• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of clinexpimmunolLink to Publisher's site
Clin Exp Immunol. Sep 2008; 153(3): 331–337.
PMCID: PMC2527366

Abnormal liver function in common variable immunodeficiency disorders due to nodular regenerative hyperplasia


Patients with common variable immunodeficiency disorders are monitored for liver function test abnormalities. A proportion of patients develop deranged liver function and some also develop hepatomegaly. We investigated the prevalence of abnormalities and types of liver disease, aiming to identify those at risk and determine outcomes. The local primary immunodeficiency database was searched for patients with a common variable immunodeficiency disorder and abnormal liver function and/or a liver biopsy. Patterns of liver dysfunction were determined and biopsies reviewed. A total of 47 of 108 patients had deranged liver function, most commonly raised alkaline phosphatase levels. Twenty-three patients had liver biopsies. Nodular regenerative hyperplasia was found in 13 of 16 with unexplained pathology. These patients were more likely to have other disease-related complications of common variable immunodeficiency disorders, in particular non-coeliac (gluten insensitive) lymphocytic enteropathy. However, five had no symptoms of liver disease and only one died of liver complications. Nodular regenerative hyperplasia is a common complication of common variable immunodeficiency disorders but was rarely complicated by portal hypertension.

Keywords: abnormal liver function tests, common variable immunodeficiency disorders, liver biopsy, nodular regenerative hyperplasia of the liver


Common variable immunodeficiency disorders (CVIDs) are a heterogeneous group of primary immunodeficiency disorders involving failure of antibody production and various cellular immune system defects. They affect between 1/10 000–20 000 of the population. Diagnosis depends upon low serum immunoglobulin (Ig) levels and inability to respond to test immunizations, with exclusion of other known causes of immunodeficiency [1]. Patients are highly susceptible to infections, due especially to encapsulated bacteria, and have a tendency to develop a range of associated complications. Patients can be divided by complication, with some patients experiencing splenomegaly, granuloma, enteropathy, autoimmunity including persistent cytopenias, polyclonal lymphoproliferative infiltration or lymphoid malignancies; others have no disease-related complications. Criteria for clinical phenotyping, based upon the presence or absence of particular disease complications, have now been devised [2]. These divide patients into five distinct types, with 83% of patients having only one clinical phenotype. The present study was devised to seek associations between complications, clinical phenotypes and underlying liver disease.

Treatment of CVIDs involves regular infusions of intravenous or subcutaneous Ig preparations. These preparations are extracted from large pools of donated plasma resulting in a risk of viral transmission including, historically, hepatitis C virus (HCV). Consequently it is standard practice to monitor patients' liver function tests (LFTs) every 8–12 weeks, to detect any signs of hepatitis [3,4]. However, not all patients with LFT rises have evidence of HCV infection [5,6].

The aim of this study was to characterize liver disease in patients with CVIDs, by assessing the prevalence of abnormal LFTs, identifying patterns of LFT derangement, looking at the cause and nature of liver disorder and defining who is at risk of developing liver disease.


The database

A database is maintained of all patients with an immunodeficiency disease who are seen and treated by the Department of Immunology at the Oxford Radcliffe Hospitals. The software package used was Filemaker 6. Clinical, radiological and laboratory data are stored on a regular basis, resulting in continuous information on patients for more than 20 years since diagnosis.

Extraction of data

The database contains the following details for each patient: a clinical summary, past medical history from childhood and details of clinic visits; blood test results, images and biopsies. Records of symptoms, infections and disease associations enable rapid searches of the database. Details include dates of onset of symptoms, diagnosis and commencement of therapy. Treatment details such as dose levels, product and infusion intervals as well as IgG trough levels are recorded. Additionally, all CVID patients have been assigned recently to one of the five phenotypes based on their profile of complications, as described above.

A search identified two groups of patients with a CVID: those with evidence of liver dysfunction (using terms of ‘abnormal LFTs’ or ‘liver biopsy’) and those without. For each patient all the LFT results were plotted over time, to search for temporal patterns in liver function derangement, in terms of alanine transaminase (ALT), alkaline phosphatase (ALP) and gamma glutamyl transpeptidase (γGT). All patients were screened for hepatitis C RNA by polymerase chain reaction. As this was a retrospective study, no accurate alcohol history was available.

The liver biopsy specimens were obtained for each patient and reviewed by a histopathologist specializing in liver histology (J. P.) without knowledge of the initial report or the patient histories. Each biopsy was assessed using Ishak scoring for fibrosis, portal inflammation, lobular inflammation, steatosis and pericellular fibrosis and comments were made about other features such as copper binding protein, granulomata and sinusoidal dilatation.

A review of the clinical phenotype of the CVID in each patient was also undertaken, to see whether any particular phenotype was more likely to develop hepatic complications.

Inclusion criteria

All patients with a CVID, as defined by the European Society for Immunodeficiencies criteria [7], with at least 1 year of follow-up, were included. Evidence of liver dysfunction was defined as abnormal LFTs for a minimum of 6 months or a liver biopsy performed in the past. ‘Abnormal’ for ALT was taken as greater than twice the upper limit of normal and for ALP was taken as greater than normal (250 iU/l) with concurrent rise in γGT.

Statistical analyses

For comparison of any two qualitative (categorical) variables, the calculation was based on the χ2 test for homogeneity using a Monte Carlo simulation of the exact distribution of the test statistic to determine the P-value. This was needed because of the relatively small cell sizes in many of the contingency tables evaluated.

Associations between the presence of nodular regenerative hyperplasia (NRH) and age of onset of symptoms, age at diagnosis, length of diagnostic delay and time on therapy were determined using Wilcoxon's rank test.


Patient selection

Of the 226 patients in the primary immunodeficiency database, 108 had a CVID. Regular LFT data was available for these patients over 1 to 25 years, with an average follow-up of 10 years. Forty-seven CVID patients had had deranged LFTs, either in the past or currently, of whom 21 had had a liver biopsy. Two patients with normal LFTs had also had liver biopsies. Of the 23 with documented liver histology, three were excluded due to previously documented HCV infection and four in whom the biopsy specimen was unavailable for review. Biopsy specimens from the remaining 16 patients were submitted to the histopathologist for blinded review. Human immunodeficiency virus screening had been undertaken in almost all (> 90%) of these 47 patients and found to be negative.

Liver function test patterns

There were 47 patients with abnormal LFTs, of whom six with known HCV infection and one with biliary obstruction due to a cholangiocarcinoma were excluded.

The most common liver enzyme to be affected in the 40 remaining patients was ALP (n = 31). Analysis of the graphs of ALP derangement revealed three distinct patterns (see Table 1): progressive elevation (n = 12), fluctuating increases (n = 15) or a transient increase only (n = 4). Of those with normal ALP (n = 9), three had abnormal ALT and γGT, two had abnormal ALT only and four had an isolated raised γGT.

Table 1
Patterns of alkaline phosphatase (ALP) derangement and frequency of patterns.

A history of alcohol excess was documented in the clinical notes in three patients, two of whom had normal LFTs and one of whom was infected with hepatitis C. In a further patient with an elevated ALP, an alcohol history was identified retrospectively as a consequence of the liver biopsy findings.

Biopsy results

Twenty biopsy slides were obtained from 16 patients, of whom four had had biopsies taken on two occasions. Thirteen patients were found to have NRH (see Table 2), two had granulomatous hepatitis only and one had previously unsuspected alcoholic liver disease. Thus, of the 108 patients with a CVID, a minimum of 13 had NRH. This may be an underestimate, as NRH may be present in patients with normal LFTs (in whom a biopsy may not be indicated). Furthermore, some of the patients with abnormal LFTs did not have a biopsy taken (n = 26) or the specimens were unobtainable (n = 4). Thus, this is a minimal prevalence of 12%.

Table 2
Results of review of liver biopsies in relation to clinical sequelae.

Only four of these cases had been diagnosed with NRH on the basis of these biopsies originally.

Histological features

Of the 13 patients with a CVID and NRH, the liver histology was almost normal in eight, with NRH as the only histological feature (patients 1, 3, 4, 5, 8, 9, 11 and 14 in Table 2). Two patients were cirrhotic (patients 7 and 10), only one of whom had co-existent NRH. The remaining five patients had moderate fibrosis; there were associated granulomata in three patients but in another two there was no apparent reason for fibrosis.

Correlation of LFT patterns with NRH on biopsy

Twelve of the 13 patients with NRH had elevated ALP with concurrent rises in γGT. The most common pattern of LFT derangement in patients with a diagnosis of NRH was that of ALP slowly increasing over years (six of 13 patients). Of the other patients with NRH, two had fluctuating ALP, four had peaks of ALP that then returned toward normal and one had normal ALP. The LFT patterns in patients with granuloma in the liver were not particularly remarkable.

Correlation of NRH with CVID complications

Common variable immunodeficiency disorder patients with NRH were more likely to have additional complications of CVIDs when compared with those patients without NRH (see Table 3). In particular, there were strong associations with the expected complications of hepatomegaly (P = 0·002), deranged LFTs (P = 0·001), granuloma (P = 0·0003) and cytopenias (P = 0·0004), but not organ-specific autoimmune conditions. NRH was associated with granuloma elsewhere in the body, not only the liver; 38% of patients with NRH had granuloma anywhere compared with only 10% of those without NRH. Of those who had granuloma elsewhere without evidence of NRH, five had normal LFTs and no symptoms (so were not biopsied) and one had abnormal LFTs but did not have a liver biopsy. Conversely, of the six patients who had granuloma in the liver (on review or on the original report), four had granulomata elsewhere: one patient (patient 6) with granuloma in the brain [8], another patient with splenic granuloma (liver biopsy unavailable) and two with skin granuloma (patients 2 and 15).

Table 3
Complications of common variable immunodeficiency disorders (CVIDs) in patients with and without biopsy-proven nodular regenerative hyperplasia (NRH)*.

Correlation of NRH with CVID disease course and treatment

Interestingly, there were no associations between NRH and age at onset (P = 0·84), age at diagnosis (P = 0·43), delay in diagnosis (P = 0·15) or length of Ig therapy (P = 0·17).

Although there were more patients (seven of 13, 54%) in the NRH group who have received three Ig products compared with those in the non-NRH group (17 of 95, 18%), the small numbers do not permit statistical significance. There were four of 13 (31%) with NRH compared with 43 of 95 (45%) who received the same Ig product throughout. The higher prevalence of NRH in those receiving three Ig products may reflect a longer time span of therapy. However, the length of Ig therapy was not associated with the development of NRH (P = 0·17), suggesting that NRH might be associated with usage of a variety of Ig products. NRH was not associated with one particular Ig product.

Correlation of biopsy results with overall clinical outcomes

Five patients with NRH had no symptoms or signs of liver disease. The other patients experienced complications including jaundice (n = 6), hepatomegaly (n = 6), pruritis (n = 3), ascites (n = 2) and oesophageal varices (n = 2) (see Table 4). Although 10 have died, this was due to complications of liver failure in only one patient. Only three patients with NRH without co-existent cirrhosis developed signs of portal hypertension (ascites or variceal haemorrhage). The prevalence of asymptomatic portal hypertension in the 13 patients with histological NRH is unknown, as these patients were not surveyed routinely by gastroscopy. In the six patients endoscoped for other reasons between 1 and 13 years (median 4 years) before the diagnosis of NRH, no varices were seen; no evidence of portal hypertension was seen endoscopically in five patients examined between 1 and 7 years (median 4 years) after the diagnosis of NRH. Of the patients who had a raised ALP but were not biopsied, none developed jaundice, variceal bleeding or liver failure. Two patients showed histological cirrhosis, of whom one died of liver failure (patient 7) and the other is alive with a histological liver picture consistent with alcoholic liver disease (patient 10). For comparison, the 95 patients without NRH are shown in Table 4.

Table 4
Comparison of clinical outcomes in patients with and without biopsy proven nodular regenerative hyperplasia (NRH).

Correlation of biopsy results with clinical phenotypes

Division of the whole cohort of patients into clinical phenotypes of CVIDs, according to the presence or absence of NRH, is shown in Table 5. In these patients, the strong association of NRH with lymphoproliferation (P = 0·0002) was not surprising as this phenotype is defined by the presence of granuloma, polyclonal lymphocytic infiltration in the lungs (lymphoid interstitial pneumonitis: LIP), persistent lymphadenopathy and unexplained hepatomegaly. However, there is an association with enteropathy (P < 0·0001) as a distinct phenotype and there is no association with the autoimmune phenotype which includes thrombocytopenia.

Table 5
Clinical phenotypes of common variable immunodeficiency disorder (CVID) patients with and without biopsy-proven nodular regenerative hyperplasia (NRH).


In this group of 108 patients with a CVID, 44% had deranged liver function lasting longer than 6 months, which was usually raised ALP alone. Recognized causes of this pattern include intrahepatic biliary obstruction (such as gallstones), infiltration by inflammatory cells to form granuloma or lymphocytic infiltration due to autoimmunity or malignancy. Non-liver causes of deranged LFTs that are known to occur in patients with CVIDs include enteropathy with villous atrophy with or without vitamin malabsorption, leading to osteomalacia and raised ALP [9]. Apart from the risk of HCV due to the use of blood products [4], the other liver diseases seen in this cohort were alcoholic (n = 1), not related pathologically to CVIDs and granulomatous hepatitis (n = 2), which is a known complication of CVIDs.

The most common abnormality on liver biopsy in this series was NRH in 13 of 16 patients. Since this analysis there has been a report with similar findings, in which 23 patients with a variety of primary antibody deficiency syndromes underwent a liver biopsy [10]. Of the 19 patients with a CVID in their study, 16 showed NRH, a similar proportion. Previously there was only one case report published of NRH in a CVID patient [11].

The prevalence of unexplained liver disease in CVIDs is low. In an unselected series, Cunningham-Rundles found that 27 patients (11·9%) had significant liver abnormalities, although this was due to HCV in 15 cases [12]. Malamut et al.[10] compared 51 patients with a variety of primary immune deficiencies (40 with CVIDs) selected for liver abnormalities with 43 patients (31 with CVIDs) with normal LFTs; they were unable to determine the prevalence. In our unselected CVIDs cohort, with a low threshold for liver biopsy, the prevalence of NRH is at least 12%, significantly higher than in the background population. The incidence of NRH in the general population, estimated from post-mortem examinations, is 0·72–2·6% [13,14].

NRH is a histological diagnosis, based on the presence of small, hyperplastic nodules in the liver parenchyma in the absence of fibrous septa around the nodules (which are found in cirrhosis). Wanless defined NRH as hepatocellular nodules of less than 3 mm diameter, not surrounded by marked fibrosis [13]. There is no inflammatory component. It is easy to miss unless sought deliberately, and reticulin staining is essential to detect NRH [15]. Clinically it is associated with a raised ALP but otherwise may be clinically silent for many years, as in our series. If complications do occur, presentation is with features of portal hypertension (hepatosplenomegaly or variceal bleeding) due to sinusoidal compression [16,17]. These complications have been reported in series of organ transplant recipients given azathioprine [18]. Exposure to azathioprine and other causes of NRH [16], such as rheumatoid arthritis, systemic lupus erythematosus, Felty's syndrome, coeliac disease, haematological malignancies and increased cellular production in bone marrow, were excluded in our patients. It has been suggested that a common underlying aetiology for these conditions might involve alterations in blood flow due to vascular injury or inflammation. NRH may be a non-specific tissue response to the parenchyma of the liver receiving varied blood flow to different areas, rather than being a specific pathological process [13].

The cause of such variation in CVIDs remains to be determined. In the present study there were no associations between NRH and delay in diagnosis or length of Ig therapy. Only three of the 108 patients had received azathioprine and no other implicated drugs had been given. Unlike the French cohort [10] there was no association with general autoimmune conditions, although that with cytopenias was significant. Our interpretation of the common finding of general lymphocytic abnormalities (79% in their series) is different, as this suggests an infiltrative process rather than an autoimmune one. The association with non-gluten sensitive enteropathy is interesting, as this also suggests an infiltrative process; this may also explain the increased finding of NRH in immunocompetent patients with coeliac disease. It was reassuring that correlation with length of time on Ig therapy was not significant, despite higher IgG trough levels in our group, suggesting that transmission of an unknown pathogen is not likely.

Common variable immunodeficiency disorder patients with NRH were more likely to have complications, particularly lymphoproliferation, granuloma, enteropathy or cytopenias than those without NRH. This suggests that the presence of a lymphocytic infiltrate, possibly as part of an inflammatory process, may play a part in the aetiology of NRH. The presence of CD8+ T cells in the liver sinusoids in some (14 of 44) patients with NRH has led to the suggestion that NRH may be the result of chronic cytotoxic T cell infiltration of the sinusoidal epithelium [19]. Association with granuloma is interesting, as NRH has been reported in 4·6% of patients with chronic granulomatous disease, another primary immunodeficiency, but no mechanism was suggested [20].

Nodular regenerative hyperplasia may remain clinically silent for many years and only 25% of patients exhibit elevated ALP, with other serum parameters (transaminases, bilirubin, albumin and clotting) being normal [16]. NRH usually develops slowly in patients with chronic systemic illness but little evidence of liver disease. The biochemical findings of a raised ALP with normal or mildly raised transaminases and normal albumin seen in our cohort are typical [16,17,21]. There has been no prospective or retrospective cohort study of patients with asymptomatic NRH, although detection in patients without clinical signs of liver disease indicates that the long-term prognosis is good. The high prevalence of portal hypertension (50%), reported in the French cohort [10], may reflect the selected nature of study population. However, only three of 51 patients in our cohort suffered liver failure and this was concomitant with viral hepatitis in two of these.

Although there is no specific treatment for NRH, besides control of the underlying condition and management of complications should they arise, screening to detect evidence of portal hypertension with endoscopy to identify varices is important. Ultrasound is less helpful, as splenomegaly is seen in CVIDs in the absence of portal hypertension. However, Doppler ultrasound may detect portal hypertension indirectly in the form of collateral vessels and a dilated portal vein. A link with hepatocellular carcinoma is possible, although unproven [22].

In conclusion, we suggest that those CVID patients with an ALP level more than 1·5 times the upper limit of normal for more than 6 months be referred to a hepatologist for further investigation. A liver biopsy (including reticulin staining) should be performed to look for NRH and granulomata, as both are associated with elevated ALP levels and are indistinguishable clinically, and the latter may respond to steroids. A liver biopsy also helps to exclude other causes of chronic liver disease and was useful in one case in making a diagnosis of alcoholic liver disease, which can also be associated with elevated ALP levels. Careful monitoring of CVID patients to detect autoimmune conditions (including cytopenias), enteropathy or lymphoproliferation should be undertaken, as those with such clinical phenotypes have been shown to have reduced survival [2]. As relying upon abnormal blood tests will miss some patients with NRH, because only about 25% of patients with NRH have raised ALP, patients with unexplained hepatomegaly should be referred. As liver biopsy carries a risk of bleeding it is inappropriate to recommend a biopsy in all patients with a CVID as a screening tool.


We are grateful to the many junior medical staff and research co-ordinators who have contributed to this database and to the patients who have allowed us to collect and store this data. Dr Martin Lee kindly provided statistical analysis. We are grateful to our colleagues, Drs R. Chapman, K. Fleming, S. Misbah, J. Lortan and E. Lopez-Granados. This work was supported by funding from: the European Union 5th Framework grant (EUROPID QLQ1-CT-2001-01395), European Union 6th Framework grant (EUROPOLICY SP23-CT-2005-006411), Baxter Healthcare, the UK Primary Immunodeficiency Association and the Jeffrey Modell Foundation.


1. Geha RS, Notarangelo LD, Casanova J-L, et al. Primary immunodeficiency diseases: an update from the international union of immunological societies primary immunodeficiency diseases classification committee. J Allergy Clin Immunol. 2007;120:776–94. [PMC free article] [PubMed]
2. Chapel H, Lucas M, Lee M, et al. Common variable immunodeficiency disorders: division into distinct clinical phenotypes. Blood. pre-published online: blood-2007-11-124545. [PubMed]
3. Healey CJ, Sabharwal NK, Daub J, et al. Outbreak of acute hepatitis C following the use of anti-hepatitis C virus – screened intravenous immunoglobulin therapy. Gastroenterology. 1996;110:1120–6. [PubMed]
4. Quinti I, Pierdominici M, Marziali M, et al. Surveillance of immunoglobulin safety – results of initial survey of 1243 patients with primary immunodeficiencies in 16 countries. Clin Immunol. 2002;104:231–6. [PubMed]
5. Bjoro K, Haaland T, Skaug K, Froland SS. The spectrum of hepatobiliary disease in primary hypogammaglobulinaemia. J Intern Med. 1999;245:517–24. [PubMed]
6. Rodrigues F, Davies EG, Harrison P, et al. Liver disease in children with primary immunodeficiencies. J Pediatr. 2004;145:333–9. [PubMed]
7. Conley ME, Notarangelo LD, Etzioni A. Diagnostic criteria for primary immunodeficiencies. Clin Immunol. 1999;93:190–7. [PubMed]
8. Misbah SA, Spickett GP, Esiri MM, et al. Recurrent intra-cranial granulomata presenting as space-occupying lesions in a patient with common variable immunodeficiency. Postgrad Med J. 1992;68:359–62. [PMC free article] [PubMed]
9. Aslam A, Misbah SA, Talbot K, Chapel H. Vitamin E deficiency induced neurological disease in common variable immunodeficiency: two cases and a review of the literature of vitamin E deficiency. Clin Immunol. 2004;112:24–9. [PubMed]
10. Malamut G, Ziol M, Suarez F, et al. Nodular regenerative hyperplasia: the main liver disease in patients with primary hypogammaglobulinemia and hepatic abnormalities. J Hepatol. 2008;48:74–82. [PubMed]
11. Ravindran J, Gillis D, Rowland R, Heddle R. Common variable immunodeficiency associated with nodular regenerative hyperplasia of the liver. Aust NZ J Med. 1995;25:741. [PubMed]
12. Cunningham-Rundles C, Bodian C. Common variable immunodeficiency: clinical and immunological features of 248 patients. Clin Immunol. 1999;92:34–48. [PubMed]
13. Wanless IR. Micronodular transformation (nodular regenerative hyperplasia) of the liver: a report of 64 cases among 2,500 autopsies and a new classification of benign hepatocellular nodules. Hepatology. 1990;11:787–97. [PubMed]
14. Colina F, Alberti N, Solis JA, Martinez-Tello FJ. Diffuse nodular regenerative hyperplasia of the liver (DNRH). A clinicopathologic study of 24 cases. Liver. 1989;9:253–65. [PubMed]
15. Geller SA, Dubinsky MC, Poordad FF, et al. Early hepatic nodular hyperplasia and submicroscopic fibrosis associated with 6-thioguanine therapy in inflammatory bowel disease. Am J Surg Pathol. 2004;28:1204–11. [PubMed]
16. Reshamwala PA, Kleiner DE, Heller T. Nodular regenerative hyperplasia: not all nodules are created equal. Hepatology. 2006;44:7–14. [PubMed]
17. Arvanitaki M, Adler M. Nodular regenerative hyperplasia of the liver. A review of 14 cases. Hepatogastroenterology. 2001;48:1425–9. [PubMed]
18. Gane E, Portmann B, Saxena R, Wong P, Ramage J, Williams R. Nodular regenerative hyperplasia of the liver graft after liver transplantation. Hepatology. 1994;20:88–94. [PubMed]
19. Ziol M, Poirel H, Kountchou GN, et al. Intrasinusoidal cytotoxic CD8+ T cells in nodular regenerative hyperplasia of the liver. Hum Pathol. 2004;35:1241–51. [PubMed]
20. Hussain N, Feld JJ, Kleiner DE, et al. Hepatic abnormalities in patients with chronic granulomatous disease. Hepatology. 2007;45:675–83. [PubMed]
21. Naber AHJ, Van Haelst U, Yap SH. Nodular regenerative hyperplasia of the liver: an important cause of portal hypertension in non-cirrhotic patients. J Hepatol. 1991;12:94–9. [PubMed]
22. Nzeako UC, Goodman ZD, Ishak KG. Hepatocellular carcinoma and nodular regenerative hyperplasia: possible pathogenetic relationship. Am J Gastroenterol. 1996;91:879–84. [PubMed]

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...