The Sequence of the Evil: A Case Report of Idiopathic Noncirrhotic Portal Hypertension Associated With Littoral Cell Angioma of the Spleen, 4 Years After the Successful Treatment of a Colon Cancer

Abstract

Idiopathic noncirrhotic portal hypertension is a rare clinical entity. Even rarer is the littoral cell angioma of the spleen. The almost simultaneous appearance of the two disorders in a 64-year-old Greek woman was a tempting stimulus for us to attempt to identify a common denominator between these two, apparently distinct, entities.

A 64-year-old woman presented to the emergency department, complaining of fatigue and debility, progressively deteriorating within the last 2 months. She had laboratory tests, from the previous days in a private laboratory, which showed anemia (Hct:20%) and low platelet count (Plt:100000). She also had a computed tomography (CT) scan of the abdomen 20 days ago, which revealed a moderately enlarged spleen interspersed with numerous hypodense areas that varied in size and did not enhance with intravenous contrast (Figure 1). She had not noticed any blood loss from the intestine.

Figure 1.

CT scan of the abdomen of our patient. Hypodense areas are seen in the spleen parenchyma and ascetic fluid surrounding the liver. CT, Computed Tomography.

She was a retired housewife. She had a history of ascending colon cancer 4 years ago, treated with the right hemicolectomy. She has had 7 cycles of chemotherapy after the surgery, and she was considered to be disease free for the last 3 years. She had diabetes mellitus type 2, arterial hypertension, and dyslipidemia.

Her physical examination was unrevealing with the exception of a moderately enlarged spleen. The basic laboratory test results are given in Table 1. The testing for infectious disease, autoimmune disorders, and thrombophilia was negative.

Table 1.

Laboratory Test Results.

General blood count		Biochemical testing
WBC	4.24 × 103/μl	UREA	49 mg/dl
RBC	2.17 × 106/μl	CRE	0.82 mg/dl
HCT	18.3%	AST	46 IU/l
HB	5.8 g/dl	ALT	46 IU/l
MCV	84.3 fl	γ-GT	176 IU/l
MCH	26.7 pg/cell	ALP	154 IU/l
DEC	1.5%	TBIL	0.36 mg/dl
INR	1.08	LDH	305 IU/l
ESR	25 mm/h	TP	5.52 g/dl
		ALB	3.7 g/dl
		GLOB	1.82 g/dl
		CK	322 IU/l
		CRP	0.34 mg/dl

WBC, White Blood Cells; RBC, Red Blood Cells; HCT, Hematocrit; HB, Hemoglobin; MCV, Mean Corpuscular Volume; MCH, Mean Corpuscular Hemoglobin; DEC, Dictyoerythrocytes; INR, International Normalized Ratio; ESR, Erythrocyte Sedimentation Rate; CRE, Creatinine; AST, Aspartate Aminotransferase; ALT, Alanine Aminotransferase; γ-GT, Gamma Glutamyltransferase; ALP, Alkaline Phosphatase; TBIL, Total Bilirubin; LDH, Lactate Dehydrogenase; TP, Total Proteins; ALB, Albumin; GLOB, Globulins; CK, Creatine Kinase; CRP, C-Reactive Protein.

The patient was admitted to the hospital and received 2 units of compact red blood cells. A new CT scan was performed, which confirmed the findings of the previous one, but also showed a small amount of ascetic fluid that was a transudate with high serum ascites albumine gradient (SAAG), few cells, and negative cytology. Although the blood cultures were negative, a heart echo was performed, which was normal. To exclude the possibility of a possible relapse of colon cancer, a colonoscopy was performed which was negative. A bone marrow biopsy was performed, and it gave normal findings. A whole-body position emission tomography (PET)/CT scan revealed no hypermetabolic areas. Gastroscopy was performed which revealed 5 esophageal varices and portal gastropathy. Three sessions of variceal ligation were performed. A triplex of the portosplenic axis revealed patent splenic and portal vein with normal flow velocities. A liver elastography excluded cirrhosis. Because we have failed to establish a definite diagnosis and fearing that a spleen lymphoma was present, we decided to proceed to splenectomy and had a liver biopsy taken. Much to our surprise, it was found that the moderately enlarged spleen bore a great number of vascular lesions, compatible with littoral cell angioma (LCA; Figure 2, Figure 3). The histology of the liver sample was compatible with nodular regenerative hyperplasia (Figure 4, Figure 5). Obliterative venopathy was not identified.

Figure 2.

Littoral cell angioma of the spleen. Anastomozing vascular spaces filled with blood. Hematoxylin and eosin stain, 4×.

Figure 3.

Littoral cell angioma of the spleen. Immunohistochemical staining of CD31 that shows endothelial differentiation, 10×.

Figure 4.

Nodular regenerative hyperplasia in the liver parenchyma. Hematoxylin and eosin stain, 10×.

Figure 5.

Nodular regenerative hyperplasia in the liver parenchyma. Reticulin stain, 20×.

The patient had an uneventful postsurgical course. The hematocrit rose to normal levels. No more variceal ligation was required. She was re-examined 6 months after the operation, and she was found perfectly healthy.

Discussion

For the practicing clinician, the simultaneous occurrence of two rare clinical entities, in a single patient, is a challenging experience. The obvious task is to try to find a link between the disorders, either by detecting similar cases, already described in the literature, or, ideally, by being able to put forward a pathogenetic mechanism combining the two, seemingly unrelated disturbances.

To begin with, we have to underline the fact that in our patient, the two diseases were convincingly confirmed on both clinical and histological grounds.

LCA of the spleen is a rare benign vascular tumor, often multifocal, first described by Falk et al in 1991.¹ Since then, 180 cases have appeared in the literature.² Its histology comprises anastomosing vascular channels with papillary projections and cyst-like spaces lined by tall endothelial cells that can often show hemophagocytosis.³ Its hallmark is the concurrent presence of two distinct markers, namely CD31 (an endothelial marker) and CD68 (characterizing monocytes/macrophages) on the membrane of the neoplastic, otherwise endothelial cells.³ In our case, the splenic endothelial cells stained positive for both these markers.

The interesting thing about LCA is that a considerable percentage of the described cases (15 of 25 cases) appear synchronously or heterochronously with malignancies of other organs, the colon included.^{4, 5} The mechanism is unknown, but it is logical to hypothesize that one or more growth factor(s) secreted by the cancer cells may have affected the development of the LCA in the spleen. Splenic microenvironment may have played a role. It is tempting to suggest that a similar mechanism was active in our patient, in view of her history of colon cancer.

Idiopathic noncirrhotic portal hypertension (INCPH) is another rare entity, tantalizing the clinicians for about a century from the era of Banti to nowadays. It has an eastern and a western variant.^{6, 7} It is clearly multifactorial, and five clusters of disorders have been associated with it. To sustain the diagnosis of INCPH, cirrhosis and other liver diseases have to be excluded and the splenic and portal veins have to be patent. The liver histology is interesting and is characterized by obliterative portal venopathy, increased number of portal vascular channels, dilated portal veins herniating into the surrounding parenchyma, sinusoidal dilatation, and periportal/perisinusoidal fibrosis.^{6, 7, 8} This complex pathology reflects the combined remodeling of small portal veins (obliterative portal venopathy) and, subsequently, the liver parenchyma (nodular hyperplastic regeneration). Items of both these vascular and parenchymal changes were present in the samples of the liver biopsy.

From the pathogenetic point of view, two mechanisms are considered prerequisites for the INCPH to develop: first, the increased blood flow via the portal vein⁹ and second, an underlying prothrombotic activity.⁸ The first mechanism is adequately accounted for, by the increased size of the spleen and the presumed increased venous blood drainage, via the splenic and portal veins. Some authors attribute the increased flow to enhanced production and secretion of NO by the endothelial cells of the splenic sinusoids.¹⁰

As far as the second mechanism is concerned, we detected no thrombophilic or prothrombotic propensity in our patient because the common tests performed in search of thrombophilia were negative. Notwithstanding this, we postulate that an underlying prothrombotic process must somehow participate. We are referring to the so-called microparticles or microvesicles, which comprise a newly discovered and presently intensively studied mechanism of cell-to-cell message transmission.¹¹ These small particles (<1 mm in diameter) that are secreted by cells, under various circumstances, contain a multitude of cell-derived substances,¹² among which tissue factor (III)¹³ is particularly relevant in our case, and are involved in prothrombotic, proinflammatory, and angiogenic processes. Their pathogenetic role is more or less firmly established in disorders such as pulmonary arterial hypertension,^{14, 15, 16} portal hypertension,¹⁷ and many inflammatory and neoplastic diseases. It is worth mentioning that endothelial cells are considered to be second only to platelets¹¹ in their capacity to release microparticles.

In short, to account for the patient's unusual combination of clinical manifestations, we propose the following pathogenetic sequence: ‘colon cancer begets LCA of the spleen and the latter begets INCPH’. Despite our meticulous search of the literature, we were able to identify only one similar case.¹⁸

Conflicts of interest

The authors have none to declare.

Funding

This study did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Author contributions

Konstantinou F, Liakopoulou, Konstantinou A, and Loverdos were the treating clinicians; they conceived and performed the investigation of this case and prepared the draft. Provatas analyzed the histopathology stains. Kokkinakis is the head of the department and observed the whole procedure.