The association of initial hemodialysis vascular access type with morbidity and mortality outcomes in ESRD patients

• Main Authors: Abdul Rahim, Nurul Nadiah Puteri, Mohamad Nor, Fariz Safhan, Seman, Mohd Ramli, Ahmad, MK, Wan Ali, Wan Ahmad Syahril Rozli, Draman, Che Rosle
• Format: Article
• Language: English
• Published: 2015
• Subjects: R Medicine (General)
• Online Access: http://irep.iium.edu.my/58968/18/vasc%20access%20and%20ktv%20ishd.pdf; http://irep.iium.edu.my/58968/; http://onlinelibrary.wiley.com/doi/10.1111/hdi.12368/full

Introduction There is growing evidence that the type of initial hemodialysis (HD) access significantly influenced patient’s outcome. The arteriovenous fistula (AVF) is well regarded as the best available initial access for providing HD. Objective The aim of this study is to determine the pattern of initial HD access in our dialysis centers and to evaluate its associated outcome of infection, all causes hospitalisations and mortality. Methods Incident ESRD patients initiated on HD in all East Pahang, Malaysia dialysis centers between 1 January 2014 and 31 December 2014 were included retrospectively. Baseline demographics were recorded and initial dialysis accesses were grouped into AVF or central venous catheters (CVC). The outcome events of infection, all cause hospitalization and mortality were analyzed. Results 184 patients were enrolled, 132 (71.7%) in CVC group and 52 (28.3%) in AVF group. Overall, all types infection occurred in 14.1%, all cause hospitalization in 15.8% and mortality in 3.3% of patients. AVF group were older, 57.9 + 13.1 compared to CVC group, 52.9 + 14.2, p = 0.03. There was no difference in gender, race, diabetes mellitus, hypertension and body mass index between the two groups. In the AVF group, the fistulas created were mainly left brachiocephalic fistula in 46.2%, followed by left radiocephalic fistula in 23.3% and right brachiocephalic fistula in 15.4%. In the CVC group, 93.2% catheters inserted were of right internal jugular catheter. The mean duration on catheter was 3.9 + 1.9 months. 88.6% in CVC group were not properly prepared for AVF. There were higher, not statistically significant, incidence of all types infections in CVC compared to AVF, 16.7% vs. 7.7%, p = 0.086. All causes hospitalisations were higher in CVC, 26 (19.7%) compared to AVF, 3 (5.8%), p = 0.013. There was no significant difference in adequacy of dialysis (kt/v) as well as patient survival. The mean serum albumin level is lower and serum phosphate is higher in CVC. Duration on catheter was associated with increased infection (5.15 + 2.1 months vs 3.68 + 1.8, p = 0.001) and increased hospitalization (5.31 + 2.1 months vs 3.59 + 1.6, p = 0.001), but no difference in mortality. Conclusion Initial use of AVF was low in our incident HD patients. Majority of those initiated on CVC were not adequately prepared either late referral to nephrologist or patient refusal. Initial use of AVF and shorter duration on CVC was associated with better outcome in terms of reduced hospitalization and infection. The survival was comparable most likely due to short duration of study. Further prospective studies need to look at ways to improve AVF preparation.