Page 217 - Read Online

Riojas et al. Vessel Plus 2024;8:6 https://dx.doi.org/10.20517/2574-1209.2023.122 Page 13 of 16

Similarly, the Massachusetts General Hospital (MGH) established a Heart Team as part of a quality
[29]
improvement program . The MGH Heart Team evaluated patients with complex CAD and other high-risk
features such as comorbidities or socioeconomic factors. Most notably, the MGH Heart Team generated a
unique and succinct structured form that contained a summary of pertinent information such as medical
history, laboratory data, angiography and other imaging information, risk scores, and a summary to mark
what the recommended treatment is based on most recent guidelines. This highly structured format
undoubtedly facilitated succinct presentations, as well as a means to track outcomes. The MGH Heart Team
noted that most of the patients were older, had more complex comorbidities, and had higher SYNTAX and
STS risk scores. Among all PCI and CABG patients, there was a low in-hospital mortality rate of 3.9%, a low
observed-to-expected ratio of 30-day mortality in the CABG group, and better adherence to guideline
recommendations for PCI, suggesting improved outcomes after Heart Team discussion based on most
current evidence .
[29]
FUTURE CONSIDERATIONS
The ultimate goal of the heart team is to provide high-quality, patient-centered care based on established
guidelines and a detailed review of the individual patient’s social and medical situation with input from the
patient, cardiologists, surgeons, and other specialists as needed. The multidisciplinary team should have
commitment from the providers and also full support from the hospital administration. This may be in the
form of accepting the time put forth as clinical time, providing the infrastructure and administrative
support to run an effective heart team, and recognizing the efforts of team members. In the future,
institutions may consider mandates or requirements for the establishment of a heart team, but this should
maintain a collegiate and collaborative atmosphere. Currently, the preprocedural assessment and Heart
Team approach is a recommended structural measure in 2023 AHA/ACC Clinical Performance and Quality
[55]
Measures for Coronary Artery Revascularization . Once a Heart Team has been established, a goal for
improvement would be to expand the access to the heart team via telemedicine to smaller branch hospitals,
including clinics and hospitals not a part of the main institution. This will not only benefit referral patterns
but, more importantly, will allow for more equitable inclusion of patients in the respective communities. In
larger metropolitan areas, this could lead to competition between major institutions. However, as long as
there is mutual respect, the competition will benefit the community and improve patient outcomes.
Another potential benefit of a telemedicine platform is that nationally and internationally renowned
providers may be able to discuss some of the more challenging scenarios. Again, this will create a
collaborative atmosphere and bear further credence to the heart team.

Finally, the rapid growth of artificial intelligence (AI) will eventually play an important role in the
[56]
evaluation of coronary artery disease and the evaluation by a heart team. At the time of writing this
manuscript, there was no specific example of AI use in heart team decision making. AI is based on using
machine learning algorithms. These algorithms are like a set of instructions for computer programs on how
to analyze data. Data are defined using a structured description, called an ontology. Machine learning
algorithms use an ontology to interpret and understand data from various sources . One benefit of
[57]
machine learning algorithms is that they can be used to process a massive amount of unstructured data in
[58]
an unbiased format . A machine learning classification model could be utilized to predict the impact of an
intervention based on the patient’s complex comorbidities and procedural risks. Then, a machine learning
regression model could be utilized to predict the outcomes after an intervention. A unique feature of
machine learning models is that they will continue to evolve as more data are gathered. Specific machine
learning algorithms could potentially be developed for each unique institution, and treatment and
evaluation recommendations would be based on major guidelines and reflective of the best outcomes within
that hospital system. Finally, a programming interface would allow a physician to interact with the model