The independent association of preoperative serum albumin on the functional maturation of radiocephalic arteriovenous fistulae

Abstract

Purpose

The aim of this study is to test the null hypothesis that preoperative albumin along with other preoperative confounders have no impact on the functional maturation of radiocephalic arteriovenous fistulae (RCAVF).

Methods

A retrospective cohort study of n = 195 individuals undergoing RCAVF formation from July 2013 to December 2015 was conducted. The null hypothesis was assessed through chi squared test. Independent association of each variable was evaluated through univariate and multivariate logistic regression model. Pearson's correlation test was also performed between scale variables to establish their causal link.

Results

Preoperative hypoalbuminaemic group of individuals demonstrated significant failure of maturation (49.3% vs. 27.2%, p = 0.002). At multivariate analysis, hypoalbuminemia remained an independent marker of fistula failure (OR 0.40, 95% CI 0.21-0.76, p = 0.004) and demonstrated a weak but a positive correlation at the endpoint of maturation (R = 0.223, p = 0.002).

Conclusions

Preoperative hypoalbuminemia (<35 mg/dL) is independently associated with 40% reduction in the functional maturation of RCAVF. Stratification of this readily available biomarker prior to RCAVF formation may require consideration subjected to further research.

Keywords

Albumin Comorbidities Functional maturation Haemodialysis Radiocephalic arteriovenous fistula Research

Introduction

In recent years, the role of various preoperative haematological and biochemical markers as an independent prognostic factor, has gained significant attention in the clinical and academic surgery (1–2–3). Autogenous radiocephalic arteriovenous fistulae (RCAVF) remain the primary vascular access choice for haemodialysis patients. Failure of functional maturation (FM) secondary to thrombosis and/or stenosis remains a major obstacle (4, 5). To date, multiple studies have assessed and evaluated the impact of different comorbidities, anatomy and technical issues on the FM of RCAVFs. Despite recognition of these factors and their stratification, a considerable proportion of RCAVFs failed to mature (6, 7). From the time of fistula formation to its FM, the fistula undergoes significant haemodynamic and pathophysiological changes. This process results in intraluminal (neo-intimal hyperplasia, activation of coagulation) and extraluminal (inflammatory and proliferative phase of wound healing, oedema, collagen and extracellular matrix deposition) alterations (8). In recent years, preoperative circulatory serum albumin as a prognostic marker of success and/or failure within general surgery has been extensively researched (3). However, insight into the role of this single protein in vascular access surgery remains unknown. Circulatory albumin (Alb) plays an important role in maintaining the oncotic pressure (colloid osmotic pressure) of the vasculature. Different levels of serum albumin (Alb), represent different pathophysiological conditions (dehydration, malnutrition, poor tissue healing, oedema) that can consequently affect wound healing process, inflammation and fistula flow rate (9). In this study, we hypothesised, that the preoperative (1-week) albumin (Alb) in addition to other variables has no impact on the endpoint of FM of RCAVFs.

Material and methods

A retrospective consecutive cohort study on 195 patients undergoing only (Brescia-Cimino) arteriovenous fistulae (autogenous RCAVF) at our unit, from July 2013 to December 2015 was conducted. The permission to access the renal registry data was granted through audit number CA13-225. Variables included patient demographics (age, gender), co-morbidities (diabetes mellitus [DM], ischaemic heart disease [IHD], congestive cardiac failure [CCF] and hypertension [HTN]), anaesthesia type (general, regional and local) Albumin (Alb), and haemoglobin (Hb).

Functional maturation was defined as the rule of 6s. Assessed clinically and with duplex ultrasonography at six weeks’ post-RCAVF formation, with a depth of not more 0.5-0.6 cm from the skin and diameter (main body of fistulae) of 6 mm with the flow rate of 600 mL/min and length of 5-6 cm for successful 2-needle dialysis (10). All comorbidities were defined in accordance to definitions provided by WHO (11). Normal albumin ranged from 35-50 g/L. Any value below this was considered as hypoalbuminemia.

All patients had a full pre-operative consultation by a vascular access consultant and a specialist dialysis nurse and postoperatively. This included full clinical and duplex examination (1 week) prior to the operation. The cephalic vein was found suitable if the tap test was positive and was continuous to the median cubital fossa and/or cephalic vein of arm directly or indirectly with a consistent diameter (12). Tap test was considered normal (positive) if, after application of tourniquet at arm, vibration would be felt at distal cephalic vein from by gentle tapping on its proximal site (wrist). The radial artery was used and assessed further with ultrasound if Allen's test was found normal. Pre-operative duplex of both vessels (radial artery and cephalic vein) assessed the internal diameter with linear transducer of 5-7 MHz, with arm position fully rested at 45-60 degrees. Haemodynamic study (flow and stenosis) was performed along with prior assessment in a non-augmented state. A cut of value of 1.8 mm ± 0.1 mm for both cephalic vein and radial artery was set if no haemodynamic issues were identified. All fistulae were created by a consultant and/or a senior surgeon with end (cephalic vein) to side (radial artery) anastomosis using 2.5 magnifying lenses with 6/0 monofilament polypropylene continuous suture from heel to base with parachute technique. All incision was similarly closed with 3/0 interrupted vicryl for subcutaneous tissue and 3/0 vicryl rapid for skin. We routinely do not prescribe antiplatelet and/or anticoagulation postoperatively.

Statistical analysis

All the statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) version 20, IBM. All continuous variables were reported as medians with the corresponding interquartile ranges (IQRs) and percentage. The data were analysed in three different ways to test the null hypothesis in a robust methodology to avoid conclusion bias. Initially, based on categorical data and endpoint of functional maturation (maturation vs. non-maturation), the data were subjected to chi-squared test and outcome was presented with percentage and p value. Second, all confounders were initially evaluated through univariate regression analysis to determine the association with the endpoint of functional maturation. Those variables that were significant (p<0.05) were then entered into a multivariate regression model. At multivariate analysis, factors that were significant, were deemed to be independently associated with fistulae maturation. Associations were reported as odds ratio (OR) with the accompanied measure of uncertainty (95% confidence interval [CI]) as well as the test of the statistical significance (p value). In addition, a Pearson's correlation test was also performed between two scale variables of albumin and fistulae maturation. This was reported with the test of the statistical significance (p value) and Pearson's correlation coefficient value R.

Results

The median age of the cohort was 68 years (IQR, 57-78) of which 73.8% (144/195) were male. Functional maturation was achieved in 64.1% (125/195) of the entire cohort. Local anaesthesia was used in most individuals (n = 193, 99%). No one received regional anaesthesia (limb block). The most common comorbidity was hypertension (n = 159, 81%) followed by diabetes mellitus (n = 73, 37%) (DM type I n = 14 vs. DM type II n = 59), ischemic heart disease (n = 41, 21%) and congestive cardiac failure (n = 9,4.6%).

Categorical data analysis (chi-squared test) on the basis of maturation versus non-maturation, demonstrated female patients to possess a higher prevalence of fistula (RCAVF) failure than male patients (50.0% vs. 30.6%, p = 0.013). This was also noted in individuals with ischaemic heart disease (IHD) (50.0% vs. 31.6%, p = 0.030) and preoperative hypoalbuminemia (≤35 g/L) in comparison to normal serum albumin levels (49.3% vs. 27.2%, p = 0.002). However, no other variable was found to be of significance (Tab. I).

TABLE I

Categorical data analysis (matured vs. non-matured) through chi-squared test

		Not matured		Matured		p
		n	%	n	%	p
Gender	Female	25	50.0%	25	50.0%	0.013
	Male	44	30.6%	100	69.4%
Age (y)	18-68 y	40	39.6%	61	60.4%	0.265
	>68 y	29	31.9%	62	68.1%
Anaesthesia	Local	68	35.4%	124	64.6%	0.668
	General	1	50.0%	1	50.0%
IHD	No IHD	48	31.6%	104	68.4%	0.030
	IHD	20	50.0%	20	50.0%
CHF	CHF	62	34.1%	120	65.9%	0.187
	No CHF	5	55.6%	4	44.4%
Hypertension	No Hypertension	11	31.4%	24	68.6%	0.603
	Hypertension	57	36.1%	101	63.9%
DM status	No DM	39	34.2%	75	65.8%	0.874
	DM type I	4	30.8%	9	69.2%
	DM type II	22	37.3%	37	62.7%
Albumin	<35	35	49.3%	36	50.7%	0.002
	≥35	31	27.2%	83	72.8%
Hb	<113 g/dL	35	39.8%	53	60.2%	0.268
	≥113 g/dL	31	32.0%	66	68.0%

IHD = ischaemic heart disease; CHF = congestive heart failure; DM = diabetes mellitus; Hb = haemoglobin.

At univariate analysis, gender (p = 0.014), IHD (p = 0.032) and serum albumin (p = 0.003) were found to be significantly associated with fistulae maturation and were therefore entered into a multivariate regression model (Tab. II). At multivariate regression analysis, 52% of the fistulas in male patients failed to mature at 6 weeks after surgery to that of females (OR 0.48, 95% CI 0.23-0.98, p = 0.043). Similarly, patients with an albumin equal to or above 35 g/dl had a 60% lower chance of fistula failure (OR 0.40, 95% CI 0.21-0.76, p = 0.004). In addition, variables of age, anaesthetic type, CHF, DM, HTN and Hb failed to reach any statistical significance (Tab. II).

TABLE II

Univariate and multivariate regression model for independent association of each confounder

		Univariate	Multivariate
		Univariate	OR	95% CI	p
Gender	Female	0.014	1 (Reference)
	Male		0.48	0.23-0.98	0.043
Age at surgery (y)	18-68 y	0.265
	>68 y
Anaesthesia	Local	0.673
	General
IHD	No IHD	0.032	1 (Reference)
	IHD		2.15	1.01-4.60	0.047
CHF	CHF	0.200
	No CHF
Hypertension	No Hypertension	0.603
	Hypertension
DM status	No DM	0.710
	DM type I
	DM type II
Albumin	<35	0.003	1 (Reference)
	≥35		0.40	0.21-0.76	0.005
Hb	<113 g/dL	0.269
	≥113 g/dL

IHD = ischaemic heart disease; CHF = congestive heart failure; DM = diabetes mellitus; Hb = haemoglobin.

The relationship between albumin and fistula maturation as a scale level was further explored in the form of Pearson correlation. There was a weak but statistically significant correlation between pre-operative albumin level and fistula formation (correlation coefficient R = 0.223, p = 0.002) (Tab. III).

TABLE III

Pearson correlation test for scale variables of albumin to the endpoint of functional maturation

		Albumin	Maturation status
Albumin	Pearson correlation	1	0.223^**
	Sig. (2-tailed)		0.002
	n	185	185
Maturation status	Pearson correlation	0.223^**	1
	Sig. (2-tailed)	0.002
	n	185	194

Correlation is significant at the 0.01 level (2-tailed).

Discussion

It is estimated that 75% of the total colloid osmotic pressure is related to the circulatory serum albumin levels. Albumin as a single polypeptide plays an important role in cell membrane transportation mechanism, diffusion, osmosis, filtration and active transportation in vasculature (13). In the study by Unver et al, hypoalbuminemia had a significant negative impact on the blood flow rate post-AVF creation and haemodialysis as compared to individuals with normal serum levels (p = 0.001) (14), attributed to decreased in the effective circulatory volume (ECV). A reduction in the oncotic pressure coupled with decrease in ECV results in subsequent decrease of cardiac output (13, 14).

Another important function of albumin is platelet function inhibition and antithrombotic effect, meaning that those patients with low albumin may be predisposed to fistulae failure due to an increased risk of thrombosis exacerbated by low flow state (reduced cardiac output) (15, 16). Aforementioned factors explain why hypoalbuminaemic group of patients maintain a significantly higher incidences of fistulae failure (49.3% vs. 27.2%, p = 0.002) in our study and the study by Kaygin et al (90.6% vs. 32.4%, p<0.001) (17). Furthermore, it appears that hypoalbuminemia (≤35 g/L) is also associated with increased mortality, morbidity and postoperative complications of haemodialysis patients (18, 19).

It is understood that albumin is a negative-phase reactant and decreases in an active inflammation. In chronic renal insufficiency, serum albumin concentrations decrease (20). This is attributed to continuous inflammation and recent research suggests that preoperative inflammation reflected in haematological markers remains an independent predictor of fistula failure in clinical and experimental research (21). It has been suggested that apart from thrombosis, stenosis secondary to neo-intimal hyperplasia remains an important factor in failure of functional maturation and neutrophil-to-lymphocyte ratio (NLR) could be reliable and a strong inflammatory indicator for such events (21, 22). In a recent study by Yilmaz et al (21), the severity and location of newly created fistula stenotic lesions (23) were evaluated against NLR and the authors identified that NLR is a strong and independent predictor of AVF stenosis (21, 22). This was based on the histological similarities of neo-intimal hyperplasia and that of atherosclerosis and their strong association with preoperative NLR which could reflect itself with low levels of serum albumin in such individuals (24).

The functional maturation of 64% in this cohort is similar to some and perhaps superior to other reports (22, 25). In our unit, we adhere to a strict policy for reporting functional maturation (“rules of 6s”). However, if we include those that had late maturation and/or intervention to gain a functional maturation the number of functional fistulas could be more than what is reported in our cohort. For any surgically created arteriovenous fistula to mature, an inflow and outflow remains crucial and judicious patient selection remains vital. However, in ischaemic heart disease, the inflow due to akinesia, dyskinesia of myocardium will result in decreased cardiac output and stroke volume (26). This perhaps could be the reason why this group of patients had significantly lower number of functional maturation in our unit (IHD) (50.0% vs. 31.6%, p = 0.030). Another outcome of our study is related to higher prevalence of fistula failure in female patients (50.0% vs. 30.6%, p = 0.013). This outcome seems to be in line with earlier reports and it is mainly related to smaller radial artery in female population, decrease elasticity, expansibility and lack of cardioprotection following menopause (27, 28).

The clinical impact of preoperative hypoalbuminaemia in vascular and specifically vascular access surgery, is poorly researched and to the best of our knowledge this is the largest study to date that has specifically investigated this biochemical marker with regards to FM of RCAVFs. Overall, it appears that the circulatory albumin could be the representative of far more than what was traditionally understood.

Strengths and limitations

The main limitation of this study is related to its retrospective nature and a prospective and/or comparative study would have been more ideal. In addition, availability of body mass index (BMI) of every individual and calculation of protein catabolic rate (PCR) could have also increased the validity of this study and its application to practice (29). However, this study could provide the platform for further research into association of various biochemical markers and that of thrombosis and/or stenosis that constitute a significant proportion of fistula failure in daily practice. This study benefits from an adequate power of a single entity fistulae but direct recognition of type and/or reason for failure (stenosis and/or thrombosis) would have significantly increased the validity of this study.

Conclusion

Preoperative hypoalbuminemia (<35 mg/dL) is independently associated with 40% reduction in the functional maturation of RCAVF. However, their contribution to thrombosis and/or stenosis is not fully understood and further research in this area is highly advocated. Furthermore, it appears that failure of functional maturation might be dependent on more than just comorbidities, anatomical factors, technical issues and patient's demographics. The evaluation of biochemical and haematological factors at micro- and macro-circulation is highly recommended.

Footnotes

Financial support: No grants or funding have been received for this study.

Conflict of interest: None of the authors has financial interest related to this study to disclose.

References

Meyer

Biochemical markers in the surgical intensive care: identifying critically ill surgical patients with complications.

Erasmus University, Rotterdam, The Netherlands:

Datawyse Universitaire Pers Maastricht

2015. Available from repub.eur.nl/pub/78340/150702_Meyer-Zainna-Carolina.pdf. Accessed November 22, 2016.

Hübner

Mantziari

Demartines

et al.

Postoperative albumin drop is a marker for surgical stress and a predictor for clinical outcome: a pilot study.

Gastroenterol Res Pract 2016 2016 8743187

Truong

Hanna

Moghadamyeghaneh

Stamos

Implications of preoperative hypoalbuminemia in colorectal surgery.

World J Gastrointest Surg 20168 5 353–362

Hayashi

Huang

Nissenson

Vascular access for hemodialysis.

Nat Clin Pract Nephrol 20062 9 504–513

Stolic

Most important chronic complications of arteriovenous fistulas for hemodialysis.

Med Princ Pract 201322 3 220–8

Masengu

Maxwell

Hanko

Investigating clinical predictors of arteriovenous fistula functional patency in a European cohort.

Clin Kidney J 20169 1 142–147

Kordzadeh

Chung

Panayiotopoulos

Cephalic vein and radial artery diameter in formation of radiocephalic arteriovenous fistula: a systematic review.

J Vasc Access 201516 6 506–511

Roy-Chaudhury

Sukhatme

Cheung

Hemodialysis vascular access dysfunction: a cellular and molecular viewpoint.

J Am Soc Nephrol 200617 4 1112–1127

Farrugia

Albumin usage in clinical medicine: tradition or therapeutic?

Transfus Med Rev 201024 1 53–63

10.

Banerjee

Fistula maturation and patency for successful dialysis.

Dial Transplant 2009 38 11 442

11.

Global Status Report on non-communicable diseases 2014. Attaining the nine global non-communicable diseases targets; a shared responsibility.

WHO; 2015. Available from http://apps.who.int/iris/bitstream/10665/148114/1/9789241564854_eng.pdf?ua = 1. Accessed November 22, 2016.

12.

Lee

Won

Jung

Clinical significance of upper-arm cephalic vein patency in autogenous radial-cephalic wrist fistulas for hemodialysis.

Eur J Vasc Endovasc Surg 201244 5 514–520

13.

Schrier

Decreased effective blood volume in edematous disorders: what does this mean?

J Am Soc Nephrol 200718 7 2028–2031

14.

Unver

Atasoyu

Evrenkaya

Effects of comorbidity from AV fistula insufficiency on fistula blood flow rate in hemodialysis patients.

Dial Transplant 200635 11 682–688

15.

Don

Kaysen

Serum albumin: relationship to inflammation and nutrition.

Semin Dial 200417 6 432–437

16.

Hockensmith

Estacio

Mehler

et al.

Albuminuria as a predictor of heart failure hospitalizations in patients with type 2 diabetes.

J Card Fail 200410 2 126–131

17.

Kaygin

Halici

Aydin

et al.

The relationship between arteriovenous fistula success and inflammation.

Ren Fail 2013 35 8 1085–1088

18.

Marckmann

Nutritional status and mortality of patients in regular dialysis therapy.

J Intern Med 1989226 6 429–432

19.

Dwyer

Larive

Leung

et al;

HEMO Study Group.

Are nutritional status indicators associated with mortality in the Hemodialysis (HEMO) Study? Kidney Int 200568 4 1766–1776

20.

Owen

Lowrie

C-reactive protein as an outcome predictor for maintenance hemodialysis patients.

Kidney Int 1998 54 2 627–636

21.

Yilmaz

Bozkurt

Cakmak

et al.

Relationship between late arteriovenous fistula (AVF) stenosis and neutrophil-lymphocyte ratio (NLR) in chronic hemodialysis patients.

Ren Fail 201436 9 1390–1394

22.

Bashar

Zafar

Ahmed

et al.

Can a Neutrophil–Lymphocyte ratio derived from Preoperative blood tests predict Arteriovenous Fistula maturation?

Ann Vasc Surg 201635:60–67

23.

Clark

Hirsch

Jindal

Veugelers

LeBlanc

Outcome and prognostic factors of restenosis after percutaneous treatment of native hemodialysis fistulas.

J Vasc Interv Radiol 2002 13 1 51–59

24.

Gagliardi

Rossi

Condino

et al.

Malnutrition, infection and arteriovenous fistula failure: is there a link?

J Vasc Access 201112 1 57–62

25.

Lok

Oliver

Overcoming barriers to arteriovenous fistula creation and use.

Semin Dial 200316 3 189–196

26.

Lip

Gibbs

Beevers

ABC of heart failure: aetiology.

BMJ 2000320 7227 104–107

27.

Beniwal

Bhargava

Kausik

Size of distal radial and distal ulnar arteries in adults of southern Rajasthan and their implications for percutaneous coronary interventions.

Indian Heart J 201466 5 506–509

28.

Winer

Sowers

Weber

Gender differences in vascular compliance in young, healthy subjects assessed by pulse contour analysis.

J Clin Hypertens (Greenwich) 20013 3 145–152

29.

Panzetta

Tessitore

Faccini

Maschio

The protein catabolic rate as a measure of protein intake in dialysis patients: usefulness and limits.

Nephrol Dial Transplant 19905 Suppl 1 125–127