OBJECTIVE: Liver transplantation is performed with increasing success and frequency all over the world. Experience with MRI of the liver allograft is, however, limited. This study was designed to correlate MRI to clinical-laboratory findings, CT, and biopsy and to evaluate the significance of the periportal collar on MRI. MATERIALS AND METHODS: Fourteen patients who had undergone orthotopic liver transplantation were studied by CT and MRI [T1-weighted imaging: gradient-echo, repetition time/echo time (TR/TE) 306/14 ms, theta 90 degrees; proton density and T2-weighted imaging: spin-echo, TR/TE 1,600/30-120 ms]. Three patients also had follow-up MR examinations 43 days, 89 days, and 5 months after transplantation. RESULTS: Magnetic resonance imaging demonstrated a perivascular collar around central portal venous branches in all 14 patients and around peripheral portal branches in 10 of the 14 patients on the initial MRI study. The perivascular collar showed low signal intensity on T1-weighted imaging and an increase in signal intensity on T2-weighted multiecho imaging. The distribution and prevalence of central and peripheral periportal collars were identical on MR and CT. Peripheral periportal collars were seen in 9 patients who had no clinical-laboratory signs of rejection. In 3 patients with biopsy-proved rejection, the periportal collar was less prominent on MR at the time of rejection when compared with MR performed when the patient had no signs of transplant rejection. CONCLUSION: A perivascular collar in a patient with liver transplantation is likely to be related to impaired lymph drainage after surgical interruption of the draining lymph vessels and lymphedema. In contrast to previous CT reports, however, a perivascular collar around peripheral portal branches does not appear to correlate to rejection, since it is frequently observed in the normal liver allograft.
OBJECTIVE: Liver transplantation is performed with increasing success and frequency all over the world. Experience with MRI of the liver allograft is, however, limited. This study was designed to correlate MRI to clinical-laboratory findings, CT, and biopsy and to evaluate the significance of the periportal collar on MRI. MATERIALS AND METHODS: Fourteen patients who had undergone orthotopic liver transplantation were studied by CT and MRI [T1-weighted imaging: gradient-echo, repetition time/echo time (TR/TE) 306/14 ms, theta 90 degrees; proton density and T2-weighted imaging: spin-echo, TR/TE 1,600/30-120 ms]. Three patients also had follow-up MR examinations 43 days, 89 days, and 5 months after transplantation. RESULTS: Magnetic resonance imaging demonstrated a perivascular collar around central portal venous branches in all 14 patients and around peripheral portal branches in 10 of the 14 patients on the initial MRI study. The perivascular collar showed low signal intensity on T1-weighted imaging and an increase in signal intensity on T2-weighted multiecho imaging. The distribution and prevalence of central and peripheral periportal collars were identical on MR and CT. Peripheral periportal collars were seen in 9 patients who had no clinical-laboratory signs of rejection. In 3 patients with biopsy-proved rejection, the periportal collar was less prominent on MR at the time of rejection when compared with MR performed when the patient had no signs of transplant rejection. CONCLUSION: A perivascular collar in a patient with liver transplantation is likely to be related to impaired lymph drainage after surgical interruption of the draining lymph vessels and lymphedema. In contrast to previous CT reports, however, a perivascular collar around peripheral portal branches does not appear to correlate to rejection, since it is frequently observed in the normal liver allograft.