Diagnosis of Sinusoidal Obstruction Syndrome/Veno-Occlusive Disease of the Liver: Problems of Interpretation

Introduction: The reported incidences of sinusoidal obstruction syndrome/veno-occlusive disease of the liver (VOD) have varied widely. In recent reviews the present incidence has been stated to be up to 10-15 %. The differences in the incidences may be due to different patient materials and transplantation methods or differences in the application and interpretation of diagnostic criteria. Modified Seattle criteria (McDonald et al 1993) or Baltimore criteria (Jones et al 1986) are used to establish the diagnosis. The diagnosis is based on a combination of clinical criteria. However, the diagnosis of VOD should only be made if there are no other conditions that could explain the clinical findings. More than one condition causing liver problems may be present at the same time, leaving space for subjective interpretation that could essentially affect the incidence of this disorder. Along with the improving opportunities for prophylaxis and treatment of VOD, it is important to have a correctpicture of its incidence. We evaluated the differential diagnostic component of VOD diagnosis in our single-center material.

Patients: The hospital notes of 300 consecutive adult patients treated with allogeneic stem cell transplantation between March 2011 and February 2015 at the Helsinki University Hospital were retrospectively reviewed. All but one had a malignant hematological disease. The most common diagnoses were AML 106, ALL 48, MM 46, NHL 34, and MDS 24. The total of 135 patients (45%) received conventional myeloabative conditioning, CyTBI (n=79) or BuCy (n=56), 151 were given treosulfan in myeloablative (n=77) or reduced (n=74) dose with fludarabine, 5 received BuFlu and 9 other regimen. Of the donors 75 % were unrelated, and 77 % of the grafts were from peripheral blood. All patients received ursodeoxycholic acid (UDCA) prophylaxis. GvHD prophylaxis consisted of CsA and a short course of MTX. ATG was given in transplantations from unrelated donor.

Methods: We identified all patients who had a total serum bilirubin concentration exceeding 34 µmol/l during the first three weeks after the transplantation. Among them we evaluated whether they had a weight increase of more than 2 % or 5 % compared to the admission weight. We also registered the clinical notes of liver enlargement, upper right abdomen pain and ascites as well as ultrasound examinations.

Results: Of the 300 patients, 54 had a serum bilirubin concentration exceeding 34 µmol/l within three weeks post-transplantation. The peak bilirubin concentration was 35-193 (median 47) µmol/l. Of these patients, 43 showed a weight gain exceeding 2 % and 29 patients exceeding 5 % of the admission weight. Thus, forty-three patients fulfilled the minimum modified Seattle criteria with the 2 % weight gain criterion and 29 with the 5 % criterion (modification, Corbacioglu et al 2012). Two patients had enlarged liver and right upper quadrant pain. No case of abnormal venous blood flow was seen. According to institutional policy, preemptive treatment with defibrotide according to bilirubin levels was started in 10 patients but in most cases discontinued within a few days based on the clinical course. Twenty-eight (65 %) of the 43 patients fulfilling the minimal VOD criteria had other clinical conditions that were regarded by the treating team as a plausible cause for the findings: infection 22, non-conditioning drug effect 4, cholecystitis or cholestasis 2, engraftment syndrome 2, GvHD 1, and hemolysis 1. In the remaining 15 patients there was no obvious alternative cause. Thirteen of these cases were mild, showing no hepatomegaly, and they had not been labelled as VOD.

Conclusion: According to the modified Seattle criteria, 15 patients (2% weight criterion) or 10 patients (5 % weight criterion), without an alternative cause to the clinical findings, had VOD. Two patients fulfilled the Baltimore criteria, but as one of them had a simultaneous septic infection, the diagnosis of VOD could not be confirmed. Therefore, depending on the criteria used and the differential diagnostic interpretations, the incidence of VOD in this material was between 5 % and 0.3 %. All cases were mild. UDCA prophylaxis and preemptive defibrotide treatment probably affected the incidence. Differences in differential diagnostic interpretations may contribute significantly to the variable incidences of VOD reported.