Fistulography of a Patent Hemodialysis Access: When Not to Treat and Implications for Establishing a Nontreatment Rate.

PURPOSE: To establish a rate of nonintervention in patients referred for hemodialysis access fistulography and to report clinical and fistulographic variables associated with nonintervention. MATERIALS AND METHODS: Encounters for fistulography were reviewed from 2001 to 2016 to determine annual rates of nontreatment over 15 years. Next, an access database was used to retrospectively identify patients undergoing fistulography from 2010 to 2016. Patients who underwent fistulography without intervention (angioplasty or stent placement) served as the nontreatment group (NTG; n = 76). Patients who underwent fistulography with intervention served as the control group (CG; n = 77). Patients with thrombosed accesses were excluded. Clinical indications for intervention and physical examination findings were correlated with fistulography. Need for subsequent percutaneous intervention was recorded. RESULTS: Annual nontreatment rates ranged from 3% to 14% (median, 10%). Preprocedure thrill was encountered in 45 patients in the NTG (59%) vs 6 in the CG (7.8%; P < .01). Aneurysm as indication for fistulography was more common in the NTG than the CG (19 [25%] vs 4 [5%]; P < .01). The NTG had a higher proportion of aneurysms noted on fistulography as well (38 [50%] vs 19 [25%]; P < .01). The CG had a higher proportion of patients needing subsequent percutaneous intervention vs the NTG (73 [96%] vs 38 [50%]; P < .001). CONCLUSIONS: A suggested nonintervention rate for hemodialysis access fistulography is 10%. Patients in the NTG were more likely to have a thrill on physical examination or to present with aneurysms as the clinical indicator. NTG patients were less likely to require subsequent percutaneous intervention.

Unmasking of Previously Asymptomatic Central Venous Stenosis following Percutaneous Transluminal Angioplasty of Hemodialysis Access.

PURPOSE: To determine the frequency of new-onset symptoms of central venous stenosis (CVS) after percutaneous transluminal angioplasty (PTA) of a hemodialysis access-related stenosis in patients with previously asymptomatic CVS and to identify risk factors for this phenomenon. MATERIALS AND METHODS: Retrospective review was performed of patients treated with PTA for an access-related stenosis (excluding central vein interventions) between 2001 and 2016 who returned within 3 months with symptoms of CVS (ie, "unmasking"): 39 patients met these criteria. A control group of 122 patients who had untreated asymptomatic CVS and did not experience unmasking was selected. Fistulograms were graded for degree of CVS. A total of 51% of the unmasked group was male, with an average age of 65 years; 57% of the control group was male, with an average age of 63 years. RESULTS: The incidence of unmasking among patients with untreated asymptomatic CVS was 4.9%. A total of 90% of the unmasked group (35 of 39) had upper-arm access, compared with 77% of the control group (94 of 122; P = .017). A total of 28% of unmasked-group patients (11 of 39) underwent thrombectomy, vs 4% of controls (5 of 122; P < .0001). A total of 54% of unmasked-group patients (21 of 39) had significant brachiocephalic vein stenosis, vs 26% of controls (32 of 122; P = .001). A total of 8% of unmasked-group patients (3 of 39) had superior vena cava stenosis, vs none of the 122 controls (P = .01). A total of 64% of unmasked-group patients (25 of 39) had extensive collateral vessels, vs 24% of controls (29 of 122; P < .0001). CONCLUSIONS: The incidence of unmasking of asymptomatic CVS is low. Prophylactic treatment of asymptomatic CVS therefore remains generally inadvisable. However, patients undergoing declotting with extensive collateral vessels might warrant treatment of asymptomatic CVS.

Central venous stenosis is more often symptomatic in hemodialysis patients with grafts compared with fistulas.

PURPOSE: To determine whether hemodialysis patients with central venous stenosis (CVS) are more frequently symptomatic if they have grafts versus fistulas. MATERIALS AND METHODS: A retrospective review was performed of 500 consecutive discrete patients, half with fistulas and half with grafts, who had fistulograms performed over a 4-year period. All fistulograms were evaluated for CVS, which was graded into quartiles. The presence of collaterals was noted and graded. Patient records were analyzed for symptoms of CVS, including face, neck, breast, or limb swelling. Statistical analysis was performed to determine the association between access type, degree of stenosis, location of stenosis, and symptoms. RESULTS: Of 500 fistulograms, 31 were excluded because of inadequate or absent central imaging. Of the remaining 469 patients, 235 had fistulas and 234 had grafts. CVS was present in 51% of patients with fistulas (119 of 237) and 51% of patients with grafts (118 of 237). When CVS was present, 29% (35 of 119) of patients with fistulas were symptomatic versus 52% (62 of 118) of patients with grafts (P = .0005). Overall, only 15% of patients with fistulas in the entire cohort were symptomatic compared with 27% of patients with grafts (P = .002). Sex, access side, and transposition did not influence symptoms; however, patients with upper arm access were more likely than patients with forearm access to be symptomatic (P < .0001), independent of access type. CONCLUSIONS: CVS is more likely to be symptomatic in patients with grafts versus fistulas, and patients with upper arm access are more likely than patients with forearm access to be symptomatic.