Reply to Condello et al.

To the Editor,

We thank Ignazio Condello, PhD, Massino Vincenzo Bonfantino, MD, and Giuseppe Nasso, MD, for their comments postulated in their letter to the editor [1] as well as for their contributions in the field of cardiac surgery. As it was stated in our article [2], our primary goal was to investigate perioperative factors and their association with fast-track failure (FTF) in a retrospective cohort study of patients undergoing minimally invasive mitral valve surgery (MIMVS). We were indeed able to show that factors associated with FTF in patients with Carpentier type I and II pathologies undergoing MIMVS are an NYHA III–IV at baseline, pre-existing chronic kidney disease, and coronary artery disease. Postoperative bleeding requiring rethoracotomy and procedure time was also identified as important factors associated with failed FT [2].

We agree with Condello et al. that the creation of a score focused on risk factors and intraoperative problems to be able to predict a successful FT following MIMVS could be useful in the future. Nevertheless, our group wants to address the fact that a predictive score may not directly be generated from our patient cohort data since certain limitations in our study [2] may be found in the potential selection bias due to the single-centre retrospective study design for example. Throughout the data set period of 2014–2020, small changes in FT protocol and patient preselection of those who underwent MIMVS were also made [2]. Conclusions from our study are limited to patients with Carpentier type I or type II MV pathologies [2]. The findings of our study cannot be extrapolated to more complex pathologies, neither are they applicable to patients treated with MV replacement nor to those with concomitant TV surgery [2]. We therefore invite and stimulate Condello et al. to produce and to validate such a predictive score in the setting of FT MIMVS using their data.

Constantinides et al. [3] determined risk factors for FTF and already incorporated them into a predictive FTF score. Their FTF score incorporated several preoperative factors and had been successfully internally validated, and after undergoing external validation and possible recalibration, it could be used as a tool to facilitate the planning and flow of cardiac surgery patients, based on the predicted probability of failure [3]. They hypothesized that the application of this score may limit FTF rates and help to reduce morbidity and cost [3].

Risk scores play an important role in clinical medicine [4]. With advances in the information technology and availability of electronic healthcare record, scoring systems of procedures can be developed [4]. Moreover, Zhang et al. [4] aimed to provide a tutorial on how to develop and validate risk scores based on a virtual dataset by using R software.

Furthermore, El Garhy et al. investigated the influence of functional impairment assessed by the Barthel index on the 3-month outcomes after TF-TAVI under general anaesthesia [5]. They found for example that a Barthel index of <80 was associated with prolonged postoperative hospital stay and was an independent predictor of FT protocol failure (OR 4; 95% CI 1.3–11; P = 0.02) [5].

In general, risk scoring is beneficial both for its predictive capabilities and its application to health management. Nevertheless, selecting the appropriate risk-scoring model for a specific patient group can be complex. In finding a model that is the best fit for an organization goals, a set of indicators that will be used to define the patient’s risk has to be established first. Furthermore, ensure data quality and standardization techniques followed by creating a meaningful methodological approach that best fits the purposes of risk scoring. Finally, consider an organization’s resource capabilities and set realistic expectations of what can be achieved. The key to successful risk scoring is to start the conversation about the goals and needs of risk scoring, which will then help to narrow down the best model to help achieve those goals.

REFERENCES