Time to expand risk evaluation systems for cardiac surgery? Looking beyond physiological parameters

Background

Cardiac surgery is common in the modern western world. The number of individuals undergoing cardiac surgery has grown over the past decades as a consequence of an increasingly aging population. With surgical advancement, more complex and high-risk patients than before are being offered surgical procedures. Irrespective of its prevalence, cardiac surgery is still considered to be a high risk procedure. Thirty-day mortality rates after cardiac surgery range from 1–3% increasing to 5–6% at one year.^1–5 Readmission rates are about 15% at 30 days after discharge,^6–8 but vary greatly after 30 days from 19% to 56%.^8,9 Since patients differ in their level of surgical complexity and risk, and because surgeons differ in their technical performance skills, efforts to reduce poor outcomes such as mortality and readmission have been implemented. This frequently includes risk assessment of patients. In some countries report-card programmes are implemented, publicly comparing hospital and surgeon performance. One of the original aims for the development of cardiac risk models was risk adjustment, allowing fair comparison of treatment outcomes among different institutions or surgeons. Subsequently, risk models were also applied in clinical decision making, advising individual patients of their operative risk.^10,11

Several risk scoring systems exist. The Society of Thoracic Surgeons (STS) risk score and the European System for Cardiac Operative Risk Evaluation (EuroSCORE) are two widely used risk assessment models that ascertain a patient’s magnitude of risk for complications such as mortality after cardiac surgery. The STS risk score was first established in 1989 to assess mortality after cardiac surgery in adults. This risk stratification model has been widely used in North America and comprises over 40 clinical parameters. In addition to mortality, an updated version of the STS score provides risk prediction of major post-operative morbidities. The STS risk score is limited to risk prediction for surgical cases of isolated coronary artery bypass grafting (CABG), valve surgery and combined procedures. ¹²

EuroSCORE was developed between 1995 and 1999. As cardiac surgery advanced over time, the original model was found to overestimate the risk of mortality for certain subgroups and was therefore updated in 2012 to EuroSCORE II. This latter risk scoring system comprises 18 clinical parameters (Supplementary Material Table 1 online) and is applicable to all major cardiac surgical procedures (except transplants). EuroSCORE II is the preferred risk score model used in most of Europe.^13,14

Both the STS risk score and EuroSCORE II only include medical and physiological factors. Furthermore, a meta-analysis showed that the existing scoring systems have a summary area under the receiver operating curve of 0.76 (95% confidence interval (CI) 0.73–0.79) for the STS score and 0.77 (0.75–0.79) for EuroSCORE II for predicting in-hospital mortality. ¹⁵ This shows that there is still room for improvement in order to make the risk assessment models more predictive and discriminatory. Indeed, there is mounting evidence that non-physiological factors, such as emotional,^16–18 behavioural,^19–21 social^22–29 and functional,^30,31 are also predictive of poor outcomes following cardiac surgery. With the present paper, we want to raise awareness about the partiality of the existing risk evaluation in cardiac surgery, as well as discuss selected non-physiological variables that have been shown to predict poor outcomes after cardiac surgery. Such non-physiological variables potentially could add accuracy and meaning to the existing scoring models if they are included above and beyond the traditional physiological parameters.

Methods

We applied the methodology of a state-of-the-art review. ³² State-of-the-art reviews aim to offer new perspectives on an issue or highlight an area in need of further research. They address more current matters in contrast to other methods of literature reviews. In this respect, state-of-the-art reviews are of considerable value for those seeking to identify potential opportunities for contemporary research. ³² According to the SALSA (Search; AppraisaL; Synthesis; Analysis) framework, the typical methodology of a state-of-the-art review is a comprehensive search of the current literature; no formal quality assessment; narrative reporting sometimes with tabular accompaniment; and providing the current state of knowledge and priorities for future investigation and research. ³² Literature searches for each factor included were performed in the PubMed database always consisting of ‘cardiac surgery’ combined with relevant terms for each factor as MeSH term and as free-text. Reviews and new literature were preferred. For the initial, general search strategy, please see Supplementary Material Table 2 online.

Emotional factors

In patients undergoing cardiac surgery the knowledge of prevalence and influence of emotional factors is lagging behind the evidence documented in other heart conditions. Indeed, the paucity of research in this group of patients regarding the influence of emotional factors on outcomes is striking. In cardiac surgery the emphasis has been on preserving cognitive function rather than mental health per se. ³³

Social factors

Socioeconomic position is usually measured by determining education, income, occupation, or a composite measure thereof. These have been found to influence disease incidence, treatment and health outcomes. Those with lower income, lower education or working in lower status occupations experience increased mortality and morbidity.^{22,25,26,28,44,45}

Living alone has been linked with poor health outcomes in earlier studies. There seem to be a number of reasons for this. Patients who are socially isolated are more likely to smoke and drink alcohol heavily,^46,47 delay in seeking treatment ⁴⁸ and demonstrate non-compliance with medical regimens. ⁴⁹ Others have found it simply to be due to lack of emotional or practical support gained through living with another person. ⁵⁰

Functional factors

The physiological factors included in EuroSCORE comprise a range of comorbidities. However, functional status as an independent predictor is not included in the score. Improved functional status and return to pre-morbid lifestyle is a major goal for most patients undergoing cardiac surgery.

Behavioural factors

Several longitudinal and cross-sectional studies demonstrate that health-risk behaviours such as smoking, alcohol consumption, a sedentary lifestyle and obesity are associated with a variety of health-risks and mortality. ⁶³

Discussion

Cardiac surgery remains an established procedure for several heart conditions and as a consequence of an increasingly older population, cardiac surgery is being performed on more complex and high-risk populations than before. Adequate risk assessment to guide the clinical decision-making is therefore of rising importance.

Several of the above mentioned factors rely on patient-reported measures. Traditionally, clinical health factors, such as blood pressure control, have been proclaimed as valid measures of cardiovascular health. ⁷⁴ Although important, these measures alone are insufficient according to the World’s Health Organization’s definition of health: ‘…a state of complete physical, mental and social well-being.’ ⁷⁵ Associations between self-reported health and morbidity and mortality have been confirmed in previous studies, where patient reported measures have been found to predict prolonged hospital stay, labour market affiliation, morbidity and mortality in cardiac patients.^76–79

For the application of patient-reported outcomes in risk assessment, it is of paramount importance to choose measures that provide valid and reliable responses. On the other hand, feasibility and practicality of collection, scoring and interpretation of patient health status data have to be taken into consideration. Comprehensive testing, using extensive research instruments, is not an option when used in risk assessment schemes. Such questionnaires are typically too lengthy to be used in clinical practice. Indeed, in risk evaluation, each factor ought to be assessed in the shortest possible way, to warrant that the overall risk assessment is achievable in day-to-day clinical practice. Short-form versions are likely most suitable. Furthermore, computer adaptive testing may be a promising approach, since the answer to a certain question determines what questions are subsequently asked. Such adaptive testing procedures can substantially reduce the questionnaire burden in respondents.

As described in this paper, several emotional, social, functional, and behavioural factors impact on mortality, readmission and complications following cardiac surgery. However, there is still some uncertainty about the true predictive value of these factors. Furthermore, it is not known what their added value would be above and beyond the physiological factors. Therefore, future research in this matter is imperative.

In Denmark, a comprehensive research project is ongoing in which we are developing a risk stratification model as a supplement to EuroSCORE. Emotional, social, functional and behavioural factors will be included through national administrative registers, clinical databases and surveys to predict mortality, readmissions and complications in patients within three months after cardiac surgery.

In conclusion, conventional risk assessment focuses on medical and physiological factors, but does not contain other relevant risk factors associated with increased risk of poor outcomes following cardiac surgery. Evidence has been established that also non-physiological factors are predictive of poor outcomes following cardiac surgery and need to be considered for inclusion in current risk evaluation models.

Supplemental Material