Graft versus host disease (GVHD) is a major complication of allogeneic bone marrow transplantation (BMT), resulting in death in the majority of steroid resistant patients. The study was designed to assess the efficacy of regional intra-arterial treatment in patients with resistant hepatic and/or gastro-intestinal GVHD.

Graft versus host disease (GVHD) is the most ominous side effect of allogeneic bone marrow or blood stem cell transplantation (BMT). GVHD causes a severe inflammatory process, which affects primarily the skin and the GI) system, including the liver. Pharmacological treatment of GVHD includes various immunosuppressive and immune-modulating drugs, including steroids, cyclosporin, tacrolimus, methotrexate and anti-lymphocyte agents. These agents can aggravate immunologic incompetence, exposing the patient to infections and secondary malignancy, as well as reducing the efficacy of graft versus leukemia / graft versus tumor (GVL / GVT) effects induced by alloreactive donor lymphocytes. In spite of this effect on the immune system, only 50-70% of the patients achieve partial control of GVHD, which can rapidly deteriorate and result in death. Despite the use of innovative immunosuppressive modalities, the prognosis of steroid resistant GVHD is usually poor. GVHD can be mostly localized to a specific organ (skin, liver, GI) with liver involvement tending to be more chronic and resistant. The bile ducts are the main target for GVHD, with their blood supply originating from the hepatic artery. This may explain the ineffective nature of oral steroids absorbed through the gut into the portal vein.

Sato et al reported that an infusion of steroids to the superior and inferior mesenteric arteries induced remission in a patient with refractory GI GVHD. This method has never been tested in patients with fulminant liver GVHD. We decided to test this approach in a larger cohort of patients with steroid resistant hepatic and intestinal GVHD.

Patients were eligible for inclusion if they developed grade 3-4 hepatic and/or GI GVHD unresponsive to treatment with IV cyclosporin 3 mg/kg and IV methylprednisolone (MP) 2 mg/kg. GVHD diagnosis was based upon clinical criteria and in some of the patients supported by biopsy. GVHD grading was graded according to the Seattle severity index.

Patients with hepatic GVHD were first treated with slow intra-arterial (hepatic artery) infusion of methotrexate (Pharmachemie, Netherland) (10mg/m2) and MP (Pharmacia and Upjohn, Belgium) (75mg/m2). The protocol was later on changed with the exclusion of methotrexate and increase of MP dose to 1000mg.

Patients with GI GVHD were treated with intra-arterial (SMA, IMA) infusion of MP (40-60 mg/vessel). In cases of upper GI GVHD intra-arterial infusion of MP was given to the gastro-duodenal artery (GDA). Later on, injections to the internal iliacs were added.

Technique Angiography was performed using standard sterile technique, local anesthesia with lidocaine 1% and conscious intravenous sedation. Visceral arteriography was performed in order to determine anatomy and identify variant arterial blood supply to the gut. A selection of 3 - 5 Fr. angiographic catheters was used at the discretion of the angiographer performing the procedure. The most frequently used catheters were 4 Fr. & 5 Fr. Cobra 2 and Rim catheters (Cook Inc., Bloomington, In.). Other catheters used included the Sos 2 (Angiodynamics Inc., Queensbury, NY.) and the 3Fr. Terumo SP co-axial catheter (Terumo Europe, N.V., Leuven, Belgium.). Once the catheter was in position, each drug was injected into the artery over approximately three minutes. After completion of the procedure, the catheter was removed and direct manual pressure over the arteriotomy maintained until hemostasis was achieved. Patients remained on complete bed rest with the involved extremity extended for 6 hours, in accordance with standard angiographic practice. Compression devices were not used.

Definitions Hepatic response: initial response - the day in which bilirubin level began to decrease, partial response - the day in which bilirubin level decreased below 70% of basal level, complete response - the day in which bilirubin level decreased below 30% of basal level.

GI response: initial response - the day in which symptom amelioration appeared (diarrhea volume and abdominal pain). Complete response - the day in which symptoms resolved.

Inclusion Criteria:

Recipient of allogeneic stem cell transplantation. AGVHD of liver and or GI tract, grade 2-4. Resistant AGVHD. Signed a written informed consent.

Exclusion Criteria:

Not fulfilling any of the inclusion criteria. Active life-threatening infection. Inability to comply with study requirements. Contraindication to arterial catheterization. Has been diagnosed with multi organ failure.