MR imaging findings of hepatic pseudolymphoma: common and rare features

doi:10.21203/rs.3.rs-4874202/v1

Purpose To summarize the common and rare MR imaging features of hepatic pseudolymphoma (HPL). Methods This retrospective study was approved by our institutional review board. 35 patients (mean age, 53.2 years) with 50 HPLs underwent enhanced MR. MR images were retrospectively reviewed and analyzed. Results There are 3 males and 32 females (8.6% vs 91.4%). 27 patients with single lesion and 8 patients with multiple lesions. Among the 50 HPLs, 90% of HPLs size ≤ 20mm, 92% were regular, and 92% were ill-defined. 7 and 5 patients had a history of malignant tumor and chronic liver disease. Tumor markers were negative in all patients. All 50 HPLs were low T1 SI, high T2 SI, high DWI SI, low ADC value, and adjacent to veins. 72% of HPLs were adjacent to the portal vein, 18% to the hepatic vein and 10% to the inferior vena cava. 72% of HPLs showed "linear sign" and "angular sign". There was a favorable agreement between "linear sign" and "angular sign" (kappa = 0.802, R < 0.001). 88% of HPLs showed “ring sign”. There were four enhancement patterns of HPLs (Table 2), wash out (46%), degressive (26%), persistent (18%) and progressive (10%). There were 2 patients with pathologically confirmed extra-hepatic pseudolymphomas. The largest HPL was 50mm in length with central cystic necrosis. Conclusions The common features of HPL are small, single, regular, adjacent to veins, low T1 SI, high T2 SI, high DWI SI and low ADC value. “Linear sign”, “angular sign” and “ring sign” could be seen as common features of HPL. The rare features of HPL included extra-hepatic lesion and central cystic necrosis.

Hepatic pseudolymphoma (HPL), also known as reactive lymphoid hyperplasia or nodular lymphoid hyperplasia, is a rare benign lymphoproliferative lesion of the liver[1, 2]. At present, no more than 150 HPL cases were reported in the literature. HPL is usually slow growing, asymptomatic, and negative for tumor markers[3–5]. It is found during physical examination or examination of other diseases in most cases[6]. Its pathogenesis is unclear and might be associated with autoimmune diseases, chronic liver diseases and tumors[7, 8].

HPL is difficult to diagnose. It is almost always misdiagnosed as malignant tumors and confirmed by pathology after biopsy or surgical resection[9–11]. MR imaging is important for the diagnosis of benign and malignant tumors. Especially when there is a small lesion and a risk of malignant tumor implantation. Studying MR imaging features is helpful for the diagnosis of HPL.

The MR imaging features of HPL have not been studied sufficiently. Most studies on HPL are case reports rather than cohort studies[10]. Previous reported cases reminded that HPL were low T1 signal intensity (SI), high T2 SI, and high DWI SI, however, these are not specific features to pseudolymphoma[9, 10]. In this study, we retrospectively analyzed MR imaging data of 50 HPLs in 35 patients. The objective is to summarize the common and rare MR imaging features of HPL.

Patients

This retrospective study was approved by our institutional review board (No. B2021-366). Between August 2013 and July 2021, a total of 35272 consecutive patients with liver disease received pathological diagnosis in our hospital. 42 patients were pathologically diagnosed with HPL (Fig. 1). The patients without immunohistochemistry (n = 2) or enhanced MR imaging (n = 5) were excluded. Two HPLs with severe artifact caused by cardiac motion were excluded. Finally, 35 patients with 50 HPLs confirmed by immunohistochemistry were included (Figure S1). There were 3 men and 32 women (mean age, 53.2 ± 1.85 years). ADC values could be accurately measured in 44 lesions. In this study, data in 12 patients have been previously reported in a smaller study[10]. We undertook this study to reach enough statistical power to allow identification of MRI features of HPL.

MR Imaging Protocol

MR imaging was performed on 3.0-T systems (UIH MR 770, GE750 MR, and Siemens Verio) and 1.5-T systems (Siemens Area). Contrast-enhanced MR imaging was performed by intravenous injection of 0.1 mmol/kg body weight of Gd-DTPA (Magnevist, Bayer Schering Pharma, Berlin, Germany), followed immediately by a 20 ml saline flush through a power injector at a rate of 2 ml/s. Taking 3.0 T UIH MR 770 scanner as an example, MR sequences include axial T2WI-FS, DWI(b values of 0, 500 mm²/s), axial pre-contrast quick3d T1WI with fat suppression and post-contrast dynamic-enhanced quick 3d T1WI at arterial phase (25–30 s), portal phase (55–60 s) and delayed phase (150-180 s). (Table S1)

Image Analysis

The common feature was defined as more than 66.7% of the total. The rare feature was defined as less than 10% of the total. "Linear sign" was defined as a continuous line of high signals connecting the mass to the adjacent portal vein on DWI (b = 500 mm²/s). "Angular sign" was defined as an angular high signal extending outward from the edge of the lesion during the peak period of enhancement. "Ring sign" was defined as the dynamically-enhanced peripheral enhancement. "Ring signs" were divided into six patterns according to the number of rings and signal characteristics (Figure S2). "Vessel-penetrating sign" was defined as the vessel partially preserved in the lesion[12]. We analyzed the location, shape, size, number, boundary, signal intensity (SI), the peak period of enhancement, and enhancement pattern of the HPL. There were four enhancement patterns in our study (wash out: portal or delayed phase wash-out; degressive: high SI in arterial phase and equal to slightly high SI in portal or delayed phase; persistent: significantly high SI in three stages; progressive: SI gradually increased). All MR images were analyzed by two senior radiologists and agreed upon in consultation when opinions were inconsistent.

Statistical Analysis

All data were analyzed using IBM SPSS 25.0 software. Data conforming to the normal distribution were expressed as`c ± s. Mann-Whitney U test was used to compare ADC values of the lesion, liver parenchyma and spleen. Kappa test was used to test the agreement between "linear sign" and "angular sign". Kappa value > 0.8 indicated excellent agreement; Differences with P < 0.05 were considered significant.

Conventional MRI features of HPL

27 patients with single lesion and 8 patients with multiple lesions. Among the 50 HPLs, 90% of HPLs size ≤ 20mm, 92% were regular, and 92% were ill-defined. 70% were in the right lobe and 78% were subcapsular. (Table 1)

All 50 HPLs showed low T1 SI and high T2 SI , and 98% of HPLs showed homogeneous T1 and T2 SI (Fig. 2-7). All HPLs were adjacent to veins, 72% to the portal vein, 18% to the hepatic vein and 10% to the inferior vena cava. In our study, the largest lesion was 50mm in length with central cystic necrosis (Fig. 2).

We found 2 patients with pathologically confirmed extra-hepatic pseudolymphomas. One patient had an extra-hepatic pseudolymphoma located in the diaphragm and the other patient had two extra-hepatic pseudolymphomas located in the hilar lymph nodes.

Table 1 MRI features of 50 HPLs in 35 patients
Variable	n=50	%
Location
Right lobe	35	70
Left lobe	10	20
Caudal lobe	5	10
Shape
Regular	46	92
Irregular	4	8
Maximum diameter
≤20mm	45	90
>20mm	5	10
Boundary
Ill-defined	46	92
Well-defined	4	8
T1
Low SI	49	98
Low with central lower SI	1	2
T2
High SI	49	98
High with central higher SI	1	2
“Linear sign”
Yes	36	72
No	14	28
Peak period of enhancement
Arterial phase	46	92
Delayed phase	4	8
“Angular sign”
Yes	36	72
No	14	28
“Ring sign”
Yes	44	88
No	6	12
“Ring sign” in arterial phase
No	39	78
Yes	11	22
“Ring sign” in portal phase
Yes	37	74
No	13	26
“Ring sign” in delayed phase
Yes	38	76
No	12	24
“Ring sign” phase
Multiple	33	66
Single	11	22
Enhancement pattern
Wash out	23	46
Degressive	13	26
Persistent	9	18
Progressive	5	10

DWI features of HPL

On DWI, all HPLs showed higher SI than liver parenchyma, 76% higher than spleen, and 72% showed "linear sign" (Fig. 3 c₁-c₃). ADC values could be accurately measured in 44 lesions. On ADC map (Fig. 4), all 44 lesions were lower than liver parenchyma and 86.4% lower than spleen (Z = -6.708, P < 0.001 and Z = -4.534, P < 0.001).

Contrast enhanced MRI features of HPL

72% showed an "angular sign" in the peak period of enhancement (Fig.3 e₁ -e₃). There was a favorable agreement between "linear sign" and "angular sign" (kappa = 0.802, R < 0.001). 92% of HPLs showed peak period of enhancement in arterial phase, which higher than liver parenchyma but lower than the portal vein.

88% of HPLs showed “ring sign”(Fig. 5-7), and 12% showed no “ring sign”. Among the 44 HPLs with “ring sign”, 33 HPLs presented "ring sign" in two or more phases (Fig. 5-6), and 11 HPLs presented "ring sign" in single phase (Fig. 7). There were four enhancement patterns of HPLs (Table 1), wash out (46%), degressive (26%), persistent (18%) and progressive (10%). Most of HPLs showed blurred boundary in the late stage of enhancement (Fig.2).

Jiang et al reported that most HPL patients were middle-aged and elderly women (male vs female, 10.74% vs 89.26%, mean age 57 years)[2], and our study also confirmed the findings; the age of HPL patients was 33-73 (53.2 ± 1.85) years, the sex distribution was 8.6% vs 91.4%. However, there were 57% of patients with previously malignant tumor, autoimmune diseases or chronic viral hepatitis in their study. Zhou et al reported that 78.6% of patients (11/14) were previously healthy[10]. In our study, the previously healthy patients were 65.7% of the total. Thus, previously healthy people account for one-half to two-thirds of all patients with HPL. In our study, most HPLs were single, less than 20mm, regular, ill-defined, and located in the right lobe. The findings are in line with previous findings[13–16].

HPL was almost always misdiagnosed as metastatic or primary malignancy, even though the tumor markers were negative[9–11]. In our study, 7 cases with tumor history were misdiagnosed with liver metastases, and the others were misdiagnosed with primary liver malignancy.

"Linear sign" and "angular sign" are the two other common features of HPL. In our study, 72% of HPLs showed "linear sign" which was defined as a high signal line connecting the mass to the adjacent portal vein on DWI. Kobayashi and Tanaka et aldescribed "linear sign" in their case reports[9, 17]. Histologically, the adjacent liver parenchyma of HPL shows lymphocytic infiltration in the portal tracts. In our study, 72% of HPLs showed "angular sign" in the peak period of enhanced enhancement. "Linear sign" was usually accompanied by "angular sign". "Linear sign" and "angular sign" had an excellent statistical agreement (kappa = 0.802). This is a newly discovered MR imaging feature. It may be useful in the differential diagnosis of HPL. ADC values may play an important role in differentiating HPL from hepatocellular carcinoma[10]. In our study, ADC values of all HPLs were lower than liver parenchyma. 86.4% ADC values were lower than spleen (median, 0.55×10^-3 mm²/s vs 0.93×10^-3 mm²/s). This is in line with the previous findings.

Adjacent to vein was another common feature of HPL in our study. Our study evaluated the distribution of HPL and veins. All 50 HPLs were adjacent to veins, 72% adjacent to the portal vein, 18% to the hepatic vein and 10% to the inferior vena cava. Lymphatic vessels distribution that may contribute to the common feature of HPL. Lymphatic vessels of liver begin in the periportal space of Mall[18–22]. Liver lymph was mainly delivered travel along the portal or hepatic veins toward to liver hilum or inferior vena cava (Figure S3). In addition, part of the liver lymph was drained into the mediastinum or toward the liver hilum.

Our study is the first to report that HPL patients with pathologically confirmed extra-hepatic pseudolymphomas. Pseudolymphoma has been found in various organs, including the orbit, skin, gastrointestinal tract, etc[23–25]. This new finding suggests that HPL could be accompanied by extra-hepatic pseudolymphomas, and that pseudolymphoma could be found simultaneously in two different organs. These two patients of extra-hepatic pseudolymphomas may also be related to lymphatic vessels distribution.

Central cystic necrosis could be found in large HPL. HPL commonly appears as a mass with homogeneous low T1 SI, high T2 SI, high DWI SI, and low ADC value[9, 10]. Our study is consistent with previous results. All 50 HPLs showed low T1 SI, high T2 SI, high DWI SI, and low ADC values. However, in our study, one HPL with a diameter of about 50mm showed central cystic necrosis. This new finding suggests that cystic necrosis could be found in large HPL, even though it is benign.

The "ring sign" is another common sign of HPL, which also known as peripheral enhancement. "Ring sign" of HPL can appear in each stage of enhancement[14, 26–28]. In our study, 66% of HPLs showed "ring sign" in multiple phases (≥2 phases), 22% in single phase, and 12% HPLs showed no "ring sign". The "ring sign" presented in multiple phases may be the characteristic MR imaging sign of HPL. It is of great value to the differential diagnosis of metastatic tumors and liver cancer. The “ring sign” of metastatic tumors is usually accompanied by cystic necrosis, while the “ring sign” of liver cancer usually occurs in the delayed stage and non-multiple phases[29].

Yoshida et al reported that "vessel-penetrating sign" could be found in larger HPL, it might be one of the HPL characteristics[12]. In our study, "vessel-penetrating sign" was observed in 24% (12/50) of HPLs.

There are few cohort studies on MR imaging enhancement patterns of HPL. Zhou et al reported that “wash in and wash out”, “degressive” and “persistent” enhancement were 60%, 35% and 5%, respectively[10]. In our study, there were four enhancement patterns of HPLs: wash out (46%), degressive (26%), persistent (18%) and progressive (10%). Thus, the enhancement pattern of HPLs is variable.

HPLs can be distinguished from malignant tumors by using image guided percutaneous needle biopsy. Liver masses with MR features suggestive of HPL could be treated conservatively in selected patients. Even though the common features proposed were not present in all HPLs, these features have the potential to aid the diagnosis of HPL.

Our study had a number of study limitations. The study was retrospective nature. The number of patients who had HPLs was small. MR imaging scanner and parameters were different. The study had no control group. There was no whole histologic sections and corresponding MR imaging of HPL.

In conclusion, the common features of HPL are small, single, regular, adjacent to veins, low T1 SI, high T2 SI, high DWI SI and low ADC value. “Linear sign”, “angular sign” and “ring sign” could be seen as common features of HPL. The rare features of HPL included extra-hepatic lesion and central cystic necrosis.

Osanai H, Kozaka K, Yoneda N, et al (2024) Reactive lymphoid hyperplasia of the liver: A case report featuring characteristic nodular and perinodular enhancement. Radiology Case Reports 19:1998–2003. https://doi.org/10.1016/j.radcr.2024.02.029
Jiang W, Wu D, Li Q, et al (2023) Clinical features, natural history and outcomes of pseudolymphoma of liver: A case-series and systematic review. Asian Journal of Surgery 46:841–849. https://doi.org/10.1016/j.asjsur.2022.08.113
Matsuyama S, Fukuda A, Omatsu R, et al (2023) A case of hepatic reactive lymphoid hyperplasia: the review of 23 cases from the literatures. Clin J Gastroenterol 16:877–883. https://doi.org/10.1007/s12328-023-01844-4
Bakdik S, Poyraz N, Kucukkartallar T, Oltulu P (2022) Liver Pseudolymphoma (Reactive Lymphoid Hyperplasia): Difficulty inRadiological and Pathological Diagnosis. CMIR 18:357–361. https://doi.org/10.2174/1573405617666210504120142
Sim KK, Fernando T, Tarquinio L, Navadgi S (2023) Hepatic reactive lymphoid hyperplasia-associated primary biliary cholangitis masquerading as a neoplastic liver lesion. BMJ Case Rep 16:e254963. https://doi.org/10.1136/bcr-2023-254963
Zen Y, Fujii T, Nakanuma Y (2010) Hepatic pseudolymphoma: a clinicopathological study of five cases and review of the literature. Modern Pathology 23:244–250. https://doi.org/10.1038/modpathol.2009.165
Inoue M, Tanemura M, Yuba T, et al (2019) A case of hepatic pseudolymphoma in a patient with primary biliary cirrhosis. Clinical Case Reports 7:1863–1869. https://doi.org/10.1002/ccr3.2378
Bai Y, Liang W (2019) Potential value of apparent diffusion coefficient in the evaluation of hepatic pseudolymphoma. Quant Imaging Med Surg 9:340–345. https://doi.org/10.21037/qims.2018.11.07
Kobayashi A, Oda T, Fukunaga K, et al (2011) MR Imaging of Reactive Lymphoid Hyperplasia of the Liver. Journal of Gastrointestinal Surgery 15:1282–1285. https://doi.org/10.1007/s11605-011-1456-3
Zhou Y, Wang X, Xu C, et al (2018) Hepatic pseudolymphoma: imaging features on dynamic contrast-enhanced MRI and diffusion-weighted imaging. Abdom Radiol 43:2288–2294. https://doi.org/10.1007/s00261-018-1468-5
Avellana Moreno R, De la Serna Esteban S, García Botella A (2020) HEPATIC REACTIVE LYMPHOID HYPERPLASIA: DIFFICULT DIFFERENTIAL DIAGNOSIS WITH HEPATOCARCINOMA. Rev Esp Enferm Dig. https://doi.org/10.17235/reed.2020.7462/2020
Yoshida K, Kobayashi S, Matsui O, et al (2013) Hepatic pseudolymphoma: imaging–pathologic correlation with special reference to hemodynamic analysis. Abdom Imaging 38:1277–1285. https://doi.org/10.1007/s00261-013-0016-6
Yang C-T, Liu K-L, Lin M-C, Yuan R-H (2017) Pseudolymphoma of the liver: Report of a case and review of the literature. Asian Journal of Surgery 40:74–80. https://doi.org/10.1016/j.asjsur.2013.07.013
Taguchi K, Kuroda S, Kobayashi T, et al (2015) Pseudolymphoma of the liver: a case report and literature review. surg case rep 1:107. https://doi.org/10.1186/s40792-015-0110-9
Sato K (2006) Reactive lymphoid hyperplasia of the liver in a patient with multiple carcinomas: a case report and brief review. Journal of Clinical Pathology 59:990–992. https://doi.org/10.1136/jcp.2005.028886
Yuan L, Zhang Y, Wang Y, et al (2012) Reactive Lymphoid Hyperplasia of the Liver: A Clinicopathological Study of 7 Cases. HPB Surgery 2012:1–7. https://doi.org/10.1155/2012/357694
Tanaka T, Saito K, Yunaiyama D, et al (2020) Diffusion-weighted imaging might be useful for reactive lymphoid hyperplasia diagnosis of the liver: A case report. WJCC 8:5313–5319. https://doi.org/10.12998/wjcc.v8.i21.5313
Trutmann M, Sasse D (1994) The lymphatics of the liver. Anat Embryol 190:. https://doi.org/10.1007/BF00234299
Moszkowicz D, Cauchy F, Dokmak S, Belghiti J (2012) Routine Pedicular Lymphadenectomy for Colorectal Liver Metastases. Journal of the American College of Surgeons 214:e39–e45. https://doi.org/10.1016/j.jamcollsurg.2012.02.015
Lupinacci RM, Paye F, Coelho FF, et al (2014) Lymphatic drainage of the liver and its implications in the management of colorectal cancer liver metastases. Updates Surg 66:239–245. https://doi.org/10.1007/s13304-014-0265-0
Ludwig J, Linhart P, Baggenstoss AH (1968) Hepatic lymph drainage in cirrhosis and congestive heart failure. A postmortem lymphangiographic study. Arch Pathol 86:551–562
Ito M, Mishima Y, Sato T (1991) An anatomical study of the lymphatic drainage of the gallbladder. Surg Radiol Anat 13:89–104. https://doi.org/10.1007/BF01623880
Mitteldorf C, Kempf W (2020) Cutaneous pseudolymphoma-A review on the spectrum and a proposal for a new classification. J Cutan Pathol 47:76–97. https://doi.org/10.1111/cup.13532
Elkholy S, Mogawer S, Farag A (2017) Nodular Lymphoid Hyperplasia of the Gastrointestinal Tract : a comprehensive review. Acta Gastroenterol Belg 80:405–410
Terada T (2012) Pseudolymphoma of the orbit. Med Oncol 29:1082–1083. https://doi.org/10.1007/s12032-011-9857-4
Kunimoto H, Morihara D, Nakane S, et al (2018) Hepatic Pseudolymphoma with an Occult Hepatitis B Virus Infection. Intern Med 57:223–230. https://doi.org/10.2169/internalmedicine.8981-17
Kwon YK (2015) Pseudolymphoma (reactive lymphoid hyperplasia) of the liver: A clinical challenge. WJH 7:2696. https://doi.org/10.4254/wjh.v7.i26.2696
Sonomura T (2015) Reactive lymphoid hyperplasia of the liver: Perinodular enhancement on contrast-enhanced computed tomography and magnetic resonance imaging. WJG 21:6759. https://doi.org/10.3748/wjg.v21.i21.6759
Chernyak V, Fowler KJ, Kamaya A, et al (2018) Liver Imaging Reporting and Data System (LI-RADS) Version 2018: Imaging of Hepatocellular Carcinoma in At-Risk Patients. Radiology 289:816–830. https://doi.org/10.1148/radiol.2018181494

No competing interests reported.

SupplementalContent.docx

MR imaging findings of hepatic pseudolymphoma: common and rare features

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

References

Additional Declarations

Supplementary Files

Status:

Version 1