IgG4-Related Hepatic Disease Unmasking Hepatic Actinomycosis

Abstract

Purpose

Hepatic actinomycosis is a rare granulomatous disease caused by an opportunistic Gram-positive bacillus of the genus Actinomyces. We report an unusual association between hepatic actinomycosis and IgG-4 related disease.

Case Description

A 70-year-old man was admitted to the University Hospital of Reunion Island for asthenia and abdominal pain. Laboratory tests showed hepatic cholestasis and inflammatory reaction without cytolysis. The computer tomography scanner (CT scan) revealed a large hepatic mass in segment IV. IgG4 related disease (IgG4-RD) was diagnosed following liver biopsy. Despite the use of corticosteroids, the lesions progressed. A new biopsy was performed and a PCR test returned positive for Actinomyces spp. We observed patient improvement after 4 months of treatment with amoxicillin and discontinuation of corticosteroids.

Conclusions

Clinical presentation, imaging characteristics, and diagnostic challenges are strikingly similar between hepatic actinomycosis and IgG4-related disease. However, distinguishing between these two disorders is essential – as the treatment of one may worsen the course of the other. Knowing and understanding the interactions between these two disorders can only be achieved through an integrated and collaborative approach involving various disciplines.

Introduction

Hepatic actinomycosis (HA) is a rare liver disorder caused by Gram-positive bacilli belonging to the genus Actinomyces. Actinomyces israelii, a commensal of the oropharynx, digestive tract and female genital tract, is one of the most commonly implicated species. Several cases of HA have been reported; the majority occur in non-immunocompromised patients [1]. Abdominal involvement accounts for 20% of all cases of actinomycosis, of which 5% are hepatic HA. Its insidious presentation makes it a diagnostic challenge. We report here a new case of association between HA and IgG4 related disease (IgG4-RD).

Case Presentation

A 70-year-old man was referred to the gastroenterology department of the University Hospital of Reunion Island in October 2022 for general deterioration, weight loss (15 kg in 3 months) and abdominal pain. His medical history included gastric adenocarcinoma treated with total gastrectomy, in remission since 1997; ulcerative fungal esophagitis in 1998; a Streptococcus anginosus liver abscess in 2008, managed with percutaneous drainage and a 6-week course of antibiotic therapy, with complete resolution; and right inguinal hernia repair before 2008. Laboratory tests revealed an inflammatory reaction with leukocytes at 9.19 X 10⁹/L, C-reactive protein (CRP) at 123 mg/L and hepatic cholestasis without cytolysis (GGT 577 IU/L, PAL 652 IU/L). An injected abdominal-pelvic computer tomography (CT) scan revealed a large hepatic mass in segment IV (6 × 6.5 × 8 cm) (Fig. 1).

Fig. 1.

Axial section abdominal CT scan with injection at arterial (A) and portal (B) phases. Infiltrative centrohepatic mass with anterior capsular retraction (asterisk), leading to dilatation of the left intrahepatic bile ducts (arrowhead). Contrast is faint and heterogeneous at portal phase

A total colonoscopy was carried out and was negative. A liver biopsy was performed, showing marked storiform fibrosis associated a lymphoplasmacytic infiltrate (more than 10 IgG4 plasma cells per field and a total IgG4/IgG ratio > 40%) (Fig. 2).

Fig. 2.

Histological section of liver biopsy taken on 9 November 2022. Hematoxylin, eosin and saffron-stained (HES) staining at X 20 magnification showing an inflammatory infiltrate composed mainly of plasma cells (A). IgG immunohistochemical staining at magnification X 20 (B) and IgG4 immunohistochemical staining at magnification X 20 (C) and X 40 (D). The IgG4/IgG ratio is over 40% with more than 10 IgG4 per field at X40 magnification

No cancerous cells or thrombotic lesions were identified. Blood tests revealed a serum IgG4 level of 7.060 g/L, and a total IgG level of 32.72 g/L (IgG4/IgG ratio of 21.5%). The diagnosis of IgG4-RD was retained and corticosteroid therapy at 1 mg/Kg/day was started on December 5th, 2022 for one month, followed by a tapering period. Liver magnetic resonance imaging (MRI) on December 13th, 2022 showed a T1 hyposignal and intermediate T2 signal extended lesion (Fig. 3).

Fig. 3.

Magnetic resonance imaging scan (MRI) showing an infiltrative, ill-limited lesion centred on hepatic segmentin IV, of 12 × 6 × 9 cm, extending to segments III, V and VIII. A mass in clear T1 hyposignal (A) is associated with anterior capsular retraction (yellow asterisk). It shows an intermediate T2 signal (B). A peritoneal infiltrating zone in contact with the hepatic capsule in T2 hypersignal invades the thoraco-abdominal wall (arrow). This lesion is in diffusion hypersignal (C). The T1 sequence injected at portal phase (D) reveals an hypointense enhancement, and ill-defined mass (white asterisk)

After one month of treatment, the patient presented no clinical improvement and the inflammatory condition remained. A new liver biopsy was performed on February 7 2023 which revealed filamentous elements suggestive of actinomycosis, associated with IgG4 rich fibro-inflammatory remodeling (30/field, total IgG4/IgG ratio of 30%) (Fig. 4).

Fig. 4.

HES liver biopsy sections at magnifications X 4 (A) and Periodic Acid Schiff staining (PAS) at magnifications X 40 (B). GRAM staining at X 630 magnification (C) and Grocott staining at X 1000 magnification (D). The biopsy shows an area of reactive fibrosis (A, red asterisk) and an intense polymorphic inflammatory infiltrate containing plasma cells (A, yellow asterisk). Adjacent to this are clusters of actinomycosis (A, large arrow and B, whole picture). Branched actinomyces filaments (C and D, arrowhead)

A laboratory-developed end-point polymerase chain reaction (PCR) targeting the tuf gene (primers [5'- > 3'] tufA485 CAAGGCCCTCGAGGGYGAC and tufA485 CAAGGCTCTGGAGGGTGACGC) of Actinomyces spp was performed and came back positive on a third sample (puncture), culture remained polymicrobial. Additional tests confirmed normal complement and conversion enzyme levels, while anti-SSA, anti-SSB antibodies and free light chain were negative. Intravenous treatment with amoxicillin 3 g per day was initiated on March 17, 2023 and was switched to oral therapy after one month. Liver MRI re-evaluation on July 18, 2023 showed partial regression of the retractile infiltrative lesion (Fig. 5).

Fig. 5.

MRI showing partial regression at 4 months of treatment of the hepatic infiltrative lesion measured at 6.5 X 5.0 cm. Fat-saturated T1 (A), and fat-saturated T2 (B) sequences show regression of the mass and adjacent peritoneal infiltration (arrow). These findings are also seen in diffusion (C) and on injected T1 with signal saturation at portal phase (D, asterisk)

A positron emission tomography (PET) scan on July 21, 2023 revealed no hypermetabolic focus suggestive of IgG4-related disease or neoplasia. On September 1, 2023 the patient was free of abdominal pain and laboratory tests were normal: GGT 102 UI/L, PAL 136 UI/L, CRP 1 mg/L, total IgG 10,400 mg/L and IgG4 1.760 g/L. The last MRI scan performed on 4th July, 2025 showed further regression of the lesion.

Discussion

This case highlights the insidious nature of HA. Actinomyces infection generally occurs when the integrity of the digestive mucosa is compromised, often as a result of abdominal or pelvic surgery or the presence of foreign material [1]. HA predominantly affects middle-aged men and presents with nonspecific symptoms, including fever, abdominal pain, and weight loss. As in our case, imaging typically reveals a solitary right hepatic lesion and may closely mimic a malignant tumor [1, 2]. MRI is the preferred modality for characterizing hepatic actinomycosis, typically demonstrating hypointense signals on T1-weighted images and hyperintense signals on T2-weighted images. The infiltrative pattern, often extending to adjacent structures such as the hepatic capsule, bile ducts, peritoneum, or even lung parenchyma, may mimic a neoplastic process, particularly cholangiocarcinoma. Increased uptake on PET has also been reported; however, this finding could not be confirmed in our case, as the PET scan was performed retrospectively [2]. Diagnosis is based on histopathological analysis, usually performed after percutaneous drainage or hepatic lobar resection. Identification of granulomatous inflammation associated with sulfur granules is pathognomonic [2]. Culture is rarely informative due to frequent contamination and the fastidious, slow-growing nature of Actinomyces species. PCR targeting the 16s ribosomal RNA gene or, preferably, the tuf gene, can help confirm the diagnosis, as in our case.

In addition to raising diagnostic uncertainties, this case also highlights the differential diagnosis with IgG4-RD. IgG4-RD is characterized by a dense lymphoplasmacytic infiltrate with an increased proportion of lymphocytes and IgG4-positive plasma cells (> 40% IgG4/IgG plasma cell ratio) [3]. Diagnostic criteria include the histological presence of a dense lymphoplasmacytic infiltrate, storiform fibrosis, increased eosinophils, and the non-obligatory presence of obliterative phlebitis. IgG4 immunostaining is essential for diagnosis, and the hepatic threshold is set at more than 10 IgG4-positive cells per high-power field (hpf) [3]. IgG4-RD liver disease typically affects men in the second half of life and presents as a solid mass, most often involving the left hepatic lobe, without biliary tract infiltration, and with elevated serum IgG4 levels. The recommended treatment is oral corticosteroid therapy. Although several reports have suggested an association between IgG4-RD and malignancy, only a few cases describing an association between IgG4-RD and HA have been reported (Table 1).

Table 1.

Comparison of literature described cases of association between IgG4-related disease and hepatic actinomycosis

Attribute	Lee et al., 2018a	Chung et al., 2019b	Song et al., 2024c	Manni et al., 2025d	Present case
Sex	Man	Man	Man	Man	Man
Age, years	67	67	72	74	70
Time to first symptoms, months	3	3	2	No information	3
Main symptoms	Abdominal discomfort, weight loss	Fever, weight loss	Anorexia, weight loss	Anorexia, weight loss	Abdominal pain, weight loss
The serum IgG4 quantitation level	No information	2,215.8 mg/dL, ratio unkown	334 mg/dL, serum IgG4/IgG ratio of 9.2%	1,680 mg/dL, IgG4/IgG ratio of 50%	7,060 g/L, IgG4/IgG ratio of 21.6%
Imaging
CT-scan	Ill-defined lobulated mass 8.5 X 7.8 cm Right hepatic lobe Segmental portal vein obliteration	Large 8.5 X 7.8 X 8.9 mass in the right hepatic lobe, segmental portal vein obliteration	10 cm hypodense lesion in the central portion of the left liver lobe. Arterial-phase peripheral hyperenhancement and progressive enhancement of the central portion, peripheral ductal dilatation	Not performed	Hypodense lesion 6 X 6.5 X 8 cm poorly delimited. Involvement of the IV hepatic segment Signs of local extension Retractile infiltrating lesion with local extension
MRI	Hypo-signal intensities T1, hyper-signal intensities T2	Hypo-signal T1 and T2-weighted hypersignal	Low signal intensity on T1 imaging, intermediate signal intensity on fat-saturated T2 imaging, diffusion restriction	Irregular margins and non-homogeneous signal on T1-weighted. Hyposignal core and hypersignal peripheral rim on T2-weighted sequences	Hypo-signal intensities T1, intermediate signal T2
PET-CT scan	Strong uptake	Nonspecific strong uptake	Not performed	Intense, heterogeneous uptake	No uptake (performed after treatment)
Diagnosis method	Pathology analysis	Pathology analysis	Pathology analysis	Pathology analysis	Pathology analysis
Pathology analysis	Lymphoplasmacytic infiltration with fibrosis 30 IgG4-positive cells per high power field Actinomycotic colonies	Abscess with abundant fibrosis and increased IgG4-positive lymphoplasma cells (about 30/high-power field) Dense lymphoplasmocytic infiltration with fibrosis and abscess formation containing actinomycotic colonies	Acute and chronic inflammation with actinomycotic sulfur granules. Storiform appearance of fibrosis. Presence of IgG-positive plasma cells, and a prominent component of IgG4-positive plasma cells	IgG4 expression, Cell aggregates of Actinomyces spp	Storiform fibrosis with lymphoplasmacytic infiltrate 30 IgG4-positive cells per high power field, IgG4/IgG ratio of 30% Actinomycotic filaments
Corticosteroid therapy	Yes, 3 months	Yes, 4 months	Yes, duration unknow	Yes, 1 year	Yes, 2 months
Surgery	Yes	Yes	No, but catheter drainage	No	No
Treatments	Ursodeoxycholic acid	No information	No information	Azathioprine 75 mg/day, ursodeoxycholic acid 300 mg three times daily Ampicillin 3 g/6 h intravenous for two weeks then orally	Amoxicillin 3 g per day intravenous for one month then orally
Duration of treatment	14 days	No information	No information	6 months (Ampicillin)	4 months
Follow-up	Three months, no recurrence	No information	No information	Three years, no recurrence	Two years, no recurrence

^aLee JH, Kim HS, Kim JS, Lee DK, Lim JH. Hepatic actinomycosis with immunoglobulin G4-related liver disease: A case report. Medicine. 2018 Jun;97(24):e11146. doi:10.1097/MD.0000000000011146; ^bChung JW, Bae YS, Lim JH, Lee DK, Yu JS. IgG4-Related Hepatic Inflammatory Pseudotumor Complicated by Actinomycosis during Steroid Therapy. Journal of the Korean Society of Radiology. 2019 Mar 1;80(2):351–8. doi:10.3348/jksr.2019.80.2.351; ^cSong H, Ju MJ, Kim SH. Unique Imaging Features in Hepatic Actinomycosis Accompanied by an IgG4-Related Inflammatory Pseudotumor: A Case Report. Investigative Magnetic Resonance Imaging. 2024 Mar 1;28(1):49–54. doi:10.13104/imri.2023.0030; ^dManni G, Pambianco M, Sicuro C, Franceschini E, Pivetti A, Bertoni L, et al. Concomitant Autoimmune Liver Disease and Hepatic Actinomycosis: A Diagnostic Challenge—Brief Report and Review of the Literature. Int J Mol Sci. 2025 Sep 26;26(19):9420. doi:10.3390/ijms26199420

CT-scan: Computed tomography scan; MRI: Magnetic resonance imaging; PET-CT scan: Positron emission tomography-computed tomography scan

Although a fibrohistocytic hepatic lesion may mimic our presentation and partially overlap with the IgG4 spectrum, the absence of histiocytic cells, multinucleated giant cells, neutrophilic infiltrates, and microabscesses, including on the initial biopsy, argues strongly against this diagnosis.

As summarized in Table 1, reported cases share a remarkably similar profile, predominantly affecting older men with constitutional symptoms and liver masses mimicking malignancy. Lack of response or clinical worsening following appropriately administered glucocorticoid therapy constitutes a strong argument for exclusion and was consistently observed across reported cases (Table 1) [4].

In our case, it is difficult to determine whether HA was secondary to IgG4-related disease or whether it was pre-existing and subsequently revealed by corticosteroid therapy. The IgG4-rich infiltrate may represent a reactive response to infection, as previously suggested in chronic infectious diseases, potentially driven by a Th2-mediated immune response promoting eosinophilia, elevated IgG4 levels, and fibrosis [5].

Available data support initial parenteral antibiotic therapy followed by oral treatment, with duration guided by clinical, biological, and radiological response, typically for 3–6 months [1].

The origin of the infection remains uncertain. Multiple dental abscesses support a possible oral source. Although recurrence at the same site is intriguing, a normal liver MRI in 2012 makes iatrogenic inoculation during abscess drainage in 2008 unlikely. Similarly, inoculation of Actinomyces spp. during the initial liver biopsy cannot be completely excluded but appears improbable.

Conclusion

This is a new case describing an association between IgG4-RD and hepatic actinomycosis. These two entities share overlapping clinical and radiological features thereby posing a significant diagnostic challenge. Accurate diagnosis requires a multimodal approach integrating clinical evaluation, imaging, and histopathological analysis.

This case highlights three key issues: (i) the importance of systematically consider an infectious etiology in presence of an infiltrative and progressive liver lesion, (ii) the need to systematically send samples for anatomopathological and specific bacteriological analysis, and (iii) the necessity for a better understanding of the interplay between infection and IgG4-RD.

Author Contributions

Clinical Investigation, A.D., J.M., K.D., M.L., and S.R.; Anatomopathological analysis, E.L., LM.D.; Microbiological analysis, AG.L., PE.G.; Proofreading radiological images, D.DB.; Writing original draft, A.D.; Writing-Review and Editing, A.D., D.DB., E.L., F.A., J.M., K.D., LM.D., M.L., PE.G., and S.R.; Supervision, S.R. All authors have read and approved the manuscript.

Funding

None.

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Ethical Statement

Written consent was obtained by the patient for the de-identified publication of his clinical case.

Competing interests

The authors declare no competing interests.