Risk scores in cardiac resynchronization therapy–A review of the literature

Boros, András Mihály; Perge, Péter; Merkely, Béla; Széplaki, Gábor

doi:10.3389/fcvm.2022.1048673

REVIEW article

Front. Cardiovasc. Med., 17 January 2023

Sec. Heart Failure and Transplantation

Volume 9 - 2022 | https://doi.org/10.3389/fcvm.2022.1048673

This article is part of the Research Topic The Endless Quest for Optimal Pacing Support in Failing Hearts View all 10 articles

Risk scores in cardiac resynchronization therapy–A review of the literature

$\r\nAndrs Mihly Boros$ András Mihály Boros¹

Péter Perge¹

Béla Merkely¹

Gábor Széplaki^1,2,3*

¹Heart and Vascular Center, Semmelweis University, Budapest, Hungary
²Heart and Vascular Centre, Mater Private Hospital, Dublin, Ireland
³Royal College of Surgeons in Ireland, Dublin, Ireland

Cardiac resynchronization therapy (CRT) for selected heart failure (HF) patients improves symptoms and reduces morbidity and mortality; however, the prognosis of HF is still poor. There is an emerging need for tools that might help in optimal patient selection and provide prognostic information for patients and their families. Several risk scores have been created in recent years; although, no literature review is available that would list the possible scores for the clinicians. We identified forty-eight risk scores in CRT and provided the calculation methods and formulas in a ready-to-use format. The reviewed score systems can predict the prognosis of CRT patients; some of them have even provided an online calculation tool. Significant heterogeneity is present between the various risk scores in terms of the variables incorporated and some variables are not yet used in daily clinical practice. The lack of cross-validation of the risk scores limits their routine use and objective selection. As the number of prognostic markers of CRT is overwhelming, further studies might be required to analyze and cross-validate the data.

Introduction

According to the most recent guidelines, cardiac resynchronization therapy (CRT) is recommended for symptomatic heart failure patients in sinus rhythm with a QRS duration ≥150 ms and left bundle branch block (LBBB) QRS morphology and with left ventricular ejection fraction (LVEF) ≤35% despite optimal medical therapy to improve symptoms and reduce morbidity and mortality (1, 2). However, mortality is still high; and approximately one-third of the patients do not respond to CRT as adequately as expected, in whom no quality of live improvement or reverse remodeling of the left ventricle is seen (3).

Consequently, there is a great need for tools that might help in optimal patient selection and provide prognostic information for the patients and their families. Ever since the first implementation of CRT, several clinical factors and biomarkers have been tested in prediction models to identify those patients who might benefit the most from the therapy (4, 5). Prediction models are useful to reveal which parameters are statistically significant in the outcome prediction by giving the hazard and odds ratios, but they are not interpretable at the level of the individual patient in the clinical practice. Therefore, risk scores have been developed that constitute predominantly categorized variables with attributed points. The sum of the points reveals the exact risk of the individual; so that, patients can be easily and quickly grouped into risk categories with meaningful information.

Several risk scores have been created in CRT in recent years; however, no literature review is available that would list the possible scores for the clinicians.

Therefore, we aimed to systematically review the risk scores in CRT and provide the calculation methods and formulas in a ready-to-use format.

Materials and methods

The literature search was performed in November 2021 and then updated in September 2022 by using the search engine PubMed.gov¹ with the input of the following equation: (((cardiac resynchronization) OR (cardiac resynchronization therapy)) OR (biventricular pacing))) AND (((prediction model)) OR (predictive model) OR (risk model) OR (score))). The flowchart of the review process is presented by Figure 1.

FIGURE 1

Figure 1. Flowchart of the review process.

Since we applied no language or publication date restrictions, the result was 1,314 possible papers. Two investigators (AB and PP) independently pre-screened the abstracts of these manuscripts by considering further inclusion criteria: original research article, and ready-to-use format. This resulted in a sum of 100 records that were further assessed by full-text review. A total of 52 papers were excluded based on the following reasons: external validation of previously described score systems (n = 18), prediction models without score systems (n = 18), machine learning algorithms without online interfaces (n = 8), miscellaneous endpoints (n = 5), and lack of CRT (n = 3). Consequently, forty-eight CRT risk scores were incorporated into the present review.

Results

To date, we identified 48 ready-to-use risk scores in heart failure patients with CRT Table 1. Summarizes the details of the models with the interpretation of the results and presents the formulas or the calculation methods of the scores Figure 2. Overviews the risk scores and helps in the selection of the appropriate risk score by considering the available data about the patient.

TABLE 1

Table 1. Risk scores in cardiac resynchronization therapy.

FIGURE 2

Figure 2. Heat map of the predictors used in the risk scores of cardiac resynchronization therapy.

The primary endpoint of the models was all-cause death or a composite of death in the majority of the cases (n = 33, 69%), otherwise, it was echocardiographic or clinical response to CRT (n = 15, 32%). The most commonly used variables in the models were ischemic etiology (n = 21, 44%), renal function (n = 21, 44%), age (n = 20, 42%), New York Heart Association classification (n = 18, 38%), LVEF (n = 15, 33%), QRS morphology (n = 15, 31%), QRS width (n = 14, 30%), atrial fibrillation (n = 13, 27%), gender (n = 13, 27%), and left ventricular dimensions (n = 12, 25%).

Discussion

The very first risk score in CRT was developed by Heist et al. (6). It investigated the immediate hemodynamic response (improved contractility as assessed by the dP/dt of the mitral regurgitation jet) to CRT by using echocardiographic and electrophysiologic parameters (6). Following that, the Charlson comorbidity index (CCI) from Charlson et al. (7), was tested in 463 heart failure patients with CRT; a CCI score ≥5, meaning several comorbidities and worse overall state, reflected a more than 3 times mortality risk (8). In parallel, the MADIT-CRT score was created by Goldenberg et al. (9) by using the data of the 1,761 patients enrolled in the Multicenter Automatic Defibrillator Implantation Trial With Cardiac Resynchronization Therapy (MADIT-CRT). The MADIT-CRT identified the most relevant routine clinical risk factors that affect mortality in CRT: gender, etiology of heart failure, the presence of left bundle-branch block and wide QRS, prior heart failure hospitalizations, and left ventricular and atrial dimensions. The MADIT-CRT score has been served as a gold standard and used as a reference in many validation studies (10–12).

The Seattle Heart Failure Model (SHFM) is a well-known risk estimation tool to predict the 1-, 2-, and 5-year mortality in chronic heart failure patients with conservative therapy (13). Perrotta et al. (14) applied the SHFM to patients who received a CRT, or a CRT-D and the model showed a good discrimination capacity in the mortality prediction. In the same year, the SHFM was validated in CRT populations by others as well (15, 16). Park et al. (17) were the first who developed a risk score, the EchoCG score, by using echocardiographic strain analysis to predict the reverse remodeling after CRT implantation. Strain analysis was included in many models later (11, 18–20). Similarly, to strain analysis, electrophysiologic modalities were also used in risk score development, such as sophisticated ECG analysis (21–23), vectorcardiography (24), or heart rate histogram analysis (25).

However, simplicity and availability are the keys to risk score development. The EAARN (26), the VALID-CRT (27), the HF-CRT (28), the CRT-SCORE (29), the AL-FINE (30), the ScREEN (31), the CRT-D Futility score (32), the MAGGIC (33), and many others can be calculated with routine laboratory and clinical parameters. Incorporating these principal concepts, machine learning algorithms can provide personalized risk predictions and online calculators are also available (34–36).

Conclusion

This is the first systematic review of risk scores in cardiac resynchronization therapy. The scores show a great diversity in terms of used predictors and endpoints. As we demonstrated, the number of the different scoring systems has drastically increased in the past few years and a very marked heterogeneity can be observed among them. Unfortunately, this makes their translation and transition into everyday clinical practice difficult. Furthermore, the majority of studies were conducted prior to the current era of quadruple HFrEF therapy. These limitations must be considered before the routine application of the score systems.

Rickard et al. have shown in a prior review that classic markers (native LBBB, non-ischemic etiology, wide QRS, female gender and sinus rhythm) predict outcomes after CRT-D (4). However, there is growing evidence available on novel risk factors for CRT response, incorporated into the numerous risk score systems. The predictors can be categorized into the following different groups: co-morbidities, clinical state, echocardiographic, electrocardiographic, routine blood markers, and novel biomarkers as shown in the present review; the overlap of the markers in the various models is minimal. Some biomarkers are not yet incorporated into the daily routine clinical practice and their widespread use is therefore limited. Moreover, the lack of cross-validation across the risk scores limits the ability to objectively determine which of them should be incorporated into daily practice.

Although all the listed risk scores have the potential to predict outcomes after CRT, more data is required to enable us to select which will be appropriate to use in the daily clinical practice to predict the prognosis of severe heart failure patients, who undergo CRT. As the number of possible predictors and combinations is overwhelming, machine learning based algorithms or the help of artificial intelligence might be required to develop a uniform CRT risk score system.

It must be emphasized that, currently, the decision of CRT implantation is based on the ejection fraction, the width of the QRS, and the presence of LBBB; none of the guidelines do endorse any risk score to be applied in the process. Therefore, risk scores are useful to give information regarding the prognosis after implantation but should not influence the implantation itself.

Author contributions

AB and GS contributed to the conception and design of the study and wrote the first draft of the manuscript. GS and BM provided the institutional background to the study. AB and PP collected data and performed the statistical analysis. All authors contributed to manuscript revision, read, and approved the submitted version.

Conflict of interest

GS reports personal fees from Abbott, Bayer, Boston Scientific, and Johnson and Johnson Medical outside the submitted work.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

^ https://www.ncbi.nlm.nih.gov/

References

1. Ponikowski P, Voors A, Anker S, Bueno H, Cleland J, Coats A, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: the task force for the diagnosis and treatment of acute and chronic heart failure of the European society of cardiology (ESC)developed with the special contribution of the heart failure association (HFA) of the ESC. Eur Heart J. (2016) 37:2129–200.