Abstract
Background:
This study aimed to evaluate patency outcomes of arteriovenous grafts (AVGs) before and after using Duplex doppler ultrasonography (DUS) in preoperative mapping and surveillance of AVG.
Methods:
In this single-centre, retrospective cohort study 212 patients receiving AVGs from January 2009 to December 2022 were included. In group 1, the creation of AVG as well as screening was based on physical examination alone. In contrast, DUS was used in the preoperative mapping and surveillance of AVG in group 2. The patients also received sulodexide as supplemental medication. Outcomes included primary and secondary patency. The Mann-Whitney U-test was used to compare the differences between groups in number of thrombectomies and preemptive percutaneous transluminal angioplasties (PTAs).
Results:
Group 1 included 90 AVGs. The mean follow-up time was 333 days (range: 1–1230 days, standard deviation: 318 days). The primary and secondary graft patency rates were 13.3%, 62.2% at 6 months; 2.2%, 52.1% at 12 months; 0%, 44.3% at 24 months and 0%, 44.3% at 36 months respectively. During the 7-year surveillance of AVG, significantly more thrombectomies were performed than preemptive PTA (p < 0.0001). Group 2 included 122 AVGs. The mean follow-up time was 584 days (range: 1–2040 days, standard deviation: 463 days). The primary and secondary graft patency rates were 54.9%, 95.9% at 6 months; 29.5%, 77.8% at 12 months; and 9.8%, 56.5% at 24 months; 2.5%, 47.1% at 36 months respectively. The primary and secondary graft patency was significantly longer (p < 0.0001, p = 0.002). During the 7-year surveillance of AVG there were significantly more preemptive PTAs performed (p = 0.0004).
Conclusions:
The primary and secondary patency of AVG were significantly improved after using DUS in preoperative mapping and surveillance. DUS surveillance led to a decrease in AVG occlusion. A potential positive effect of sulodexide on patency rate of AVG needs more research.
Introduction
The role of arteriovenous graft (AVG) is increasing among the dialysis population, mainly in ageing patients. It is due to a higher primary failure rate in arteriovenous fistula (AVF) in elderly patients,1,2 whose upper limb superficial veins are often poor. However, grafts are also used to correct long-term functional autologous access and its complications. Unfortunately, a disadvantage of AVG is access graft survival. 3 The lifespan of the AVG is limited mainly by the development of vascular access stenosis due to neointimal hyperplasia (NIH), 4 which is mostly located in the venous anastomosis. 5 The failure of the vascular access due to progressive vascular access stenosis significantly contributes to patient morbidity and costs to the health care system. 6 A timely revealed AVG stenosis followed by preemptive angioplasty might protect the graft from occlusion. Therefore several surveillance methods are used for the early detection of AVG stenosis. Owing to these facts we have decided to retrospectively investigate the AVG survival rate at the Vascular Access Center Olomouc, Czech Republic and compare the AVG survival rate before and after the development of local standards for Duplex doppler ultrasonography (DUS) surveillance and adjuvant medication supplemented by sulodexide (SDX) and analyse their impact on the AVG survival rate.
Materials and methods
Study design
Data regarding all AVGs created at University Hospital Olomouc, Czech Republic in years 2009–2022 were analysed. The patients were divided into two groups based on different preoperative examination and surveillance of AVGs.
Group 1
Group 1 included patients with an AVG created in 2009–2015. The patients underwent only physical examination alone before AVG creation. Neither DUS vessel mapping nor phlebography of the upper limbs were performed. After the creation of AVG the patients were screened by clinical monitoring based on physical examination (inspection, palpation and auscultation) of the AVG once a month to detect signs that suggest the presence of pathology. DUS surveillance was not performed. Adjuvant medication included aspirin at a dose of 100 mg taken orally once a day.
Group 2
Group 2 included patients with an AVG created in 2016–2022. The patients underwent physical examination before AVG creation, furthermore DUS vessel mapping of the upper limbs was performed. Phlebography of the upper limbs was also performed to rule out central vein stenosis. In 2016 our department became a Vascular Access Center and local standards for DUS surveillance of AVG were developed. The patients, in addition to traditional screening, were examined by DUS accordance to the local DUS protocol. DUS examination of the AVG was performed before the first cannulation, then in a month and then every 3 months in each patient. Adjuvant medication included aspirin at a dose of 100 mg together with sulodexide (SDX) at a dose of 250 lipasemic units (LSU) taken two times a day after graft placement until definitive graft failure, both taken orally.
When there was suspicion of significant stenosis based on physical examination in group 1 and DUS findings in group 2, AVG fistulography was indicated. If the stenosis was significant, percutaneous transluminal angioplasty (PTA) was performed. However, if the stenosis was resistant to PTA or in the presence of early restenosis, stentgraft placement was indicated. AVG occlusion was treated by thrombectomy, always in hybrid mode in the operating room with digital subtraction angiography (DSA) performed under the mobile C-arm, with suitable treatment of stenosis – PTA alone or together with stentgraft placement or resection of the venous anastomosis. Since 2012, AVG thrombectomies have been performed in the hybrid operating room at our centre.
The cause of AVG dysfunction was recorded in both studied groups and included stenosis of venous anastomosis, outflow vein, central vein, arterial anastomosis, mural thrombosis in the body of AVG and other general factors as hypotension, coagulopathy and excessive compression after haemodialysis.
From the studied groups of 262 AVGs, patients taking different antiplatelet and anticoagulant medication (n = 15) were excluded so as to compare patients with the same medication affecting AVG patency. Other excluded patients included those dialysed through a central venous catheter instead of a functional AVG (n = 9) due to eventual better AVG patency resulting from avoiding puncture zone stenosis. Patients with an AVG infection (n = 35) were also excluded due to a different etiology of vascular access loss.
In our study we set two end points for evaluating AVG patency. The first was the date of any intervention aiming to restore or maintain blood flow and the second was the date of definitive failure. From these two end points, the primary and secondary patency were calculated. During the 7-year surveillance of AVG we also recorded a number of early detected stenoses followed by preemptive percutaneous transluminal angioplasty (PTA) and thrombotic events with sequential thrombectomy of the AVG in both groups. If there was no definitive failure of the AVG, patient status was obtained from medical records at the most recent follow-up.
Ultrasound examination
Since 2016 the GE Logiq V2 ultrasound system has been used for AVG ultrasonographical surveillance as well as in preoperative vessel mapping. Regular DUS examination included access flow volume (Qa) measurement (directly in the graft), the inflow artery, inflow anastomosis, graft, outflow anastomosis and outflow vein. The following parameters were evaluated at outflow anastomosis: diameter, peak systolic velocity (PSV) and the height of intimal hyperplasia. Significant stenosis was considered a combination of the following findings: flow volume decrease by >25% compared to the previous value, PSV increase >2× together with residual diameter <2 mm. In case of incomplete criteria of the significant stenosis, the patient was reexamined in 4 weeks. Only DUS findings of significant stenosis, mostly with a clinical correlation of AVG hypofunction, were referred to fistulography and percutaneous transluminal angioplasty (PTA). DUS examination was performed before the first AVG cannulation, then in a month and then every 3 months. In patients with repeated access failure, oedema of the extremity bearing the vascular access or DUS borderline finding, DUS examination was performed in 4 weeks.
Study definitions
The following definitions refer to terminology previously published by Sidawy et al. 7
Primary patency was defined as the time from access creation until any intervention (endovascular or surgical) to maintain or restore blood flow.
Secondary patency was defined as the interval from time of access placement to access abandonment.
Statistical analysis
The survival data were analysed using log-rank (Cox-Mantel) test with a level of significance 0.05. The differences in survival were shown using Kaplan-Meier graphs. The statistical analysis was performer using statistical software.
Percentage and frequency were used to express categorical variables. Non-normally distributed continuous variable age was reported as median and range (minimum-maximum). The Chi-square test or Fisher’s exact test were used to compare patient’s groups in categorical parameters. The Mann-Whitney U-test was used to compare the differences between groups in age, number of thrombectomies and PTAs. The normality of the distributions was tested using the Shapiro-Wilk test. Primary and secondary patency rates with 95% CI (confidence interval) were calculated using Kaplan–Meier’s analysis . . .. Log-rank test was used to compare the groups in primary a secondary patency. Statistical analysis was performed using IBM SPSS Statistics version 22. A two-tailed p-value <0.05 was considered statistically significant.
Results
From January 2009 to December 2022, 212 AVGs were created (124 men, 88 women) at the Vascular Access Centre Olomouc, Czech Republic. Baseline patient characteristics are shown in Table 1; there was no significant difference in baseline characteristics between both groups.
Baseline patient characteristics.
Values for categorical variables are given as number (percentage); values for continuous variable are given as median (minimum-maximum).
Group 1
This group included 90 AVGs (51 men, 39 women) created in 2009–2015. The mean follow-up time was 333 days (range: 1–1230 days, standard deviation: 318 days). Primary and secondary graft patency rates are depicted in Figure 2. The primary and secondary graft patency rates were 13.3% (95% CI: 6.3%–20.4%) and 62.2% (95% CI: 51.9%–72.4%) at 6 months; 2.2% (95% CI: 0.0%– 5.3%) and 52.1% (95% CI: 41.1%–63.1%) at 12 months; 0% and 44.3% (95% CI: 32.6%–56.1%) at 24 months and 0%, 44.3% (95% CI: 32.6%–56.1%) at 36 months respectively.
During the 7-year surveillance of the AVG, which was based on physical examination, there were 27 early detected stenoses of the AVG with sequential preemptive PTA performed and 138 thromboses of the AVG recorded with subsequent thrombectomy. The most common stenosis location was the AVG venous anastomosis in 58.28%. Further causes of AVG dysfunction are listed in Figure 1. Multiple stenotic lesions were found in 10 cases, mostly in the combination of venous anastomosis with central vein stenosis. Sixteen stentgrafts were placed into the venous anastomosis in this group. We recorded significantly more thrombectomies than preemptive PTAs (p < 0.0001) performed during the AVG surveillance (Figure 3(b)).
Bar graph of AVG dysfunction causes.
Group 2
This group included 122 AVGs (73 men, 49 women) created in 2016–2022. The mean follow-up time was 584 days (range: 1–2040 days, standard deviation: 463 days). Primary and secondary graft patency rates are depicted in Figure 2. The primary and secondary graft patency rates for AVGs were 54.9% (95% CI: 46.%–63.7%) and 95.9% (95% CI: 92.3%–99.4%) at 6 months; 29.5% (95% CI: 21.4%–37.6%) and 77.8% (95% CI: 69.9%–85.7%) at 12 months; and 9.8% (95% CI: 4.6%–15.1%) and 56.5% (95% CI: 45.7%–67.2%) at 24; 2.5% (95% CI: 0.0%–5.2%) and 47.1% (95% CI: 34.8%–59.5%) at 36 months respectively. The primary and secondary graft patency was significantly higher (p < 0.0001, p = 0.002) compared to Group 1, Figure 2).
Primary (a) and secondary (b) patency of arteriovenous grafts in both studied groups.
During the 7-year surveillance of the AVG based on physical examination together with DUS surveillance there were 94 early detected stenosis of AVG with sequential preemptive PTA performed and 72 thrombosis of AVG recorded with following thrombectomy. The most common stenosis location was the AVG venous anastomosis in 72.88% . Further causes of AVG dysfunction are listed in Figure 1. Multiple stenotic lesions were found in eight cases, mostly in combination of venous anastomosis with outflow vein stenosis. Thirty-six stentgrafts were placed into the venous anastomosis in this group. We recorded significantly more preemptive PTA than thrombectomies (p = 0.0004) performer during the AVG surveillance (Figure 3(a)).
Quartile box plots of the number of PTAs (a) performed due to timely-revealed stenosis of the AVG and thrombectomies (b) due to thrombosis of the AVG during 7-years in both groups.
Discussion
The present study shows a significant improvement in primary and secondary graft patency (p < 0.0001, p = 0.002) in group 2 with the use of DUS in vessel mapping and surveillance and adjuvant medication presented by aspirin and sulodexide. AVG dysfunction is most commonly caused by a haemodynamically significant stenosis at the venous anastomosis, 8 which was also documented in our study in both groups. The venous anastomosis represents a critical part of vascular access and is the main focus during surveillance of the AVG. A combination of the following findings in our DUS surveillance protocol is considered to be significant for stenosis: flow volume decrease by >25% compared to the previous value, PSV increase >2× together with residual diameter <2 mm. However, there are different nonuniform criteria for the evaluation of stenosis significancy published9–12 which lead to different results in venous anastomosis stenosis treatment and its impact on AVG survival rate. Malik et al. reported an improvement in cumulative AVG patency due to regular ultrasonographic screening 13 ; Mauro et al., found that in patients with AVG, DUS performed at 3 months and then every 6 months postoperatively improved 5-year secondary patency in the surveillance group compared with the historical control group. 14 By contrast, there was no evidence that AVG surveillance by flow or DUS reduced thrombosis or improved AVG survival. 15 In our study we recorded improved AVG survival after DUS surveillance. In the group with DUS surveillance there were also significantly more preemptive PTAs (p = 0.0004) performed due to timely-revealed stenosis. When the stenosis of venous anastomosis was resistant to PTA or in the presence of early restenosis, stentgraft placement was indicated. In the case of early AVG thrombosis (30 days after creation) due to stenotic venous anastomosis, anastomosis resection was performed. On the other hand, an overevaluation of DUS detected stenosis (which might be borderline or clinically insignificant) leads to overindication to preemptive PTA and might impair AVG survival rate. 16 To avoid this problem, the Kidney Foundation Dialysis Outcomes Quality Initiative (KDOQI) recommends no intervention in case stenosis is detected by surveillance in the absence of clinical indicators. 17 Unnecessary PTA could stimulate progression of stable stenotic lesions.18,19 Therefore a routine surveillance of the AVF or AVG is not recommended by KDOQI in asymptomatic patients to improve access patency. 17 In our centre, a vascular surgeon selects the type and location of vascular access, creates it and also provides DUS surveillance of it.
There were not only differences in AVG surveillance between both groups in our study, but also in preoperative examination. Using DUS in preoperative vessel mapping and phlebography of central veins before the AVG creation might affect the primary patency of AVGs created after 2016. The use of phlebography led to a decrease in central vein stenosis in group 2 (Figure 1). Ultrasound preoperative mapping (UPM) allows direct visualisation of vessel quality and selection of the best vessels for access creation. UPM may reveal pathological findings that might be missed by physical examination alone and provides conditions for vascular access creation with a high probability of success without early complications or primary failure. Vascular access was created following the physical examination, without UPM and phlebography in group 1.
Since 2016 the adjuvant medication in patients with AVG has been aspirin together with SDX. Aspirin had also been used in the historical group, so we would like to point out the possible positive effect of SDX use on AVG patency rate alone. SDX is a mixture of glycosaminoglycans composed of 80% low-molecular mass heparin and 20% dermatan sulfate. 20 Due to numerous biological effects, including reduction of venous and arterial thrombogenesis, anti-inflammatory effects and endothelial protection, SDX has a wide clinical application in vascular diseases.21,22 There is only one observational study analysing the effect of SDX on NIH in a rat AVF model. 23 In this study, SDX decreased NIH, possibly through regulation of the angiopoietin/Tie system, which is dysregulated in the process of intimal hyperplasia. The development of stenosis due to NIH with subsequent thrombosis is a main cause in the reduction of the AVG lifespan. 6 We have started with SDX use in patients with AVG considering that in AVGs the pathogenesis is similar to NIH formation in AVFs with the exception of histological findings of the prominence of macrophages which indicates an additional inflammatory response likely due to the foreign graft material.4,24
Patients with AVG received 250 LSU of SDX taken orally two times a day until definitive failure of the AVG. We assume that SDX might improve the AVG patency rate, but we need more research in this field; there is a need for randomised and prospective studies analysing the effect of SDX in AVGs to confirm or disconfirm our hypothesis. There has been no study of the sulodexide effect on AVG patency published yet and we are the first centre pointing to a possible positive effect of sulodexide on AVG patency rate.
Limitations of this study include the differences of both groups presented by using different graft types, different indication types and locations of the AVG while using a different feeding artery and different superficial or deep vein for AVG creation, which might affect the patency of the AVG. We have also noticed better cooperation with dialysis centres and improved patient compliance in group 2, which affect the lifespan of the AVG and related patency rate.
We note there could be an overindication to fistulography due to DUS findings of AVG stenosis at the beginning of DUS surveillance, so we think the results may be biased and the study should be repeated in a few years to compare the groups with DUS surveillance of AVGs with improved skills, and see the possible changes of AVG patency rate.
Conclusion
In conclusion, the primary and secondary patency of AVGs were significantly improved after using DUS in preoperative mapping and surveillance. Preemptive treatment of early detected stenosis of AVG revealed by DUS surveillance led to a significant decrease of AVG occlusion. A potential positive effect of sulodexide on the AVG patency rate needs further research.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
The study was carried out in accordance with the Declaration of Helsinki and approved by the Ethical Committee of Faculty of Medicine and Dentistry, Palacky University Olomouc and University Hospital Olomouc, Czech Republic.
Informed consent
Written consent was obtained.
