Clinical monitoring and surveillance of hemodialysis arteriovenous grafts to prevent thrombosis

INTRODUCTION — Hemodialysis requires access to blood vessels capable of providing rapid extracorporeal blood flow. These requirements are best met by primary arteriovenous (AV) fistulas and synthetic AV grafts. (See "Approach to the adult patient needing vascular access for chronic hemodialysis".)

Issues surrounding monitoring and surveillance of hemodialysis AV grafts in the attempt to prevent thrombosis and AV graft failure are presented in this topic review. Similar issues for hemodialysis AV fistulas are discussed separately. (See "Clinical monitoring and surveillance of the mature hemodialysis arteriovenous fistula".)

RATIONALE FOR MONITORING AND SURVEILLANCE OF ARTERIOVENOUS GRAFTS — The assessment of hemodialysis arteriovenous (AV) grafts for stenosis is performed using clinical monitoring and noninvasive surveillance methods. Unfortunately, the ability to detect stenotic lesions does not necessarily mean that clinical outcomes are improved, such as a decreased rate of thrombosis and/or improved AV graft survival [1].

More than 90 percent of thrombosed AV grafts have a stenotic lesion, suggesting that such an anatomic abnormality is required for AV graft thrombosis. Noninvasive monitoring methods have a relatively high positive predictive value (PPV) for detecting >50 percent AV graft stenosis. A randomized, controlled trial [2] found that abnormalities detected by clinical monitoring had a 70 percent PPV for hemodynamically significant AV graft stenosis compared with Duplex ultrasound, which had an 80 percent PPV. However, in general, fewer than one-half of AV grafts with >50 percent stenosis are destined to thrombose in the absence of preemptive angioplasty. Any type of program to preemptively identify and treat stenosis necessarily results in a substantial number of superfluous angioplasties. To complicate matters, the vascular injury resulting from angioplasty may actually stimulate neointimal hyperplasia and accelerate AV graft restenosis and potentially graft loss [3].

CLINICAL MONITORING — Clinical monitoring refers to assessments that can be performed by physical examination of the access or by using readily available information that is collected in the course of treating dialysis patients [4]. A number of studies have observed that a hemodynamically significant (>50 percent) stenosis is present in approximately 70 to 90 percent of patients identified to have abnormalities of clinical monitoring [2,5-9]. The major advantages of clinical monitoring are it is free, it does not require additional equipment or personnel, and it can be performed by multiple staff members.

Clinical monitoring falls into the following three broad categories:

●Abnormalities of physical examination, such as absent thrill, discontinuous bruit, or edema in the extremity distal to the AV graft. In one study, an abnormality of physical examination had an 80 percent positive predictive value (PPV) for significant AV graft stenosis [2]. (See "Physical examination of the arteriovenous graft".)

●Problems noted during the dialysis session, such as difficulty with cannulation, aspiration of clots, inability to achieve the target dialysis blood flow, or prolonged bleeding from the needle puncture sites. An abnormality noted during the dialysis session had a 66 percent PPV for significant AV graft stenosis in one study [2]. The PPV was 76 percent for prolonged bleeding from the needle puncture site, 58 percent for difficulty in cannulation, but only 30 percent for aspiration of clots.

●An unexplained decrease in the delivered dialysis dose (Kt/V) on a constant hemodialysis prescription. An unexplained (>0.2 units) decrease in Kt/V had a 69 percent PPV for significant AV graft stenosis in the abovementioned study [2].

SURVEILLANCE — To overcome the limitations of clinical monitoring, a number of noninvasive surveillance methods have been developed. Unlike clinical monitoring, each of these methods requires specialized equipment or specially trained staff. Each method relies on the observation that progressive AV graft stenosis results in an increase in intragraft pressure (normalized for systemic blood pressure) and/or a decrease in access blood flow. Access surveillance is more consistent and reproducible compared with clinical monitoring, but surveillance programs require additional costs that are not reimbursed in the United States. Thus, to justify this added expense, these approaches must demonstrate that their use is associated with improved AV graft patency.

The three main types of access surveillance are [4]:

●Intra-access blood flow monitoring

●Static dialysis venous pressure

●Duplex ultrasound

Flow monitoring is generally favored over static dialysis venous pressure, which is the least used method in the United States because it is highly operator dependent.

Although flow monitoring is not as accurate as duplex ultrasound, information can be obtained at the bedside in the hemodialysis unit, whereas duplex ultrasound is generally performed elsewhere (eg, hospital or outpatient vascular laboratory). Thus, when we elect to use surveillance, we select flow monitoring.

Intra-access blood flow monitoring — Intra-access blood flow can be measured using several different methods. The published literature suggests that all the methods of flow monitoring yield largely similar values. Dialysis access blood flow measurements are typically performed monthly.

Among the available methods, the most commonly used flow-monitoring technique is ultrasound dilution, which quantifies access blood flow by injecting ice-cold saline via the dialysis needle after reversing the dialysis lines [10]. A sensor measures the rate of increase in temperature following the injection and uses this information to calculate the access flow rate. Access flows <600 mL/min or a >25 percent decrease in access flow from the baseline value are suspicious for the presence of hemodynamically significant stenosis. Two prospective studies found that patients with such abnormal access blood flows had an 87 to 100 percent positive predictive value (PPV) for significant stenosis [11,12]. In the Hemodialysis Access Surveillance Evaluation (HASE) study, the per-patient visit thrombotic event rate was reduced in those randomly assigned to monthly AV access (AV grafts or AV fistula) flow measurement compared with standard care alone (0.12/patient versus 0.23/patient) [13]. The rate of and time to first thrombotic event, total number of catheters used, and total number of angiographic procedures was similar between the groups.

When flow monitoring is used, an angiogram should be obtained if the flow is <600 mL/min or >25 percent lower from baseline. However, given the variability of flow measurements, low values should be confirmed before sending a patient for an angiogram.

Other methods for obtaining AV graft blood flow include dilution based on urea or thermal techniques, glucose pump infusion techniques, differential conductivity, duplex ultrasound, and magnetic resonance angiography.

Duplex ultrasonography — Duplex ultrasonography is probably the most reproducible and accurate AV graft surveillance method. Although portable ultrasound units are available for use in a variety of other clinical situations, limitations in the dialysis unit include a lack of staff in the hemodialysis unit trained to use them and lack of reimbursement of the procedure in this setting, at least in the United States.

Duplex ultrasonography measures the peak systolic velocity (PSV) on either side of an area of visual stenosis. The ratio of PSVs measured across the stenotic lesion is calculated. A PSV ratio >2.0 is suspicious for significant stenosis [2,7]. In addition, duplex ultrasound can be used to estimate the access blood flow as corroborating evidence for the PSV ratio. In one prospective study, a PSV ratio >2.0 was associated with a significant stenosis in 80 percent of cases [2].

Static venous dialysis pressure — With static venous dialysis pressure, the intragraft pressure is measured using a manometer connected to the dialysis needle prior to turning on the dialysis pump. The reading is normalized to the systemic systolic blood pressure [14,15]. A ratio of intragraft to systemic pressure <0.4 is considered normal, whereas a ratio >0.6 is suspicious for significant stenosis. Values between 0.4 and 0.6 require repeated measurements for trend analysis. A large, prospective study found that abnormal static dialysis venous pressures had a 92 percent PPV for significant stenosis [14].

EFFECTIVENESS — As discussed above, the majority of thrombosed AV grafts have an underlying stenosis. Detection of significant stenosis in a timely fashion and preemptive treatment could reduce the frequency of AV graft thrombosis, and, since thrombosis is the major cause of graft failure, such an approach could improve AV graft longevity. (See 'Rationale for monitoring and surveillance of arteriovenous grafts' above.)

The bulk of the evidence from observational studies and randomized trials suggests that surveillance with preemptive angioplasty for stenosis does not decrease the frequency of subsequent AV graft thrombosis or improve AV graft patency.

The impact of AV graft surveillance on access outcomes was evaluated in six trials. Surveillance methods included duplex ultrasound in four studies, access flow in two studies, and static dialysis venous pressure in one study (one study included two different surveillance methods) [1,2,11,16-18]. In contrast to the observational studies, none of these trials showed any benefit for AV graft surveillance in reducing the rate of AV graft thrombosis, and only one of these observed an increase in AV graft patency [1,2,11,16-18]. Earlier observational studies had noted an approximately 40 to 80 percent reduction in AV graft thrombosis after introducing monitoring or surveillance protocols that included clinical monitoring [6,8,9,14], static dialysis venous pressures [14], and duplex ultrasound [19].

A meta-analysis of 12 trials found no effect for AV graft surveillance (compared with historic controls) on the risk of AV graft thrombosis (relative risk [RR] 0.94, 95% CI 0.77-1.16; 446 participants) or access loss (RR 1.08, 95% CI 0.83-1.40) [20]. at best, AV graft surveillance with preemptive angioplasty could decrease AV graft thrombosis up to 23 percent, which is a more modest benefit than was noted in the observational studies. However, it could increase AV graft thrombosis by as much as 16 percent.

Possible reasons for these results include:

●Benefits may be limited to relatively new AV grafts. This is supported by an observational study for which the benefit of AV graft surveillance and preemptive angioplasty was greatest in AV grafts less than three months of age [21].

●Thrombosis-free AV graft survival may be most likely observed in AV grafts without a previous intervention [22].

●There may be an insufficient opportunity to noninvasively detect the stenosis due to the rapid progression of the lesion in AV grafts with early thrombosis. This is supported by the 20 to 25 percent false-negative rate observed with AV graft surveillance [23,24].

●Benefits may be very short lived due perhaps to restenosis after angioplasty [3,12,25]. Whether or not stenting or stent-grafting might improve long-term AV graft patency is discussed in detail elsewhere. (See "Techniques for angioplasty of the arteriovenous hemodialysis access".)

OUR APPROACH — Evidence does not suggest that surveillance with preemptive angioplasty has any significant benefit with respect to decreasing the frequency of AV graft thrombosis or long-term patency. Thus, in agreement with the National Kidney Foundation Kidney Disease Outcomes Initiative (KDOQI) Clinical Practice Guideline for Vascular Access: 2019 Update, we rely primarily on clinical monitoring and do not use any AV graft surveillance methods [26].

Our dialysis staff performs routine physical examination of the AV graft weekly, noting any abnormalities during the dialysis session. Abnormalities on physical examination, problems with dialysis, or any unexplained (>0.2 units) persistent decreases in the delivered dialysis dose (Kt/V) prompt an angiogram.

If a hemodialysis unit is not able or willing to do clinical monitoring, the next best option is flow monitoring. The threshold for referral for AV graft is a flow <600 mL/min or >25 percent decrease from baseline. These abnormal values should be confirmed before proceeding to an angiogram.

However, the proficiency and motivation of dialysis staff to perform such clinical monitoring may vary. When clinical monitoring alone is used, suboptimal detection of AV graft stenosis can occur, as evidenced in several randomized trials that found that clinical monitoring alone was associated with an angioplasty rate ranging from 0 to 0.64 events per patient-year, which is low compared with other surveillance methods.

Indications for referral — If a clinician chooses to continue AV graft surveillance in selected dialysis patients, referral for a diagnostic angiography with possible angioplasty should be considered in the following circumstances [27]:

●With all techniques, findings with prospective trend analysis consistent with stenosis rather than a single, isolated value

●Persistent abnormalities with any of the monitoring or surveillance techniques

●AV graft flow rate of <600 mL/min or >25 percent reduction of access flow from the baseline value

●AV graft static venous pressure ratio >0.5

●On duplex ultrasound, an arterial segment static pressure ratio >0.75, a peak systolic velocity (PSV) ratio of >2 across a stenotic lesion

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dialysis" and "Society guideline links: Hemodialysis vascular access".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Preparing for hemodialysis (The Basics)")

SUMMARY AND RECOMMENDATIONS

●More than 90 percent of thrombosed arteriovenous (AV) grafts have a stenotic lesion, suggesting that such an anatomic abnormality is required for AV graft thrombosis. Noninvasive assessment is preferred for efficient screening to identify those AV grafts most likely to be stenotic. Assessment of hemodialysis AV grafts for stenosis is performed using clinical monitoring and noninvasive surveillance. (See 'Clinical monitoring' above.)

●Clinical monitoring includes identification of abnormalities on physical examination, noting problems during the dialysis session, and/or unexplained decreases in the delivered dialysis dose (Kt/V) on a constant hemodialysis prescription. (See 'Clinical monitoring' above.)

●The three main methods of access surveillance are intra-access blood flow monitoring, duplex ultrasound, and static dialysis venous pressure. (See 'Surveillance' above.)

●Evidence suggests that surveillance with preemptive angioplasty for AV graft stenosis does not decrease the frequency of AV graft thrombosis or improve long-term patency. (See 'Effectiveness' above.)

●Our dialysis staff performs physical examination of the AV graft weekly. Abnormalities on physical examination or an unexplained (>0.2 units) persistent decrease in Kt/V prompts an angiogram. If a hemodialysis unit is not able or willing to do clinical monitoring, the next best option is flow monitoring. The threshold for referral for AV graft is flow <600 mL/min or >25 percent decrease from baseline. These abnormal values should be confirmed before proceeding to an angiogram. (See 'Our approach' above.)