Decreased access resistance in haemodialysis patients with upper arm arteriovenous fistulae

Sir, We have read with interest the recent publication of Basile et al. [ 1 ], who identified high vascular access flow (Qa) as a risk factor for high output failure in haemodialysis patients. Furthermore, they reported that the cardiopulmonary recirculation (CPR, Qa/CO ratio) is higher in upper arm arteriovenous fistulae (AVFs) than in forearm AVFs. We confirm these findings and made relevant additional observations in our haemodialysis population (1) by validating the transonic CO measurement technique, (2) by including a group of patients with PTFE graft and (3) by calculating the access resistances. Because quantification of CO by the transonic technique has not yet been validated against another modality, we compared this technique to echocardiography (using standard transthoracic two-dimensional and Doppler echocardiographic recordings). The measurements were performed during the first hour of a dialysis session, in random order and shortly after each other. We compared the techniques in 35 stable haemodialysis patients. The correlation coefficient was high ( r = 0.95, P < 0.001), and Bland–Altman analysis showed no regression of the differences versus the means ( r = 0.03, Figure 1 ), indicating that measurements obtained with both techniques do not differ significantly. Bland–Altman plot of CO measured by transonic and Doppler echo techniques. Secondly, we wished to address the idea that the finding that CO is higher in upper arm AVF is indeed explained by the AVF and not by a higher systemic blood pressure or a lower systemic resistance. In the study of Basile et al. [ 1 ] no blood pressure values were presented. We measured CO, Qa and mean arterial pressure (MAP, contralateral arm) in 32 patients with upper arm AVFs, 39 with forearm AVFs and 15 with PTFE grafts. Subsequently, we calculated total peripheral vascular resistance (TPVR = MAP/CO), access resistance (AR = MAP/Qa) and systemic vascular resistance (SVR = MAP/(CO − Qa). Similar to the findings of Basile et al. [ 1 ], the relation between CO and Qa could be described by a third-order polynome ( y = 3.2 + 2.0 x + 0.44 x2 − 0.20 x3 ; R2 = 0.45, P < 0.001). As shown in table 1 , Qa, CO, CPR and Qa adjusted for CO (by analysis of covariance) were significantly higher in the AVF upper arm group than in both the AVF forearm and the PTFE groups, whereas AR was significantly lower (one-way ANOVA, studentized Newman–Keuls post hoc test). In contrast, MAP and SVR were almost numerically equal in the three groups, being far from significantly different. Shunt flows and resistances of haemodialysis patients with various shunt types Significances versus AVF upper arm group: *P < 0.001; †P < 0.01; ¶P < 0.02. a Qa adjusted by analysis of covariance for variations related to variations in CO. Qa = access flow, CO = cardiac output, CPR = cardiopulmonary recirculation, MAP = mean arterial pressure, TPVR = total peripheral vascular resistance, AR = access resistance, SVR = systemic vascular resistance. Shunt flows and resistances of haemodialysis patients with various shunt types Significances versus AVF upper arm group: *P < 0.001; †P < 0.01; ¶P < 0.02. a Qa adjusted by analysis of covariance for variations related to variations in CO. Qa = access flow, CO = cardiac output, CPR = cardiopulmonary recirculation, MAP = mean arterial pressure, TPVR = total peripheral vascular resistance, AR = access resistance, SVR = systemic vascular resistance. We validated the transonic CO technique, and we confirm and extend the findings presented by Basile et al. [ 1 ] by showing that the higher CO in upper arm AVF is a result of the lower TPVR, caused by a lower AR and not by a lower SVR. Conflict of interest statement . None declared.