An unusual cause of hematemesis: Goiter

Abstract

Downhill varices are located in the upper part of the esophagus and are usually related to superior vena cava obstruction. Bleeding from these varices is extremely rare. We describe a 77-year-old patient with hematemesis due to downhill varices as a result of recurrent goiter. A right lobe thyroidectomy was carried out with disappearance of the varices.

Full text

portal vein when both the superior vena cava and the
azygos vein are occluded. In contrast to the high risk of
hemorrhage from uphill varices in portal hypertension,
bleeding from downhill varices is extremely rare. Although
an increased variceal wall tension is the ultimate factor
causing bleeding in both types of varices, several factors
may underlie this difference in bleeding tendency. First,
in patients with uphill varices, coagulation capacity may
be reduced due to concomitant liver disease with an
inherently increased bleeding tendency. Second, exposure
to esophagogastric reflux damages distal rather than
proximal varices. Third, because distal uphill varices
predominantly distend at subepithelial levels compared
to the submucosal location of downhill varices in the
midthoracic and proximal esophageal wall, variceal rupture
is much more likely to occur near the esophagogastric
junction[2]. We report a rare case of bleeding downhill
varices in the absence of superior vena cava obstruction.
Detailed diagnostic work-up showed that the downhill
varices were caused by goiter. The varices disappeared
after subtotal thyroidectomy.
CASE REPORT
A 77-year-old female was admitted to the hospital because
of hematemesis. The patient had a one-year history of
dysphagia and weight loss. The patient had recurrent
goiter after a subtotal thyroidectomy in 1979 for multi-
nodular goiter. Her medical history also revealed chronic
obstructive pulmonary disease (COPD) and smoking. The
medication consisted of inhalation of salbutamol and
ipratropium bromide. Physical examination revealed a pale
non-icteric woman with normal vital signs and a normal
voice without stridor. There was a large, firm, nodular
mass on the right side of the neck with a horizontal
thyroidectomy scar. There were no dilated veins and no
bruits audible over the mass. Physical signs of liver disease,
Graves’ disease and superior vena cava syndrome were
absent.
Laboratory findings including thyroid function tests
were normal except for 7 mmol/L hemoglobin (normal:
7.5-10.0 mmol/L) and 0.31% hematocrit (normal:
0.37%-0.47%), respectively. At emergency gastroscopy,
the descending part of the duodenum and duodenal bulb
were unremarkable. The stomach showed no abnormality
except for a few adherent streaks of blood and a small
amount of dark blood without clots. Careful gastric
mucosal examination after use of a water jet was again
unremarkable. The distal esophagus appeared normal
CASE REPORT
An unusual cause of hematemesis: Goiter
Astrid AM van der Veldt, Mohammed Hadithi, Marinus A Paul, Fred G van den Berg, Chris JJ Mulder,
Mikael E Craanen
www.wjgnet.com
Astrid AM van der Veldt, Mohammed Hadithi, Chris JJ Mul-
der, Mikael E Craanen, Department of Gastroenterology, VU
University Medical Center, Amsterdam, The Netherlands
Marinus A Paul, Department of Surgery, VU University Medical
Center, Amsterdam, The Netherlands
Fred G van den Berg, Department of Radiology, VU University
Medical Center, Amsterdam, The Netherlands
Correspondence to: Dr. Astrid AM van der Veldt, PO Box 7057,
Amsterdam 1007 MB,
The Netherlands. aam.vanderveldt@vumc.nl
Telephone: +31-20-4440613
Received: 2006-03-07 Accepted: 2006-05-22
Abstract
Downhill varices are located in the upper part of the
esophagus and are usually related to superior vena
cava obstruction. Bleeding from these varices is
extremely rare. We describe a 77-year-old patient with
hematemesis due to downhill varices as a result of
recurrent goiter. A right lobe thyroidectomy was carried
out with disappearance of the varices.
© 2006 The WJG Press. All rights reserved.
Key words: Downhill; Esophageal; Varices; Goiter; He-
matemesis
van der Veldt AAM, Hadithi M, Paul MA, van den Berg FG,
Mulder CJJ, Craanen ME. An unusual cause of hematemesis:
Goiter.
World J Gastroenterol
2006; 12(33): 5412-5415
http://www.wjgnet.com/1007-9327/12/5412.asp
INTRODUCTION
Bleeding from distal esophageal varices is a frequent
complication of portal hypertension. The portal
blood drains into the superior vena cava by gastric and
esophageal collaterals, also called uphill varices, referring
to the upward direction of blood fl ow to the superior vena
cava. In contrast to uphill varices, downhill varices have
a retrograde blood flow and are located in the proximal
esophagus. Downhill varices are rare and usually caused
by superior vena cava obstruction due to bronchogenic
carcinoma and mediastinal tumors, etc[1,2]. They serve as
collaterals either to bypass superior vena cava obstruction
via the azygos vein or to drain the superior system to the
PO Box 2345, Beijing 100023, China World J Gastroenterol 2006 September 7; 12(33): 5412-5415
www.wjgnet.com World Journal of Gastroenterology ISSN 1007-9327
wjg@wjgnet.com © 2006 The WJG Press. All rights reserved.

without signs of refl ux esophagitis or Mallory-Weiss lesion.
However, proximal grade Ⅱ-Ⅲ downhill esophageal
varices were visualized 6 centimeters below the upper
esophageal sphincter. Importantly, one of the varices
showed a small fibrin plug indicating recent bleeding.
Since bleeding stopped spontaneously, no treatment was
given except for blood transfusion and a diet of semi-
solid food. A second gastroscopy after 48 h showed again
grade Ⅱ-Ⅲ varices without signs of (re)bleeding (Figure
1A). CT of the thorax showed a substernal goiter and
a slight compression of the trachea without any other
abnormalities. CT angiography demonstrated a dilated
venous plexus around the esophagus connected with a
thyroid vein (Figures 2, 3A). The venous plexus extended
over 7.4 centimeters from the seventh cervical vertebra
to the third thoracic vertebra and drained into the azygos
vein. There was compression of the right internal jugular
vein near the junction of the brachiocephalic vein and
a dilatation of the more cranial part. Ultrasound and
Doppler examination demonstrated a craniocaudal blood
fl ow. Flebography excluded superior vena cava obstruction.
Arteriography of the aortic arch excluded the presence
of arteriovenous malformations. Since goiter seemed to
cause the downhill varices, a thyroidectomy of the right
lobe was carried out sparing the right recurrent laryngeal
nerve and both parathyroids. Intraoperatively, a dilated
right internal jugular vein was seen. The thyroid specimen
weighed 85 g and was histopathologically diagnosed
as multi-nodular goiter. The patient had a satisfactory
van der Veldt AAM
et al.
Bleeding downhill esophageal varices and goiter 5413
www.wjgnet.com
recovery and was euthyroid after surgery with dysphagia
complaints improved during follow-up. Twelve months
after surgery, endoscopy and CT angiography showed
almost complete disappearance of the varices (Figures 1B,
3B). Twenty months postoperation the patient was still
non-symptomatic.
DISCUSSION
Downhill varices are usually associated with superior vena
cava obstruction due to bronchogenic carcinoma, different
types of mediastinal tumor and fi brosis, venulitis, surgical
caval ligation and thyroid masses[1,2]. Occasionally, as in
our patient, downhill varices may develop without superior
vena cava obstruction[3,4]. Relatively few case reports have
been published on bleeding downhill varices in relation to
thyroid pathology (Table 1). In a study of 1051 patients
with cervical and retrosternal goiter, 3% of patients
developed non-bleeding downhill varices[13]. Lagemann[14]
performed barium swallows in 50 patients with recurrent
thyroid enlargement and demonstrated that more than
50% of the patients have non-bleeding downhill varices.
Blood from the thyroid plexus fl ows through the inferior
thyroid veins (also called thyroid ima veins) into the
brachiocephalic vein. In case of obstruction of the inferior
thyroid veins, blood flows via the deep esophageal veins
leading to esophageal varices. The esophageal varices
can drain into collaterals to the brachiocephalic, azygos,
hemiazygos and accessory hemiazygos veins, all of which
finally drain into the superior vena cava. In the present
case, goiter caused compression of the internal jugular
vein. Blood fl ow over the thyroid plexus draining into the
inferior thyroid veins might bypass compression of the
internal jugular vein. However, in this patient downhill
varices developed and bypassed this compression,
suggesting that the function of inferior thyroid veins is
insufficient. Both previous thyroidectomy and recurrent
goiter are possible explanations, since inferior thyroid veins
can be occluded either by primary or recurrent thyroid
tumors or by surgical ligation during thyroidectomy and
fi brogenesis or mediastinitis secondary to surgery. In the
present case, the downhill varices drained into the azygos
vein as illustrated in Figure 4. Hemorrhage of downhill
varices is an emergency. However, the experience with
treatment is limited because of its rare bleeding propensity.
BA
Figure 1 Endoscopic view demonstrating downhill varices (A) and almost
complete disappearance of the downhill varices (B) before and 12 mo after
thyroidectomy of the right lobe.
BA
Figure 3 CT angiography demonstrating downhill varices one day after
hemorrhage (A) and disappearance of the downhill varices 12 mo after
thyroidectomy of the right lobe (B). * indicates varices around the esophagus; T,
trachea.
Figure 2 3D reconstruction of CT angiography
with view at the dorsal wall of the trachea (T)
demonstrating a venous plexus of downhill
varices (V) on the wall between the esophagus
and trachea connected with a thyroid vein at
the goiter (G) of the right thyroid lobe; *, out of
plane, cut off level thick slice.
G
V
T
*
TT
*
Goiter

www.wjgnet.com
In contrast to uphill varices, endoscopic sclerotherapy
is not generally performed to treat downhill varices up
to 5 cm below the upper esophageal sphincter, because
retrograde flow of sclerosant through the azygos vein
could result in spinal cord and vertebral body infarction[15].
Fatal pulmonary embolism of cyanoacrylate used for
endoscopic embolization of downhill varices has also
been reported[11]. Therefore, downhill varices should be
recognized and distinguished from uphill varices. The
use of a Sengstaken-Blakemore tube can be lifesaving[8].
Endoscopic band ligation has been shown to be effective
in preventing recurrent bleeding of downhill varices[2].
Finally, defi nitive treatment is performed to eliminate the
cause of venous obstruction. As in this patient, surgery
can successfully relieve obstruction. For goiter- related
downhill varices jodium therapy can also be effective[8].
In conclusion, downhill varices although rare, can cause
upper gastrointestinal bleeding and should be suspected
in any patient with evidence of thyroid enlargement or
having a history of thyroid surgery, even though signs of
superior vena cava obstruction are absent. Management
of the underlying cause, as in this case by thyroidectomy,
can effi ciently lead to recovery and disappearance of the
esophageal varices.
ACKNOWLEDGMENTS
The authors thank Ms. Bertholet for illustrating the
manuscript.
REFERENCES
1 Felson B, Lessure AP. “Downhill” varices of the esophagus.
Dis Chest 1964; 46: 740-746
2 Savoy AD, Wolfsen HC, Paz-Fumagalli R, Raimondo M. En-
doscopic therapy for bleeding proximal esophageal varices: a
case report. Gastrointest Endosc 2004; 59: 310-313
3 Palmer ED. The sources of upper gastrointestinal bleeding.
Nebr State Med J 1967; 52: 490
4 Serin E, Ozer B, Gumurdulu Y, Yildirim T, Barutcu O, Boya-
Table 1 Case reports on bleeding downhill varices associated with thyroid pathology
Year Author Aetiology SVCO Treatment Outcome
1960 Sundermann and Kämmerer[5] Recurrent goiter No Thyroidectomy Barium swallow after six years:
disappearance of varices
1976 Barber et al[6] Goiter No Thyroidectomy Endoscopy after four months:
disappearance of varices
1978 Johnson et al[7] Carcinoma Yes Thyroidectomy Barium swallow after two months:
disappearance of varices
1982 Fleig et al[8] Recurrent goiter No Sengstaken-Blakemore
tube
Endoscopy after two weeks:
still varices
1982 Kelly et al[9] Goiter No Thyroidectomy Venogram after one month:
disappearance of varices
1986 Takahashi et al[10] Recurrent goiter No Thyroidectomy Thyroid arteriography after three weeks:
disappearance of varices
1998 Tsokos et al[11] Recurrent goiter No Sclerotherapy Death caused by pulmonary embolism
of cyanoacrylate used for sclerotherapy
Endoscopy and venography after 4 mo:
2006 Bédard and Deslauriers[12] Posterior mediastinal goiter Yes Resection of mediastinal
mass
disappearance of varices and SVCO syndrome
2006 van der Veldt et al Recurrent goiter No Thyroidectomy Endoscopy and CT angiography after twelve
months: almost complete disappearance of varices
current report
SCVO: Superior vena cava obstruction.
Figure 4 Venous blood fl ow of the downhill varices in relation to goiter. 1: internal
jugular vein; 2: dilated right internal jugular vein; 3: goiter with compression of the
right internal jugular vein; 4: thyroid plexus draining into the esophagus varices; 5:
inferior thyroid vein; 6: brachiocephalic vein; 7: varices around the esophagus; 8:
azygos vein; 9: superior vena cava; Block: Occlusion of the inferior thyroid vein,
possibly as a result of previous surgery or recurrent goiter.
11
2
3
44
55
Block
6
7
8
9
5414 ISSN 1007-9327 CN 14-1219/ R World J Gastroenterol September 7, 2006 Volume 12 Number 33

cioglu S. A case of Castleman’s disease with “downhill” vari-
ces in the absence of superior vena cava obstruction. Endoscopy
2002; 34: 160-162
5 Sundermann A, Kammerer J. Retrosternale struma und osoph-
agusvarizen. Munch Med Wochenschr 1960; 102: 2133-2136
6 Barber PV, Edwards JD, Enoch BA, Shafiq M, Vallon AG.
Retrosternal goitre presenting as bleeding oesophageal vari-
ces. Br Med J 1976; 1: 564
7 Johnson LS, Kinnear DG, Brown RA, Mulder DS. ‘Downhill’
esophageal varices. A rare cause of upper gastrointestinal
bleeding. Arch Surg 1978; 113: 1463-1464
8 Fleig WE, Stange EF, Ditschuneit H. Upper gastrointestinal
hemorrhage from downhill esophageal varices. Dig Dis Sci
1982; 27: 23-27
9 Kelly TR, Mayors DJ, Boutsicaris PS. “Downhill” varices; a
cause of upper gastrointestinal hemorrhage. Am Surg 1982; 48:
35-38
10 Takahashi K, Ishitobi K, Kodera K, Kaneda S, Yazawa M,
Mimura T. Bleeding esophageal varices caused by Graves’ hy-
pervascular cervical goiter. Jpn J Surg 1986; 16: 363-366
11 Tsokos M, Bartel A, Schoel R, Rabenhorst G, Schwerk WB.
Fatal pulmonary embolism after endoscopic embolization of
downhill esophageal varix. Dtsch Med Wochenschr 1998; 123:
691-695
12 Bedard EL, Deslauriers J. Bleeding “downhill” varices: a rare
complication of intrathoracic goiter. Ann Thorac Surg 2006; 81:
358-360
13 Schmidt KJ, Lindner H, Bungartz A, Hofer VC, Diehl K. Me-
chanical and functional complications in endemic struma (au-
thor’s transl). MMW Munch Med Wochenschr 1976; 118: 7-12
14 Lagemann K. Upper oesophageal varices due to thyroid
enlargement. Fortschr Geb Rontgenstr Nuklearmed 1973; 118:
440-445
15 Heller SL, Meyer JR, Russell EJ. Spinal cord venous infarction
following endoscopic sclerotherapy for esophageal varices.
Neurology 1996; 47: 1081-1085
S- Editor Wang J L- Editor Wang XL E- Editor Ma WH
van der Veldt AAM
et al.
Bleeding downhill esophageal varices and goiter 5415
www.wjgnet.com