Prognostic factors for radiocephalic arteriovenous fistula maturation in patients with prior placement of a central venous catheter and relationship with inflammation

Aim: A mature and functional arteriovenous fistula (AVF) is considered the best modality for vascular access(VA) for hemodialysis (HD) treatment but the incidence of early failure is high, especially in patients start their HD with a central venous catheter. The aim of this study was to evaluate the prognostic value and association of certain patient characteristics and specific inflammatory markers with early failure of AVF in patients who started their HD therapy with a CVC and a first autogenous radiocephalic AVF (RCAVF) was created after vascular consultation. Material and Methods: A retrospective review of 168 patients with end-stage renal disease who underwent RCAVF creation by the same surgeon by using the same surgical technique and whose primary vascular access for HD treatment was obtained via CVC at the time of access consultation was performed. The patients enrolled into this study were categorized into two groups as Group 1: patients with early failure (n=46) and Group 2: patients with no failure (n=122). Demographic characteristics, medical comorbidities, preoperative doppler ultrasound mapping results, laboratory parameters, postoperative follow-up details of these patients were collected. Primary patency of all patients, early failure rate, maturation failure rate, duration of CVC was calculated. Results: Female gender was found to be a significant risk factor in early failure of RCAVF (69.5% vs 36.1%; p=0.001). The number of patients whose diameter of cephalic vein< 2 mm were significantly higher in EF group (78.3% vs 22.1 ; p=0.028). The duration of CVC access of group 1 was significantly longer than group 2 (6.8 ± 3.6 months vs 2.3 ± 1.7 months, respectively; p<0.05). Overall maturation failure rate was 12.5% and primary patency at 1 year was 72.6%. Levels of C-Reactive protein (7.2 ± 9.6 vs 3.1 ± 3.3 mg/L, respectively; p=0.001) and neutrophil lymphocyte ratio (2.91± 0.30 vs 2.17 ± 0.22, respectively; p<0.05) was significantly lower at group 2 at one year. Conclusion: In patients whose VA for HD treatment was provided by CVC, small cephalic vein diameter, female gender and systemic inflammation may play a role in early failure of RCAVF.


Introduction
An increase in the global incidence of end-stage renal disease (ESRD) has led to the increasing demand for hemodialysis [1,2], which is the most common method for treating ESRD. A mature and functional arteriovenous fistula (AVF) is considered the best modality for vascular access (VA) when compared to arteriovenous grafts (AVG), and central venous catheters (CVC) [3][4][5], and a radiocephalic AVF (RCAVF) at the level of the wrist is the first choice for VA creation; however recent studies have shown high failure rates of up to 46%, with one-year patencies range from 52% to 83% . [6] Surgeons often confront with smaller-caliber vessels, and construction of an AVF is more likely to result in early failure, leading to increased morbidities, related to reoperations, longer hospitalization, and increased costs. [7,8] Early failure of AVF also delay the establishment of permanent dialysis access. It is, therefore, important to identify patients who will have a high likelihood of early AVF failure.
Early failure is defined as any fistula not used for dialysis due to loss patency (thrombosis, etc.) or lack of maturation.
A fistula is considered mature when it is thought to be appropriate for cannulation with minimal complications, and to deliver the prescribed blood flow throughout the HD procedure. In other words, when a VA is cannulated successfully with two needles over a period of at least 6 HD sessions during 30 days, and delivering the prescribed blood flow throughout the HD procedure (at least 350 ml/ min), the VA is finally considered adequate for HD (functional and successfully used). [6,9] In clinical practice, as many as 60%-80% of the incident, patients start their hemodialysis therapy with a CVC due to being unable to wait for the maturation of AVFs or having a condition in which AVF development is not feasible. [10,11] Among ESRD patients initiating hemodialysis with a CVC, the time at which they switch to a mature AVF is influenced by successful AVF maturation which depends on several factors including patient comorbidities and demographics; diameters of cephalic vein and radial artery; peri-operative and postoperative factors. [12][13][14][15] However, these studies are a mixed picture (i.e., not limited to RCAVFs) and included conflicting results and the evidence derived from these articles is not consistent Studies do not provide a solid platform for the planning of RCAVF formation, and does not assist in the process of informed consent (percentage likelihood of success 125  Öz Amaç: Matüre ve fonksiyonel bir arteriovenöz fistül (AVF), hemodiyaliz (HD) tedavisi tedavisinde vasküler erişim için en iyi modalite olarak kabul edilir; ancak erken başarısızlık oranı HD tedavisine santral venöz kateter (SVK) ile başlayan hastalarda yüksektir. Bu çalışmanın amacı, HD tedavilerine SVK ile başlayan ve daha sonra ilk kez radiosefalik AVF (RSAVF) oluşturulan hastalarda belirli hasta özelliklerinin ve spesifik inflamatuar belirteçlerin AVF'nin erken başarısızlığı açısından prognostik değeri ve ilişkisini incelemekti.
Anahtar kelimeler: otojen radyosefalik arteriovenöz fistül; erken başarısızlık; sefalik ven; inflamasyon and/or failure). For example, many authors have agreed that duplex vein mapping increases utilization of AVF [16,17], but as vein diameter is dynamic (subject to constriction from changes in venous sympathetic tone), intraoperative measurements may differ from mapped vein diameters.
Furthermore, Central Venous Pressure, positioning of the arm, hydration status, ambient room temperature, caffeine intake, and medications may contribute to misleading scans. As a result, though widely recommended, even duplex mapping may not improve functional AVF patency. [18] Besides, several authors have noted the pivotal role of inflammation in neointimal hyperplasia, which is a foundation of AVF nonmaturation. [19][20][21] The relation between C-reactive protein (CRP) levels and the development of intimal hyperplasia [22]; thrombosis due to disproportionate intimal hyperplasia resulting in access thrombosis [23] has previously claimed. Likewise, neutrophile lymphocyte ratio (NLR) is a robust inflammatory indicator and associated with both coronary atherosclerosis and restenosis. [24] Given the undeniable role of inflammation in AVF stenosis as well as the histopathological similarity of AVF stenosis with atherosclerosis, a relationship between NLR and AVF stenosis/ maturation was questioned. [25,26] Furthermore, previous studies have reported that CVC placement contributes to chronic inflammation independent of infection. [27,28] The correlation of NLR with the AVF stenosis, as well as the role of NLR and CRP as a predictor of access failure and their pathogenic role in NIH is not clearly understood. Moreover, there are limited data about long-term serial changes in inflammatory marker levels and their relationship to access type in hemodialysis patients, and the contribution of access type to the inflammatory status of hemodialysis patients is not well described. [29,30] The objective of this paper was to report our findings from the last 10 years in a university hospital located in Antalya. Since recent evidence has highlighted a failure of the literature to identify factors associated with maturation [31], we aimed to test the hypothesis that certain patients' characteristics (age, gender, vessel diameters, and medical comorbidities) affect the maturation of AVF. We also aimed to test the prognostic value and association of specific inflammatory markers (white cell count, neutrophil-lymphocyte ratio, C-reactive protein) with early failure of AVF in patients who started their HD therapy with a CVC and a first autogenous RCAVF was created after consultation to our department.

Study Design, Setting and Patient Selection
We performed a retrospective chart review of all patients with and whose VA for HD treatment was obtained via CVC at the time of access consultation and whose preoperative and intraoperative vessel diameters were recorded. Patients who had a life expectancy less than 12 months and to avoid factors influencing CRP levels, patients who had any sign of infection (fever, leukocytosis, cellulitis) or received PTA within 1 month before or after blood sampling, as well as those with rheumatic disease or cancer, were excluded in this study.
We also excluded the AVFs that required 2-stage operations.
The patients enrolled in this study were categorized into two groups as group 1: patients with early failure (n=46) and group 2: patients with no failure (n=122).
Each patient must be followed up at the vascular surgery and/ or nephrology clinics for at least one year or until AVF failure.
Data on the survival and prognostic predictors of AVF were extracted from the hospital's electronic database.
In our tertiary care university hospital, patients were regularly seen by the nephrologists, and the decision to start dialysis treatment was made based on the severity of the worsening of renal function. A detailed history and physical examination was undertaken from every patient, including age, gender, history, cause of chronic kidney disease, and the presence of comorbidities/risk factors and noted. The latter included diabetes mellitus (defined as the use of insulin or oral hypoglycemic agents), hypertension (systolic blood pressure >140 mmHg, diastolic blood pressure >90 mmHg, or use of anti-hypertensives), dyslipidemia (defined as the use of antilipidaemic agents, e.g., statins, ezetimibe, etc.), coronary artery disease (defined as the history of angina, myocardial infarction, or coronary intervention, including angioplasty and/or bypass grafting), and peripheral arterial disease (PAD) (defined as the history of intermittent claudication, critical limb ischemia, or revascularization of the lower limbs).
The common causes of ESRD among the patients included in this study were diabetes mellitus, hypertension, and glomerulonephritis. AVF was not created in the presence of calcifications of the feeding artery wall [32], and anastomosis was not created distal to stenosis above 50% in the radial artery. We did not attempt to construct RCAVF with RA diameters below 1.5 mm. When the decision concerning RA suitability is doubtful, we look at venous mapping results and often decide for the RCAVF formation attempt when there is a large, distensible CV present with normal Doppler venous waveform, and wellestablished phenomena of respiratory filling Criteria for venous size as a predictor of RCAVF outcome fluctuated even more than RA diameter cut-offs across published studies. Minimal CV internal diameters associated with RCAVF outcomes in the range of 1.6-2.6 mm were reported. [33][34][35][36] Similar to arterial preference, we did not use CV below 1.5 mm. [34] Venous outflow was assessed accurately to exclude venous outflow stenosis and accessory veins.

Preoperative Vascular Evaluation
Evaluation of vein compressibility and thrombus exclusion was performed before tourniquet placement.
Since the threshold diameters for both RA and CV diameters for a suitable RCAVF were 2.0 mm, and diameters between 1.6-1.9 were defined as "grey zone", so we decided to compare the groups taking 2.0 mm as a threshold. [34][35][36][37] Evaluation of the dominant arm was performed solely when the non-dominant arm evaluation was unsatisfactory. [38]

Laboratory Tests
All laboratory studies were performed by Başkent University Laboratories (Alanya, Antalya-TR) using automated methods.

Surgical Technique
All patients were scheduled for a primary AVF creation between the radial artery and cephalic vein (Figure 1).

Figure 1: Autogenous radio-cephalic arteriovenous fistula
All the patients gave their informed consent before surgery.
The creation of AVF was performed under local anesthesia (2% Lidocaine-Xylocaine). A longitudinal 3-4 cm skin incision was used, as this was found to give good access to both vein and artery. While evaluated during preoperative planning, a vein diameter of less than 2.5 mm on the preoperative ultrasound duplex did not preclude the surgeon from exploring the vessel in the operating room. Intraoperatively, hydrodilatation maneuver was routinely done with a 4-fr infant feeding catheter for a vein with a diameter <2 mm. During the hydrodilatation, the continuity of the cephalic vein and any recognizable resistance changes were followed closely. If the surgeon felt that it responded adequately, the vein was then used for an AV fistula. If it did not, the patient received a prosthetic graft or another type of autogenous fistula, which is out of the scope for this study. Following clamping, for the standard arteriotomy, the radial artery was incised 6 mm. An end-to-side anastomosis was created between the cephalic vein and the radial artery using continuous polypropylene sutures (7/0 Prolene) with the aid of 2.5x magnifying loupes.
A palpable date thrill was taken as an indicator of successful AVF creation.

Anticoagulation
Following exploration of the arteries and veins and before placing the clamp, 5,000 IE heparin was routinely administered intravenously to all patients during AVF creation.

Follow-up
Antibiotherapy and antiaggregant treatment were not used routinely during the postoperative term.
Postoperative surveillance was scheduled at two weeks and then every month for an additional three to six months to monitor the AVF outcomes and possible complications.
All AVFs were assessed clinically 6-8 weeks postoperatively for the presence of a strong thrill over a sufficiently dilated (e.g., 8-10 cm length and >5-6 mm diameter) vein with a superficial course. Clinical criteria were used for the detection of nonfunctioning AVFs. The inability to cannulate the AVF or to obtain sufficient dialysis blood flow within 6 weeks with three sessions per week after fistula creation was classified as maturation failure, regardless of whether it is patent. If the AVF was considered mature, CVC was removed from the patient; otherwise, CVC was continued to be the route of VA for HD and an additional surgical, or endovascular intervention would be performed to promote fistula maturation or patency. US was performed for all patients with nonmaturating AVFs.

End-Points
The primary endpoint was fistula maturation and functioning AVF, which was defined by the determination of both vascular surgeon and nephrologist. We aimed to evaluate our patients' characteristics that have been reported to be associated with AVF non-maturation and loss of patency in the literature.
Primary patency, success rate, assisted primary patency, and primary failure rates were also primary endpoints.
We examined the relationship between demographic characteristics including age, gender, diabetes, hypertension, peripheral vascular disease, coronary artery disease, bodymass index, smoking, the type and duration of CVC used; preoperative and intraoperative physical examination and measurements of vessels diameters were also used.

Statistical Analysis
Data are given as percentages and means ± SD. Rates were

Results
A total of 168 patients were recruited in the study. The demographics and preoperative vessel diameter measurements of all patients included in the study are listed in Table 1. Of the total 168 patients, 76 (45.2%) were female, and female gender was found to be a significant risk factor in EF of RCAVF (69.5% vs 36.1%; p=0.001). The most common comorbidity was HT (n=123, 73.2%), followed by DM (n=91, 54.2%). There were no significant differences between the groups in terms of age, BMI, DM, HT, KAH, PVD and smoking habits, as shown in Table 1.
Even there was a tendency, the radial artery diameters were not significantly higher in NF group than in EF group (p=0.074). The number of patients whose CV diameter < 2 mm were significantly higher in EF group (78.3% vs 22.1 ; p=0.028) ( Table 1). Considering the all study group, the overall maturation failure rate was calculated as 12.5%. Primary patency at 1 year was 72.6%. Kaplan-Meier survival analysis of RCAVF primary patency is shown in figure 2. Regarding outcomes for the secondary endpoints, no statistically significant difference among the groups was found in levels of albumin, calcium, intact PTH, serum TG, LDL, blood hemoglobin, WBC and PLT count depicted in table 2. CRP level was higher than the normal range (normal range <3.0 mg/l) at the first 30-days and did not differ between the groups. CRP levels of group 2 were significantly lower than group 1(7.2 ± 9.6 vs. 3.1 ± 3.3 mg/L, respectively; p=0.001) at 1-year. Likewise, the same pattern was followed by the NLR.  reported that females were 36% less likely than males to use an AVF at dialysis initiation. [48] The exact mechanism of different AVF outcomes between genders is unclear, but it has been suggested that difference in vascular diameter, reactivity and impaired ability of venous dilatation to arterial pressure being the possible explanations.
[47] We did not find age as a prognostic factor for early failure. There has been conflicting results in literature about age. Some studies identified old age as a poor prognostic indicator [49,50], whereas others did not. [51,52] DM is one of the most common causes of ESRD (53), the second most common cause in our study, but it was not associated with adverse outcome of fistula maturation during the first three months of its creation, namely resulting in early failure [54], but it may have a negative impact on late AVF survival [55] since DM promotes platelet aggregation [56] and vascular calcification. [57] We perform a detailed preoperative vascular evaluation as well as a detailed careful physical examination. We believe that physical examination plays a pivotal role in making a proper decision to create RCAVF. When the decision concerning RA suitability is doubtful, we look at venous mapping results and often decide for the RCAVF formation attempt when there is a large, distensible CV present with normal Doppler venous waveform and well established phenomena of respiratory filling. In a study by Wells et al, US was considered unnecessary in majority of patients who fulfilled the clinical criteria for AVF creation. [58] Two subsequent randomized controlled trials also did not find additional advantage of vein mapping over clinical assessment in patients with favourable anatomy, in terms of early AVF failure and cumulative AVF survival rate. [59,60] Wong et al admitted that preoperative vein mapping may improve AVF maturation rates but the difference did not reach statistical significance and suggested that larger clinical trial is needed to confirm the clinical benefit . [18] On the other hand, Hossain et al. reported that the primary failure rate in the ultrasound group was 18% compared with 47% (P < 0.001) in the group of patients who did not undergo ultrasound examination. In patients without preoperative ultrasound, there were higher rates of new access creation (31% vs 9%; P< .001) and fistula abandonment (66% vs 39%; P < .001). [61] There is also an ongoing debate about the threshold of vessel diameters. Wong et al reported that cephalic vein diameter less than 1.6 mm was associated with early radiocephalic AVF failure. [62] Mendes et al found low AVF success rate of 16% in vein diameter of 2 mm or less, as compared to 76% of those >2 mm in a cohort of 44 patients. [36] On the other hand, Lee etal., did not find vein size to be statistically significant in predicting fistula maturation, and AVF can be successfully created in mean vein diameter of <2 mm in more than 70% of patients.
[63] Eslami MH et al, a larger target vein diameter was the most predictive variables predicting early failure. [64] The conflict about the use of US is because of veins dynamic status: subject to constriction from changes in venous sympathetic tone, intraoperative measurements may differ from mapped vein diameter and postoperative ultrasound protocols do not take dynamic enlargement with access augmentation (occlusion of the outflow) into account. The use of vein diameter and to a lesser degree, arterial diameter has been tested as a predictor for fistula maturation with reasonable success. There is an increasing agreement that a minimal arterial diameter >2 mm and venous >2 mm should be considered as a cutoff point, as anything less than that is likely to be associated with nonmaturation. [65,66] In our study, not the radial artery diameter but the CV vein diameter found to play a significant role in early failure. A diameter of CV< 2mm was found to be important in early failure.
It is well known that the low resistance circuit, resulting from the creation of the anastomosis between the artery and the vein, triggers an immediate increase in blood flow and elevation of blood pressure in the veins. Elevation of blood flow rate is responsible for a rapid increase in wall shear stress (WSS) and venous tensile stress induced by the velocity gradient on the luminal vessel surface. [67] On the other hand, WSS changes are the major determinants of vessel dilatation and remodeling. In rodent models of venous thrombosis created by ligation of inferior vena cava to induce venous hypertension and altered WSS, thrombus initiation is associated with a rapid vein wall inflammatory reaction involving early endothelial activation and neutrophil infiltration, similarly to observations conducted in the arterial side. [68] In studies studying local hemodynamic conditions in AVF using computational fluid Dynamics, Ene-Lordache and Ramuzzi suggested that despite the significant increase in flow rate, in selected locations, the WSS is oscillating and in average it is low in magnitude. [69] The same team also showed that, while the flow is almost laminar in the proximal arterial limb, in the venous segment leading the velocity field is highly unstable and multidirectional [70] leading a transitional laminar to turbulent-like flow developing in areas of the juxta-anastomotic vein. The presence of disturbed WSS patterns (unstable in direction and magnitude) may induce different physical stimuli in endothelial cells that actually lead to the proliferation of neointimal cells and to induction of a proinflammatory state preventing vessel wall dilatation and outward remodeling of arterial and venous vessels that take place when endothelial cells are exposed to unidirectional WSS directed along vessel axis.
The relationship between high levels of CRP and HD was previously described. [21] CRP serum concentration increases in cases of inflammation, infection and tissue damage. Kaygin et al. found a threefold increase in serum CRP levels in unsuccessful AVF cases and a positive correlation. [21] Wali et al. [71] have stated that AVF insufficiency appears as a result of platelet activation and intimal hyperplasia which is caused by the secretion of mediators because of primary and/ or secondary defects in vascular endothelium due to mucoid or myxoid degeneration, mural calcification, inflammatory reaction or erythrocyte/macrophage infiltration on the vascular wall. Chou et al [22] identified CRP level as an independent risk factor for fistula thrombosis. These investigators suggested that CRP level strongly predicts access thrombosis events in maintenance hemodialysis patients, possibly because CRP is a marker of intimal hyperplasia in AVFs.
Morena et al. [72] have stated that due to mineral metabolism deterioration in HD patients and due to inflammation, thrombosis risk increases in the group where CRP, Calcium, PTH increases. In addition, some studies have identified CRP level as a risk factor for the development of access thrombosis. [73,74] In our study, we did not find a difference in terms of calcium and pth.
The patients enrolled in our study started their HD therapy with a CVC due to being unable to wait for the maturation of AVF. A history of CVC placement or prolonged use of CVC was a poor prognostic predictor of AVF survival [75,76], but the mechanisms by which preexisting CVCs affect AVF maturation remain elusive. Systemic inflammation, a common condition occurring in the setting of CVC placement [27,77], has been proposed as a pathogenetic mechanism underlying neointimal hyperplasia [78], which is a foundation of AVF failure.
Available evidence suggests that CRP is an objective measure of a patient's inflammatory state and that it accurately reflects the generation of proinflammatory cytokines, such as IL-6 and tumor necrosis factor-alpha. [29,30] There are limited data about long-term serial changes in inflammatory marker levels and their relationship to access type in hemodialysis patients, and the contribution of access type to the inflammatory status of hemodialysis patients is not well described. [29,30] Our study corroborated the findings of previous studies [77] that CVCs in comparison to fistulas have a greater state of inflammation defined by CRP levels in incident hemodialysis patients.Banerjee et al [79], reported that CRP levels decreased over time in cases of an AVF, with the highest inflammatory state 30 days after access placement, and found a significant decrease in CRP levels when there was a change from a CVC to an AVF was associated with decreased CRP levels compared with patients who used CVCs at both times. As also consistent with our study, Goldstein et al. [77], who investigated the levels of inflammatory markers at the time of dialysis initiation and again 6 months later, found that patients with persistent CVC use from dialysis initiation through 6 months had consistently high inflammatory levels over the period, whereas the levels of inflammatory markers were attenuated in patients who changed from a catheter to an AVF. Their findings are consistent with our study as we also obtained a significant decrease in CRP levels after changing the VA from CVC to AVFnot in 30 days but in 1 year-. This finding was strengthened further by our findings on the association of the inflammatory state reflected by NLR.
NLR level increased with CVC and stay high when a mature functioning AVF was obtained. In patients whose CVC was removed, a dramatic decrement in NLR was observed in one year. The clinical trials showed that IL-6, pentraxin and complement system had roles in AVF dysfunction [80,81]. Yilmaz et al. [25] reported that in chronic HD patients with established AVF access, patients who developed late stenosis were found to have higher level of NLR. An increased level of NLR reflects inflammation . [24,[80][81][82][83] Spark et al. Evaluated NLR to predict mortality in patients with chronic critical limb ischemia. They found that an elevated NLR along with a high troponin level (>0.1) was the only independent predictor of mortality in those patient. [84] In a study of 83 patients who underwent infrapopliteal percutaneous interventions for critical limb ischemia, Chan et al. [85] reported that those with NLR > 5.25 had an increased risk of death.
This finding that change in CRP levels and NLR are associated with change in access type adds additional support to the body of the observational evidence, suggesting that the catheter itself contributes to an increase in inflammatory marker levels in hemodialysis patients.

Study Limitation
Nevertheless, this study was limited by being a retrospective study. Hence, some data might have been unavailable, such as blood flow measurements and the results of other inflammatory marker blood tests. This study was conducted with a homogeneous cohort of ESRD patients from a single institution. Hence, the results might not be the same in other settings where people have different reference ranges for WBC counts or dissimilar material types of CVC are used.

Conclusion
As a conclusion, early failure of RCAVF is an obstacle we have to overcome. Certain clinical factors including female gender, anatomical factors including a diameter of CV < 2mm was found significant in early failure. A history of CVC placement or prolonged use of CVC is a poor prognostic predictor of AVF survival in which systemic inflammation plays an important role. A significant decrease in CRP levels was observed after changing the VA from CVC to AVF-not in 30 days but in 1 year-. This finding was strengthened further by our findings on the association of the inflammatory state reflected by NLR.