Albumin and magnetic resonance imaging-liver volume to identify hepatitis B-related cirrhosis and esophageal varices.

OBJECTIVE

To investigate whether liver lobe volume and albumin (ALB) could predict the presence and severity of liver cirrhosis, and esophageal varices.

METHODS

Seventy-one cirrhotic patients with hepatitis B and 21 healthy individuals were enrolled in this study. All the participants underwent abdominal enhanced magnetic resonance imaging to measure each liver lobe volume, and biochemical workup for testing ALB and Child-Pugh class. All cirrhotic patients underwent upper gastrointestinal endoscopy to show the presence of cirrhotic esophageal varices. Right liver lobe volume (RV), left medial liver lobe volume (LMV), left lateral liver lobe volume (LLV), and caudate lobe volume (CV) were measured using enhanced magnetic resonance imaging. The ratios of RV to ALB (RV/ALB), LMV to ALB (LMV/ALB), LLV to ALB (LLV/ALB) and CV to ALB (CV/ALB) were calculated. Statistical analyses were performed to determine whether and how the combination of liver lobe volume measured using magnetic resonance imaging and albumin could predict the presence and severity of liver cirrhosis, and the presence of esophageal varices.

RESULTS

RV, LMV, LLV and CV decreased (r = -0.51-0.373; all P < 0.05), while RV/ALB increased (r = 0.424; P < 0.05), with the progress of Child-Pugh class of liver cirrhosis. RV, LMV, CV, LLV/ALB and CV/ALB could identify presence of liver cirrhosis; LLV and LMV could distinguish Child-Pugh class A from B; RV, LMV, LLV, CV, RV/ALB and LLV/ALB could distinguish class A from C; RV and LLV/ALB could differentiate B from C; and RV, RV/ALB and CV/ALB could identify presence of esophageal varices (all P < 0.05). Among these parameters, CV/ALB could best identify the presence of liver cirrhosis, with an area under receiver operating characteristic curve (AUC) of 0.860, a sensitivity of 82.0% and a specificity of 83.0%. LLV could best distinguish class A from B, with an AUC of 0.761, a sensitivity of 74.4% and a specificity of 73.1%. RV could best distinguish class A from C, with an AUC of 0.900, a sensitivity of 90.3% and a specificity of 84.5%. LLV/ALB could best distinguish class B from C, with an AUC of 0.900, a sensitivity of 93.8% and a specificity of 81.5%. RV/ALB could best identify esophageal varices, with an AUC of 0.890, a sensitivity of 80.0% and a specificity of 83.5%.

CONCLUSIONS

The combination of liver lobe volume and ALB has potential to identify presence and severity of cirrhosis, and presence of esophageal varices.