Dipartimento di Ingegneria informatica, automatica e gestionale

Background and Aims: The delivery of an adequate haemodialysis (HD) dose (i.e., a target single-pool Kt/V (spKt/V) of 1.4 per session) is often challenging, especially in patients with huge urea distribution volumes (V). On the other hand, patients with an extremely small V, such as children, are at higher risk of disequilibrium syndrome, which is caused by an excessive or overly rapid correction of urea levels, particularly during the first HD session. The treatment prescription often neglects many elements contributing to the final delivered HD dose, whereas a systematic and comprehensive evaluation of all adjustable parameters could improve the management of patients, not only facilitating the achievement of treatment goals, but also avoiding short- and long-term complications. We aimed at developing a web app capable of predicting the delivered HD dose, given a wide range of parameters, and using renowned kinetic equations relevant to urea clearance in HD. Method: The platform used to develop our app is based on Angular, an open-source framework to develop web applications. Specifically, it can be used to realise single-page apps and it is only for frontend. This fits our case, since a backend is not required due to the absence of huge databases or complex operations. The application is grounded on a stepper component, a paradigm to guide the user step-by-step during the data entry process. It consists of three phases, namely (1) patient, (2) haemodialyser, (3) target spKt/V. The last section is also for output. A modification of any input triggers a real-time output change. Results: We developed a free access user-friendly web app, through which HD professionals can simulate a treatment prescription and obtain an estimate of the delivered dose in terms of spKt/V. The app is structured in three sections and all calculations are based on validated formulas and kinetic equations applicable to HD: (1) patient’s V is calculated using either the Watson or the Morgenstern formula for adult and paediatric patients, respectively. Effective inlet blood flow (Qe), corresponding to the blood water flow, is derived from blood flow (Qb), considering the patient’s haematocrit and plasma protein levels (Fig. 1A) (2) haemodialyser can be selected from a provided list, which should include almost all filters on the market. Mass transfer-area coefficient (KoA) “in vitro” is calculated using the Michaels equation from urea clearance values reported on available product data sheets. KoA “in vitro” is converted to expected KoA “in vivo”, as suggested by Daugirdas (Fig. 1B). (3) expected spKt/V is calculated based upon several parameters (Fig. 1C): - previously established V and KoA - ultrafiltration volume (UF) and duration of the HD session (t) - Qb and dialysate flow (Qd) and their reciprocal direction (co-current or counter-current) - estimated recirculation. The output “expected spKt/V” changes in real time, after any modification of the input parameters. The result can thus be compared to the treatment goal (i.e., 1.4 for a standard HD session; 0.55 for a first treatment, aiming at reducing the risk of disequilibrium syndrome), as suggested by international guidelines and expert opinions. Conclusion: We developed a user-friendly web app, which will be freely accessible online by HD professionals, aimed at predicting the delivered HD dose (spKt/V) considering a wide range of patient- and prescription-related parameters. The application of such a tool in everyday clinical practice, particularly in cases requiring a more accurate definition of the actual delivered dose, could improve the management of patients through a personalisation of the treatment. This could enhance prescription adjustments aimed at achieving an adequate spKt/V in patients receiving an insufficient HD dose; on the other hand, it could help preventing disequilibrium syndrome in patients with