Is salvage Plug-Assisted Retrograde Transvenous Obliteration (PARTO) safe and effective for bleeding gastric varices ?- A preliminary single-center experience.

Background: Vascular plug-assisted retrograde transvenous obliteration (PARTO) obliterates the gastric varices and portosystemic shunt, thus resulting in a lower rebleeding rate than endoscopic glue/sclerotherapy.

Aims: To evaluate the safety and efficacy of PARTO as salvage therapy in liver cirrhosis with gastric variceal bleed (GVB) after failed endotherapy. We assessed the clinical success rate and changes in liver function at 6- months.

Materials And Methods: Patients who underwent salvage PARTO after failed endotherapy for GVB (between December 2021 and November 2022) were searched and analyzed from the hospital database. Clinical success rate and rebleed rate were obtained at six months. Child-Pugh score (CTP) and Model for end-stage liver disease (MELD) score were calculated and compared between baseline and 6-month follow-up.

Results: Fourteen patients (n = 14, Child-Pugh class A/B) underwent salvage PARTO. Nine had GOV-2, and five had IGV-1 varices. The mean shunt diameter was 11.6 ± 1.6 mm. The clinical success rate of PARTO was 100% (no recurrent gastric variceal hemorrhage within six months). No significant deterioration in CTP (6.79 ± 0.98 vs. 6.21 ± 1.52; p = 0.12) and MELD scores (11.5 ± 4.05 vs. 10.21 ± 3.19; p = 0.36) was noted at 6 months. All patients were alive at 6 months. One patient (n = 1, 7.1%) bled from esophageal varices after three days of PARTO and was managed with variceal banding. 21.4% (3/14) patients had progression of esophageal varices at 6 months requiring prophylactic band ligation. Three patients (21.4%) had new onset or worsening ascites and responded to low-dose diuretics therapy.

Conclusions: PARTO is a safe and effective procedure for bleeding gastric varices without any deterioration in liver function even after six months. Patient selection is critical to prevent complications. Further prospective studies with larger sample size are required to validate our findings.