Mem Inst Oswaldo Cruz, Rio de Janeiro, 105(4) July 2010
Schistosoma mansoni: magnetic resonance analysis of liver fibrosis according to WHO patterns for ultrasound assessment of schistosomiasis-related morbidity
IPrograma de Pós-Graduação em Ciências da Saúde: Infectologia e Medicina Tropical, Faculdade de Medicina
IIDepartamento de Diagnóstico por Imagem, Instituto Hermes Pardini, Belo Horizonte, MG, Brasil
IIIServiço de Radiologia, Hospital das Clínicas, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brasil
For the last two decades, ultrasound (US) has been considered a surrogate for the gold standard in the evaluation of liver fibrosis in schistosomiasis. The use of magnetic resonance imaging (MRI) is not yet standardised for diagnosing and grading liver schistosomal fibrosis. The aim of this paper was to analyse MRI using an adaptation of World Health Organization (WHO) patterns for US assessment of schistosomiasis-related morbidity. US and MRI were independently performed in 60 patients (42.1 ± 13.4 years old), including 37 men and 23 women with schistosomiasis. Liver involvement appraised by US and MRI was classified according to the WHO protocol from patterns A-F. Agreement between image methods was evaluated by kappa index (k). The correlation between US and MRI was poor using WHO patterns k = 0.14; confidence interval (CI) 0.02; 0.26. Even after grouping image patterns as "A-D", "Dc-E" and "Ec-F", the correlation between US and MRI remained weak (k = 0.39; CI 0.21; 0.58). The magnetic resonance adaptation used in our study did not confirm US classification of WHO patterns for liver fibrosis.
The gold standard method in evaluating schistosomal hepatic fibrosis was wedge liver biopsy, a safe procedure performed during abdominal surgery, but not justified in non-surgical patients. An alternative procedure, percutaneous liver biopsy, has low sensitivity because it retrieves small and fragmented samples with few portal tracts (Bogliolo 1957, Cheever 1968, Maharaj et al. 1986).
For the last 20 years, abdominal ultrasound (US) has become the best tool for the evaluation of liver fibrosis in schistosomiasis mansoni (Lambertucci et al. 1996, Richter et al. 2001). It is an indirect method of diagnosis and classification of the disease (Abdel-Latif et al. 1981, Cerri et al. 1984, Fataar et al. 1984, Homeida et al. 1988, Pinto-Silva et al. 1994, Gerspacher-Lara et al. 1998, Lambertucci et al. 2001). US in hepatosplenic schistosomiasis shows characteristic alterations, such as echogenic thickening of the walls of the portal vein and its branches, corresponding to periportal fibrosis and echogenic enlargement of the gallbladder wall. However, US has disadvantages. Training is necessary for the examiner to apply the World Health Organization (WHO) protocol for US assessment of schistosomiasis (Niamey Working Group 2000) and it is a subjective procedure and, therefore, examiner-dependent.
Magnetic resonance imaging (MRI) is a very sensitive imaging technique and, unlike US, is less dependent on the examiner. MRI has been described in hepatosplenic schistosomiasis mansoni (Patel et al. 1993, Willemsen et al. 1995, Lambertucci et al. 2002, 2004, Bezerra et al. 2004, 2007, 2008, Silva et al. 2006), but there is no standardisation of the method in the evaluation of liver periportal fibrosis. The aim of this paper was to analyse MRI using an adaptation of the WHO patterns for US assessment of schistosomiasis-related morbidity.
PATIENTS, MATERIALS AND METHODS
Patients - The study was approved by the Ethical Review Board of the Universidade Federal de Minas Gerais (UFMG), Brazil, and all study subjects signed the informed consent for participation. Sixty adult patients (37 men and 23 women) with a diagnosis of schistosomiasis from the Infectious Diseases Outpatient Clinic, UFMG, were studied from 2004-2006. Schistosomiasis was defined by microscopic evidence of the infection (positive parasitological stool examination or rectal biopsy) and/or characteristic US signs of periportal fibrosis in a patient exposed to stream waters of an endemic area. Patients included in the study did not have a history of previous surgery for portal hypertension or evidence or vestiges of other causes of chronic liver disease, such as cirrhosis, congestive heart failure and toxic or viral hepatitis. Ages ranged from 19-77 years old (mean 42.1 ± 13.4). Baseline characteristics of the patients are shown in Table I.
Methods - Routine outpatient evaluation included clinical examination, abdominal US, blood cell count, assessment of serum albumin, prothrombin, aspartate aminotransferase and alanine aminotransferase levels and serology for hepatitis B and C.
US was performed using a real-time device (ALOKA SSD 1700, ALOKA CO, Japan) with an electronic convex 3.5 MHz transducer, according to the WHO protocol for assessment of schistosomiasis-related morbidity (Niamey Working Group 2000). Hepatic fibrosis was classified from patterns A-F, defined as follows: A: normal liver structure; B: "starry sky" (diffuse echogenic foci); C: highly echogenic "ring echoes", which correspond to the "pipe stems" seen in a scan perpendicular to the one where rings are seen; D: highly echogenic "ruff" around portal bifurcation and main stem; Dc: combined D and C patterns; E: highly echogenic "patches" extending from the main portal vein and branches into the parenchyma; Ec: combined E and C patterns; F: highly echogenic "bands" and "streaks", extending from the main portal vein and its bifurcation to the liver surface, causing retraction of the organ surface.
MRI was obtained using a 1.5-T unit (GE Sigma unit, General Electric, USA). Axial and coronal 7 mm slice thickness images were acquired in T1 and T2-weighted sequences before and after gadopentetate dimeglumine administration (0.1 mmol/kg) (Magnevist, Bayer Pharmaceuticals, Germany). MRI analysis was adapted from the WHO protocol proposed for US assessment of schistosomiasis-related morbidity (Fig. 1).
Two radiologists (one for US and another for MRI) independently examined the images. They were blinded regarding patients' clinical states.
Statistical analysis - Agreement between image methods was evaluated by kappa index (k). The degree of agreement was classified as follows: 0.81-1, almost perfect; 0.61-0.8, strong; 0.41-0.6, moderate; 0.21-0.4, fair; 0-0.2, poor.
Of the 60 patients in the study, US and MRI identified periportal thickening in 54 (90%) and 47 (78.3%) cases, respectively. MRI fibrosis was characterised by hypointense periportal bands on T1-weighted images, which were hyperintense on T2-weighted images (Fig. 2). Periportal enhancement after injection of paramagnetic contrast material was observed in 41 out of 47 (87.2%) patients.
US image patterns were classified according to WHO orientations as follows (Table II): A, n = 6 (10%); D, n = 7 (11.7%); Dc, n = 6 (10%); E, n = 9 (15%); Ec, n = 24 (40%) and F, n = 8 (13.3%). MRI patterns were classified as follows: A, n = 13 (21.7%); D, n = 1 (1.7%); Dc, n = 25 (41.7%); Ec, n = 12 (20%) and F, n = 9 (15%). The correlation between US and MRI was poor using WHO patterns in schistosomiasis [k = 0.14; confidence interval (CI) 0.02; 0.26]. Even after grouping image patterns as "A-D", "Dc and E" and "Ec and F", the correlation between US and MRI remained fair (k = 0.39; CI 0.21; 0.58).
In the present study, US and MRI presented a poor correlation in classifying liver fibrosis according to the WHO patterns for US assessment of schistosomiasis-related morbidity. After grouping image patterns, the correlation barely improved to fair. Therefore, the magnetic resonance adaptation used in our study did not confirm US classification of WHO patterns for liver fibrosis.
MRI identified schistosomal fibrosis as hyperintense periportal bands on T2-weighted images and hypointense on T1-weighted images with enhancement after contrast, in accordance with previous reports and studies (Patel et al. 1993, Willemsen et al. 1995, Lambertucci et al. 2002, 2004, Bezerra et al. 2004, 2007, 2008).
Efforts to establish procedures for US evaluation of schistosomiasis have been published since the early 1990s (Cairo Working Group 1992, Niamey Working Group 2000). In 2000, the WHO published a protocol for US performance and analysis in the evaluation of schistosomiasis-related morbidity (Niamey Working Group 2000), based on previous studies. In the protocol, recommendations to assess periportal fibrosis, portal hypertension and liver and spleen enlargement were established. However, the quantitative assessment of periportal thickness of portal vein walls, particularly of the secondary branches, proved to be difficult in everyday use. In addition, it is considered time-demanding and complex, leading to low reproducibility (Thomas et al. 1997, Yazdanpanah et al. 1997, Hoffmann et al. 2001, Richter et al. 2001, Ruiz et al. 2002, King et al. 2003).
Besides the quantitative evaluation, an additional qualitative method for assessing periportal fibrosis was proposed, comparing the observed liver texture with a series of standard reference patterns. The qualitative proposal was expected to define morbidity promptly. Even though the method is simple to use, small nonspecific alterations were mistakenly interpreted as abnormal (for example, patients with depleted liver glycogen due to prolonged fasting were diagnosed as having periportal thickening). Major observer-related discordances have also been reported.
In a patient without portal hypertension and with intense periportal thickening diagnosed by US, MRI described a normal liver (Silva et al. 2006). In fact, MRI identified fat tissue around the main portal vein that was seen as fibrosis by US. Therefore, in patients with periportal thickening without portal hypertension, US must be confirmed by MRI.
MRI has been used in a variety of liver diseases and its value has been proven. In schistosomiasis, there were only reports of cases or series of cases describing characteristic alterations, but without classification of liver fibrosis. This study is the first to classify schistosomal periportal fibrosis intensity using MRI. The use of WHO US patterns as a guide did not result in sufficient agreement to be recommended. Our results indicate that new patterns should be constructed to better reflect MRI findings.
Abdel-Latif Z, Abdel-Wahab MF, El-Kady NM 1981. Evaluation of portal hypertension in cases of hepatosplenic schistosomiasis using ultrasound. J Clin Ultrasound 9: 409-412.
Bezerra AS, D'Ippolito G, Caldana RP, Cecin AO, Ahmed M, Szejnfeld J 2007. Chronic hepatosplenic schistosomiasis mansoni: magnetic resonance imaging and magnetic resonance angiography findings. Acta Radiol 48: 125-134.
Bezerra AS, D'Ippolito G, Caldana RP, Leopoldino DD, Batista GR, Borges DR, Lopes Filho GJ, Ahmed M 2008. Differentiating cirrhosis and chronic hepatosplenic schistosomiasis using MRI. AJR Am J Roentgenol 190: W201-207.
Bezerra ASA, D'Ippolito G, Caldana RP, Cecin AO, Szejnfeld J 2004. Avaliação hepática e esplênica por ressonância magnética em pacientes portadores de esquistossomose mansônica crônica. Radiol Bras 37: 313-321.
Bogliolo L 1957. The anatomical picture of the liver in hepato-splenic schistosomiasis mansoni. Ann Trop Med Parasitol 51: 1-14.
Cairo Working Group 1992. The use of diagnostic ultrasound in schistosomiasis - attempts at standardization of methodology. Acta Trop 51: 45-63.
Cerri GG, Alves VA, Magalhães A 1984. Hepatosplenic schistosomiasis mansoni: ultrasound manifestations. Radiology 153: 777-780.
Cheever AW 1968. A quantitative pos-mortem study of schistosomiasis mansoni in man. Am J Trop Med Hyg 17: 38-64.
Fataar S, Bassiony H, Satyanath S, Vassileva J, Hanna RM 1984. Characteristic sonographic features of schistosomal periportal fibrosis. AJR Am J Roentgenol 143: 69-71.
Gerspacher-Lara R, Pinto-Silva RA, Serufo JC, Rayes AAM, Drummond SC, Lambertucci JR 1998. Splenic palpation for the evaluation of morbidity due to schistosomiasis mansoni. Mem Inst Oswaldo Cruz 93 (Suppl. I): 245-248.
Hoffmann H, Esterre P, Ravaoalimalala VA, Ehrich JH, Doehring E 2001. Morbidity of schistosomiasis mansoni in the highlands of Madagascar and comparison of current sonographical classification systems. Trans R Soc Trop Med Hyg 95: 623-629.
Homeida M, Abdel-Gadir AF, Cheever AW, Bennett JL, Arbab BM, Ibrahium SZ, Abdel-Salam IM, Dafalla AA, Nash TE 1988. Diagnosis of pathologically confirmed Symmers' periportal fibrosis by ultrasonography: a prospective blinded study. Am J Trop Med Hyg 38: 86-91.
King CH, Magak P, Salam EA, Ouma JH, Kariuki HC, Blanton RE 2003. Measuring morbidity in schistosomiasis mansoni: relationship between image pattern, portal vein diameter and portal branch thickness in large-scale surveys using new WHO coding guidelines for ultrasound in schistosomiasis. Trop Med Int Health 8: 109-117.
Lambertucci JR, Andrade LM, Pinto-Silva RA 2002. Magnetic resonance imaging of the liver in hepatosplenic schistosomiasis mansoni. Rev Soc Bras Med Trop 35: 679-680.
Lambertucci JR, Cota GF, Pinto-Silva RA, Serufo JC, Gerspacher-Lara R, Drummond SC, Antunes CM, Nobre V, Rayes A 2001. Hepatosplenic schistosomiasis in field-based studies: a combined clinical and sonographic definition. Mem Inst Oswaldo Cruz 96 (Suppl.): 147-150.
Lambertucci JR, Gerspacher-Lara R, Pinto-Silva RA, Barbosa MM, Teixeira R, Barbosa HF, Serufo JC, Rezende DF, Drummond SC, Rayes AA 1996. The Queixadinha Project: morbidity and control of schistosomiasis in an endemic area in the northeast of Minas Gerais, Brazil. Rev Soc Bras Med Trop 29: 127-135.
Lambertucci JR, Silva LC, Andrade LM, de Queiroz LC, Pinto-Silva RA 2004. Magnetic resonance imaging and ultrasound in hepatosplenic schistosomiasis mansoni. Rev Soc Bras Med Trop 37: 333-337.
Maharaj B, Maharaj RJ, Leary WP, Cooppan RM, Naran AD, Pirie D, Pudifin DJ 1986. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancet 1: 523-525.
Niamey Working Group 2000. Ultrasound in schistosomiasis. A practical guide to the standardized use of ultrasonography for the assessment of schistosomiasis-related morbidity. World Health Organization/TDR/SCH/ULTRASON/document, Geneva, Switzerland. Available from: http://www.who.int/tdrold/publications/publications/ultrasound.htm.
Patel SA, Castillo DF, Hibbeln JF, Watkins JL 1993. Magnetic resonance imaging appearance of hepatic schistosomiasis, with ultrasound and computed tomography correlation. Am J Gastroenterol 88: 113-116.
Pinto-Silva RA, Abrantes WL, Antunes CM, Lambertucci JR 1994. Sonographic features of portal hypertension in schistosomiasis mansoni. Rev Inst Med Trop Sao Paulo 36: 355-361.
Richter J, Domingues ALC, Barata CH, Prata AR, Lambertucci JR 2001. Report of the Second Satellite Symposium on Ultrasound in Schistosomiasis. Mem Inst Oswaldo Cruz 96 (Suppl.): 151-156.
Ruiz R, Candia P, Garassini M, Tombazzi C, Certad G, Bruces AC, Noya O, Alarcón de Noya B 2002. Schistosomiasis mansoni in low transmission areas. Abdominal ultrasound. Mem Inst Oswaldo Cruz 97 (Suppl. I): 153-159.
Silva LCS, Pereira ACF, Queiroz LC, Andrade LM, Antunes CMF, Lambertucci JR 2006. Disagreement between ultrasound and magnetic resonance imaging in the identification of schistosomal periportal fibrosis. Mem Inst Oswaldo Cruz 101 (Suppl. I): 279-282.
Thomas AK, Dittrich M, Kardorff R, Talla I, Mbaye A, Sow S, Niang M, Yazdanpanah Y, Stelma FF, Gryseels B, Doehring E 1997. Evaluation of ultrasonographic staging systems for the assessment of Schistosoma mansoni induced hepatic involvement. Acta Trop 68: 347-356.
Willemsen UF, Pfluger T, Zoller WG, Kueffer G, Hahn K 1995. MRI of hepatic schistosomiasis mansoni. J Comput Assist Tomogr 19: 811-813.
Yazdanpanah Y, Thomas AK, Kardorff R, Talla I, Sow S, Niang M, Stelma FF, Decam C, Rogerie F, Gryseels B, Capron A, Doehring E 1997. Organometric investigations of the spleen and liver by ultrasound in Schistosoma mansoni endemic and nonendemic villages in Senegal. Am J Trop Med Hyg 57: 245-249.