Impact of a preoperative evaluation on the outcomes of an arteriovenous fistula

Autogenous arteriovenous fistulas (AVFs) are the preferred vascular access for chronic hemodialysis because of better outcomes, a lower complication rate once matured, and reduced costs compared with prosthetic arteriovenous grafts (AVGs) or central venous catheters [1, 2, 3, 4, 5, 6]. Nevertheless, their primary failure rates, reported to be between 10% and 50%, are quite high due to maturation failure and stenotic complications [7, 8, 9]. Several preoperative factors have been shown to predict the risk of primary AVF failure, primarily the diameters of the artery and vein. Comorbidities associated with primary AVF failure include advanced age, diabetes mellitus, and systemic atherosclerosis [10, 11]. As part of the preoperative planning, duplex ultrasound vascular mapping to assess anatomical suitability is recommended before vascular access creation for the accurate measurement of vessel diameter, and its routine use can increase the placement of an autogenous vascular access and the proportion of patients undergoing dialysis with an AVF [5, 12]. Although duplex ultrasound plays an integral part in both the preoperative planning of AVFs and their subsequent evaluation [13], it remains to be established whether routine use of preoperative duplex ultrasound can improve AVF outcomes [14].

The aim of our present retrospective single-center study was to determine the possible predictors of primary AVF failure and examine the impact of preoperative evaluation on AVF outcomes in chronic kidney disease patients receiving hemodialysis.

Of the 539 chronic kidney disease patients in our current study series, physical examination alone was deemed satisfactory for AVF creation in 351 patients (65.1%). In the remaining 188 patients (34.9%) with insufficient physical examination findings, additional duplex ultrasound was performed; in these patients, initial placement of an AVF was proposed in 118 patients, whereas an AVG was proposed in 70 patients (Fig. 1). Thus, in response to physical examination and additional duplex ultrasound findings, AVF creation was preoperatively proposed in 469 patients (87.0%). The demographics, characteristics, and AVF outcomes of these patients are summarized in Table 1. Because we used different inclusion criteria for AVF creation and did not adjust for baseline differences between patients with physical examination alone and patients with additional duplex ultrasound, the percentages of female gender and diabetes mellitus were significantly higher in patients assessed by duplex ultrasound (P = 0.014 and P = 0.002, respectively) and the rate of AVF creation on the most distal site was significantly higher in patients assessed by physical examination alone (P = 0.001). Although the primary failure rate was significantly higher in patients assessed by duplex ultrasound (P = 0.001), ultrasound information changed the clinical plan, increasing AVF use for dialysis, in 92 of the 188 patients (48.9%) with insufficient physical examination findings (Fig. 1).

Table 1
Demographics, clinical characteristics, and outcomes in chronic kidney disease patients undergoing AVF creation

• Click for larger image

Female gender and diabetes mellitus were the risk factors significantly associated with primary AVF failure in univariate (P = 0.001 and P = 0.006, respectively) and multivariate (P = 0.001 and P = 0.028, respectively) analyses (Table 2). The site of AVF creation and insufficient physical examination findings were not associated with primary AVF failure (Table 2). Kaplan-Meier survival analysis showed that patients assessed by physical examination alone had significantly higher functional primary patency (P = 0.002) and AVF survival rates (P = 0.001) than patients assessed by duplex ultrasound (Fig. 2). The cumulative functional primary patency rates at 1, 2, 3, and 4 years were 84.2%, 79.6%, 73.2%, and 68.6%, respectively, in patients assessed by physical examination alone and 72.7%, 68.8%, 57.3%, and 53.8%, respectively, in patients assessed by duplex ultrasound. The cumulative AVF survival rates at 1, 2, 3, and 4 years were 89.0%, 86.3%, 83.5%, and 82.2%, respectively, in patients assessed by physical examination alone and 78.2%, 77.2%, 70.7%, and 67.3%, respectively, in patients assessed by duplex ultrasound. In the Cox analysis, after adjustment for all independent variables, insufficient physical examination findings (i.e., need for additional duplex ultrasound) was the only risk factor significantly associated with both functional primary patency and AVF survival, whereas diabetes mellitus was significantly associated with functional primary patency and female gender was significantly associated with AVF survival (Tables 3, 4).

Fig. 2
Kaplan-Meier survival analysis of patients undergoing AVF creation (including maturation failure). Cumulative functional primary patency (A) and cumulative AVF survival (B) between patients assessed by physical examination alone and patients additionally assessed by duplex ultrasound. AVF, arteriovenous fistula.

• Click for larger image

Table 2
Factors associated with the risk of primary AVF failure

• Click for larger image

Table 3
Factors associated with the risk of functional primary patency

• Click for larger image

Table 4
Factors associated with the risk of AVF survival

• Click for larger image

The long-term patency of a functioning vascular access is extremely important for quality of life and longevity in chronic kidney disease patients receiving hemodialysis. The AVF is the preferred access for hemodialysis [12]. However, AVF failure has become more common because patient age is increasing and more patients have diabetes mellitus or vascular disease [14]. It is well recognized that AVFs as the access for hemodialysis are less prevalent among older patients and female patients and those with obesity, diabetes mellitus, or cardiovascular disease [18]. In addition to these preoperative risk factors, AVF failure has been attributed to the use of inadequate vessels for surgery, and the guidelines of the Kidney Disease Outcomes Quality Initiative recommend routine use of ultrasound for mapping in all AVF patients while acknowledging the lack of level I evidence to support this recommendation [12].

Physical examination usually yields more information from venous than arterial assessment, whereas duplex ultrasound gives somewhat more information about the artery than the vein. Physical examination alone is insufficient in a considerable proportion of patients, and ultrasound information changes the clinical plan in a substantial number of patients who could receive an AVF rather than an AVG. In the current study, suitability for AVF creation was initially determined by physical examination. However, in patients for whom physical examination alone was not sufficient to evaluate suitable sites for AVF creation due to patient habitus and compromised vasculature, additional duplex ultrasound was performed to increase AVF use in the dialysis and to reduce unnecessary exploration; we found that physical examination alone was insufficient in 34.9% and that duplex ultrasound made a relevant contribution to 48.9% of these patients.

Duplex ultrasound is a noninvasive, effective, and safe method for establishing morphologic and functional parameters or characteristics of vessels that could help surgeons to improve AVF maturation rates [19]. It can identify suitable veins that cannot be determined by physical examination alone [19]. In recent years, some authors have recommended the routine use of duplex ultrasound prior to AVF creation. Others, however, have found no evidence that duplex ultrasound decreases AVF failure rates, providing evidence that selected patients with insufficient physical examination benefit from duplex ultrasound while duplex ultrasound is not needed when adequate vessels are defined by physical examination [15, 20]. There is no general consensus on the role of duplex ultrasound prior to AVF creation, and controversy persists regarding the optimal criteria of a suitable site for AVF on duplex ultrasound. Although some authors reported that vein size is the major predictor of a successful AVF [21, 22], the optimal preoperative venous diameter for AVF creation has not been validated in prospective randomized trials. In our present study, we used a different cutoff of minimum venous diameter proposed for initial placement of an AVF between patients assessed by physical examination alone and patients assessed by additional duplex ultrasound. We remained aggressive in attempting to place an AVF in all suitable patients, and additional duplex ultrasound was performed in patients with unsatisfactory physical examination, using a minimum external venous diameter of 1.6 mm without a tourniquet as a suitable site for AVF creation. This difference resulted in a higher primary failure rate in patients assessed by duplex ultrasound, but ultrasound information changed the clinical plan, increasing AVF use for dialysis, in patients with unsatisfactory physical examination findings.

Different outcomes of AVF have been used in previous studies examining the effect of preoperative evaluation [14]. In our current study, we evaluated AVF outcomes using primary AVF failure, functional primary patency, and AVF survival. Primary AVF failure considers all AVFs that remain inadequate for dialysis after creation [16]. Because primary failure can take account of AVFs in patients not already on dialysis, it is probably the best measure for assessing the effect of preoperative evaluation on AVF outcomes [14]. The aim of our present study was to examine the impact of preoperative evaluation on AVF outcomes in patients who require an AVF. Therefore, we modified the definition of primary AVF failure to an actual site of vascular access creation different from the planned site assessed by preoperative physical examination or duplex ultrasound, in addition to all AVFs remaining inadequate for dialysis after creation, and found that female gender and diabetes mellitus were significantly associated with primary AVF failure.

We acknowledge that our study had several limitations. Although there are definite criteria of anatomical suitability for AVF creation in our institution, this was a retrospective review that relied on available patient information and it was thus vulnerable to insufficient data availability and accuracy. Patients were selected for initial placement of an AVF according to physical examination findings and additional use of duplex ultrasound was also determined based on initial physical examination findings, suggesting a possible selection bias. Furthermore, our definitions of AVF outcomes were limited to our clinical assessment after an AVF placement. Finally, our current findings were obtained from at a single center, resulting in a small sample size and a short study period that limited the overall relevance of our results.

Despite these potential limitations, we found that chronic kidney disease patients assessed by physical examination alone had significantly better AVF outcomes, and insufficient physical examination, as judged by the need for additional duplex ultrasound, was the only risk factor significantly associated with both functional primary patency and AVF survival. Duplex ultrasound information could increase AVF use in dialysis and reduce unnecessary exploration in such patients with insufficient physical examination findings.

In conclusion, routine use of duplex ultrasound is not necessary in chronic kidney disease patients with satisfactory physical examination findings. Given that female gender and diabetes mellitus were significantly associated with primary AVF failure, duplex ultrasound could be of particular benefit in these specific patients who have an insufficient physical examination. Future prospective trials with larger cohorts will lead to a better understanding of the role of preoperative duplex ultrasound in chronic kidney disease patients receiving hemodialysis.