Portal Hypertension in Coeliac Disease: A Controversial Association ()
1. Introduction
Celiac disease (CD), or gluten-sensitive enteropathy (GSE), is an autoimmune disorder characterized by reversible small bowel mucosal inflammation with villous atrophy affecting patients with a specific genetic predisposition (HLA DR3-DQ2 and HLA DR4-DQ8). The mucosal lesion develops after ingestion of dietary gluten and recovers when gluten-containing cereals, wheat, rye, and barley are withdrawn from the diet [1].
Gluten-induced immune effects are not limited to the intestine alone, but other organs such as the skin, brain, and bones are also affected. The liver is also commonly involved in patients with CD [2]. Liver dysfunctions in CD include asymptomatic cytolysis, autoimmune hepatitis, nodular regeneration of the liver, portal thrombosis, Budd-Chiari syndrome, and cirrhosis [3]-[6].
However, with a surge in interest and research on CD, patients with idiopathic portal hypertension (PHT) were being found to have positive serology and diagnostic duodenal biopsy changes for CD [7].
The aim of our study is to report the rate of PHT in CD, to investigate the coexistence of CD in patients with PHT of various etiologies, as well as to analyze the clinical profiles of patients with CD associated with PHT, and to describe the impact of gluten-free diet (GFD) on their evolution.
2. Materials And Methods
2.1. Study Design and Population
This was a retrospective and descriptive study conducted in the Department of Hepato-Gastroenterology at Hassan II University Hospital in Fez, Morocco. The study covered the period between January 2009 and July 2022 and included 173 patients diagnosed with celiac disease (CD). Among these, 16 patients (9.24%) also presented with portal hypertension (PHT), which was the focus of this study.
2.2. Data Collection
Data were collected from three main sources: hospitalization records from the Department of Hepato-Gastroenterolog, electronic medical records via the hospital information system “Hosix”, a standardized data collection sheet designed for this study to gather detailed epidemiological, clinical, paraclinical, therapeutic, and follow-up data.
2.3. Inclusion and Exclusion Criteria
Inclusion criteria: All patients with confirmed celiac disease who developed portal hypertension either before or during the course of their illness.
Exclusion criteria: Patients with portal hypertension of known cirrhotic origin (compensated or decompensated) were excluded to focus on non-cirrhotic or less common etiologies.
2.4. Variables Studied
Epidemiological data: Age, sex, date of admission, family and personal medical history, and risk factors for PHT.
Clinical data: Diagnosis of PHT was based on clinical features (e.g., ascites, splenomegaly, digestive bleeding, and collateral circulation), radiological findings (Doppler ultrasound, abdominal CT, and MRI) showing portal vein dilation, porto-systemic shunting, and ascites, and endoscopic evidence of esophageal/gastric varices or portal hypertensive gastropathy.
Etiological investigation: Included screening for alcohol intake, HBsAg, anti-HCV antibodies, lipid profile, fasting blood glucose, HbA1c, protein electrophoresis (PEP), autoimmune antibodies (AMA, ASMA, LKM, ANA), total IgG, TTG-IgA, and iron/copper metabolism parameters (ferritin, serum iron, transferrin saturation, cupremia, ceruloplasmin, urinary copper).
For patients presenting with portal vein thrombosis or hepatic vein (suprahepatic) thrombosis, a thrombophilia work-up was performed. This included the measurement of protein C, protein S, factor V, and homocysteine levels, as well as screening for the JAK2 mutation. In young female patients, antiphospholipid antibodies were also tested. Liver biopsy: In selected cases with inconclusive non-invasive workup and no contraindications, an ultrasound-guided percutaneous liver biopsy was performed.
2.5. Statistical Analysis
All collected data were recorded in Microsoft Excel, qualitative variables were expressed as number and percentage (n (%)), quantitative variables were described using mean ± standard deviation or median and range, depending on distribution. Results were summarized using tables and graphs for clarity.
2.6. Ethical Considerations
No prior information or consent was obtained from the patients before their inclusion in the study. However, all personal data were anonymized in accordance with current regulations and ethical standards.
3. Results
We included 16 (9.24%) cases of CD and PHT association from 173 cases of CD followed in our unit. The mean age of our patients was 32 years (range: 16-62 years), with a female predominance of 115 (66.19%).
The main symptoms of CD discovery were chronic diarrhea in 80 (46.3%) cases, deterioration of general condition in 114 (65.7%) cases and anemia in 31 (17.9%) cases. The CD was diagnosed during screening in 30 (17.34 %) cases and was incidentally discovered in 29 (16.76%) cases. CD was detected during the etiological assessment of PHT in 8 (50%) cases (Table 1).
Table 1. Summarizes the modes of detection of celiac disease.
Mode of discovery |
Frequency (%) |
Deterioration of general condition |
114 (65.89%) |
Chronic diarrhea |
80 (46.3%) |
Anemia |
31 (17.9%) |
Screening |
30 (17.34%) |
Incidentally discovered |
29 (16.76%) |
Serology showed positive IgA tissue transglutaminase antibodies (TTG-IgA) in 97 (56.06%), TTG-IgG in 65 (37.57%), and IgA endomysial antibodies (EMA) in 11 (6.35%) out of 173 patients.
Abnormalities detected during upper endoscopy that are characteristic of celiac disease, but not specific to the disease, include scalloped folds and fissuring in the duodenum in 145 (83.60%) cases.
Intestinal biopsy analysis is therefore used to confirm the diagnosis of celiac disease; all of our patients underwent a duodenal biopsy, the abnormalities founded were lymphocytic exocytosis with villous atrophy in all of patients 173 (99.9%).
For the diagnosis of PHT, ascites was present in 8 (50%), splenomegaly in 5 (31.25%), collateral circulation in 2 (12.5%) and digestive bleeding in 11 (68.75%) patients. Radiological findings (Doppler ultrasound, abdominal CT, and MRI) showing portal vein dilation in 9 (56.25%), porto-systemic shunting 5 (31.25%), and ascites 10 (62.5%) cases. Upper endoscopy found esophageal variceal in 9 (56.25%) patients, isolated gastric variceal in 2 (12.5%) patients, and hypertension gastropathy in 7 (43.75%) patients.
The causes of PHT were portal thrombosis (PT) in 5 (31.25%) cases, Budd-Chiari syndrome (BCS) in 3 (18.75%) cases, and one patient had combined PT and BCS. Primary sclerosing cholangitis in one case (6.25%), postviral C cirrhosis in one case (6.25%), cirrhosis on autoimmune hepatitis in one case (6.25%), and cryptogenic cirrhosis in 5 (31.25%) cases (Table 2).
Table 2. summarizes the different etiologies of PHT.
Causes of PHT |
Frequency (%) |
Portal thrombosis |
5 (31.25%) |
Budd chiari syndrome 3 (18.75%) |
None |
Primary sclerosing cholongitis |
1 (6.25%) |
Postviral C cirrhosis 1 (6.25%) |
None |
Cirrhosis on autoimmune hepatitis |
1 (6.25%) |
Cryptogenic cirrhosis 5 (31.25%) |
None |
Patients with portal thrombosis and/or Budd-Chiari syndrome, 8 (50%) cases, underwent a thrombophilia workup, which included testing for Factor V Leiden, protein S, protein C, and homocysteine levels. The results revealed hyperhomocysteinemia in one (12.5%) case and a decreased protein C level in another case (12.5%); the remaining, 6 (75%), patients had normal thrombophilia profiles. No patient tested positive for the JAK2 mutation, and all female patients had negative antiphospholipid antibody results.
In a patient with a high likelihood of autoimmune hepatitis, a liver biopsy revealed portal and periportal inflammatory infiltrates rich in plasma cells, confirmed by CD138 immunohistochemistry. This was accompanied by portal fibrosis with fibrous septa (F2), as demonstrated by trichrome staining. Hepatic lobules showed normal-thickness hepatocyte plates, highlighted by reticulin staining, with regular hepatocyte morphology.
For cirrhotic patients, 5 (31.25%), with inconclusive etiological assessments, liver biopsies was performed in 2 patients (40%) and was indicated inactive chronic liver disease with moderate to severe fibrosis (F2-F3). No biopsy findings suggested celiac hepatitis.
All our patients have been put on a gluten-free diet (GFD).
Patients with portal thrombosis and Budd-Chiari syndrome were put on anticoagulant therapy (ACT), follow-up imaging (duplex ultrasound or abdominal CT angiography) demonstrated patency of the portal vein and hepatic veins under ACT and GFD. Patient with cirrhosis on autoimmune hepatitis was put on corticosteroids and azathioprine, the patient’s progress was characterized by stable laboratory results, with no evidence of cytolysis under corticosteroid therapy combined with a GFD. The patient with cirrhosis was put on GFD and clinical-biological monitoring. The patient’s course was marked by clinical and biological stability, with no cirrhosis decompensation under a GFD.
4. Discussion
CD is a common chronic autoimmune disorder affecting the small intestine, with a prevalence of 1%. It can be associated with autoimmune liver diseases such as autoimmune hepatitis, primary biliary cirrhosis, and primary sclerosing cholangitis. Currently, an increasing number of CD cases are being reported in association with various other liver diseases [8]. In Sweden, a large population-based study showed that individuals with CD have a 2- to 6-fold increased risk of developing liver disease later in life, and that a pre- existing autoimmune liver disease increases the risk of subsequently developing CD by 4- to 6-fold [9]. In our study, PHT was observed in 9.24% of patients with CD, which is consistent with data reported in the literature, including a study conducted in India that found 2 out of 14 (14%) patients with CD had non-cirrhotic idiopathic portal hypertension (NCIPH) [10].
Among patients with portal PHT, most cases are attributed to cryptogenic or idiopathic chronic liver disease. A study conducted in Sweden comparing patients with cryptogenic chronic liver disease (cCLD) to the general population showed that the prevalence of celiac disease (CD) was 15 times higher in the cCLD group than in the general population. Another Indian study revealed that 10% of individuals with non-cirrhotic idiopathic portal hypertension (NCIPH) had CD confirmed with duodenal biopsy. The diagnosis of CD preceded that of PHT in 17% of cases, compared to 50% in our study [11] [12].
In other cases, PHT was associated with multiple etiologies in conjunction with CD, including portal thrombosis, Budd-chiari syndrome, autoimmune hepatitis, and other diseases.
Rare cases of an association between CD and portal vein thrombosis have been reported in the literature [13]. A recent meta-analysis [14] reported a 25% increased risk of developing venous thrombosis in patients with CD compared to the general population. A recent review revealed that venous thrombosis is significantly more frequent in the context of CD, affecting approximately 80% of patients, with hepatic vein thrombosis (Budd-Chiari syndrome) observed in 30.91% of cases. This risk is higher in patients who do not adhere to a gluten-free diet (GFD) [15]. In our study, we observed portal vein thrombosis and/or Budd-Chiari syndrome in 8 patients, accounting for 50% of cases.
As previously mentioned, CD can coexist with other autoimmune liver diseases [8], and these hepatic conditions may be more severe [16]. In 1988, the first reported association between CD and primary sclerosing cholangitis (PSC) was described [17]. Since then, several studies have been published. A study involving 61 patients with PSC found a CD prevalence of 1.6%. Another larger study conducted in Sweden suggested that the prevalence of PSC in patients with CD is 4 to 8 times higher than in individuals without CD [18]. In our study, we report one case of PSC in a patient with CD, representing 6.25%.
As for autoimmune hepatitis (AIH), the first case was described by Lindberg et al. [19] in 1979, and since then, several associations between CD and AIH have been reported in the literature [20] [21]. In prospective studies, the prevalence of AIH in patients with CD ranges from 0.6% to 0.8% [22]; our study revealed a similar rate (0.58%).
AIH and CD share common features, including a high frequency of autoantibodies, the same type of inflammatory infiltrate in target organs, and common HLA phenotypes such as DR3 and DQ2. CD may precede or coexist with liver disease. The impact of a gluten-free diet (GFD) on AIH is limited, as AIH is generally not influenced by dietary changes [23].
Other non-autoimmune conditions, such as viral infections, may also coexist with CD. A study involving 259 patients with chronic hepatitis C revealed that the prevalence of CD in these patients was 1.2%, compared to 0.4% in non-infected individuals—a 3-fold increased risk [24]. Another study including 534 patients with chronic hepatitis C showed a CD prevalence of 1.3% [25]. In our study, this association was observed in only one patient, representing a rate of 0.58%.
Patients with cirrhotic liver disease (CLD) and CD have an 8-fold increased risk of death [26]. This highlights the need for systematic screening for CD before establishing a diagnosis of cryptogenic cirrhosis [27]. Such screening should be performed in at-risk patients or those presenting with one or more risk factors, including: a family history of gluten-sensitive enteropathy or dermatitis herpetiformis, presence of HLA-DQ2 or HLA-DQ8 haplotypes, type 1 diabetes, early-onset osteoporosis, or osteomalacia [28] [29].
5. Conclusions
This study highlights the significant prevalence of PHT in patients diagnosed with CD, with a notable occurrence rate of 9.24%. The findings suggest a complex interplay between CD and various liver conditions, including portal thrombosis, Budd-Chiari syndrome, and autoimmune hepatitis, underscoring the necessity for comprehensive evaluation in patients presenting with liver dysfunction.
This retrospective analysis of 16 patients provides valuable insights into the clinical manifestations and diagnostic challenges associated with CD, particularly in the context of PHT. Furthermore, the impact of a GFD on the management and progression of liver-related symptoms in these patients showcases the importance of dietary intervention in improving clinical outcomes. These results emphasize the need for clinicians to remain vigilant for signs of CD in patients with unexplained PHT, as early diagnosis and treatment may significantly alter the disease trajectory and enhance patient quality of life. Future research should focus on elucidating the underlying mechanisms linking CD and liver pathology, as well as establishing standardized screening protocols for at-risk populations.