Hepatic involvement in larva migrans of Toxocara canis: Report of a case with pathological and radiological findings

Digestive and Liver Disease 38 (2006) 511–514

Brief Clinical Observation

Hepatic involvement in larva migrans of Toxocara canis: Report of a case with pathological and radiological findings N. Leone a,∗ , M. Baronio a , L. Todros a , E. David b , F. Brunello a , S. Artioli c , M. Rizzetto a b

a Department of Gastroenterology, Azienda Ospedaliera S. Giovanni Battista, Turin, Italy Department of Biomedical Science and Human Oncology, Azienda Ospedaliera S. Giovanni Battista, Turin, Italy c Operative Unit of Infectious Diseases, ASL 5, Regione Liguria, Italy

Received 15 June 2005; accepted 18 July 2005 Available online 15 September 2005

Abstract Patients with the clinical syndrome of visceral larva migrans as a result of Toxocara species, have typical lesions in the liver consisting of granulomas that contain numerous eosinophils and often Charcot–Leyden crystals. This syndrome is rarely taken into account in patients with cholestatic syndrome, especially when hypereosinophilic reaction is absent. We report the case of a 47-year-old immunocompetent woman who presented with abdominal pain, in whom multiple focal liver lesions were discovered. She had come in contact with dogs. Diagnosis of toxocariasi was done. A good clinical response has been obtained by treating with thiabendazole. We present the findings of various imaging studies of the patient. This report shows that visceral larva migrans may be the cause of a chronic liver disease and should be suspected also in patients without fever and hypereosinophilia with cryptogenic cholestatic and focal liver lesions. © 2005 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. Keywords: Cholestatic syndrome; Granulomatous hepatitis; Hepatic liver lesions; Toxocariasis; Visceral larva migrans

1. Introduction Toxocara canis is a common nematode of dogs worldwide. Human infestation is common, mainly among children, in both developing and industrialised countries [1]. Generally, infection with T. canis is clinically covert; however, severe manifestation of visceral larva migrans (VLM) was observed, involving liver, respiratory tract, eyes, myocardium and central nervous system [2–4]. Humans are infected with T. canis by swallowing invasive eggs. In the upper small bowel the eggs release larvae, which can penetrate the wall of the gut, gaining access to portal venous circulation and hepatic tissue. Larvae of T. ∗ Corresponding author at: Department of Gastroenterology, Ospedale Molinette, C.so Bramante 88, 10126 Turin, Italy. Tel.: +39 011 6336481/75; fax: +39 011 6634213. E-mail address: [email protected] (N. Leone).

canis move actively through the liver and cause necrosis, interstitial oedema, haemorrhage and eosinophilic exudates [3]. Larva survival in the liver may last several years while remaining in a dormant state, triggering no response or only a weak response from the immunological system. It can also cause serious ocular damage by migrating into the retina (ocular larva migrans). Dogs infected with adult worms release in their faeces a large number of infecting larvae which begin a somatic migration though the inner organs, releasing immunogenic compounds globally known as excretory/secretory antigens. Clinical manifestation of toxocariasis frequently includes a marked hypereosinophilia, asthenia, abdominal pain or asthma. VLM is rarely included in the differential diagnosis of secondary liver localisation. We report a case of a woman with hepatic involvement in VLM.

1590-8658/$30 © 2005 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dld.2005.07.007

512

N. Leone et al. / Digestive and Liver Disease 38 (2006) 511–514

2. Case report A 47-year-old immunocompetent woman was admitted to our facility for desultory abdominal pain without fever, started 2 weeks earlier. She was not taking medications and was not abusing alcohol. Her past medical history included a cholecystectomy, one isolated episode of acute arthritis quickly regressed with steroid therapy and a cryptogenic acute hepatitis without jaundice, respectively, 10 and 4 years before. Upon physical examination skin rash, abdominal tenderness or palpable liver, spleen or abdominal mass were all absent; laboratory parameters were normal, except mild increase of inflammatory markers (CRP 3.3 mg/dl, ESR 29 mm, serum fibrinogen 487 mg/dl). Serological studies for hepatitis B and C virus, herpes simplex, Epstein–Barr virus and Cytomegalovirus were negative. Microscopic examination of a stool specimen revealed neither parasitic larvae nor eggs. In the blood smear eosinophil counts ranged from 120 to 154 mm−3 . Levels of serum immunoglobulin were normal. A lung X-ray did not demonstrate parenchymal or pleural abnormality. Abdominal ultrasound (US) revealed five small hypoechoic liver lesions of approximately 10 mm in diameter, with irregular form and poorly defined margins, in the side sector of the right lobe. Colour/power-Doppler examination demonstrated intra-lesion arteries with low or normal diastolic resistance, no obstruction of the hepatic veins and a normal blood flow velocity in the portal vein. No intra-hepatic or extra-hepatic biliary dilatation was seen; liver, spleen, pancreas and kidneys were of normal size. Noncontrast computed tomography (CT) demonstrated multiple low-density areas with poorly defined margins (Fig. 1). These lesions were slightly enhanced at the early phase after injection of a contrast medium. T2-weighted magnetic resonance images (MRI) showed multiple mass lesions with high signal intensities (Fig. 2); these lesions were slightly enhanced at the early phase after the administration of gadolinium

Fig. 1. Enhanced CT shows hepatomegaly and low-density small areas with poorly defined margins (arrows).

Fig. 2. MRI of liver before treatment: two nodules in the hepatic right lobe with high signal intensities on T2-weighted images (arrows).

diethylenetriaminepenta-acetic acid (Gd-DTPA). Both the imaging techniques set the suspicion of necrotic areas or metastatic lever lesions. Thoracic and encephalic CT, Rx mammography, US of thyroid, colonoscopy, gastroscopy and skeletal scintigraphy were all negative for primitive neoplasm. A US-guided fine needle biopsy, performed on a small lesion, showed only necrotic cells, so laparoscopic biopsy was done. Focal necrosis with fibrosis but not atypical cells was observed in the examined sections. The surrounding liver had an accentuation of the portal connective tissue without signs of cirrhosis. The result of a specific enzyme-linked immunosorbent assay (ELISA) test to detect antibodies against Lehismania, Entamoeba histolitica, Echinococcus granulosus and Toxoplasma gondii was negative. The patient was discharged in good clinical condition with the indication to repeat a spiral abdominal CT to follow the evolution of the focal lesions. The CT scan was done after 2 months and showed an increase in the size and number of poorly delineated hypodense mass lesions in the right hepatic lobe. For better clarification of focal changes the patient underwent surgical resection. The hepatic surface on laparoscopy was smooth, without nodules and plaques. The histological examination showed an unaltered liver architecture without eosinophilic granulomatous reaction and the presence of Charcot–Leyden crystals. The sinusoids were dilated; there were focal lobular lesions including foci of hepatocellular necrosis associated with infiltrations of chronic inflammatory cells. In a field, a nematode of 25 ␮m × 400 ␮m was observed, consistent with larva migrans of T. canis (Fig. 3). An ELISA test, performed on peripheral blood, confirmed the diagnosis of hepatic involvement in VLM. The patient was put on thiabendazole 500 mg twice/day for 15 days and a second cycle with the same dose was repeated after 15 days. On follow-up visits 3 and 6 months after discharge, the patient was doing well and the focal mass lesions in the liver were reduced in number and size on CT scan.

N. Leone et al. / Digestive and Liver Disease 38 (2006) 511–514

Fig. 3. Histological examination (haematoxylin and eosin stain, magnification ×200) shows an unaltered liver architecture, foci of hepatocellular necrosis associated with infiltrations of chronic inflammatory cells. A nematode of 25 ␮m × 400 ␮m was observed (arrows), consistent with larva migrans of T. canis.

3. Discussion The two main clinical presentations of toxocariasis are VLM and ocular larva migrans (OLM). T. canis is the main aetiological agent of VLM. This syndrome, which occurs mostly in preschool children, may also be related to Toxocara catis and less commonly to Ascaris lumbricoides, Capillaria hepatica and Ascaris suum [3,4]. The prevalence of T. canis antibodies in the healthy population shows territorial variability [5]; in Northern Italy the reported seropositivity for VLM was 4% [6]. Stool examination is of no value; the diagnosis is based on the clinical features and positive results on serological tests. Several studies have evidenced that factors such as dog ownership, rural residence and low sociological status carry an increased risk of T. canis infection [5–14]. In most patients the infection with T. canis is clinically covert. The reason for the development of the symptomatic form of VLM is not entirely clear, but massive alimentary infestation and certain types of immune response are incriminating factors. The surface of T. canis is now recognised as a dynamic structure which turns over quite rapidly and serves as a renewable source of large quantities of antigens. The major host responses to these antigens include a marked eosinophilia and hyperglobulinemia. Both of these responses are apparently ineffective at ridding the body of infective larvae. Both eosinophils and IgE antibodies are manifestations of the Th2 subset of T helper cells and the cytokines that they secrete (mainly IL-1 and ␥-interferon). Further, there is reason to believe that the antigens released from T. canis larvae favour the induction of this cellular population. There is mounting evidence that the chronic production of parasite antigen and its continued stimulation of the host immune system with a concomitant production of eosinophils can lead to systemic complications. The liver is one of the most common sites for such lesions and hepatic involvement is common due to portal venous drainage of visceral organs [15,16].

513

VLM has a wide clinical spectrum. Heavy infections leading to clinical suspicion of the disease can cause pulmonary, ocular and neurologic symptoms, as well as abdominal pain, fever, hepatomegaly and marked eosinophilia. However, mild infections are often asymptomatic, and eosinophil counts and Toxocara antibody titres may be normal or only mildly elevated. This is the case of our patient in whom the diagnosis of Toxocara or other parasitic infection was not initially suspected. Imaging findings of multiple low-density areas in the liver, in the absence of systemic symptoms, had initially directed our suspect for the secondary nature of the lesions. We have searched for the causative organism with serology for Toxocara only later, when the pathologist showed a nematode strongly suggestive for VLM. So the feature of our case is represented by the absence of eosinophilic granulomatous lesions in the liver, whose identification should suggest the diagnosis of a parasitic infection, and the detection of the parasite in the tissue, which is instead rarely done. Various imaging studies are applied to confirm a diagnosis of VLM. Imaging appearance depends on the stage or activity of the underlying disease. Bhatia and Sarin [19] reported that their ultrasonographic findings changed from hypoechoic mass lesions to lesions with hyperechoic rims as the disease advanced. The hepatic lesions in our patient were homogeneously hypoechoic masses with poorly defined margins. The ‘bead sign’, described as a characteristic finding of VLM on US by Hirata et al. [20], refers to hypoechoic nodules with a central spot echo. It signifies the presence of periportal granuloma. The absence of granulomas in the liver tissue of our patient may be the reason for the absence of this sign. There are only five reports about liver manifestations of VLM on US, CT and MRI [17–19,21,22]. Jain et al. [17] reported the presence of heterogeneous mass lesions on noncontrast CT; Ishibashi et al. [21] reported that undetectable lesions with noncontrast CT were detectable in the delayed phase on contrast CT. In our report, CT showed low-density areas with slight enhancement after the administration of contrast medium. The mass lesions were detected as low-intensity signal areas on T1 and high-intensity on T2-weighted images. In previous studies the lesions were not detected on T1-weighted images. Bhatia and Sarin [19] stated that CT and MRI did not provide additional information for the diagnosis or management of patients with VLM. They are useful if US does not detect the lesions. In our case they did not have any specificity in the diagnosis of multiple liver nodules such as abscesses, granulomas and hypovascular metastases. In conclusion, we presented a case of VLM producing multiple focal liver lesions with no associated eosinophilia. VLM should be included in the differential diagnosis of multiple liver nodules, especially in cases of eosinophilia, but not alone. Conflict of interest statement None declared.

514

N. Leone et al. / Digestive and Liver Disease 38 (2006) 511–514

References [1] Overgaauw PA. Aspects of Toxocara epidemiology: human toxocarosis. Crit Rev Microbiol 1997;23:215–31. [2] Taylor MR, Keane CT, O’Connor P, Mulvihill E, Holland C. The expanded spectrum of toxocaral disease. Lancet 1988;26:692–5. [3] Brandborg LL. Parasitic disease of the liver. In: Zahim D, Boyer TD, editors. Hepatology. Philadelphia: Sunders; 1982. p. 1010– 35. [4] Gayotto LCDC, DaSilva LC. Ascariasis, visceral larva migrans, capillariasis, strongyloidiasis and pentastomiasis. In: McIntyre N, Benhamou JP, Bircher J, editors. Oxford textbook of clinical hepatology. Oxford: Oxford Medical Publications; 1991. p. 730–9. [5] Holland CV, O’Lorcain P, Taylor MR, Kelly A. A sero-epidemiology of toxocariasis in school children. Parasitology 1995;110:535–45. [6] Genchi C, Di Sacco B, Gatti S, Sangalli G, Scaglia M. Epidemiology of human toxocariasis in northern Italy. Parassitologia 1990;32:313–9. [7] Speiser TM, Tonz O. Toxocariasis in Swiss children. Schweiz Med Wochenschr 1983;113:1500–7. [8] Ljungstrom I, van Knapen F. An epidemiological and serological study of Toxocara infection in Sweden. Scand J Infect Dis 1989;21:87–93. [9] Uhlikova M, Hubner J. Seroprevalence of Toxocara canis infection in the Czech Republic. Cent Eur J Public Health 1998;6:195–8. [10] Gundlach JL, Sadzikowski AB, Tomczuk K. Occurrence of antibodies against Toxocara canis in the sera of humans. Med Wet 1996;52:516–7. [11] Fenoy S, Cuellar C, Guillen JL. Seroprevalence of toxocariasis in children and adults in Madrid and Tenerife, Spain. J Helminthol 1996;70:109–13.

[12] Avdiuhina TI, Lysenko A. How many patients with visceral toxocariasis in Russia? Med Parazitol 1994;1:12–6. [13] Van Gemund JJ, Buijs J, van Dongen PA, van den Bergh JP. Seroprevalence of Toxocara infection in young children in the city of the Hague. Trop Geogr Med 1989;41:294–6. [14] Kimming P, Naser K, Frank W. Seroepidemiologic studies of human toxocariasis. Zentralbl Hyg Umweltmed 1991;191:406–22. [15] Rayes AA, Lambertucci JR. Visceral larva migrans and pyogenic liver abscess. Am J Gatroenterol 1999;94:1116. [16] Kayes SG. Human toxocariasis and the visceral larva migrans syndrome: correlative immunopathology. Chem Immunol 1997;66:99–124. [17] Jain R, Sawhney S, Bhargava DK, Panda SK, Berry M. Hepatic granulomas due to visceral larva migrans in adults: appearance on US and MRI. Abdom Imaging 1994;19:253–6. [18] Hayashi K, Tahara H, Yamashita K, Kuroki K, Matsushita R, Yama Hori T, et al. Hepatic imaging studies on patients with visceral larva migrans due to probable Ascaris suum infection. Abdom Imaging 1999;24:465–9. [19] Bhatia V, Sarin SK. Hepatic visceral larva migrans: evolution of the lesion, diagnosis, and role of high-dose albendazole therapy. Am J Gastroenterol 1994;89:624–7. [20] Hirata T, Yamasaki K, Li Y, Majima Y, Tsuji M. Demonstration of hepatic granuloma due to visceral larva migrans by ultrasonography. J Clin Ultrasound 1990;18:429–33. [21] Ishibashi H, Shimamura R, Hirata Y, Kudo J, Onizuka H. Hepatic granuloma in toxocaral infection: role of ultrasonography in hypereosinophilia. J Clin Ultrasound 1992;20:204–10. [22] Azuma K, Yashiro N, Kinoshita T, Yoshigi J, Ihara N. Hepatic involvement of visceral larva migrans due to Toxocora canis: a case report—CT and MR findings. Radiat Med 2002;20:89–92.