Discussion
Based on our studies, we propose that QRS axis is an important factor for risk stratification in population with PR prolongation. The extent to which this population with PR prolongation and QRS axis ≤37° is at higher risk of death compared with the population without PR prolongation. These associations were independent of P-wave duration and other known risk factors.
Our results were similar to previous studies, indicated that PR prolongation is not an independent risk factor for poor prognosis among adults without CVD2,12. Some studies of PR interval examining general populations only reported prolongation of the PR interval has been associated with increased risks of AF and HF compared with a normal PR interval6,8,21. For all-cause death, results are conflicting. But from the pathophysiological point of view, diastolic mitral regurgitation (MR) was indeed observed in individuals with PR prolongation22. Furthermore, normalized PR interval may eliminate this hemodynamic change, leading to a decrease in left atrial pressure and higher LV preload23. Thus, accurate risk stratification on the potential significance of this change in this population is warranted.
It is worth noting that PR prolongation is an important risk factor for early death in some special population. In a recent study in patients with HCM, PR interval ≥200ms was associated with HCM-related death, including the combined end point of sudden death or lethal arrhythmia10. In sub-analyses of two trials on cardiac resynchronization therapy (CRT) in patients with HF, individuals with a prolonged PR interval in the control group had a higher risk of all-cause death than those with normal PR interval24,25. Diastolic dysfunction (DD) is a prominent clinical feature both in HCM or HF patients26,27 and superposition of PR prolongation and DD would probably produce greater damage28. One possible mechanism would be mechanical LV dyssynchrony associated with DD deteriorated this abnormality of atrioventricular coupling29. Several studies investigated the value of restored AV coupling for clinical benefit in patients with prolonged PR interval. In these studies, CRT reduced the all-cause mortality risk of HF patients, but this benefit was confined to individuals with PR prolongation30,31. Our study may complement the mechanism proposed above for the interaction between LV function and PR interval.
QRS axis as a parameter directly obtained from ECG reports could be a surrogate marker of LV diastolic20. But we found that prolonged PR and QRS axis do not had the strong association with death when considered separately. This result may show that neither abnormal QRS axis nor prolonged PR interval can be used separately as a criterion to assess prognosis of population without CVD. The previous study also suggested that taking into account both abnormal QRS axis and prolonged PR interval may be helpful in further identifying abnormal LV function32. And we detected statistically significant interaction between the two. With QRS axis left shift, population with PR prolongation were at continuously increased risk of all-cause death. Therefore, we proposed that QRS axis can be an important factor for risk stratification in population with PR prolongation and compared the prognostic difference across populations stratified by their combination. This result is also in accordance with our conjecture. People had worse outcomes when they suffer from both QRS less than median and PR prolongation.
Previous studies have indicated that P-wave duration is a more sensitive marker for assessing the cardiovascular risk than PR segment because significant correlations between PR prolongation and death were observed only in high P/PR ratio group15. Prolonged P-wave duration indeed reflects both structural as well as electrophysiologic abnormalities in the atria33. The clinical implications of the two are partly overlapping yet not entirely identical. Our sensitivity analyses highlight a still independent prognostic value for the combination of both PR prolongation and QRS ≤37° in the low P/PR ratio population. Notably, MR associated with prolongation PR segment also induces atrial chronic stretch and conduction abnormalities and causes P-wave prolongation34,35. Their causal relationship cannot be determined due to the lack of some relevant studies. Our data provide a new light that P-wave duration might serve as another factor for risk stratification other than QRS axis in people with PR prolongation. The ultimate goal of our study presented here is to improve risk assessment for populations with PR prolongation by ECG. For those at high risk, close follow-up and aggressive early intervention may be required.