Background: Confluent porto-mesenteric thrombosis (PMT) is a complex condition which can result in catastrophic variceal bleeding or refractory ascites secondary to portal hypertension (PHT). Management of PHT is challenging and often requires a multivisceral transplant (MVT).  While short and medium term outcomes of MVT have improved, long term outcomes remain suboptimal.
We report a single center experience of non-transplant management of confluent PMT, with specific focus on results of adopting a multi-D approach since 2012. End points were patient survival and re-bleeds.

Method: Retrospective review of patients with PMT in a single center over a 10-year period. A dedicated multi-D team comprising of hepatology, transplant surgery, hematology and interventional radiology (IR) was developed over the last 5 years. All imaging was reviewed by TX surgery and IR to confirm confluent PMT. Patients who underwent MVT, non-confluent PMT or isolated superior mesenteric vein (SMV) thrombosis were excluded for this analysis.

Results: Twenty patients met inclusion criteria (Table 1). 7 patients underwent ‘modified’ single stage Sugiura procedures (key Table1) and 13 had IR procedures for revascularization with additional thrombectomy in acute situations. 3 patients had percutaneous mesocaval shunts, while the other 10 had extended TIPS procedures. 3 patients re-bled after percutaneous procedures and 1 after a Sugiura. 7 of the 13 patients managed percutaneously had at least one re-intervention for complications. Reintervention was required in 7 out of 13 patients who had a percutaneous approach and in 1 patient post-Sugiura. 4 (2 surgical and 2 IR) of the 20 patients have died. Causes of death were: multiorgan failure in 2 patients with refractory re-bleed (1 acute, with intra-peritoneal bleeding post extended TIPS) and in 1, 9 months after a failed Sugiura. 1 patient had an infected TIPS and recurrent fungemias and 1 death was late after a Sugiura from unrelated cardiac causes.

Conclusions: A multi-D approach may allow some patients with confluent PMT  to avoid MVT.  The considerable morbidity and even mortality in our experience emphasizes the need for very careful patient selection by an experienced team; while no patient in this cohort has required MVT, we believe rapid access to MVT may be a critical component of this approach. Despite the need for reinterventions in some patients, the avoidance of MVT and overall outcomes appear to justify our approach.