6 - Demographical and Procedural Factors Impacting Patency after Percutaneous Thrombectomies for Clotted Dialysis Access

Purpose: The aim of this study was to identify demographical, morphologic, and procedural factors associated with increased patency after percutaneous thrombectomy for patients presenting with clotted hemodialysis fistulas and grafts.

Materials and Methods: Retrospective chart review was completed on 99 patients who underwent 250 declot procedures between August 2019 and August 2024 at a safety net hospital. Patency was defined as the number of days between the first thrombectomy procedure and any subsequent dialysis access related procedure, including repeat thrombectomy, fistulagram, placement of a temporary or tunneled catheter, and abandonment of access due to failure of thrombolysis. Morphological data collected included access type and access location. Demographical information collected included age, sex, gender, race, BMI, preferred language, local zip code, type of insurance, and housing status. Procedure related data included type of intervention, length of procedure, adverse outcomes, radiation dose, and total number of procedures. Interventions were grouped into pharmacomechanical, i.e. balloon maceration with intraluminal alteplase, or rheolytic, named Jeti or AngioJet. Generalized linear model was used to determine impact of these factors on primary patency. P-value of < 0.05 was considered statistically significant.

Results: The final cohort involved 37 patients who underwent 134 thrombectomies with a mean age of 58 and female to male ratio of 11:26. The median patency was 45 days and patients underwent a median of 1 fistulagram and 3 thrombectomies. Pharmacomechanical was the most recurrent intervention (18, 50%) and brachioaxillary graft the most common access location (16, 42%). Black/African American was the most common race (15, 41%). Multivariate analysis showed brachiocephalic fistula (p = 0.001) and pharmacomechanical intervention (p = 0.05) were independent predictors of longer patency after thrombectomy. Zip codes including 94102 (p = 0.05), 94107 (p = 0.02) and 94110 (p = 0.05) were also independently associated with longer primary assisted patency.  

Conclusion: The use of pharmacomechanical intervention and brachiocephalic location of the fistula were associated with longer patency after a declot procedure. These findings suggest that the use of pharmacomechanical thrombolysis may offer a patency advantage over rheolytic methods and therefore result in more intervention-free days for these complex patients. Further research examining the use of purely mechanical thrombectomy methods in this population is needed.