Endoscopic and histopathological hints on infections in patients of common variable immunodeficiency disorder with gastrointestinal symptoms

Common variable immunodeficiency (CVID) is the commonest symptomatic primary immunodeficiency with a prevalence ranging from 1/50,000 to 1/200,000. Affected individuals typically manifest with hypogammaglobulinaemia, poor vaccination response, and recurrent infections. The pathogenesis is multifactorial, including several humoral and cell-mediated immunity defects involving B cells, T cells, natural killer cells, macrophages and monocytes, and genetics as well [1, 2]. Given the underlying immunodeficiency heterogeneity, clinical manifestations include recurrent infections, autoimmunity, inflammation, and malignancies. The gastrointestinal (GI) tract which is the largest lymphoid organ is also fairly involved. Common GI symptoms include diarrhea, bloating and abdominal pain. Up to 20% ~ 60% of CVID patients suffer from chronic diarrhea and even malabsorption in severe cases, leading to weight loss, anemia, iron deficiency, loss of minerals and fat-soluble vitamins [1‐5]. Importantly, the presence of enteropathy elevates the risk with a 4-fold increase mortality [6]. Why some patients have GI involvement while others do not, remains elusive. Studies on small bowel histopathology in CVID revealed various findings. Most patients, but remarkably not all, lack plasma cells. The major bowel histopathology could be categorized into three patterns: (i) a celiac-like histology, mainly with villous atrophy and increased numbers of intraepithelial lymphocytes (IELs). (ii) IBD-like changes, (iii) nodular lymphoid hyperplasia (NLH) [4, 7, 8]. NLH has been reported in 18% ~ 53% of CVID patients and likely to be underestimated, since asymptomatic subjects may not undergo endoscopy, radiography or biopsy procedures [9‐11]. All these conditions might be associated with chronic diarrhea, indicating abnormal immune response in GI tract. There are likely diverse immune-mediated mechanisms resulting in these histological changes, occasionally driven by infections.

CVID patients are usually susceptible to respiratory infections. Meanwhile, recurrent intestinal infections are also common, caused by Giardia lamblia, Campylobacter species, Salmonella species, cytomegalovirus, norovirus, Clostridium difficile, especially in patients with undetectable serum IgA [3, 7, 10, 12, 13]. Intravenous immunoglobulin (IVIG) therapy is insufficient to improve GI symptoms and does not accelerate the response to antibiotics and the eradication of pathogens. IVIG can only substitute IgG, while IgA and IgM, the major secretory antibodies at mucosal surface, cannot be replaced [3, 14, 15]. That made the CVID-associated intestinal infections more challenging.

Endoscopy is helpful for identifying GI mucosal lesions and providing hints on the underlying pathogenic mechanisms. Previous endoscopic studies showed various findings in CVID patients including inflammatory changes (edema, erosion, or ulcer), villous atrophy, NLH, adenoma and even malignancy. NLH may be observed in general population, especially in the ileum of young people, whereas NLH in CVID patients is usually more obvious and widespread, involving the proximal small intestine as well as the distal ileum, even proximal colon. NLH has also been reported to increase the risk of GI tumors (mainly GI lymphoma) and recurrent infections (particularly by Giardia lamblia) [16, 17]. So far, there was insufficient evidence to illustrate the correlation among clinical manifestations, infections, endoscopic and histopathological features in CVID patients.

Therefore, we performed a retrospective study in CVID patients with GI symptoms who underwent endoscopy and biopsy, with the aim to reveal the endoscopic and histopathological characteristics as well as their correlation with clinical and immunological features, especially with infections.

In our CVID cohort, 23% (20/87) of patients experienced diarrhea, with a high malabsorption rate as determined by D-xylose absorption test. Quinti I reported chronic diarrhea was observed in 22.4% of CVID patients and resulted in a significant malabsorption in 8.1% [5]. Chapel H reported 9% CVID enteropathy, which was defined as biopsy-proven lymphocytic infiltration in lamina propria and interepithelial mucous with villous atrophy, insensitive to gluten withdrawal [6]. However, CVID enteropathy lacks universal definition, with estimated prevalence of 9% ~ 34% [22]. Endoscopy could give a general clue for villous atrophy with confirmation by histopathology [7, 19]. Previous studies found intestinal villous atrophy in approximately 30% ~ 50% of CVID patients with GI symptoms [21‐23]. We found he duodenal villous atrophy was over 85%, which was higher than previous reported. IM Andersen categorized CVID enteropathy into severe and non-severe types based on whether there were weight loss, malnutrition, and severe GI-loss [22]. In our cohort, 85% of patients were under weight, and 52% presented with nutritional anemia, suggesting severe disease manifestation. Some disparities in findings might arise due to sample size limitations, disease severity, and regional factors. Other conditions displaying villous atrophy need to be differentiated including celiac disease, autoimmune enteropathy, tropical sprue, protracted viral or bacterial infection, giardiasis, T-cell lymphoma, food protein hypersensitivity, and graft-versus-host disease [20, 24]. CVID patients have hypoimmunoglobins, profoundly decreased plasma cells and NLH in mucosa, and resistance to gluten-free diet [21, 23]. So, the clinical history, biopsy histopathology and therapy response could help to get accurate diagnosis.

Further differentiation is required when CVID enteropathy presents with hypoalbuminemia from conditions like protein loss enteropathy (PLE). PLE can be caused by different mechanisms: increased lymphatic pressure, mucosal erosions, and increased mucosal permeability. It has been reported that PLE has lower infection and lymphoproliferation rate, higher serum levels of IgG, and mildly decreased to normal serum levels of IgA (> 0.5 g/L) than CVID. However, PLE can occur during CVID and requires higher IgG replacement therapy dosage [25]. There are 10 patients with hypoalbuminemia in our study, who all had obvious low IgA levels. Nine of them lacked plasma cells in gut biopsy, and the other one with mild hypoalbuminemia had no etiology of PLE. All above support their CVID diagnosis.

NLH shows pseudopolypoid appearance with multiple or occasionally innumerable nodules measuring 2–3 mm and usually not exceeding 10 mm in diameter of duodenum or ileum mucosa under endoscopy, sometimes through the whole small bowel [4, 10, 11, 23, 26]. Biopsy could confirm the diagnosis of NLH in microscopic view, however, it is dependent on the biopsy site and hard to evaluate the gross degree. Endoscopy is a good compensatory tool for the general evaluation. NLH had been reported in following conditions: CVID, selective IgA deficiency syndrome, giardiasis, H. pylori infection (gastric-NLH), food hypersensitivity, HIV, familial adenomatous polyposis, and GI malignancy, especially lymphoma [9, 17, 26]. Our data suggests that diffuse and obvious NLH may indicate infections, especially with Giardia and bacteria. A previous study on infections in 252 CVID patients showed that 47% had GI symptoms, 14% had Giardia lamblia infection and 19% had other GI bacterial infections [13]. Giardia lamblia could be one of the antigenic stimulators and associated with NLH in patients with or without immunodeficiency syndromes, leading to watery diarrhea, steatorrhea, and malabsorption [17, 27‐29]. It is reported that diffuse NLH of the bowel associated with CVID and refractory giardiasis markedly improved after successfully treating giardiasis [30]. The pathogenesis of NLH is still unknown. Infection maybe a trigger of mucosal immune response and disturbance. Repetitive stimulation of infectious agents probably lead to the hyperplasia of lymphoid follicles [9]. Lymphoproliferation was also present elsewhere in CVID, such as splenomegaly and lymphadenopathy. Similarly, H. pylori infection could cause gastric nodularity which could be normal after H. pylori eradication treatment. Increased IELs might also be an immune compensation and dysregulation. In all, infections could cause both acute diarrhea and chronic immune dysregulation in CVID enteropathy. Our study showed that diffuse and obvious NLH might be an endoscopic clue for infections in these patients. Although NLH may be related to high risk of malignancy, we do not find in our study.

The reason for the variance in CVID manifestations, with some patients developing GI symptoms and others not, remains unclear. T-cell dysfunction and autoimmunity against intestinal tissue, absence of mucosal plasma cells and defective antibody production, especially mucosal IgA, have been reported [3]. On the other hand, infections maybe also an important factor such as chronic norovirus [22]. Early diagnosis and IVIG replacement therapy (0.4 to 0.5 g/kg/month) can reduce the incidence of respiratory tract infection. However, IVIG did not improve diarrhea, especially in patients with lower serum IgA titers. Different options have been used to treat GI symptoms, for example, antibiotics such as metronidazole or ciprofloxacin, 5-aminosalicylic acid, and immunosuppressive agents such as corticosteroids, azathioprine, and infliximab [2, 12]. Endoscopy is a good general evaluation providing useful information for mucosal change and possible co-infections. The endoscopic and histopathological assessment should be performed in CVID with GI symptoms to facilitate diagnosis and treatment.

Our study has several limitations. Firstly, it was a single center retrospective study with potential biases due to a limited number of patients. Secondly, we utilized the ESID/PAGID (1999) criteria for CVID, while more recent updates have been made. However, these were all phenotypic diagnostic criteria, and there will be no definite CVID diagnoses, until the genetic or other etiopathogenetic causes can be defined for all patients [31, 32]. It was reported that the probable disease-causing mutations were found in 30% CVID patients by whole exome sequencing (WES) [33]. The gene analyses (WES, whole genome sequencing, or panel) was not performed routinely in our clinic. We did not know whether the patients had monogenic diseases, which might be done in the future. Thirdly, we lacked consistent laboratory data across time, but ensured expert reassessment of endoscopic and histopathological findings. Finally, the improvement was evaluated by clinical symptoms. We could not evaluate endoscopic and histopathological features from patients without GI symptoms or those post-treatment.

CVID frequently affects the GI tract, the biggest immune and environmental interface. The precise role of infections remains uncertain but is believed to potentially trigger immune regulation and impact on disease progression. Endoscopic evaluation, especially on small intestine, is crucial for CVID patients with GI symptoms. The diffuse and obvious NLH may be a sign for GI infection. Comprehensive endoscopy and histopathology assessment can offer vital diagnostic clues for CVID enteropathy and infections, facilitate prompt treatment, and ultimately improve patient quality of life. Further research is needed to refine treatments for this patient group.