Objective: Cardiovascular complications are the leading cause of death in hemodialysis patients. One of these complications is pulmonary hypertension which complicates chronic renal failure in 12-45% of cases and causes an increase in all-cause mortality of 2-3 fold in the dialysis population. Another complication is left ventricular hypertrophy which is a strong predictor of morbidity and mortality in patients with end-stage renal disease (ESRD). Arteriovenous fistulas (AVFs) increase cardiac output and lead to significant increases in both the mass and the diameter of the left ventricular wall in the long-term duration. There is evidence that AVF creation is a major risk factor for developing a new onset congestive heart failure. Our study aimed to assess the relationship between blood flow volume of AVF and its effect on the pulmonary artery pressure (PAP) and left ventricular (LV) dimension and function, as well as to find out the cut-off value of the AVF blood flow volume above which the cardiac functions start to get affected. Results: After 3 months of creating AVF, transthoracic echocardiography showed significant changes in the ejection fraction (P-value of 0.029), fractional shortening (P-value of 0.009), and the E/A ratio (P-value of 0.005). Also, there was a correlation between AVF blood flow volume and LV internal dimensions. Left ventricular end-diastolic diameter (LVEDD) and left ventricular end-systolic diameter (LVESD) increased significantly in patients with AVF blood flow volume of 1010 mL/min or above and 1120 mL/min or above, respectively. Surprisingly, pulmonay artery systolic pressure (PASP) showed no statistically significant change in our study. Conclusion: High flow AVF adversely affects cardiac dimensions and functions with a cut off value of 1010 mL/min or above.
VASCULAR DISEASE MANAGEMENT 2020;17(8):E170-E177.
Key words: Arteriovenous fistula, hemodialysis, pulmonary artery pressure, left ventricular dimensions
AVFs have become the gold standard for patients with chronic renal failure who require chronic hemodialysis since 1966 when Brescia and associates constructed a fistula between the radial artery and the cephalic vein. It carries lower infection and mortality rates, cheaper, and better longevity than any other vascular access.1
Most of the signs and symptoms (local, peripheral, and central) typical of an AVF are caused by the short circuit between the high-pressure arterial system and the low-pressure venous system and are inversely related to its hemodynamic resistance.2
It should be clear that the exact effects of AVFs on cardiac functions is difficult to determine as patients with stage VI renal disease requiring chronic dialysis have, in almost every case, volume overload due to water and salt retention. Arterial sclerosis, hypertension, as well as chronic anemia, can lead to pressure overload which causes a secondary increase in cardiac output. In addition, those patients can have significant pre-existing cardiac disease. However, deteriorating cardiac functions soon after AVF creation has been noticed favoring that AVF may affect certain cardiac functions.3,4
Although some authors assume that cardiac failure in chronic renal failure patients with AVF occurs only in individuals with previously established chronic cardiac conditions, there is evidence that AVF creation is a major risk factor for developing new onset congestive heart failure.5
Recent studies have used echocardiography to assess cardiac dimensions and functions before and after AVF creation. Different study styles were used including cross-sectional, Cohort prospective, and retrospective studies to assess the relation and effects of AVF creation on cardiac dimensions and functions. Cross-sectional studies were done just to prove high-flow fistulas effects on the heart, whereas cohort studies were used to follow up cardiac dimensions and functions after fistula creation and sometimes after fistula closure in patients who underwent kidney transplantation.6,7
Since recent studies have demonstrated that high-flow AVF have several undesirable effects on cardiac dimensions and functions, and since vascular surgeons can modify these fistulas surgically to decrease the flow inside them, this study was designed to find out the cut-off values of the AVF blood flow volume above which cardiac functions start to get affected.
Our study aims to assess the relationship between blood flow volume of arteriovenous fistula (AVF) and its effect on the pulmonary artery pressure (PAP) and the left ventricular (LV) dimensions and functions.
Also, we aim to find out the cut off values of the AVF blood flow volume above which the cardiac dimensions and functions start to get affected.
A single-center study cohort that included 50 patients with ESRD scheduled for replacement therapy in the form of maintenance hemodialysis therapy via AVF, selected from Internal Medicine and Nephrology outpatient clinics and inpatient wards during the period from May 2017 until April 2019. All patients with chronic obstructive pulmonary disease (COPD), interstitial lung disease, pulmonary thromboembolism, or patients with primary cardiac conditions such as cardiac failure, valvular heart disease, and ischemic heart disease were excluded. Also, patients with connective tissue diseases such as scleroderma, lupus erythematosus, rheumatoid arthritis, and Sjogren’s syndrome and patients with primary pulmonary hypertension were excluded.
All patients were subjected to initial evaluation which included full history taking, general examination, and cardiac examination. Laboratory investigations included complete blood count, serum albumin, kidney function tests (serum creatinine, serum urea, sodium, potassium), as well as calcium and phosphorus levels. Also, all patients were subjected to a standard 12-lead ECG and a transthoracic echocardiogram (TTE) once at the beginning of the study (before or immediately after AVF creation) and once after 3 months. Echocardiographic studies were performed by an operator who is blinded to the study and double checked for interobserver variation. All patients underwent M-mode, two-dimensional, pulsed-wave, and color-flow Doppler echocardiographic examination while in the left lateral decubitus position. Images were obtained from parasternal short and long axis, apical four-chamber, and parasternal modified views. All patients were subjected to echocardiographic assessment of pulmonary artery pressure and left ventricular functions and dimensions.
Pulmonary artery systolic pressure (PASP) was estimated from the tricuspid regurgitation (TR) peak velocity. Provided that there is no valve obstruction, peak TR velocity depends on the pressure gradient between the right ventricle and the right atrium [difference between peak right ventricular systolic pressure (RVSP) and right atrial (RA) pressure]. Therefore, estimated RVSP is equal to this pressure difference, determined from the peak TR velocity, plus the estimated RA pressure. If there is no obstruction across the pulmonic valve, the RVSP will be similar to the PASP:
PASP= 4× (peak TR velocity) 2 + estimated RA pressure.
Inferior vena cava sonography was performed in the supine position with 2-dimensional guide M-Mode echocardiography. Diameters were measured from sub-xiphoidal long axis view.9
LV ejection fraction was measured using the M-mode and 2D methods. LV dimensions were measured using the M-mode.10 Pulsed wave Doppler on the mitral inflow was recorded from apical four-chamber view. Standard measurements include E/A ratio, and deceleration time (DT). Tissue Doppler myocardial velocities were recorded just below the mitral annulus from a TTE apical approach. Standard measurements were the early myocardial velocity (E') and atrial myocardial velocity (A'). An E'/A' ratio more than 1.0 was to be considered normal, with a reduced ratio indicating impaired early diastolic relaxation.11
A Duplex sonography on AVF to measure blood flow rate after 3 months of fistula creation was also performed on all patients in the study. A fistula was considered successful if the mean blood flow rate was 780 mL/min ± 401.12. For the color Doppler ultrasound evaluation of access flow, the diameter of the feeding artery is determined by B-mode ultrasonography in a transverse plane from inner edge to inner edge. The cross-sectional area is calculated by equipment software. At the same site, Doppler spectra for calculation of time averaged velocity (TAV) are obtained in a longitudinal plane with an insonating angle maintained at ≤60°. The sample volume size must be sufficiently large to include the entire luminal cross section. Access flow is determined by equipment software using the formula13:
TAV (cm x 8 -1) x cross-sectional area (r2π; cm2) x 60 = flow volume (mL/min)
Data were collected, tabulated and all the results were subjected to adequate statistical analysis using the Chi-square test, the one-way ANOVA test, and the Kruskal Wallis test. P-value >0.05: Non significant (NS); P-value <0.05: Significant (S); P-value <0.01: highly significant (HS).
It was a prospective, single-center study; the mean age of our studied patients were 48.72 ± 13.40 years with 56% of our population being males. 26% had diabetes mellitus (DM), 68% had hypertension and 40% were smokers. None of the subjects studied had ischemic heart disease nor had previous pulmonary embolism. All patients were in sinus rhythm with only 6% having left ventricular hypertrophy before creating the AV fistula. Regarding AV fistula access, it was either distal (radio-cephalic) or proximal (brachio-cephalic or brachio-basilic). Thirty-three patients (66%) had their fistula created distally (radio-cephalic), while 17 patients had a proximal fistula (brachio-cephalic). There were no cases with brachio-basilic AVFs in our study. The mean blood flow volume was 1072.00 ± 424.77 mL/min with a maximum flow of 2560 mL/min.
Table 2 compares the echocardiographic parameters before creating the AVF and after 3 months of creating it. It shows a significant change in ejection fraction (P-value of 0.029), the fractional shortening (P-value of 0.009), and the E/A ratio (P-value of 0.005). Although the ejection fraction showed a significant change as a value, it remained within normal range and classifying it into normal and abnormal showed no significant change. Also the E/A ratio which reflects the diastolic function showed no significant change when classified into normal and diastolic dysfunction grade I. The study did not include patients with diastolic dysfunction grade II or grade III. The inter-ventricular septum during diastole (IVSd) and during systole (IVSs) along with the left ventricular posterior wall diameter (LVPWD) showed a minor increase that was not statistically significant. The left atrial (LA) diameter also showed a minor increase.
With regard to the correlation between the blood flow volume in the AVF and the changes that happened in the echocardiographic data after 3 months, Table 2 shows that only the LVEDD and the LVESD showed statistically significant change with P-values of 0.004 and 0.000, respectively. With regard to the ejection fraction, although the change that happened was statistically non-significant, the P-value was 0.191.
In a roc curve made to find the cut-off value of the blood flow volume that above which the LVEDD is affected, a value of 1010 mL/min or above was found to affect the LVEDD with specificity of 73.3% and sensitivity of 80%.
Similarly, the roc curve made in correlation to LVESD showed a value of 1120 mL/min or above to affect the LVESD with specificity of 95.45% and sensitivity of 64.29%.
With regard to the relationship between the site of the fistula and the change in the echocardiographic data before and after creating the AVF, Table 3 shows that only the LVEDD and the LVESD showed a significant relationship with P-values of 0.002 and 0.007 respectively, indicating that the proximal fistulas affected the dimensions more than the distal fistulas.
By comparing normal AVF flow patients (28 patients) to high AVF flow patients (22 patients), the latter group had more LVEDD [14 patients (63.6%) vs 6 patients (21.4%); P-value: 0.002], and LVESD [18 patients (81.8%) vs 10 patients (35.7%); P-value: 0.001). No statistically significant difference was detected between both groups regarding EF or PAP changes or any other echocardiography parameters as shown in Table 4.
As shown in Table 5, a directly proportional relationship was found between the blood flow volume of the AVF and the LVEDD in the follow up echocardiography (P-value: 0.030). Also, the LVESD showed a directly proportional relationship with the blood flow volume of the AVF (P-value: 0.009). An inversely proportional relationship was found between the blood flow volume of the AVF and the ejection fraction in the follow up echo with a P-value of 0.023.
In a sub-analysis of the relationship between the blood flow volume and the change in the echocardiographic data before and after creating the AV fistula according to gender or age, it showed that neither gender nor age category had any contribution in our study.
An AVF is currently considered the gold standard access for hemodialysis, as it has lower risk for infection, lower tendency to thrombotic occlusion, greater blood flow, reduced treatment time and is less expensive to maintain than alternative vascular access methodologies. AVF creation causes significant hemodynamic changes in cardiovascular parameters and can result in progressive left and right heart failure. It has been noticed that there is worsening in cardiac functions soon after AVF creation. This favors the idea of the AVF to have on certain cardiac functions.14
The current literature suggests that the creation of AVF can cause or exacerbate the following conditions: congestive heart failure, left ventricular hypertrophy, pulmonary hypertension, right ventricular dysfunction, coronary artery disease, and valvular dysfunction.15 According to the guidelines of The National Kidney Foundation’s Kidney Disease Outcomes Quality Initiative NKF-K/DOQI 2015, an access can be defined functional when the flow is >600 mL/min.
In our study, the target was to find the cut-off value of the AVF blood flow volume (Qa) above which, the LV dimensions and functions are affected. Although there are more recent and more accurate methods for assessment of LV dimensions and function, such as strain and speckle tracking; however, we used conventional methods that are more familiar to most of the hospital’s echo lab.
Although our study showed no statistically significant difference in the LV dimensions (LVEDD 51.44 ± 4.26 mm before fistula creation and 51.50 ± 4.24 mm three months after fistula creation (P-value 0.884)), and (LVESD 33.88 ± 4.39 mm before fistula creation and 34.66 ± 5.09 mm three months after fistula creation (P-value 0.070)), LVEDD and the LVESD showed statistically significant correlation with Qa (P-values of 0.004 and 0.000, respectively). This trivial difference of LV dimensions before and after AVF creation may be explained by the short period between the two echocardiography studies. The roc curve used to find the values above which the LV dimensions are affected found that a value of 1010 mL/min or above was found to affect the LVEDD with a specificity of 73.3% and a sensitivity of 80%, and a value of 1120 mL/min or above was found to affect the LVESD with specificity of 95.45% and sensitivity of 64.29%.
These results meet in part the studies that prove the effect of the AVF on LV internal dimensions. Iwashima et al conducted a study on sixteen patients with chronic renal failure who underwent echocardiographic studies before and 3, 7, and 14 days after the AVF operation. They found a significant change in LVEDD (48.3 ± 1.0 mm) before fistula creation, compared to (49.8 ± 0.8 mm) after 3 days, (50.3 ± 0.6 mm) after 7 days and (50.3 ± 0.6 mm) after 14 days (P-value <0.01). On the contrary the LVESD showed no significant change (30.3 ± 1.0 mm) before fistula creation, compared to (30.3 ± 1.0 mm) after 3 days, (30.0 ± 1.0 mm) after 7 days and (30.3 ± 1.1 mm) after 14 days.16
It is well proven that a high flow AVF can affect LV dimensions as early as 3 months. Martínez-Gallardo et al performed a prospective study involving 562 CKD patients stages 4 or 5 pre-dialysis which showed that 17% developed at least one episode of HF (incidence of 19/1000 person-years) during the follow-up. The multivariate logistic regression revealed that, in addition to traditional risk factors in developing pre-dialysis HF, the presence of a functioning AVF was the most important. In 47 out of 95 patients who developed CHF, a functioning AVF had previously been created, 92% of which were upper arm native AVF, with a median of 51 days between the surgical procedure and CHF episode.16
Movilli et al conducted a prospective observational study on 35 patients with functional AVF and 25 with closed AVF. At 6 months follow up by echocardiography, patients with closed AVF (after renal transplantation) showed that LVEDD significantly decreased (51 ± 4 vs 49 ± 4 mm).7
In a prospective observational study conducted on 8 patients with functional AVF and 17 patients with closed AVF (after renal transplantation), Unger et al found that patients with closed AVF showed significant reduction in LVEDD (29.5 ± 3.4 vs 26.2 ± 3.2 mm, P-value: 0.017) at 21 months follow up.17
In a prospective interventional study conducted by Van Duijnhoven et al. on 22 patients with functional AVF in University Hospital Maastricht, patients with persistent AVF showed no difference in echocardiographic data. After 3 months, patients with closed AVF showed significant reduction in LVEDD (51.5 ± 5.8 vs 49 ± 5.4 mm, P-value: <0.01).18
In a retrospective case-controlled study conducted on 38 patients, Cridlig et al found that patients with persistent AVF had larger LVEDD (52.1 ± 7.1mm vs 48.5 ± 6.0mm, P-value: 0.02) after 65 months of follow-up when compared with those with closed AVF.19
Comparing the echocardiographic data before and after fistula creation, our study showed that the only significant changes were: (a) the ejection fraction which was 62.82 ± 7.31% before fistula creation and was 60.98 ± 8.17% after fistula creation (P-value of 0.029), (b) the fractional shortening which was 34.57 ± 5.18% before fistula creation and 33.18 ± 5.67% after fistula creation (P-value of 0.009), and (c) the E/A ratio which was 1.02 ± 0.39 before fistula creation and 1.20 ± 0.43 after fistula creation (P-value of 0.005). These changes were statistically significant only in terms of absolute values as the parameters mentioned remained within normal range throughout the study. This might be due to short period of follow up, and the values might have decreased below normal if the patients were followed up for a longer period of time.
This goes in agreement with Elbaz et al,20 a cross-sectional study conducted on 55 patients with chronic renal failure (CRF) on regular HD via native AVF. The study divided the patients into two groups: group I which included patients with AVF flow more than 1200 (mL/min) which was considered high flow fistula and group II which included patients with AVF flow from 500 (mL/min) to 1200 (mL/min) which was considered normal flow fistula. It showed significant difference in ejection fraction (EF) (50.21% ± 6.12% in group I vs 56.87% ± 3.75% in group II), fraction shortening (FS) (28.7 ± 11.9% in group I vs 35.06 ± 11.68% in group II).
Also, in the Iwashima et al16 study, the EA ratio showed similar results 14 days after fistula creation (0.79 ± 0.04 vs 0.93 ± 0.06, P-value <0.001). By dividing the patients into normal and diastolic dysfunction grade I, it showed no statistically significant results and this again might be due to short period of follow up. The study included zero patients with diastolic dysfunction grade II or III.
In a cross-sectional study, Saleh et al studied 100 patients with chronic ESRD receiving hemodialysis sessions at Ain Shams University dialysis unit. AVF blood flow volume (Qa) was assessed using Color Doppler ultrasonography and the flow volume was obtained. Accordingly, the study population was categorized into 2 groups based on AVF Qa, where group A (non-High Flow group) with Qa <2000 mL/min [76 patients], and group B (High Flow group) with Qa >2000 mL/min [24 patients]. The high flow group demonstrated a significant lower LV EF 57.32 ± 6.19% compared to the non-high flow group) 62.90 ± 5.76% (P-value of 0.001).21
Our study also showed that among the 50 patients studied a positive correlation was found between the LV internal dimensions with the blood flow volume in the follow-up echocardiography with P-value of 0.030 for the LVEDD and a P-value of 0.009 for the LVESD. A negative correlation was found between the blood flow volume and the ejection fraction with a P-value of 0.023. This was in agreement with Saleh et al who found a significant positive correlation between Qa and LV dimensions (LVEDD P-value of 0.003) & (LVESD P-value of 0.000) and a significant negative correlation between Qa and LV EF (P-value of 0.001).21
Surprisingly, our study showed no statistically significant correlation between pulmonary artery pressure and AV fistula creation (before and after 3 months of fistula creation). Also, no relation was found between PAP and the blood flow volume. Again, this might be due to a short follow-up period, and further follow-up might have found significant changes. In a study conducted on fifty patients who were followed up by echocardiography before fistula creation and 6 months after fistula creation, Akbar et al found a statistically significant positive correlation between fistula flow and pulmonary artery pressure in the follow-up echocardiography, as well as between fistula flow and PAP changes (P-value: <0.05).6
In our study, the mean PAP before fistula creation was 30.35 mm Hg and decreased to 26.92 mm Hg after 3 months of fistula creation, a decrease not statistically significant. Similarly, Akbar et al found that the mean PAP before fistula creation was 25.2 mm Hg. It was 21.3 mm Hg at the second determination, a decrease not statistically significant (P-value: 0.2). This might be due to the clinical state of the patients when they did the first echocardiograph as most of renal failure patients suffer from a clinical state of volume overload.6
Although the LA diameter didn’t show a statistically significant change in our study, it showed a minor increase comparing it before and after fistula creation (39.25 ± 4.73 mm vs 39.72 ± 3.88 mm). This again goes partially in agreement with Iwashima et al where the LA diameter showed statistically significant increase after 14 days of fistula creation (39.4 ± 1.5 mm vs 40.9 ± 1.4 mm).16
Also, the PW thickness and the IVSD showed minor increase that was not statistically significant after fistula creation (9.86 ± 1.82 mm vs 10.10 ± 1.56 mm and 10.02 ± 1.65 mm vs 10.20 ± 1.75 mm, respectively). Similarly, according to Movilli et al, patients with closed AVF showed significant reduction in IVS and PW thickness (11.8 ± 2.1 mm vs 11.0 ± 2.2 mm, and 10.8 ± 1.7 mm vs 10.0 ± 1.9 mm, P-value: 0.001) 6 months after fistula closure.7
The site of the fistula also showed a statistically significant difference. Patients with proximal fistula had a greater percentage of positive change in the LVEDD and the LVESD (70.6% and 82.4% of the total number of patients with proximal fistula respectively) in agreement with most of the studies mentioned above.
Also, in agreement with all of the studies mentioned above, neither gender nor age showed a significant relation between the echocardiographic data and the AVF blood flow volume.
Limitations of the Study
1. Relatively small study population studied in a single center. Our conclusion should be validated by larger multicenter study.
2. Short period of follow up compared to other studies, longer period of follow up might be needed in order to determine the exact effects of AVF on the LV dimensions, EF and PASP.
It is recommended that every patient planned for AVF creation must do a baseline echocardiography before the procedure, and preferably when patient is not in a volume overload state. Three months after AVF creation, Duplex ultrasonography for the AVF is recommended, measuring the diameter of the AVF and the AVF blood flow volume. Although follow-up echocardiography 3, 6, and 12 months after creating the AVF and comparing it with the first echocardiography is recommended to anticipate any change in cardiac dimensions and functions and to predict high cardiac output failure, early detection of early changes in LV dimensions associating high flow AVFs is of utmost importance before developing more deterioration.
If the Duplex ultrasonography at 3 months shows a higher cut-off value for blood flow, careful cardiac evaluation should be done for cardiac status with precise measurement of LV internal dimensions and cardiac output. If there are any changes in LV dimensions, it would be recommended to intervene with this fistula either by ligation or by banding. n
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. They report no conflicts of interest regarding the content herein. The authors did not receive any specific funding.
Address for correspondence: Ehab Elfekky
Department of Cardiology, Faculty of Medicine, Ain Shams University
Department of Cardiology, Faculty of Medicine, Ain Shams University
Approval of the Ain Shams university ethical committee and approval of echocardiography scientific group of cardiology department faculty of medicine Ain Shams University was obtained according to the ethical guidelines of the 1975 declaration of Helsinki as revised in 2008. Informed approval consent was obtained from all patients included in this research.
Availability of data and material:
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
All authors contributed to the research equally. All authors read and approved the final manuscript. AAL selected patients according inclusion criteria, take history, performed general and local physical examinations and arranged for echocardiography and duplex ultrasound. EME did echocardiography and duplex ultrasounds and also wrote chapters of method and results. ANA did echocardiography and duplex ultrasound and also wrote other chapters of manuscript. OAD supervised and coordinated the work and revised the final edition of the manuscript.
1. Lok CE. Fistula first initiative: advantages and pitfalls. Clin J Am Soc Nephrol. 2007;2(5):1043-1053.
2. Sumner DS. Hemodynamics and pathophysiology of arterivenous fistulae. In: Rutherford RB, editors: Vascular surgery. 5th Edition. Saunders 2000: USA; 1400-1415.
3. Amin M, Fawzy A, Hamid MA, Elhendy A. Pulmonary hypertension in patients with chronic renal failure: role of parathyroid hormone and pulmonary artery calcifications. Chest. 2003;124(6):2093-2097.
4. Nakhoul F, Yigla M, Gilman R, Reisner SA, Abassi Z. The pathogenesis of pulmonary hypertension in haemodialysis patients via arterio-venous access. Nephrol Dial Transplant. 2005;20(8):1686-1692.
5. Beigi AA, Sadeghi AMM, Khosravi AR, Karami M, Masoudpour H. Effects of the arteriovenous fistula on pulmonary artery pressure and cardiac output in patients with chronic renal failure. J Vasc Access. 2009;10(3):160-166.
6. Movilli E, Viola BF, Brunori G, et al. Long-term effects of arteriovenous fistula closure on echocardiographic functional and structural findings in hemodialysis patients: a prospective study. Am J Kidney Dis. 2010;55(4):682-689.
7. Griffin BP, Callahan TD, Menon V. Transthoracic—Intracardiac Pressure measurements. Topol and Griffin's Manual of Cardiovascular Medicine. 2013;p1095-1096.
8. Agarwal R, Bouldin JM, Light RP, Garg A. Inferior vena cava diameter and left atrial diameter measure volume but not dry weight. Clin J Am Soc Nephrol. 2011;6(5):1066-1072.
9. Otto CM, Schwaegler RG, Freeman RV. Left and right ventricular systolic function. Echocardiography Review Guide. 2016;p98-122.
10. Otto CM, Schwaegler RG, Freeman RV. Ventricular diastolic filling and function. Echocardiography Review Guide. 2016;p224.
11. Robbin ML, Chamberlain NE, Lockhart ME, et al. Hemodialysis arteriovenous fistula maturity: US evaluation. Radiology. 2002;225(1):59-64.
12. Nonnast-Daniel B, Martin RP, Lindert O, et al. Colour Doppler ultrasound assessment of arteriovenous haemodialysis fistulas. Lancet. 1992;339(8786):143-145.
13. Rao N, Worthley M, Disney P, Faull R. Dramatic improvement in decompensated right heart failure due to severe tricuspid regurgitation following ligation of arteriovenous fistula in a renal transplant recipient. Semin Dial. 2014;27(2):E24-E26.
14. Alkhouli M, Sandhu P, Boobes K, Hatahet K, Raza F, Boobes Y. Cardiac complications of arteriovenous fistulas in patients with end-stage renal disease. Nefrologia. 2015;35(3):234-245.
15. Iwashima Y, Horio T, Takami Y, et al. Effects of the creation of arteriovenous fistula for hemodialysis on cardiac function and natriuretic peptide levels in CRF. Am J Kidney Dis. 2002;40(5):974-982.
16. Martínez-Gallardo R, Ferreira-Morong F, García-Pino G, Cerezo-Arias I, Hernández-Gallego R, Caravaca F. Congestive heart failure in patients with advanced chronic kidney disease: association with preemptive vascular access placement. Nefrologia. 2012;32(2):206-212.
17. Unger P, Velez-Roa S, Wissing KM, et al. Regression of left ventricular hypertrophy after arteriovenous fistula closure in renal transplant recipients: A long-term follow-up. Am J Transplant. 2004;4(12):2038-2044.
18. van Duijnhoven EC, Cheriex EC, Tordoir JH, Kooman JP, van Hooff JP. Effect of closure of the arteriovenous fistula on left ventricular dimensions in renal transplant patients. Nephrol Dial Transplant. 2001;16(2):368-372.
19. Cridlig J, Selton-Suty C, Alla F, et al. Cardiac impact of the arteriovenous fistula after kidney transplantation: a case-controlled, match-paired study. Transpl Int. 2008;21(10):948-954.
20. Elbaz TZ, Ahmed SF, Zahra AM. The relationship between arteriovenous fistula and cardiac abnormalities in hemodialysis patients. AAMJ. 2013;11(3):321-335.
21. Saleh MA, El Kilany WM, Keddis VW, El Said TW. Effect of high flow arteriovenous fistula on cardiac function in hemodialysis patients. Egypt Heart J. 2018;70(4):337-341.
22. Teodorescu V, Gustavson S, Schanzer H. Duplex ultrasound evaluation of hemodialysis access: A Detailed Protocol. Int J Nephrol. 2012; Article ID 508956.