Rodrigo da Fonseca Monjardim1; Danilo Manuel Cerqueira Costa1; Ricardo Francisco Tavares Romano1; Priscila Silveira Salvadori1; Jaime de Vargas Conde dos Santos1; Augusto Castelli Von Atzingen2; David Carlos Shigueoka3; Giuseppe D'Ippolito4
ABSTRACT
OBJECTIVE: To evaluate the diagnostic capacity of abdominal computed tomography in the assessment of hepatic steatosis using the portal phase with a simplified calculation method as compared with the non-contrast-enhanced phase. MATERIALS AND METHODS: In the present study, 150 patients were retrospectively evaluated by means of non-contrast-enhanced and contrast-enhanced computed tomography. One hundred patients had hepatic steatosis and 50 were control subjects. For the diagnosis of hepatic steatosis in the portal phase, the authors considered a result of < 104 HU calculated by the formula [L - 0.3 × (0.75 × P + 0.25 × A)] / 0.7, where L, P and A represent the attenuation of the liver, of the main portal vein and abdominal aorta, respectively. Sensitivity, specificity, positive and negative predictive values were calculated, using non-contrast-enhanced computed tomography as the reference standard. RESULTS: The simplified calculation method with portal phase for the diagnosis of hepatic steatosis showed 100% sensitivity, 36% specificity, negative predictive value of 100% and positive predictive value of 75.8%. The rate of false positive results was 64%. False negative results were not observed. CONCLUSION: The portal phase presents an excellent sensitivity in the diagnosis of hepatic steatosis, as compared with the non-contrast-enhanced phase of abdominal computed tomography. However, the method has low specificity.
Keywords: Fatty liver; Steatosis; Liver disease; Tomography; Multidetector computed tomography; Contrast media.
RESUMO
OBJETIVO: Comparar a capacidade diagnóstica para esteatose hepática utilizando-se a fase portal com método simplificado de cálculo com a fase sem contraste na tomografia computadorizada de abdome. MATERIAIS E MÉTODOS: Foi realizado estudo retrospectivo em 150 pacientes submetidos a tomografia computadorizada de abdome sem e com contraste intravenoso, 100 deles com esteatose hepática e 50 controles. Para diagnóstico de esteatose hepática na fase portal considerou-se um resultado < 104 UH aplicando-se a fórmula [L - 0,3 × (0,75 × P + 0,25 × A)] / 0,7, onde L, P e A representam a atenuação hepática, da veia porta e da aorta, respectivamente. Foram calculados sensibilidade, especificidade, valores preditivos positivos e negativos, utilizando-se a tomografia computadorizada sem contraste intravenoso como padrão de referência. RESULTADOS: O método simplificado de cálculo para o diagnóstico de esteatose hepática na fase portal mostrou sensibilidade de 100%, especificidade de 36%, valor preditivo negativo de 100% e valor preditivo positivo de 75,8%. A taxa de falso-positivos foi 64%. Não foram encontrados falso-negativos. CONCLUSÃO: A utilização da fase portal apresenta elevada sensibilidade para o diagnóstico de esteatose hepática, quando comparada à fase sem contraste da tomografia computadorizada de abdome. Por outro lado, o método apresenta baixa especificidade.
Palavras-chave: Fígado gorduroso; Esteatose hepática; Hepatopatias; Tomografia; Tomografia computadorizada multidetectores; Meios de contraste.
INTRODUCTION Hepatic steatosis (HS) has been considered a clinically significant condition due the increased risk for progression to necroinflammatory alterations, even in cases where the disease is not associated with alcohol consumption (in this case known as nonalcoholic steatohepatitis – NASH), and to cirrhosis(1). Data from the United States of America estimates that 40 million adults are affected by NASH(1). Other studies indicate that the prevalence of HS in the general population may reach 24%(2,3). In cases where NASH is early diagnosed, the treatment may avoid progression to a final and irreversible stage of hepatic cirrhosis(4). Additionally, HS patients may be disqualified as eligible living liver donor candidates for hepatic transplant because the steatotic liver is more vulnerable to injuries from anesthetic procedures and ischemic reperfusion(5–7). In this context, up to 30% of living liver donor candidates are rejected for presenting hepatic steatosis(8). Percutaneous liver biopsy has been the reference standard for diagnosis and staging of HS(4), but it is an invasive method with possible complications. On the other hand, imaging methods such as ultrasonography (US), computed tomography (CT) and magnetic resonance imaging (MRI) have been extensively utilized, not only for the comprehensive assessment of abdominal disorders, but specifically in the diagnosis of HS(1,4,9). US, despite being widely available and without contraindications, relies on subjective and hardly reproducible criteria which do not allow a reliable quantification of steatosis(4). MRI is currently considered as being the most effective noninvasive method for the diagnosis of HS(4), but it is still a highly expensive method and is not widely available in our country. CT, on its turn, has been the method most widely utilized in the evaluation of abdominal disorders and is an important tool in the diagnosis of HS(10,11). Several factors have contributed to such a wide utilization, among them the development of multidetector CT apparatuses (MDCT), allowing faster images acquisition, with greater comfort for the patient and high diagnostic accuracy. However, the harmful effects of ionizing radiation inherent to the method have been a constant source of preoccupation(12,13). Among the several strategies for radiation dose reduction in CT scans, a thorough review of CT protocols has been recommended with the purpose of avoiding unnecessary scan phases(14). The contrast-enhanced portal phase performed between 60 and 80 seconds after intravenous contrast injection is isolatedly considered as being the most useful phase in the greatest majority of indications for upper abdomen CT, provided the reliability and diagnostic accuracy of the method is assured(9,14,15), leading to a tendency to abolish the non-contrast-enhanced phase. On the other hand, according to several authors, the CT criteria for diagnosis of HS are dependent on such phase(10,11,16). Such a fact has stimulated the undertaking of studies which attempted to establish the value of the portal phase isolatedly in the diagnosis of hepatic steatosis(9,10,17). The most recent of those studies demonstrated that, with the isolated utilization of the portal phase, it is possible to obtain accuracy similar to that obtained with the criteria adopted at non-contrastenhanced CT in the diagnosis and quantification of HS(9). The authors proposed a mathematical equation which takes into consideration the average density of the portal vein and the aorta based on measurements obtained at three different sites and the average density of the liver obtained from eight measurements (two for each one of the four segments of the right lobe). Such a strategy, in spite of having demonstrated good results, is time consuming for the investigator, making the method somewhat tiresome and certainly more complex than the simple measurements obtained with the non-contrast-enhanced phase. Additionally, the several variables involved in the process, such as concentration, contrast medium injection velocity and acquisition time may influence the results(16), lead to the necessity of validating the previous results and prove the usefulness of a simplified method for HS diagnosis by means of contrast-enhanced CT. Considering the prevalence of HS, the present study was aimed to evaluate the diagnostic capability of the portal phase with a simplified calculation method, as compared with the non-contrast-enhanced phase of abdominal CT. MATERIALS AND METHODS The present study was approved by the Committee for Ethics in Research of the authors' institution, and the application of a term of free and informed consent was waived. The abdominal CT images were retrospectively analyzed regardless of clinical indication, in the period between May, 2011 and February, 2012, until the arbitrarily established number of 100 HS patients, called "cases", and 50 HS-free patients, called "controls", was reached, utilizing non-contrast-enhanced CT as a reference imaging method. In the group of cases in the sample, 49 patients were men (mean age = 51.6 years ± 14.3) and 51 were women (mean age = 54.7 years ± 14.0). Among the "controls" 26 patients were men (mean age = 57.6 years ± 18.2) and 24 were women (mean age = 53.2 years ± 18.4). Only patients above the age of 18 and with spontaneous requests for contrast-enhanced abdominal CT were included in the present study. Exclusion criteria were the following: a) heterogeneous liver; b) patients with innumerable lesions in both liver lobes; c) inappropriate contrast medium injection (for example: contrast medium extravasation); d) images with artifacts (for example: respiratory artifacts) which would make density measurements inaccurate or unreliable; e) incomplete images. Non-inclusion criteria were the following: a) patients under the age of 18; b) indication for non-contrast-enhanced abdominal CT; c) patients with contraindication for iodinated contrast medium; d) splenectomized patients. The scans were performed in a Brilliance 64® model CT apparatus (Philips Medical Systems; Best, The Netherlands), with volumetric acquisition before and 70 seconds after intravenous injection of the hydrosoluble iodinated contrast medium, by means of an automatic injection pump at a rate of 3–4 ml/s and 1.5–2 ml/kg of body weight, with a maximum volume of 150 ml. The technical parameters commonly utilized at CT scans were the following: a) collimation of 64 × 0.625 mm; b) gantry rotation time = 0.5 s; c) reconstruction with 1 and 3 mm slice thickness; d) pitch = 0.891; e) 120 kVp; f) mAs depending upon automatic modulation of the radiation dose (DoseRight ACS®). All the images were interpreted by a single observer with three-year experience in imaging diagnosis, on a Synapse® PACS/3D (FujiFilm; USA) workstation. At the moment of the interpretation, none of the clinical data were available to the observer. The differentiation between the two patient groups was based on the diagnostic criteria for HS established for non-contrastenhanced CT, utilizing the analysis of hepatic and splenic attenuation values in Hounsfield units (HU), considering that a liver attenuation value lower than the splenic attenuation value indicates the diagnosis of steatosis(4,17,18). The diagnosis of HS with contrast-enhanced CT was obtained by means of the method proposed by Kim et al.(9), utilizing the formula [L – 0.3 × (0.75 × P + 0.25 × A)] / 0.7, where L corresponds to hepatic attenuation, P corresponds to portal vein attenuation, and A corresponds to abdominal aorta attenuation. The measurements were performed in the portal venous contrast phase, i.e., between 60 and 80 seconds after the initiation of the intravenous contrast medium injection. In the original study, the variables L, P and A were obtained from the averages of eight measurements of the liver, three of the portal vein and three of the aorta, respectively, which makes the process quite burdensome and hardly practical. Some studies have proposed that the utilization of a single liver attenuation measurement does not present significant difference in relation to the average obtained after multiple measurements(11,19). With such premise in mind and with the objective of validating a simplified method for the diagnosis of HS relying on the portal phase isolatedly, a pilot study was initially undertaken with five patients without HS (liver attenuation value higher than splenic attenuation value in the non-contrast-enhanced phase), five with mild HS (difference between liver and spleen attenuation values > –10 HU), and five with severe HS (difference between liver and spleen attenuation values < –10 HU)(4), comparing the values resulting from the utilization of the above mentioned formula with all mean values of the variables (original method) with a single measurement for each variable (simplified method). The analysis of the pilot study was performed by two independent investigators blinded with respect to the results from the non-contrast-enhanced phase. A almostperfect agreement was observed between the original and the simplified methods of calculation (kappa = 0.842; p = 0.001). The initial results justified the continuity of the present study, applying the previously proposed formula(9) with a single measurement for each anatomical site (right hepatic lobe, portal vein and aorta), thus simplifying the method (Figure 1).