Twenty-five-year milestone for the ECHSA database: time to focus on patient survival rather than procedural success

The accompanying paper in this issue of the Journal by Protopapas et al. [1] documents a 25-year milestone for the European Congenital Heart Surgeons Association (ECHSA) congenital surgical procedural database. The paper presents a simple and unadjusted review of congenital heart surgery procedural outcomes collected between 1999 and 2024. The authors document improved surgical procedural mortality over time. Mortality information is broken down by 30 days vs hospital mortality, European vs ‘other continent patients’ as well as by benchmark procedures. Although mortality has decreased over time, there has been a trend towards longer length of hospital stay.

A distinguishing innovative feature of the ECHSA database from its outset has been on-line accessibility for surgeons and centres. The creators made a bold move in the early years of the internet to do things differently from the traditional methods employed pre-internet by the Society of Thoracic Surgeons (STS) adult database that compiled lengthy interval reports. This same dated methodology was employed when the STS congenital database was developed.

Times have changed since the creation of both the ECHSA and STS congenital databases. Prenatal diagnosis is widely available today. Birth rates and birth weights have both continued to decline. Interventional catheter and hybrid methods of palliation have become widely used particularly for high-risk patients. And high-risk patients have been clearly characterized, especially smaller premature patients with genetic and extracardiac anomalies such as trachea-oesophageal fistula and diaphragmatic hernia.

In addition to changes in patient characteristics over the last 25 years, hospital administrators, insurance companies and government agencies responsible for paying for expensive complex procedures are focusing more intently on outcomes data than ever before. Hospital and surgeon rankings based on outcomes data have assumed far greater importance than ever imagined by those who created databases in the 1990’s with the principal goal of facilitating clinical research studies which would allow refinement of evolving techniques of the era.

The coalescence of these factors, namely increasing patient complexity, availability of hybrid palliative triage and above all, ever increasing negative consequences for procedural outcomes that appear suboptimal, has made it abundantly clear that the time has arrived for the ECHSA database to make another bold move. The time has arrived for the database to distinguish itself in the same way that its founders achieved distinction at its outset. It is time to focus on patient survival and not surgical procedural success. Surgeons today who tout their numbers collected by a procedural database like the STS or ECHSA database sound much like the interventional cardiologists in the early years of new procedures in the cath lab when ‘technical success’ in the lab was loudly proclaimed, instead of providing the ultimate outcome for the patient, family and payor. But we are way past the foundational years of congenital heart surgery.

Patient survival is not the same as long-term survival. This is a distinction that seems difficult for many to grasp. Patient survival is simply survival to discharge from hospital or to 30 days after completion of a 2 ventricle in-series circulation, or to end-state palliation for those with a single ventricle, usually an in-series circulation. Patient survival as a metric does not allow congenital teams to divert high-risk small premature patients with genetic or non-cardiac anomalies on to a hybrid palliative track or ‘comfort care’ track with no ultimate repair. In a patient survival database, all patients entering a programme with congenital heart disease are accounted for. In contrast, in a procedural database triage allows the demise of high-risk patients to be camouflaged.

Prenatal diagnosis or diagnosis early in infancy mean that enrolment in a diagnosis/patient-based database is eminently doable in today’s environment of the universal Electronic Medical Record. This is not too difficult. That excuse has been held up for too long and is no longer valid.

In conclusion, while undoubtedly Protopapas et al. are correct that much of what has been achieved in the last 25 years in terms of improved outcomes for patients with congenital heart disease has been a result of better techniques and better team coordination and performance, going forwards the advances will be more subtle and will only be captured accurately with a patient-based rather than surgical procedural-based database. It is time for those who manage the ECHSA database to make a bold move like they did in the 1990’s when developing their surgical procedural database. It is time for a patient/diagnosis centric database. If neither ECHSA or STS will make the first move, it may be up to another group.

Conflict of interest: none declared.

REFERENCE