Background

QRS complex prolongation is not only an established prognostic marker in patients with heart failure, (HF) but it also identifies patients for whom standard treatment includes cardiac resynchronization therapy (CRT). In contrast, the course of QRS progression over time on clinical status and whether it can be predicted has been incompletely explored.    

Aim

This analysis aimed to provide pilot data that could help to identify features associated with QRS progression and deteriorating clinical status.

Methods and Results

The present retrospective analysis relates to 100 consecutive patients attending in our hospital-based HF clinic between April and August 2021. Each patient with a least one previous attendance including an ECG was included and a maximum of 4 previous visits were assessed per patient. This retrospective approach offered a mean time between visits of 227 days.  Mean (± SD) age and left ventricular ejection fraction were 58±13 years and 36±10%. 

In total 240 datasets were created from at least two consecutive visits including clinical status, electrocardiographic (ECG), laboratory and echocardiographic data. Datasets were stratified into tertiles based on change in QRS duration (mm/month). These were labelled as ‘QRS regression’ (>-0,34 ms/month), ‘QRS stability’ (-0.33 – 0.49ms/month and ‘QRS progression’ (>0.50ms/month).

The incidence of the combined endpoint of worsening symptomatic HF (deterioration in NYHA class by at least one grade) and HF hospitalisation was significantly higher in the QRS progression group compared with the stable group (Log Rank test: progression v stability p=0.013). In addition, patients with QRS progression had a higher baseline plasma NT-pro-BNP levels (p=0.008) and higher baseline heart rate (p=0.007). 

To explore whether patients at higher risk of QRS progression could be identified at baseline, we built a prediction model based upon baseline NTproBNP levels and baseline heart. We determined that an NT-pro-BNP >837 pg/ml and a heart rate >83/bpm were reliable thresholds for greater risk of QRS progression (Figure 1 A + B). Both of these variables were independent of guideline-directed medical HF therapy (including beta-antagonist use and dose). 

An interaction analysis revealed that a combination of these two risk factors was associated with a 14-fold increased risk of QRS progression compared with neither or either alone, suggesting synergism between these variables (HR: 14.2 [95% 6.9 – 53.6];p<0.0001, p for interaction=0.016). 

Conclusion

In the current pilot study, we demonstrate that QRS progression is associated with future risk of clinical deterioration of HF. In addition, we show that higher baseline NTproBNP plasma levels combined with higher baseline heart rate indicate a group at markedly increased risk of progression of QRS width, independently of HF therapy. These parameters might therefore identify people potentially benefitting from intensified follow up, optimisation of guideline directed medical therapy and could also help coordinate the application of a more personalised approach to device therapy