J Hepatobiliary Pancreat Sci (2014) 21:669–675 DOI: 10.1002/jhbp.112
ORIGINAL ARTICLE
Diagnostic accuracy of transient hepatic attenuation differences on computed tomography scans for acute cholangitis in patients with malignant disease Toshiya Sugishita · Ryota Higuchi · Satoru Morita · Takehiro Ota · Masakazu Yamamoto
Published online: 12 May 2014 © 2014 Japanese Society of Hepato-Biliary-Pancreatic Surgery
Abstract Background Few reports have assessed the accuracy of transient hepatic attenuation differences (THAD) in the arterial phase of computed tomography (CT) in diagnosing acute cholangitis (AC). We examined the diagnostic accuracy of THAD in AC in patients with malignant disease. Methods We retrospectively examined the records of 123 consecutive patients (73 men, 50 women; mean age, 69 [range, 43–91] years) who underwent dynamic CT 3 days before biliary drainage for malignant hepatobiliary disease between 2006 and 2011. We examined the diagnostic ability of THAD for AC and assessed the relationship between THAD and AC severity. Results Acute cholangitis was present in 45% of patients. The sensitivity and specificity of THAD for AC were 93% and 39%, respectively. Diffuse, hemi-hepatic, and segmental THAD for AC were seen in 84%, 4%, and 9%, respectively. In all, 64% of patients were diagnosed with severe AC and 29% with mild AC, depending on the intensity of THAD, but the patterns and degree of intensity of THAD and AC severity were not related. Conclusion Although the pattern and the degree of intensity of THAD and AC severity were not related, THAD might be useful in diagnosing AC. Thus, THAD should be considered a diagnostic criterion for AC. Keywords Cholangitis · CT · Diagnosis · Guideline · THAD
T. Sugishita · R. Higuchi (*) · S. Morita · T. Ota · M. Yamamoto Department of Surgery, Institute of Gastroenterology, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan e-mail: [email protected]
Introduction Acute cholangitis (AC) is a fatal biliary disease. The basic treatment is biliary drainage at the appropriate time depending on severity [1–5]. Therefore, an accurate diagnosis of the severity of AC is critical. Conventionally, transient hepatic attenuation differences (THAD) is used for diagnosing cholangitis [6–16]. The importance of THAD for diagnosing AC has also been discussed in the Tokyo Guideline 2013 (TG13) [2]. THAD are areas of parenchymal enhancement visible during the hepatic arterial phase of computed tomography [10]. However, few reports have discussed its diagnostic accuracy and the correlation between THAD and the severity of AC, especially with regard to biliary malignancy. Consequently, in this study, we examined the diagnostic ability of THAD for AC in patients with malignant disease. Methods In this retrospective study, we examined the records of 123 consecutive patients who had visited our institute between 2006 and 2011. We included patients who had undergone dynamic CT within 3 days before undergoing biliary drainage for hepato-pancreato-biliary neoplasm between 2006 and 2011. We excluded patients who had cholecystitis or any other source of infection. We defined AC as a local inflammation of the bile duct (fever ≥37.5°C, C-reactive protein [CRP] level ≥1.0 mg/dl, or white blood count [WBC] ≥10,000/μl), cholestasis that improved on biliary drainage, or positive culture of the bile or blood. The institutional review board of Tokyo Women’s Medical University approved this retrospective study, and the need for patients’ informed consent was waived. All patients underwent multidetector computed tomography (MDCT) with one of three scanners: 4 row, Aquilion
670
(a)
J Hepatobiliary Pancreat Sci (2014) 21:669–675
(c)
(b)
Fig. 1 (a) Diffuse type. Diffuse transient hepatic attenuation differences (THAD) in which the whole liver shows heterogeneous enhancement in a patient with pancreatic head cancer and acute cholangitis (AC) (arrow). (b) Hemi-hepatic type. Hemi-hepatic and diffuse THAD in which the right liver shows slightly stronger enhancement than the left liver without the portal vein obstruction in a patient with perihilar cholangiocarcinoma without AC (arrow). (c) Segmental type. Segmental THAD in which the liver shows segmental enhancement in a patient with pancreatic head cancer without AC (arrow)
4, Toshiba Medical Systems, Tochigi, Japan; 16 row, Aquilion 16, Toshiba Medical Systems; or 16 row, Light Speed Ultra, GE Healthcare, USA. Images of the late arterial, portal venous, and equilibrium phases were obtained at 40, 60 and 110 s, respectively, after the nonionic contrast medium (100–150 ml at rate of 3–4 ml/s) was injected. Omnipaque 300 (Daiichi Sankyo, Tokyo, Japan) and Iopamiron 370 (Bayer, Osaka, Japan) were the contrast material used. Images were reconstructed at a thickness of 5.0 mm. We classified enhanced THAD patterns in the liver parenchyma in the arterial phase as diffuse (Fig. 1a), hemihepatic (Fig. 1b), and segmental (Fig. 1c) types. We classified the degree of intensity of THAD according to two patterns: severe (an obvious difference in intensity) and mild (a mild difference in intensity). We also defined portal vein inflow obstruction as the absence of portal vein flow due to major portal-branch obstruction. The CT findings were interpreted in consensus by one radiologist (SM) and two hepatobiliary surgeons (ST and RH). We studied the diagnostic ability of THAD for AC according to the relationship between patterns, intensity of THAD, and the severity of AC. The severity of AC was diagnosed according to TG13 [2]. Statistical analysis Continuous variables were expressed as means (minimum– maximum values). Associations between continuous variables were analyzed using the Welch 2-sample t-test. Associations among qualitative variables were analyzed using Fisher’s exact test in the R environment (R version 2.15.2; The R Foundation for Statistical Computing, Vienna, Austria) to test the independence of a contingency table. All P-values
ORIGINAL ARTICLE
Diagnostic accuracy of transient hepatic attenuation differences on computed tomography scans for acute cholangitis in patients with malignant disease Toshiya Sugishita · Ryota Higuchi · Satoru Morita · Takehiro Ota · Masakazu Yamamoto
Published online: 12 May 2014 © 2014 Japanese Society of Hepato-Biliary-Pancreatic Surgery
Abstract Background Few reports have assessed the accuracy of transient hepatic attenuation differences (THAD) in the arterial phase of computed tomography (CT) in diagnosing acute cholangitis (AC). We examined the diagnostic accuracy of THAD in AC in patients with malignant disease. Methods We retrospectively examined the records of 123 consecutive patients (73 men, 50 women; mean age, 69 [range, 43–91] years) who underwent dynamic CT 3 days before biliary drainage for malignant hepatobiliary disease between 2006 and 2011. We examined the diagnostic ability of THAD for AC and assessed the relationship between THAD and AC severity. Results Acute cholangitis was present in 45% of patients. The sensitivity and specificity of THAD for AC were 93% and 39%, respectively. Diffuse, hemi-hepatic, and segmental THAD for AC were seen in 84%, 4%, and 9%, respectively. In all, 64% of patients were diagnosed with severe AC and 29% with mild AC, depending on the intensity of THAD, but the patterns and degree of intensity of THAD and AC severity were not related. Conclusion Although the pattern and the degree of intensity of THAD and AC severity were not related, THAD might be useful in diagnosing AC. Thus, THAD should be considered a diagnostic criterion for AC. Keywords Cholangitis · CT · Diagnosis · Guideline · THAD
T. Sugishita · R. Higuchi (*) · S. Morita · T. Ota · M. Yamamoto Department of Surgery, Institute of Gastroenterology, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan e-mail: [email protected]
Introduction Acute cholangitis (AC) is a fatal biliary disease. The basic treatment is biliary drainage at the appropriate time depending on severity [1–5]. Therefore, an accurate diagnosis of the severity of AC is critical. Conventionally, transient hepatic attenuation differences (THAD) is used for diagnosing cholangitis [6–16]. The importance of THAD for diagnosing AC has also been discussed in the Tokyo Guideline 2013 (TG13) [2]. THAD are areas of parenchymal enhancement visible during the hepatic arterial phase of computed tomography [10]. However, few reports have discussed its diagnostic accuracy and the correlation between THAD and the severity of AC, especially with regard to biliary malignancy. Consequently, in this study, we examined the diagnostic ability of THAD for AC in patients with malignant disease. Methods In this retrospective study, we examined the records of 123 consecutive patients who had visited our institute between 2006 and 2011. We included patients who had undergone dynamic CT within 3 days before undergoing biliary drainage for hepato-pancreato-biliary neoplasm between 2006 and 2011. We excluded patients who had cholecystitis or any other source of infection. We defined AC as a local inflammation of the bile duct (fever ≥37.5°C, C-reactive protein [CRP] level ≥1.0 mg/dl, or white blood count [WBC] ≥10,000/μl), cholestasis that improved on biliary drainage, or positive culture of the bile or blood. The institutional review board of Tokyo Women’s Medical University approved this retrospective study, and the need for patients’ informed consent was waived. All patients underwent multidetector computed tomography (MDCT) with one of three scanners: 4 row, Aquilion
670
(a)
J Hepatobiliary Pancreat Sci (2014) 21:669–675
(c)
(b)
Fig. 1 (a) Diffuse type. Diffuse transient hepatic attenuation differences (THAD) in which the whole liver shows heterogeneous enhancement in a patient with pancreatic head cancer and acute cholangitis (AC) (arrow). (b) Hemi-hepatic type. Hemi-hepatic and diffuse THAD in which the right liver shows slightly stronger enhancement than the left liver without the portal vein obstruction in a patient with perihilar cholangiocarcinoma without AC (arrow). (c) Segmental type. Segmental THAD in which the liver shows segmental enhancement in a patient with pancreatic head cancer without AC (arrow)
4, Toshiba Medical Systems, Tochigi, Japan; 16 row, Aquilion 16, Toshiba Medical Systems; or 16 row, Light Speed Ultra, GE Healthcare, USA. Images of the late arterial, portal venous, and equilibrium phases were obtained at 40, 60 and 110 s, respectively, after the nonionic contrast medium (100–150 ml at rate of 3–4 ml/s) was injected. Omnipaque 300 (Daiichi Sankyo, Tokyo, Japan) and Iopamiron 370 (Bayer, Osaka, Japan) were the contrast material used. Images were reconstructed at a thickness of 5.0 mm. We classified enhanced THAD patterns in the liver parenchyma in the arterial phase as diffuse (Fig. 1a), hemihepatic (Fig. 1b), and segmental (Fig. 1c) types. We classified the degree of intensity of THAD according to two patterns: severe (an obvious difference in intensity) and mild (a mild difference in intensity). We also defined portal vein inflow obstruction as the absence of portal vein flow due to major portal-branch obstruction. The CT findings were interpreted in consensus by one radiologist (SM) and two hepatobiliary surgeons (ST and RH). We studied the diagnostic ability of THAD for AC according to the relationship between patterns, intensity of THAD, and the severity of AC. The severity of AC was diagnosed according to TG13 [2]. Statistical analysis Continuous variables were expressed as means (minimum– maximum values). Associations between continuous variables were analyzed using the Welch 2-sample t-test. Associations among qualitative variables were analyzed using Fisher’s exact test in the R environment (R version 2.15.2; The R Foundation for Statistical Computing, Vienna, Austria) to test the independence of a contingency table. All P-values
