Original Paper Received: October 3, 2013 Accepted: February 2, 2014 Published online: April 26, 2014
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
Clinical Outcomes after Arteriovenous Fistula Creation in Chronic Kidney Disease Mardiana Lee a Matthew A. Roberts c Maree-Ross Smith a Jason Chuen b Peter F. Mount a
Departments of a Nephrology and b Vascular Surgery, Austin Health, Heidelberg, and c Department of Nephrology, Eastern Health, Melbourne, Vic., Australia
Key Words Arteriovenous fistula · Haemodialysis · Chronic kidney disease
Abstract Background: Optimal timing of arteriovenous fistula (AVF) surgery in chronic kidney disease (CKD) is uncertain. Methods: A single-centre retrospective study of pre-dialysis CKD patients having first AVF surgery. Results: The median estimated glomerular filtration rate (eGFR) at the time of AVF surgery in 100 patients was 15 ml/min/1.73 m2, with patients classified as having an early AVF if eGFR was >15 ml/ min/1.73 m2 (n = 46) or a late AVF if eGFR was ≤15 ml/ min/1.73 m2 (n = 54). In the eGFR ≤15 group, 81% of patients started haemodialysis (HD), compared with 63% of the eGFR >15 patients (p = 0.04). The median time to starting HD was 30.3 months in the eGFR >15 group compared to 10.7 months for the eGFR ≤15 group (log rank p = 0.018). There were no differences in the requirements for a dialysis catheter (eGFR >15 24% vs. eGFR ≤15 11%, p = 0.20) or additional access procedures between the two groups. Conclusions: AVF surgery with an eGFR >15 ml/min/1.73 m2 was associated with a higher risk of AVF non-use and a more prolonged time to the need for HD.
© 2014 S. Karger AG, Basel
© 2014 S. Karger AG, Basel 0253–5068/14/0373–0163$39.50/0 E-Mail [email protected] www.karger.com/bpu
Introduction
Commencement of haemodialysis (HD) with a native arteriovenous fistula (AVF) for vascular access is associated with superior outcomes compared to either an arteriovenous graft or a central venous catheter (CVC) [1– 5]. Most guidelines, therefore, recommend that chronic kidney disease (CKD) patients with a plan for HD be referred to a surgeon to prepare for AVF creation if the estimated glomerular filtration rate (eGFR) is in the range of 15–30 ml/min/1.73 m2 [6–10]. While the advantages of an AVF over a CVC are well established, optimal timing of AVF creation for CKD patients remains uncertain. In part this relates to difficulty in predicting the times required for both AVF maturation and for CKD progression to the need for HD [11, 12]. The risk of creating an AVF late relates primarily to the risks associated with CVC use [3, 13]. The risks of performing AVF surgery early are less well studied but might include risks associated with surgery itself [14], the risk of death before HD is performed [15], and the potential for an AVF to contribute to cardiac dysfunction [16]. Most evidence guiding recommendations regarding the timing of AVF creation is derived from retrospective data comparing the outcomes of patients commencing HD with or without an AVF [2–5]. In contrast, there is a paucity of studies examining the outcomes of CKD paDr. Peter Mount Department of Nephrology Austin Health, Studley Road Heidelberg, VIC 3084 (Australia) E-Mail Peter.Mount @ austin.org.au
tients undergoing AVF surgery that assesses outcomes irrespective of whether or not the patient was subsequently treated with HD. The aim of the present study was to determine the outcomes in our centre of patients undergoing AVF surgery for pre-dialysis CKD. In particular, we sought to determine the time from first AVF creation to the requirement for HD or renal replacement therapy (RRT) and the clinical factors that influenced this.
Statistics Continuous variables were analysed using a t test or MannWhitney test depending upon data distribution. Comparison of proportions was performed using χ2 tests, with Fisher’s exact test used when appropriate. Kaplan-Meier analysis was performed on the time to use of the AVF. Observation time commenced at the date of first AVF creation and patients were followed until they commenced HD (or RRT) with censoring at whichever came first: (1) first episode of HD or RRT, or (2) patient death, or (3) the 31st December 2011. Differences between groups were assessed using the log-rank test. Statistical analysis was performed using Stata Version 11.0 (StataCorp, College Station, Tex., USA).
Subjects and Methods This was a retrospective study of all pre-dialysis CKD patients having a first AVF created at Austin Health in anticipation of a future need for HD from 1/1/2007 to 31/12/2009, with follow-up until 31/12/2011. The study was approved by the Human Research Ethics Committee of Austin Health and conducted consistent with the principles of the Declaration of Helsinki. Participants and Setting Patients were included if they had CKD and underwent creation of an AVF in anticipation of requiring dialysis. Patients who had previously commenced any form of RRT were excluded. Data was extracted by review of medical records. All patients had venous ultrasound mapping performed prior to assessment in a Vascular Access Clinic, attended by two or more vascular surgeons, an advanced trainee in nephrology and an advanced trainee in vascular surgery. Final decisions regarding optimal timing of referral and surgery were made by the treating nephrologist. At our centre, the time from nephrologist referral to surgical assessment is usually 3–6 weeks and the time from surgical assessment to performing AVF surgery is usually less than 8 weeks. Baseline Characteristics Baseline patient data collected included demographic information, co-morbidities and measures of kidney function at the time of AVF surgery. The eGFR (ml/min/1.73 m2) was determined from serum creatinine by the 4-variable modified MDRD equation [17]. The rate of decline in eGFR (ΔeGFR) was calculated as the difference between the eGFR at the time of surgery and the nearest measurement of eGFR ≤6 months prior to surgery, divided by the time in months and expressed as ml/min/1.73 m2/month. The presence of severe proteinuria was defined as: >1,000 mg protein/24 h; a spot urinary protein/creatinine ratio >100 mg/mmol, or a spot urinary albumin/creatinine ratio >60 mg/mmol.
Outcomes Primary outcomes were: (1) time from first AVF creation procedure to first episode of HD, and (2) time from first AVF creation procedure to first episode of any form of RRT. Secondary outcomes included: (1) need for a CVC at initiation of HD, (2) the need for a subsequent surgical or radiological vascular access procedure prior to commencing HD or in the first 3 months after HD start, and (3) patient status at the end of the study and at the time of commencing RRT. Patient status was defined as: on HD; on peritoneal dialysis; with a functioning kidney transplant; CKD not receiving RRT, or deceased.
164
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
Results
Baseline Characteristics Between 1/1/2007 and 31/12/2009, 100 pre-dialysis CKD patients underwent a surgery to create a first AVF, all of which were upper limb fistulae. The most common procedure was the creation of a forearm radiocephalic AVF, which was performed in 71 patients. The remainder of the patients had upper arm AVF surgery, with 21 brachiocephalic AVFs and 9 brachiobasilic AVFs. The median period of follow-up from the time of AVF creation until either the need for HD or the census date (31/12/2011) was 16 months (interquartile range 6–34 months). The median eGFR at the time of AVF surgery was 15 ml/min/1.73 m2. On the basis of eGFR in relation to the median, patients were classified as having had early AVF surgery (eGFR >15) or late AVF surgery (eGFR ≤15). There were no differences between these two groups in age, aetiology of CKD or major co-morbidities (table 1). Mean serum creatinine at the time of AVF surgery was 414 μmol/l in the eGFR ≤15 group and 277 μmol/l in the eGFR >15 group (p < 0.001). There were a lower proportion of men in the eGFR ≤15 group (48%) than the eGFR >15 group (76%) (p = 0.004). The rate of decline in kidney function (ΔeGFR ml/min/1.73 m2) in the 6 months prior to AVF surgery had been slower in the eGFR >15 group (p = 0.003). Requirement for HD or RRT after AVF Surgery Over 2–5 years of follow-up, 73/100 patients started HD and 79/100 started RRT (table 2). The 6 patients who started RRT but not HD were managed with peritoneal dialysis. No patient received a preemptive kidney transplant. Patients in the eGFR >15 group were less likely than the eGFR ≤15 group to require either HD (63 vs. 81%, p = 0.04) or RRT (65 vs. 91%, p = 0.003) (table 2). At 12 months, 26% of patients in the eGFR >15 group had commenced HD, compared to 56% in the eGFR ≤15 Lee/Roberts/Smith/Chuen/Mount
Table 1. Baseline patient characteristics eGFR >15 (n = 46) Age Range Gender, male
eGFR ≤15 (n = 54)
63.1±14.2 26.7–81.6 35 (76%)
63.8±13.5 29.9–85.1 26 (48%)
Race Caucasian Asian Aboriginal
45 (98%) 1 (2%) 0
48 (89%) 4 (7%) 2 (4%)
Cause of ESKD Diabetic nephropathy Renovascular Glomerulonephritis Polycystic Reflux Other Unknown
14 (30%) 8 (17%) 10 (22%) 2 (4%) 2 (4%) 6 (13%) 4 (9%)
14 (26%) 3 (6%) 18 (33%) 4 (7%) 1 (2%) 9 (17%) 5 (9%)
Kidney parameters at AVF creation Creatinine, μmol/l Range eGFR, ml/min/1.73 m2 Range ΔeGFR, ml/min/1.73 m2/month Proteinuria >1 g/day*
p value 0.81 0.004 0.19
0.46
277±48 414±101 183–400 264–729 19.9±3.1 11.8±2.4 16–30 5–15 –0.06±1.4 –0.95±1.5 29 (64%) 41 (80%)
Comorbidities Diabetes IHD PVD CVA Cardiac failure Hypertension Dyslipidaemia
27 (59%) 26 (57%) 10 (22%) 11 (24%) 17 (37%) 43 (93%) 38 (83%)
22 (41%) 24 (44%) 12 (22%) 13 (24%) 19 (35%) 51 (94%) 37 (69%)
Smoking Never Current Past
18 (39%) 6 (13%) 22 (48%)
25 (46%) 10 (19%) 19 (35%)
Malignancy Never Active Past
32 (70%) 5 (11%) 9 (20%)
39 (74%) 10 (19%) 4 (7%)
Medications Antiplatelet therapy Anticoagulant therapy
30 (65%) 9 (20%)
33 (61%) 5 (10%)
15 group (table 2; fig. 1a). The corresponding times for 50% of patients receiving any form of RRT was 7.7 months for the eGFR ≤15 group compared to 30.3 months for the eGFR >15 group (table 2; fig. 2a). There was no difference in time to commencement of HD or RRT by median age, diabetes status or presence of severe proteinuria (fig. 1, 2). Although there was a gender imbalance between the early and late groups, no association of gender with time to HD (log rank p = 0.39) or time to RRT (log rank p = 0.71) was demonstrated. Vascular Access at Commencement of Dialysis Of the 73 patients who started HD, 61 (84%) used an AVF at HD commencement whereas 12 (16%) patients used a CVC. Patients in the eGFR >15 group appeared at least as likely to require a CVC at the commencement of HD as the eGFR ≤15 group (24 vs. 11%, p = 0.20) (table 2). A further 3 patients who commenced HD with an AVF used a CVC within 3 months of starting HD (2 from the early AVF group, 1 from the late AVF group). Requirement for Additional Vascular Access Procedures A high requirement for additional vascular access procedures was observed regardless of eGFR at the time of AVF creation (table 3). Overall, 37 patients required 44 additional procedures in addition to their initial AVF surgery but prior to having any HD (39 surgical procedures, 5 radiological procedures). The risk of having an additional surgical or radiological vascular access procedure prior to starting HD was not different between the eGFR >15 (39%) and eGFR ≤15 (35%) groups (p = 0.83). Additional vascular access interventions were also common in the first 3 months after starting HD, with half (15/29) in the eGFR >15 group and a third (13/44) in the eGFR ≤15 AVF group requiring a procedure to improve or restore AVF function within 3 months from commencing dialysis. At our institution, this was predominantly radiological intervention (table 3). Characteristics of Patients Not Requiring HD after AVF Creation Over the period of follow-up, 27/100 patients did not require HD following AVF creation. The duration of follow-up for these 27 patients ranged from 26.4 to 59.3 months. In comparing the patients who did not start HD (HD–) with those that started HD (HD+), the HD– patients had a lower serum creatinine and a higher eGFR at Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
165
log rank p = 0.018 GFR POPLQ GFR !POPLQ
0.75 0.50 0.25 0
0.75 0.50 0.25 0
12
24
36
48
60
Number at risk GFR POPLQ 54 GFR !POPLQ 46
24 34
16 26
9 13
5 3
0 0
log rank p = 0.14 1RGLDEHWHV Diabetes
1.00 0.75
0
b
Time after first AVF creation (months)
0.50 0.25
12
24
31 27
22 20
48
60
8 14
3 5
0 0
log rank p = 0.39 Proteinuria JGD\ Proteinuria !JGD\
1.00
0
36
Time after first AVF creation (months)
Number at risk Age \HDUV 50 Age !\HDUV 50
HD-free survival
0
a
HD-free survival
log rank p = 0.85 Age \HDUV Age !\HDUV
1.00
HD-free survival
HD-free survival
1.00
0.75 0.50 0.25 0
0
c Number at risk 1RGLDEHWHV 51 Diabetes 49
12
24
36
48
60
32 26
25 17
15 7
7 1
0
d
Time after first AVF creation (months) 0 0
Fig. 1. Time from first AVF creation to first treatment with HD. a Analysis according to eGFR greater than, compared with less than
or equal to, the median value of 15 ml/min/1.73 m2; log rank p =
12
24
36
48
60
Time after first AVF creation (months)
Number at risk Proteinuria JGD\ 26 Proteinuria !JGD\ 70
13 43
9 32
6 16
2 6
0 0
0.018. b Comparison of age greater than, compared with less than or equal to, the median value of 67.3 years. c Diabetes versus no diabetes. d Effect of severe proteinuria (>1,000 mg/24 h equivalent).
Table 2. RRT outcomes after AVF creation
HD started (n = 73) Time from AVF creation to HD start, months Median (interquartile range) RRT started (n = 79) Time from AVF creation to RRT start, months Median (interquartile range) Used catheter at HD commencement Proportion (95% confidence interval) not requiring HD at 1 year after AVF creation 2 years after AVF creation 3 years after AVF creation
166
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
eGFR >15 (n = 46)
eGFR ≤15 (n = 54)
29 (63%)
44 (81%)
30.3 (11.9– ) 30 (65%) 30.3 (9.6–44.1) 7 (24%) 74% (59–84) 57% (41–69) 43% (28–57)
10.7 (4.2–31.6) 49 (91%) 7.7 (3.0–18.7) 5 (11%) 44% (31–57) 30% (18–42) 24% (14–36)
Lee/Roberts/Smith/Chuen/Mount
p value 0.04 0.018 0.002 0.001 0.15 0.018
0.75
RRT-free survival
RRT-free survival
1.00
log rank p < 0.001 GFR POPLQ GFR !POPLQ
1.00
0.50 0.25 0
0.75 0.50 0.25 0
0
a
12
24
36
48
60
0
b
Time after first AVF creation (months)
Number at risk GFR POPLQ 54 GFR !POPLQ 46
18 33
12 25
5 12
3 3
0 0
12
24
27 24
19 18
1.00
RRT-free survival
0.75 0.50 0.25
36
48
60
Time after first AVF creation (months)
Number at risk Age \HDUV 50 Age !\HDUV 50
log rank p = 0.40 1RGLDEHWHV Diabetes
1.00
RRT-free survival
log rank p = 0.77 Age \HDUV Age !\HDUV
6 11
1 5
0 0
log rank p = 0.78 Proteinuria JGD\ Proteinuria !JGD\
0.75 0.50 0.25 0
0 0
c Number at risk 1RGLDEHWHV 51 Diabetes 49
12
24
36
48
60
0
d
Time after first AVF creation (months) 28 23
22 15
11 6
5 1
0 0
Fig. 2. Time from first AVF creation to first RRT. a Analysis ac-
cording to eGFR greater than, compared with less than or equal to, the median value of 15 ml/min/1.73 m2; log rank p < 0.001. b Com-
12
24
36
48
60
Time after first AVF creation (months)
Number at risk Proteinuria JGD\ 26 Proteinuria !JGD\ 70
12 37
8 28
6 11
2 4
0 0
parison of age greater than, compared with less than or equal to, the median value of 67.3 years. c Diabetes vs. no diabetes. d Effect of severe proteinuria (>1,000 mg/24 h equivalent).
Table 3. Additional vascular access procedures performed before or within 3 months of the initiation of HD
Procedures performed before HD start New AVF created Surgical revision of AVF Diagnostic angiogram Radiological intervention Procedures performed within 3 months after HD start New AVF created Surgical revision of AVF Diagnostic angiogram Radiological intervention
Outcomes after AVF Creation in CKD
eGFR >15 (n = 46)
eGFR ≤15 (n = 54)
11 11 1 2
5 12 1 1
2 2 1 10
1 1 3 7
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
167
Table 4. Comparisons between patients not having HD (HD–) and having HD (HD+) after AVF creation
Age, years Gender, male eGFR, ml/min/1.73 m2 Serum creatinine, μmol/l ΔeGFR, ml/min/1.73 m2/month – preceding 6 months Diabetes Secondary VA procedures pre HD
the time of surgery (p = 0.02) (table 4). There were no significant differences found between HD– and HD+ patients in terms of age, sex, diabetes or the rate of decline of renal function prior to surgery (ΔeGFR) (table 4). Interestingly, 56% of HD– patients had at least one additional vascular access procedure prior to receiving any HD and this was higher than in the HD+ group (30%) (p = 0.03) (table 4). Patient Status at End of Study At the end of the study there were more patients in the eGFR >15 group than the eGFR ≤15 group that were alive independent of RRT [16/46 (35%) vs. 5/54 (9%)] (p = 0.003) (fig. 3). Of the 21 patients in this category, 19 still had a plan for future RRT, whereas 2 patients from the early AVF group changed their management plan to conservative care. At the end of the study, patients in the eGFR >15 group were less likely than the eGFR ≤15 group to be on any form of dialysis (HD or PD combined) [21/46 (46%) vs. 42/54 (78%), p = 0.002]. By study end, 7 patients had received a kidney transplant, 3 from the early group and 4 from the late group, and all these grafts remained functional. There were 6 deaths from the eGFR >15 group (13%) and 3 from the eGFR ≤15 group (6%) (p = 0.29). None of these deaths occurred before the need for RRT.
Discussion
The proportion of patients commencing HD with an AVF of 84% in this study compares favourably with rates reported elsewhere [18, 19], although the median eGFR of 15 ml/min/1.73 m2 at the time of surgery might be seen as relatively late AVF creation in relation to some existing recommendations (see table 5) [7, 8, 10, 20, 21]. These results were achieved by the use of a multidisciplinary vascular access clinic overseen by a renal vascular access 168
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
HD– (n = 27)
HD+ (n = 73)
p value
63.9±13.6 17 (63%) 17.4±4.7 317±110 –0.25±1.31 10/27 (37%) 15/27 (56%)
63.3±13.9 44 (60%) 14.8±4.9 364±102 –0.64±1.55 39/73 (53%) 22/73 (30%)
0.83 1.00 0.02 0.02 0.25 0.18 0.03
% 80
eGFR >15 H*)5 31
60
40
*
18
16 11
20 3 0
HD
4
3 PD
KT
5
6
No RRT
3 D
Fig. 3. Patient status at the end of the study (31/12/2011). Comparisons are shown between eGFR >15 and eGFR ≤15 AVF groups showing the % for each outcome for that group. Patients were classified as being on HD, peritoneal dialysis (PD), with a functioning kidney transplant (KT), alive and not requiring renal replacement (no RRT) or deceased (D). The actual number of patients in each category is indicated above the bars. Patients in the eGFR >15 group were more likely to be in the no RRT category (p = 0.003) and trended to be less likely to be in either the HD or PD group (p = 0.08). * p = 0.003.
coordinator. This multidisciplinary approach has previously been reported by others to achieve favourable vascular access outcomes [22, 23]. The present study also found a substantial proportion of patients undergoing AVF creation did not start HD for a prolonged interval, and many patients underwent subsequent procedures prior to commencing HD. A previous larger study also found that the time from AVF creation to the initiation of HD was highly variable and there was a subgroup of patients not requiring HD over a prolonged period of follow-up after AVF creation Lee/Roberts/Smith/Chuen/Mount
Table 5. Published recommendations on the timing of AVF surgery for CKD patients
Organisation
Recommendation
Kidney Disease Outcomes Quality Initiative [6]
A fistula should be placed at least 6 months before the anticipated start of HD treatments (B)
British Renal Association [21]
Guideline 3.1 – Planning of vascular access We suggest that planning for access should commence when patients enter CKD stage 4 (2C) Guideline 3.2 – Creation of vascular access We recommend that the exact timing of placement of vascular access will be determined by rate of decline of renal function, co-morbidities and by the surgical pathway (1C)
Canadian Society of Nephrology [8]
Establish AV fistulae when the patient has an estimated GFR of 15–20 ml/min and progressive kidney disease (Grade D, opinion)
European Best Practice Guidelines [10]
Potential chronic HD patients should be ideally referred to the nephrologist and/or surgeon for preparing vascular access when they reach the stage 4 of their CKD (GFR rate 15 group might have contributed to the higher CVC use in this group. Overall, these data suggest that while it is highly desirable to create an AVF at least several months before needing HD, cre-
Outcomes after AVF Creation in CKD
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
169
ation of an AVF years before HD is required probably does not further reduce the need for CVC insertion. It is known that some patients require multiple procedures for the successful establishment of vascular access for HD [11, 24]. The present study found that 37% of patients underwent at least one additional vascular access procedure prior to commencement of HD. Furthermore, 38% of patients who commenced HD had further vascular access intervention in the first 3 months after commencing HD. An advantage of earlier AVF creation might be that it allows more time for an AVF to develop; however, in some patients earlier AVF creation could increase the time available to develop stenosis or thrombosis, leading to additional access procedures before receiving HD. Furthermore, some patients could subsequently decide to have conservative care for ESRD or die from another cause before using their AVF. In these patients, creation of an AVF too early could lead to unnecessary procedures for an AVF that is never used. Most current guidelines recommend referral of CKD patients with a plan for HD to a surgeon for AVF assessment and creation when the eGFR is 15– 30 ml/min/1.73 m2, although the evidence level for these recommendations is recognised to be low quality (table 5) [7, 8, 10, 20, 21]. The present study suggests that in many patients creation of an AVF when the eGFR is ≥20 ml/min/1.73 m2 could be too early, exposing the patient to problems associated with vascular access long before the actual need for HD. Our data suggest that in some patients it might be safe to delay the creation of an AVF until the eGFR is ≤15 ml/min/1.73 m2. It is simplistic, however, to conclude that an arbitrary eGFR cut-off of 15 ml/min/1.73 m2, or another value, should be the primary consideration in determining the optimal timing of AVF surgery. This decision should be based on an estimate of the likelihood of HD being performed in a chosen time frame, based on all the available clinical data for an individual patient. O’Hare et al. [12] found that the mean annual GFR decline for a large group of CKD 4/5 patients was –2.3 ml/min/1.73 m2, emphasising that a higher eGFR at the time of AVF surgery will increase the risk and interval of AVF non-use. In the present study, the rate of decline in renal function in the 6 months prior to AVF surgery showed substantial variability and was slower in the eGFR >15 group; however, the ΔeGFR did not differ between the patients who did or did not receive HD. Overall, the available data indicate that the optimal timing for AVF surgery should take account both the eGFR and the predicted rate of decline of kidney function.
170
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
Whilst the risks of unnecessary or excessively early AVF surgery are not well defined, short- and long-term complications of AVF creation can be identified that should be considered when determining the timing of surgery. AVF creation leads to an immediate left to right shunt reducing systemic vascular resistance and increasing cardiac output, with the potential to precipitate or exacerbate a variety of cardiovascular problems including high output cardiac failure and pulmonary hypertension [25, 26]. As commonly seen in patients with severe CKD, the patients in our study had a high baseline rate of cardiac disease, indicating that this population may be particularly susceptible to any adverse cardiovascular impact of AVF surgery. Other important complications to consider with AVF surgery include the risk of hand ischaemia caused by steal syndrome [25–27] and the risk of immediate peri-operative complications. In conclusion, this study found in a cohort of pre-dialysis CKD patients undergoing AVF surgery with a median eGFR of 15 ml/min/m2 a low rate of CVC use at the time of commencing HD. In the patients with eGFR >15, the time to needing HD was longer and the risk of not doing HD over 2–5 years of follow-up was higher. This suggests that adherence to some current guidelines could result in AVF creation earlier than is clinically optimal in a substantial proportion of patients. A better understanding of the factors that influence the likelihood that HD will be needed within a period of time is required to avoid unnecessary procedures and their sequelae.
Acknowledgements The authors would like to thank Prof. David Power and Mr. Gary Fell for departmental support for this project.
Disclosure Statement The authors have no conflicts to interest to disclose.
References
1 Ravani P, Palmer SC, Oliver MJ, Quinn RR, MacRae JM, Tai DJ, Pannu NI, Thomas C, Hemmelgarn BR, Craig JC, Manns B, Tonelli M, Strippoli GF, James MT: Associations between hemodialysis access type and clinical outcomes: a systematic review. J Am Soc Nephrol 2013;24:465–473. 2 Xue JL, Dahl D, Ebben JP, Collins AJ: The association of initial hemodialysis access type with mortality outcomes in elderly Medicare ESRD patients. Am J Kidney Dis 2003; 42: 1013–1019.
Lee/Roberts/Smith/Chuen/Mount
3 Polkinghorne KR, McDonald SP, Atkins RC, Kerr PG: Vascular access and all-cause mortality: a propensity score analysis. J Am Soc Nephrol 2004;15:477–486. 4 Bradbury BD, Fissell RB, Albert JM, Anthony MS, Critchlow CW, Pisoni RL, Port FK, Gillespie BW: Predictors of early mortality among incident US hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Clin J Am Soc Nephrol 2007; 2:89–99. 5 Moist LM, Trpeski L, Na Y, Lok CE: Increased hemodialysis catheter use in Canada and associated mortality risk: data from the Canadian Organ Replacement Registry 2001–2004. Clin J Am Soc Nephrol 2008;3:1726–1732. 6 Vascular Access Work G: Clinical practice guidelines for vascular access. Am J Kidney Dis 2006;48(suppl 1):S248–S273. 7 Sidawy AN, Spergel LM, Besarab A, Allon M, Jennings WC, Padberg FT Jr, Murad MH, Montori VM, O’Hare AM, Calligaro KD, Macsata RA, Lumsden AB, Ascher E, Society for Vascular S: The Society for Vascular Surgery: clinical practice guidelines for the surgical placement and maintenance of arteriovenous hemodialysis access. J Vasc Surg 2008; 48(5 suppl):2S–25S. 8 Jindal K, Chan CT, Deziel C, Hirsch D, Soroka SD, Tonelli M, Culleton BF, Canadian Society of Nephrology Committee for Clinical Practice G: Hemodialysis clinical practice guidelines for the Canadian Society of Nephrology. J Am Soc Nephrol 2006;17(3 suppl 1):S1–S27. 9 Hoggard J, Saad T, Schon D, Vesely TM, Royer T, American Society of D, Interventional Nephrology CPC, Association for Vascular A: Guidelines for venous access in patients with chronic kidney disease. A Position Statement from the American Society of Diagnostic and Interventional Nephrology, Clinical Practice Committee and the Association for Vascular Access. Semin Dial 2008;21:186–191.
Outcomes after AVF Creation in CKD
10 Tordoir J, Canaud B, Haage P, Konner K, Basci A, Fouque D, Kooman J, Martin-Malo A, Pedrini L, Pizzarelli F, Tattersall J, Vennegoor M, Wanner C, ter Wee P, Vanholder R: EBPG on Vascular Access. Nephrol Dial Transplant 2007;22(suppl 2):ii88–ii117. 11 Huijbregts HJ, Bots ML, Wittens CH, Schrama YC, Moll FL, Blankestijn PJ, group Cs: Hemodialysis arteriovenous fistula patency revisited: results of a prospective, multicenter initiative. Clin J Am Soc Nephrol 2008;3:714– 719. 12 O’Hare AM, Bertenthal D, Walter LC, Garg AX, Covinsky K, Kaufman JS, Rodriguez RA, Allon M: When to refer patients with chronic kidney disease for vascular access surgery: should age be a consideration? Kidney Int 2007;71:555–561. 13 Weber CL, Djurdjev O, Levin A, Kiaii M: Outcomes of vascular access creation prior to dialysis: building the case for early referral. ASAIO J 2009;55:355–360. 14 Schinstock CA, Albright RC, Williams AW, Dillon JJ, Bergstralh EJ, Jenson BM, McCarthy JT, Nath KA: Outcomes of arteriovenous fistula creation after the Fistula First Initiative. Clin J Am Soc Nephrol 2011; 6: 1996– 2002. 15 Oliver MJ, Quinn RR, Garg AX, Kim SJ, Wald R, Paterson JM: Likelihood of starting dialysis after incident fistula creation. Clin J Am Soc Nephrol 2012;7:466–471. 16 Blake PG, Quinn RR, Oliver MJ: The risks of vascular access. Kidney Int 2012;82:623–625. 17 Mathew TH, Australasian Creatinine Consensus Working G: Chronic kidney disease and automatic reporting of estimated glomerular filtration rate: a position statement. Med J Aust 2005;183:138–141. 18 Centers for M, Medicaid S, Kinney R: 2005 Annual Report: ESRD Clinical Performance Measures Project. Am J Kidney Dis 2006; 48(4 suppl 2):S1–S106.
19 Lopez-Vargas PA, Craig JC, Gallagher MP, Walker RG, Snelling PL, Pedagogos E, Gray NA, Divi MD, Gillies AH, Suranyi MG, Thein H, McDonald SP, Russell C, Polkinghorne KR: Barriers to timely arteriovenous fistula creation: a study of providers and patients. Am J Kidney Dis 2011;57:873–882. 20 FistulaFirst. Fistula First Change Concepts. www.fistulafirst.org/Professionals/FFBI ChangeConcepts.aspx. 21 Fluck R, Kumwenda M: Renal Association Clinical Practice Guideline on vascular access for haemodialysis. Nephron Clin Pract 2011; 118(suppl 1):c225–c240. 22 Allon M, Bailey R, Ballard R, Deierhoi MH, Hamrick K, Oser R, Rhynes VK, Robbin ML, Saddekni S, Zeigler ST: A multidisciplinary approach to hemodialysis access: prospective evaluation. Kidney Int 1998;53:473–479. 23 Dwyer A, Shelton P, Brier M, Aronoff G: A vascular access coordinator improves the prevalent fistula rate. Semin Dial 2012; 25: 239–243. 24 Voormolen EH, Jahrome AK, Bartels LW, Moll FL, Mali WP, Blankestijn PJ: Nonmaturation of arm arteriovenous fistulas for hemodialysis access: A systematic review of risk factors and results of early treatment. J Vasc Surg 2009;49:1325–1336. 25 MacRae JM, Pandeya S, Humen DP, Krivitski N, Lindsay RM: Arteriovenous fistula-associated high-output cardiac failure: a review of mechanisms. Am J Kidney Dis 2004; 43:e17– e22. 26 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;20:1686–1692. 27 Beathard GA, Spergel LM: Hand ischemia associated with dialysis vascular access: an individualized access flow-based approach to therapy. Semin Dial 2013;26:287–314.
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
171
Copyright: S. Karger AG, Basel 2014. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
Clinical Outcomes after Arteriovenous Fistula Creation in Chronic Kidney Disease Mardiana Lee a Matthew A. Roberts c Maree-Ross Smith a Jason Chuen b Peter F. Mount a
Departments of a Nephrology and b Vascular Surgery, Austin Health, Heidelberg, and c Department of Nephrology, Eastern Health, Melbourne, Vic., Australia
Key Words Arteriovenous fistula · Haemodialysis · Chronic kidney disease
Abstract Background: Optimal timing of arteriovenous fistula (AVF) surgery in chronic kidney disease (CKD) is uncertain. Methods: A single-centre retrospective study of pre-dialysis CKD patients having first AVF surgery. Results: The median estimated glomerular filtration rate (eGFR) at the time of AVF surgery in 100 patients was 15 ml/min/1.73 m2, with patients classified as having an early AVF if eGFR was >15 ml/ min/1.73 m2 (n = 46) or a late AVF if eGFR was ≤15 ml/ min/1.73 m2 (n = 54). In the eGFR ≤15 group, 81% of patients started haemodialysis (HD), compared with 63% of the eGFR >15 patients (p = 0.04). The median time to starting HD was 30.3 months in the eGFR >15 group compared to 10.7 months for the eGFR ≤15 group (log rank p = 0.018). There were no differences in the requirements for a dialysis catheter (eGFR >15 24% vs. eGFR ≤15 11%, p = 0.20) or additional access procedures between the two groups. Conclusions: AVF surgery with an eGFR >15 ml/min/1.73 m2 was associated with a higher risk of AVF non-use and a more prolonged time to the need for HD.
© 2014 S. Karger AG, Basel
© 2014 S. Karger AG, Basel 0253–5068/14/0373–0163$39.50/0 E-Mail [email protected] www.karger.com/bpu
Introduction
Commencement of haemodialysis (HD) with a native arteriovenous fistula (AVF) for vascular access is associated with superior outcomes compared to either an arteriovenous graft or a central venous catheter (CVC) [1– 5]. Most guidelines, therefore, recommend that chronic kidney disease (CKD) patients with a plan for HD be referred to a surgeon to prepare for AVF creation if the estimated glomerular filtration rate (eGFR) is in the range of 15–30 ml/min/1.73 m2 [6–10]. While the advantages of an AVF over a CVC are well established, optimal timing of AVF creation for CKD patients remains uncertain. In part this relates to difficulty in predicting the times required for both AVF maturation and for CKD progression to the need for HD [11, 12]. The risk of creating an AVF late relates primarily to the risks associated with CVC use [3, 13]. The risks of performing AVF surgery early are less well studied but might include risks associated with surgery itself [14], the risk of death before HD is performed [15], and the potential for an AVF to contribute to cardiac dysfunction [16]. Most evidence guiding recommendations regarding the timing of AVF creation is derived from retrospective data comparing the outcomes of patients commencing HD with or without an AVF [2–5]. In contrast, there is a paucity of studies examining the outcomes of CKD paDr. Peter Mount Department of Nephrology Austin Health, Studley Road Heidelberg, VIC 3084 (Australia) E-Mail Peter.Mount @ austin.org.au
tients undergoing AVF surgery that assesses outcomes irrespective of whether or not the patient was subsequently treated with HD. The aim of the present study was to determine the outcomes in our centre of patients undergoing AVF surgery for pre-dialysis CKD. In particular, we sought to determine the time from first AVF creation to the requirement for HD or renal replacement therapy (RRT) and the clinical factors that influenced this.
Statistics Continuous variables were analysed using a t test or MannWhitney test depending upon data distribution. Comparison of proportions was performed using χ2 tests, with Fisher’s exact test used when appropriate. Kaplan-Meier analysis was performed on the time to use of the AVF. Observation time commenced at the date of first AVF creation and patients were followed until they commenced HD (or RRT) with censoring at whichever came first: (1) first episode of HD or RRT, or (2) patient death, or (3) the 31st December 2011. Differences between groups were assessed using the log-rank test. Statistical analysis was performed using Stata Version 11.0 (StataCorp, College Station, Tex., USA).
Subjects and Methods This was a retrospective study of all pre-dialysis CKD patients having a first AVF created at Austin Health in anticipation of a future need for HD from 1/1/2007 to 31/12/2009, with follow-up until 31/12/2011. The study was approved by the Human Research Ethics Committee of Austin Health and conducted consistent with the principles of the Declaration of Helsinki. Participants and Setting Patients were included if they had CKD and underwent creation of an AVF in anticipation of requiring dialysis. Patients who had previously commenced any form of RRT were excluded. Data was extracted by review of medical records. All patients had venous ultrasound mapping performed prior to assessment in a Vascular Access Clinic, attended by two or more vascular surgeons, an advanced trainee in nephrology and an advanced trainee in vascular surgery. Final decisions regarding optimal timing of referral and surgery were made by the treating nephrologist. At our centre, the time from nephrologist referral to surgical assessment is usually 3–6 weeks and the time from surgical assessment to performing AVF surgery is usually less than 8 weeks. Baseline Characteristics Baseline patient data collected included demographic information, co-morbidities and measures of kidney function at the time of AVF surgery. The eGFR (ml/min/1.73 m2) was determined from serum creatinine by the 4-variable modified MDRD equation [17]. The rate of decline in eGFR (ΔeGFR) was calculated as the difference between the eGFR at the time of surgery and the nearest measurement of eGFR ≤6 months prior to surgery, divided by the time in months and expressed as ml/min/1.73 m2/month. The presence of severe proteinuria was defined as: >1,000 mg protein/24 h; a spot urinary protein/creatinine ratio >100 mg/mmol, or a spot urinary albumin/creatinine ratio >60 mg/mmol.
Outcomes Primary outcomes were: (1) time from first AVF creation procedure to first episode of HD, and (2) time from first AVF creation procedure to first episode of any form of RRT. Secondary outcomes included: (1) need for a CVC at initiation of HD, (2) the need for a subsequent surgical or radiological vascular access procedure prior to commencing HD or in the first 3 months after HD start, and (3) patient status at the end of the study and at the time of commencing RRT. Patient status was defined as: on HD; on peritoneal dialysis; with a functioning kidney transplant; CKD not receiving RRT, or deceased.
164
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
Results
Baseline Characteristics Between 1/1/2007 and 31/12/2009, 100 pre-dialysis CKD patients underwent a surgery to create a first AVF, all of which were upper limb fistulae. The most common procedure was the creation of a forearm radiocephalic AVF, which was performed in 71 patients. The remainder of the patients had upper arm AVF surgery, with 21 brachiocephalic AVFs and 9 brachiobasilic AVFs. The median period of follow-up from the time of AVF creation until either the need for HD or the census date (31/12/2011) was 16 months (interquartile range 6–34 months). The median eGFR at the time of AVF surgery was 15 ml/min/1.73 m2. On the basis of eGFR in relation to the median, patients were classified as having had early AVF surgery (eGFR >15) or late AVF surgery (eGFR ≤15). There were no differences between these two groups in age, aetiology of CKD or major co-morbidities (table 1). Mean serum creatinine at the time of AVF surgery was 414 μmol/l in the eGFR ≤15 group and 277 μmol/l in the eGFR >15 group (p < 0.001). There were a lower proportion of men in the eGFR ≤15 group (48%) than the eGFR >15 group (76%) (p = 0.004). The rate of decline in kidney function (ΔeGFR ml/min/1.73 m2) in the 6 months prior to AVF surgery had been slower in the eGFR >15 group (p = 0.003). Requirement for HD or RRT after AVF Surgery Over 2–5 years of follow-up, 73/100 patients started HD and 79/100 started RRT (table 2). The 6 patients who started RRT but not HD were managed with peritoneal dialysis. No patient received a preemptive kidney transplant. Patients in the eGFR >15 group were less likely than the eGFR ≤15 group to require either HD (63 vs. 81%, p = 0.04) or RRT (65 vs. 91%, p = 0.003) (table 2). At 12 months, 26% of patients in the eGFR >15 group had commenced HD, compared to 56% in the eGFR ≤15 Lee/Roberts/Smith/Chuen/Mount
Table 1. Baseline patient characteristics eGFR >15 (n = 46) Age Range Gender, male
eGFR ≤15 (n = 54)
63.1±14.2 26.7–81.6 35 (76%)
63.8±13.5 29.9–85.1 26 (48%)
Race Caucasian Asian Aboriginal
45 (98%) 1 (2%) 0
48 (89%) 4 (7%) 2 (4%)
Cause of ESKD Diabetic nephropathy Renovascular Glomerulonephritis Polycystic Reflux Other Unknown
14 (30%) 8 (17%) 10 (22%) 2 (4%) 2 (4%) 6 (13%) 4 (9%)
14 (26%) 3 (6%) 18 (33%) 4 (7%) 1 (2%) 9 (17%) 5 (9%)
Kidney parameters at AVF creation Creatinine, μmol/l Range eGFR, ml/min/1.73 m2 Range ΔeGFR, ml/min/1.73 m2/month Proteinuria >1 g/day*
p value 0.81 0.004 0.19
0.46
277±48 414±101 183–400 264–729 19.9±3.1 11.8±2.4 16–30 5–15 –0.06±1.4 –0.95±1.5 29 (64%) 41 (80%)
Comorbidities Diabetes IHD PVD CVA Cardiac failure Hypertension Dyslipidaemia
27 (59%) 26 (57%) 10 (22%) 11 (24%) 17 (37%) 43 (93%) 38 (83%)
22 (41%) 24 (44%) 12 (22%) 13 (24%) 19 (35%) 51 (94%) 37 (69%)
Smoking Never Current Past
18 (39%) 6 (13%) 22 (48%)
25 (46%) 10 (19%) 19 (35%)
Malignancy Never Active Past
32 (70%) 5 (11%) 9 (20%)
39 (74%) 10 (19%) 4 (7%)
Medications Antiplatelet therapy Anticoagulant therapy
30 (65%) 9 (20%)
33 (61%) 5 (10%)
15 group (table 2; fig. 1a). The corresponding times for 50% of patients receiving any form of RRT was 7.7 months for the eGFR ≤15 group compared to 30.3 months for the eGFR >15 group (table 2; fig. 2a). There was no difference in time to commencement of HD or RRT by median age, diabetes status or presence of severe proteinuria (fig. 1, 2). Although there was a gender imbalance between the early and late groups, no association of gender with time to HD (log rank p = 0.39) or time to RRT (log rank p = 0.71) was demonstrated. Vascular Access at Commencement of Dialysis Of the 73 patients who started HD, 61 (84%) used an AVF at HD commencement whereas 12 (16%) patients used a CVC. Patients in the eGFR >15 group appeared at least as likely to require a CVC at the commencement of HD as the eGFR ≤15 group (24 vs. 11%, p = 0.20) (table 2). A further 3 patients who commenced HD with an AVF used a CVC within 3 months of starting HD (2 from the early AVF group, 1 from the late AVF group). Requirement for Additional Vascular Access Procedures A high requirement for additional vascular access procedures was observed regardless of eGFR at the time of AVF creation (table 3). Overall, 37 patients required 44 additional procedures in addition to their initial AVF surgery but prior to having any HD (39 surgical procedures, 5 radiological procedures). The risk of having an additional surgical or radiological vascular access procedure prior to starting HD was not different between the eGFR >15 (39%) and eGFR ≤15 (35%) groups (p = 0.83). Additional vascular access interventions were also common in the first 3 months after starting HD, with half (15/29) in the eGFR >15 group and a third (13/44) in the eGFR ≤15 AVF group requiring a procedure to improve or restore AVF function within 3 months from commencing dialysis. At our institution, this was predominantly radiological intervention (table 3). Characteristics of Patients Not Requiring HD after AVF Creation Over the period of follow-up, 27/100 patients did not require HD following AVF creation. The duration of follow-up for these 27 patients ranged from 26.4 to 59.3 months. In comparing the patients who did not start HD (HD–) with those that started HD (HD+), the HD– patients had a lower serum creatinine and a higher eGFR at Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
165
log rank p = 0.018 GFR POPLQ GFR !POPLQ
0.75 0.50 0.25 0
0.75 0.50 0.25 0
12
24
36
48
60
Number at risk GFR POPLQ 54 GFR !POPLQ 46
24 34
16 26
9 13
5 3
0 0
log rank p = 0.14 1RGLDEHWHV Diabetes
1.00 0.75
0
b
Time after first AVF creation (months)
0.50 0.25
12
24
31 27
22 20
48
60
8 14
3 5
0 0
log rank p = 0.39 Proteinuria JGD\ Proteinuria !JGD\
1.00
0
36
Time after first AVF creation (months)
Number at risk Age \HDUV 50 Age !\HDUV 50
HD-free survival
0
a
HD-free survival
log rank p = 0.85 Age \HDUV Age !\HDUV
1.00
HD-free survival
HD-free survival
1.00
0.75 0.50 0.25 0
0
c Number at risk 1RGLDEHWHV 51 Diabetes 49
12
24
36
48
60
32 26
25 17
15 7
7 1
0
d
Time after first AVF creation (months) 0 0
Fig. 1. Time from first AVF creation to first treatment with HD. a Analysis according to eGFR greater than, compared with less than
or equal to, the median value of 15 ml/min/1.73 m2; log rank p =
12
24
36
48
60
Time after first AVF creation (months)
Number at risk Proteinuria JGD\ 26 Proteinuria !JGD\ 70
13 43
9 32
6 16
2 6
0 0
0.018. b Comparison of age greater than, compared with less than or equal to, the median value of 67.3 years. c Diabetes versus no diabetes. d Effect of severe proteinuria (>1,000 mg/24 h equivalent).
Table 2. RRT outcomes after AVF creation
HD started (n = 73) Time from AVF creation to HD start, months Median (interquartile range) RRT started (n = 79) Time from AVF creation to RRT start, months Median (interquartile range) Used catheter at HD commencement Proportion (95% confidence interval) not requiring HD at 1 year after AVF creation 2 years after AVF creation 3 years after AVF creation
166
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
eGFR >15 (n = 46)
eGFR ≤15 (n = 54)
29 (63%)
44 (81%)
30.3 (11.9– ) 30 (65%) 30.3 (9.6–44.1) 7 (24%) 74% (59–84) 57% (41–69) 43% (28–57)
10.7 (4.2–31.6) 49 (91%) 7.7 (3.0–18.7) 5 (11%) 44% (31–57) 30% (18–42) 24% (14–36)
Lee/Roberts/Smith/Chuen/Mount
p value 0.04 0.018 0.002 0.001 0.15 0.018
0.75
RRT-free survival
RRT-free survival
1.00
log rank p < 0.001 GFR POPLQ GFR !POPLQ
1.00
0.50 0.25 0
0.75 0.50 0.25 0
0
a
12
24
36
48
60
0
b
Time after first AVF creation (months)
Number at risk GFR POPLQ 54 GFR !POPLQ 46
18 33
12 25
5 12
3 3
0 0
12
24
27 24
19 18
1.00
RRT-free survival
0.75 0.50 0.25
36
48
60
Time after first AVF creation (months)
Number at risk Age \HDUV 50 Age !\HDUV 50
log rank p = 0.40 1RGLDEHWHV Diabetes
1.00
RRT-free survival
log rank p = 0.77 Age \HDUV Age !\HDUV
6 11
1 5
0 0
log rank p = 0.78 Proteinuria JGD\ Proteinuria !JGD\
0.75 0.50 0.25 0
0 0
c Number at risk 1RGLDEHWHV 51 Diabetes 49
12
24
36
48
60
0
d
Time after first AVF creation (months) 28 23
22 15
11 6
5 1
0 0
Fig. 2. Time from first AVF creation to first RRT. a Analysis ac-
cording to eGFR greater than, compared with less than or equal to, the median value of 15 ml/min/1.73 m2; log rank p < 0.001. b Com-
12
24
36
48
60
Time after first AVF creation (months)
Number at risk Proteinuria JGD\ 26 Proteinuria !JGD\ 70
12 37
8 28
6 11
2 4
0 0
parison of age greater than, compared with less than or equal to, the median value of 67.3 years. c Diabetes vs. no diabetes. d Effect of severe proteinuria (>1,000 mg/24 h equivalent).
Table 3. Additional vascular access procedures performed before or within 3 months of the initiation of HD
Procedures performed before HD start New AVF created Surgical revision of AVF Diagnostic angiogram Radiological intervention Procedures performed within 3 months after HD start New AVF created Surgical revision of AVF Diagnostic angiogram Radiological intervention
Outcomes after AVF Creation in CKD
eGFR >15 (n = 46)
eGFR ≤15 (n = 54)
11 11 1 2
5 12 1 1
2 2 1 10
1 1 3 7
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
167
Table 4. Comparisons between patients not having HD (HD–) and having HD (HD+) after AVF creation
Age, years Gender, male eGFR, ml/min/1.73 m2 Serum creatinine, μmol/l ΔeGFR, ml/min/1.73 m2/month – preceding 6 months Diabetes Secondary VA procedures pre HD
the time of surgery (p = 0.02) (table 4). There were no significant differences found between HD– and HD+ patients in terms of age, sex, diabetes or the rate of decline of renal function prior to surgery (ΔeGFR) (table 4). Interestingly, 56% of HD– patients had at least one additional vascular access procedure prior to receiving any HD and this was higher than in the HD+ group (30%) (p = 0.03) (table 4). Patient Status at End of Study At the end of the study there were more patients in the eGFR >15 group than the eGFR ≤15 group that were alive independent of RRT [16/46 (35%) vs. 5/54 (9%)] (p = 0.003) (fig. 3). Of the 21 patients in this category, 19 still had a plan for future RRT, whereas 2 patients from the early AVF group changed their management plan to conservative care. At the end of the study, patients in the eGFR >15 group were less likely than the eGFR ≤15 group to be on any form of dialysis (HD or PD combined) [21/46 (46%) vs. 42/54 (78%), p = 0.002]. By study end, 7 patients had received a kidney transplant, 3 from the early group and 4 from the late group, and all these grafts remained functional. There were 6 deaths from the eGFR >15 group (13%) and 3 from the eGFR ≤15 group (6%) (p = 0.29). None of these deaths occurred before the need for RRT.
Discussion
The proportion of patients commencing HD with an AVF of 84% in this study compares favourably with rates reported elsewhere [18, 19], although the median eGFR of 15 ml/min/1.73 m2 at the time of surgery might be seen as relatively late AVF creation in relation to some existing recommendations (see table 5) [7, 8, 10, 20, 21]. These results were achieved by the use of a multidisciplinary vascular access clinic overseen by a renal vascular access 168
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
HD– (n = 27)
HD+ (n = 73)
p value
63.9±13.6 17 (63%) 17.4±4.7 317±110 –0.25±1.31 10/27 (37%) 15/27 (56%)
63.3±13.9 44 (60%) 14.8±4.9 364±102 –0.64±1.55 39/73 (53%) 22/73 (30%)
0.83 1.00 0.02 0.02 0.25 0.18 0.03
% 80
eGFR >15 H*)5 31
60
40
*
18
16 11
20 3 0
HD
4
3 PD
KT
5
6
No RRT
3 D
Fig. 3. Patient status at the end of the study (31/12/2011). Comparisons are shown between eGFR >15 and eGFR ≤15 AVF groups showing the % for each outcome for that group. Patients were classified as being on HD, peritoneal dialysis (PD), with a functioning kidney transplant (KT), alive and not requiring renal replacement (no RRT) or deceased (D). The actual number of patients in each category is indicated above the bars. Patients in the eGFR >15 group were more likely to be in the no RRT category (p = 0.003) and trended to be less likely to be in either the HD or PD group (p = 0.08). * p = 0.003.
coordinator. This multidisciplinary approach has previously been reported by others to achieve favourable vascular access outcomes [22, 23]. The present study also found a substantial proportion of patients undergoing AVF creation did not start HD for a prolonged interval, and many patients underwent subsequent procedures prior to commencing HD. A previous larger study also found that the time from AVF creation to the initiation of HD was highly variable and there was a subgroup of patients not requiring HD over a prolonged period of follow-up after AVF creation Lee/Roberts/Smith/Chuen/Mount
Table 5. Published recommendations on the timing of AVF surgery for CKD patients
Organisation
Recommendation
Kidney Disease Outcomes Quality Initiative [6]
A fistula should be placed at least 6 months before the anticipated start of HD treatments (B)
British Renal Association [21]
Guideline 3.1 – Planning of vascular access We suggest that planning for access should commence when patients enter CKD stage 4 (2C) Guideline 3.2 – Creation of vascular access We recommend that the exact timing of placement of vascular access will be determined by rate of decline of renal function, co-morbidities and by the surgical pathway (1C)
Canadian Society of Nephrology [8]
Establish AV fistulae when the patient has an estimated GFR of 15–20 ml/min and progressive kidney disease (Grade D, opinion)
European Best Practice Guidelines [10]
Potential chronic HD patients should be ideally referred to the nephrologist and/or surgeon for preparing vascular access when they reach the stage 4 of their CKD (GFR rate 15 group might have contributed to the higher CVC use in this group. Overall, these data suggest that while it is highly desirable to create an AVF at least several months before needing HD, cre-
Outcomes after AVF Creation in CKD
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
169
ation of an AVF years before HD is required probably does not further reduce the need for CVC insertion. It is known that some patients require multiple procedures for the successful establishment of vascular access for HD [11, 24]. The present study found that 37% of patients underwent at least one additional vascular access procedure prior to commencement of HD. Furthermore, 38% of patients who commenced HD had further vascular access intervention in the first 3 months after commencing HD. An advantage of earlier AVF creation might be that it allows more time for an AVF to develop; however, in some patients earlier AVF creation could increase the time available to develop stenosis or thrombosis, leading to additional access procedures before receiving HD. Furthermore, some patients could subsequently decide to have conservative care for ESRD or die from another cause before using their AVF. In these patients, creation of an AVF too early could lead to unnecessary procedures for an AVF that is never used. Most current guidelines recommend referral of CKD patients with a plan for HD to a surgeon for AVF assessment and creation when the eGFR is 15– 30 ml/min/1.73 m2, although the evidence level for these recommendations is recognised to be low quality (table 5) [7, 8, 10, 20, 21]. The present study suggests that in many patients creation of an AVF when the eGFR is ≥20 ml/min/1.73 m2 could be too early, exposing the patient to problems associated with vascular access long before the actual need for HD. Our data suggest that in some patients it might be safe to delay the creation of an AVF until the eGFR is ≤15 ml/min/1.73 m2. It is simplistic, however, to conclude that an arbitrary eGFR cut-off of 15 ml/min/1.73 m2, or another value, should be the primary consideration in determining the optimal timing of AVF surgery. This decision should be based on an estimate of the likelihood of HD being performed in a chosen time frame, based on all the available clinical data for an individual patient. O’Hare et al. [12] found that the mean annual GFR decline for a large group of CKD 4/5 patients was –2.3 ml/min/1.73 m2, emphasising that a higher eGFR at the time of AVF surgery will increase the risk and interval of AVF non-use. In the present study, the rate of decline in renal function in the 6 months prior to AVF surgery showed substantial variability and was slower in the eGFR >15 group; however, the ΔeGFR did not differ between the patients who did or did not receive HD. Overall, the available data indicate that the optimal timing for AVF surgery should take account both the eGFR and the predicted rate of decline of kidney function.
170
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
Whilst the risks of unnecessary or excessively early AVF surgery are not well defined, short- and long-term complications of AVF creation can be identified that should be considered when determining the timing of surgery. AVF creation leads to an immediate left to right shunt reducing systemic vascular resistance and increasing cardiac output, with the potential to precipitate or exacerbate a variety of cardiovascular problems including high output cardiac failure and pulmonary hypertension [25, 26]. As commonly seen in patients with severe CKD, the patients in our study had a high baseline rate of cardiac disease, indicating that this population may be particularly susceptible to any adverse cardiovascular impact of AVF surgery. Other important complications to consider with AVF surgery include the risk of hand ischaemia caused by steal syndrome [25–27] and the risk of immediate peri-operative complications. In conclusion, this study found in a cohort of pre-dialysis CKD patients undergoing AVF surgery with a median eGFR of 15 ml/min/m2 a low rate of CVC use at the time of commencing HD. In the patients with eGFR >15, the time to needing HD was longer and the risk of not doing HD over 2–5 years of follow-up was higher. This suggests that adherence to some current guidelines could result in AVF creation earlier than is clinically optimal in a substantial proportion of patients. A better understanding of the factors that influence the likelihood that HD will be needed within a period of time is required to avoid unnecessary procedures and their sequelae.
Acknowledgements The authors would like to thank Prof. David Power and Mr. Gary Fell for departmental support for this project.
Disclosure Statement The authors have no conflicts to interest to disclose.
References
1 Ravani P, Palmer SC, Oliver MJ, Quinn RR, MacRae JM, Tai DJ, Pannu NI, Thomas C, Hemmelgarn BR, Craig JC, Manns B, Tonelli M, Strippoli GF, James MT: Associations between hemodialysis access type and clinical outcomes: a systematic review. J Am Soc Nephrol 2013;24:465–473. 2 Xue JL, Dahl D, Ebben JP, Collins AJ: The association of initial hemodialysis access type with mortality outcomes in elderly Medicare ESRD patients. Am J Kidney Dis 2003; 42: 1013–1019.
Lee/Roberts/Smith/Chuen/Mount
3 Polkinghorne KR, McDonald SP, Atkins RC, Kerr PG: Vascular access and all-cause mortality: a propensity score analysis. J Am Soc Nephrol 2004;15:477–486. 4 Bradbury BD, Fissell RB, Albert JM, Anthony MS, Critchlow CW, Pisoni RL, Port FK, Gillespie BW: Predictors of early mortality among incident US hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Clin J Am Soc Nephrol 2007; 2:89–99. 5 Moist LM, Trpeski L, Na Y, Lok CE: Increased hemodialysis catheter use in Canada and associated mortality risk: data from the Canadian Organ Replacement Registry 2001–2004. Clin J Am Soc Nephrol 2008;3:1726–1732. 6 Vascular Access Work G: Clinical practice guidelines for vascular access. Am J Kidney Dis 2006;48(suppl 1):S248–S273. 7 Sidawy AN, Spergel LM, Besarab A, Allon M, Jennings WC, Padberg FT Jr, Murad MH, Montori VM, O’Hare AM, Calligaro KD, Macsata RA, Lumsden AB, Ascher E, Society for Vascular S: The Society for Vascular Surgery: clinical practice guidelines for the surgical placement and maintenance of arteriovenous hemodialysis access. J Vasc Surg 2008; 48(5 suppl):2S–25S. 8 Jindal K, Chan CT, Deziel C, Hirsch D, Soroka SD, Tonelli M, Culleton BF, Canadian Society of Nephrology Committee for Clinical Practice G: Hemodialysis clinical practice guidelines for the Canadian Society of Nephrology. J Am Soc Nephrol 2006;17(3 suppl 1):S1–S27. 9 Hoggard J, Saad T, Schon D, Vesely TM, Royer T, American Society of D, Interventional Nephrology CPC, Association for Vascular A: Guidelines for venous access in patients with chronic kidney disease. A Position Statement from the American Society of Diagnostic and Interventional Nephrology, Clinical Practice Committee and the Association for Vascular Access. Semin Dial 2008;21:186–191.
Outcomes after AVF Creation in CKD
10 Tordoir J, Canaud B, Haage P, Konner K, Basci A, Fouque D, Kooman J, Martin-Malo A, Pedrini L, Pizzarelli F, Tattersall J, Vennegoor M, Wanner C, ter Wee P, Vanholder R: EBPG on Vascular Access. Nephrol Dial Transplant 2007;22(suppl 2):ii88–ii117. 11 Huijbregts HJ, Bots ML, Wittens CH, Schrama YC, Moll FL, Blankestijn PJ, group Cs: Hemodialysis arteriovenous fistula patency revisited: results of a prospective, multicenter initiative. Clin J Am Soc Nephrol 2008;3:714– 719. 12 O’Hare AM, Bertenthal D, Walter LC, Garg AX, Covinsky K, Kaufman JS, Rodriguez RA, Allon M: When to refer patients with chronic kidney disease for vascular access surgery: should age be a consideration? Kidney Int 2007;71:555–561. 13 Weber CL, Djurdjev O, Levin A, Kiaii M: Outcomes of vascular access creation prior to dialysis: building the case for early referral. ASAIO J 2009;55:355–360. 14 Schinstock CA, Albright RC, Williams AW, Dillon JJ, Bergstralh EJ, Jenson BM, McCarthy JT, Nath KA: Outcomes of arteriovenous fistula creation after the Fistula First Initiative. Clin J Am Soc Nephrol 2011; 6: 1996– 2002. 15 Oliver MJ, Quinn RR, Garg AX, Kim SJ, Wald R, Paterson JM: Likelihood of starting dialysis after incident fistula creation. Clin J Am Soc Nephrol 2012;7:466–471. 16 Blake PG, Quinn RR, Oliver MJ: The risks of vascular access. Kidney Int 2012;82:623–625. 17 Mathew TH, Australasian Creatinine Consensus Working G: Chronic kidney disease and automatic reporting of estimated glomerular filtration rate: a position statement. Med J Aust 2005;183:138–141. 18 Centers for M, Medicaid S, Kinney R: 2005 Annual Report: ESRD Clinical Performance Measures Project. Am J Kidney Dis 2006; 48(4 suppl 2):S1–S106.
19 Lopez-Vargas PA, Craig JC, Gallagher MP, Walker RG, Snelling PL, Pedagogos E, Gray NA, Divi MD, Gillies AH, Suranyi MG, Thein H, McDonald SP, Russell C, Polkinghorne KR: Barriers to timely arteriovenous fistula creation: a study of providers and patients. Am J Kidney Dis 2011;57:873–882. 20 FistulaFirst. Fistula First Change Concepts. www.fistulafirst.org/Professionals/FFBI ChangeConcepts.aspx. 21 Fluck R, Kumwenda M: Renal Association Clinical Practice Guideline on vascular access for haemodialysis. Nephron Clin Pract 2011; 118(suppl 1):c225–c240. 22 Allon M, Bailey R, Ballard R, Deierhoi MH, Hamrick K, Oser R, Rhynes VK, Robbin ML, Saddekni S, Zeigler ST: A multidisciplinary approach to hemodialysis access: prospective evaluation. Kidney Int 1998;53:473–479. 23 Dwyer A, Shelton P, Brier M, Aronoff G: A vascular access coordinator improves the prevalent fistula rate. Semin Dial 2012; 25: 239–243. 24 Voormolen EH, Jahrome AK, Bartels LW, Moll FL, Mali WP, Blankestijn PJ: Nonmaturation of arm arteriovenous fistulas for hemodialysis access: A systematic review of risk factors and results of early treatment. J Vasc Surg 2009;49:1325–1336. 25 MacRae JM, Pandeya S, Humen DP, Krivitski N, Lindsay RM: Arteriovenous fistula-associated high-output cardiac failure: a review of mechanisms. Am J Kidney Dis 2004; 43:e17– e22. 26 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;20:1686–1692. 27 Beathard GA, Spergel LM: Hand ischemia associated with dialysis vascular access: an individualized access flow-based approach to therapy. Semin Dial 2013;26:287–314.
Blood Purif 2014;37:163–171 DOI: 10.1159/000360273
171
Copyright: S. Karger AG, Basel 2014. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.
