Vascular access and survival of patients with hemodialysis: features of cause-effect relationship

Aim: to analyze features of the causal relationship between the vascular access type at the time of hemodialysis (HD) start and survival rates of patients, taking into account the cause of chronic kidney disease (CKD) and comorbidity.Materials and methods. The retrospective analysis included 604 HD patients divided into three groups: «AVF» (n = 336) patients started and continued HD with AVF; «CVC-AVF» (n = 152) patients started HD with CVC and later successfully converted to AVF; «CVC» (n = 116) patients who started and continued HD with CVC only. Patients with other types of conversions were not included in the analysis. The mean follow-up period since the beginning of HD was 38 [interquartile range 19; 48] months.Results. Unadjusted survival rate after 5 years in the AVF group was 61% [95%CI 51.8; 71.9], that in the CVC-AVF group - 53.9% [95%CI 42.5; 67], and that in the CVC group - 31.6% [95% CI 21.4; 41.4]. Survival rate in the CVC group varied from that in the AVF (p < 0.0001) and CVC-AVF (p < 0.0001) groups. CVC-AVF and CVC groups patients had significantly worse comorbidity than that of AVF group patients. After adjustment for comorbidity, age, sex, and cause of CKD, the survival rate in the groups after 5 years came to the following: 56.7% [95%CI 51.1; 62.8] in the AVF group, 51.7% [95%CI 42.5; 61.7] in the CVC-AVF group, 33.3% [95%CI 24; 42.8] in the CVC group. The results in the AVF group differed significantly from that in the CVC group (p < 0.001), but not from that in the CVC-AVF group (p = 0.425). The results in the CVC-AVF group are also statistically significantly varied from that in the CVC group (p = 0.009). Diabetes mellitus and systemic diseases were important risk factors. In the 5 years’ time period the survival rate of the group of patients with diabetes mellitus within in the AVF group adjusted (for sex, age, cause of CKD and comorbidity) was 38.1% [95% CI 29; 47.1], that in the CVC-AVF group - 29.7% [95% CI 18.9; 41.2] and that in the CVC group - 20.3% [95% CI 11.6; 31.8]. The results in the AVF group statistically significantly differed from that in the CVC group (p = 0.001), and from that in the CVC-AVF group (p = 0.011). The results in the CVC-AVF group are also statistically significantly varied from that in the CVC group (p = 0.021). In the 5 years’ time period the adjusted survival rate within the patients in the AVF group with systemic processes, was 34.2% [95% CI 18.8; 50.3], that in the CVC-AVF group - 23.9% [95% CI 10.5; 40.3], and that in the CVC group - 20.5 % [95% CI 7.3; 38.5]. We did not note statistically significant differences between the groups (p > 0.05 in all cases).Conclusion. The HD beginning with the use of CVC does not increase the risk of death in case of successful conversion to AVF. The use of CVC as the only vascular access is associated with a significant increase in the adjusted risk of death. Within the patients with diabetes mellitus, the use of CVC is associated with a deterioration of the adjusted survival rate even with subsequent successful conversion to functional AVF. Patients with systemic processes (vasculitis, myeloma, HIV-associated nephropathy, renal neoplasms, etc.) have low predicted survival rate disregarding the type of vascular access (there are no significant differences between the types of vascular access). The differences in survival rates are determined not only by the types of vascular access, but also by the comorbid background