays except for
demonstration of natural/spontaneous shunts11
and in better demonstration of shuntable veins
for surgery.
The presence of a primary upper abdominal
venous thrombosis should stimulate extensive
Hypercoagulability encompasses two clinical
situations, the presence of laboratory
abnormalities or clinical conditions, such as cancer,
abdominal infections, pregnancy, or the
postoperative period (particularly splenectomy in
Fig. 3 : This section shows portal vein thrombosis with
thrombus extending into major branches with splenomegaly
and hypodensity (intarction) at periphery, and hilar splenic
collaterals due to thrombus.
patients with a platelet disorder)12,13. The primary
hypercoagulable states include: antihrombin III
deficiency, protein C deficiency, protein S
deficiency, disorders of fibrinolytic system,
dysfibrinogenaemia, factor XII deficiency, and
lupus anticoagulant12.
When mesenteric venous thrombosis complicates
such a situation, there are chances of bowel
infarction in about 50% of cases 14 . Clinical
presentation of mesenteric venous thrombosis, as
in our case, is often of a subacute nature with
history of abdominal pain, low grade pyrexia, and
leukocytosis. Signs of hypovolaemia may be
evident, but shock is rare unless intestinal infarction
has occurred. Other symptoms such as nausea,
vomiting, and less frequently, changes in bowel
habits, haematemesis, and malena may be
present14,15. In our case, extension of thrombosis
to splenic vein resulted in splenic infarction which
also contributed to the overall clinical picture.
It has been observed that many morphological
and epidemiological features are common to noncirrhotic portal fibrosis (NCPF) and extrahepatic
portal - splenic venous obstruction (EHPSO).
Though hepatic lesions in EHPSO are much less
frequent and milder than in NCPF16. For reasons
Journal, Indian Academy of Clinical Medicine
Vol. 2, No. 3
July-September 2001
Fig. 4 : CECT revealing PVT, splenic vein thrombosis, and
splenomegaly with displacement of left kidney inferioly.
as yet not determined, the portal venous system both in its intrahepatic and extrahepatic course seems to be susceptible to the development of
thrombosis and the location of thrombi determines
the establishment of either NCPF or EHPSO,
conditions which present with sustained portal
hypertension. In EHPSO, there is early
development of symptoms; while in NCPF, the
onset of portal hypertension is late.
Primary myeloproliferative disorders in a full-blown
or latent form, or at an early stage - are a major
cause of portal vein thrombosis. Latent primary
myeloproliferative disorders, in which the
abnormalities of peripheral blood are absent, can
be specifically recognised by the spontaneous
formation of erythroid colonies in cultures of bone
marrow progenitor cells17. The present case should
have been investigated on these lines also.
We feel that for the management of portal or
mesenteric vein thrombosis, the guidelines
proposed by Valla D (1999) are worth following:
“(i) imaging investigation for portal vein
thrombosis in patients presenting with abdominal
pain; (ii) immediate anticoagulation therapy in
patients with recent portal or mesenteric vein
thrombosis until re-permeation or for 6 months;
(iii) exhaustive investigation of both local and
systemic causes of venous thrombosis; (iv)
prevention of first or recurrent bleeding in patients
Journal, Indian Academy of Clinical Medicine
Vol. 2, No. 3
Fig. 5 : This section shows left pleural effusion with splenic
hypodensity due to splenic infarction.
with gastric or oesophageal varices using
pharmacological and/or endoscopic therapy as
recommended for patients who are in good
condition but have cirrhosis; (v) permanent
anticoagulation for patients in whom a
prothrombotic disorder has been documented or
is strongly suspected because of personal or
familial history of recurrent thrombosis”.
In patients of PVT, receiving anticoagulation, no
increase in risk of or the severity of gastrointestinal
bleeding has been found2. Splenectomy may be
curative particularly if isolated splenic vein
thrombosis is detected. The surgical management
of PVT patients may be difficult because of the
absence of a patent vein to use for making a portal
- systemic shunt. If mesenteric vein thrombosis is
the main source of symptoms, following initial
supportive care to stabilize the patient’s condition,
an operation may be performed to resect infarcted
or severely ischaemic bowel. Reconstructive venous
surgery is not generally possible. Anticoagulation,
in postoperative period is recommended except
in patients who have underlying disease processes
that would make this too hazardous.
Acknowledgements : The authors are grateful
to Dr. CP Singh, the Medical Superintendent of
Dr. Ram Manohar Lohia Hospital, New Delhi for
permission to publish this case. Special thanks are
July-September 2001
extended to the radiodiagnosis department of the
institution, particularly to Dr. BKS Chauhan,
Consultant and Head of this Department.
Valla D. Portal vein thrombosis and prothrombotic
disorders. J Gastroenterol Hepatol 1999; 14: 1051-2.
Condat B, Valla D, Erlinger S. Acute portal venous
thrombosis: Characteristics and outcome. Hepatology
1998; 28: 455A.
Weltin G, Taylor KJW, Carter AR, Taylor CR. Duplex
Doppler: Identification of cavernous transformation of
portal vein. AJR 1985; 144: 999-1001.
Miller VE, Berland LL. Pulsed Doppler duplex sonography
and CT of portal vein thrombosis. AJR 1985; 145: 73-6.
Yang YY, Chan CC, Wang SS et al. Portal vein thrombosis
associated with hereditary protein C deficiency. A report
of 2 cases. J Gastroenterol Hepatol 1999; 14: 1119-23.
Dilawari JB, Chawla YK. Extrahepatic portal venous
obstruction in India. Gut 1988; 29: 554-5.
Koshy A, Bhasin DK, Kapur KK. Bleeding in extrahepatic
portal vein obstruction. Indian J Gasteroenterol 1984;
3: 13-4.
Neho LJ, Sherlock S. Aetiology, presentation and natural
history of extrahepatic portal vein obstruction. QJ Med
1979; 192: 627-39.
Mathieu D, Vasile N, Grenier P. Portal thrombosis :
Dynamic CT features and course. Radiology 1985; 154:
10. Zirinsky K, Markisz JA, Rubenstein WA et al. MR imaging
of portal venous thrombosis : correlation with CT and
sonography. AJR 1988; 150: 283-8.
11. Dilawari JB, Raju GS, Chawla YK. Development of large
spleno-adreno-renal shunt after endoscopic
sclerotherapy. Gastroenterology 1989; 97: 421-6.
12. Schafer AI. The hypercoagulable states. Ann Intern Med
1985; 102: 814-26.
13. Gordon DH, Schaffner D, Bennet JM, Ischwartz SI. Postsplenectomy thrombocytosis : its association with splenic,
portal, and/or renal vein thrombosis in patients with
myeloproliferatie disorders. Arch Surg 1978; 113: 7135.
14. Grendell JH, Ockner RK. Mesenteric venous thrombosis.
Gastroenterology 1982; 82: 358.
15. Williams DA. Venous mesenteric thrombosis and
coexisting benign gastric ulcer. South Med J 1981; 74:
16. Arora R, Mohanty M, Nundy S, Nayak NC.
Phlebothrombosis as a common pathogenic
denominator in non-cirrhotic portal fibrosis and
extrahepatic portal splenic venous obstruction. Indian J
Med Res 1984; 79: 392-403.
17. Valla D, Casadevall N, Huisse MG et al. Etiology of portal
vein thrombosis in adults : a prospective evaluation of
primary myeloproliferative disorders. Gastroenterology
1988; 94: 1063-9.
14th Annual conference of
will be held on October 20 - 21
at B.P. Koirala Institute of Health Sciences, Dharan, Nepal.
The central theme of this years’ conference is “Lipids and Endothelial Dysfunction”.
Our invitation is cardially to all those who are working in areas of research
which have a bearing on atherosclerosis.
For details please contact :
Prof. S Dwivedi / Prof. KK Sharma
Conference Secretariat, Department of Medicine
B.P. Koirala Institute of Health Sciences, Dharan, Nepal.
Tel. : 977-25-25555 Ext. 2048/2052, Fax : 977-25-20251,
E-mail : [email protected][email protected][email protected]/
Website : www.bpkihs.edu
Journal, Indian Academy of Clinical Medicine
Vol. 2, No. 3
July-September 2001