Assessing cardiac autonomic function via heart rate variability analysis requires monitoring respiration: reply Free

Confounders of heart rate variability

We surely agree with Wessel et al. that heart period fluctuations are linked to many physiological oscillations. As already explained in the 1996 heart rate variability (HRV) standards,¹ these oscillations include those of respiration,² blood pressure,³ etc. The shifts in the respiration-related HRV components into the nominal low-frequency band are also known⁴ as is the fact that the low-frequency HRV modulations cannot be taken as a single reliable expression of sympathetic activity.¹ It is difficult to consider respiration as an overwhelming determinant of short-term HRV. Adding additional provocations to constant controlled breathing still leads to profound HRV changes.⁵ Breathing frequency is also only one and probably a simpler descriptor of respiration. In clinical well-defined populations, the value of respiration assessment⁶ also appears somewhat different from that of novel HRV methods.

Moreover, a distinction is needed between the use of HRV in physiological and clinical, e.g. risk stratification, studies. Different confounders including respiration should be controlled in physiological investigations. However, the power of clinical risk assessment studies increases by considering the autonomic homeostatic maintenance comprehensively. Adjusting HRV measurements for confounders that are also under autonomic control might affect their predictive value. Our position statement⁷ provided an update on some recent approaches to properly complement previously listed HRV parameters¹ in large clinical studies. We have not aimed at reviewing the vast literature on the interactions between HRV and other signals, but it would certainly be important to clinically validate the multivariate indices in the same way.

We do not see very clearly the association between the somewhat limited clinical applications of the new HRV methods⁷ and the respiration–HRV coupling. Multisignal analyses including, among others, electrocardiogram, blood pressure, pulse oximetry, and respiration likely improve physiological quantification of autonomic reflexes, especially under provocations,⁸ but the limited spread of novel HRV methods is probably caused by other factors. While the interpretation of the standard, e.g. spectral HRV analyses, reflects well-established physiological models easily comprehended by clinical researchers, such transparent models are not necessarily widely available for the HRV methods that we reviewed⁷ and perhaps even less available for novel methods that we excluded since they have not been used in large studies. Applications of the non-standard HRV methods also require tools that are not readily available within commercial electrocardiographic equipment. This limits the new methods to specific laboratories that combine engineering knowledge with clinical experience. Outside such centres, the contacts between clinical and engineering researchers are sadly limited.⁹ On the one hand, collections of clinical data are too frequently either not provided to engineering teams or provided with limited background. On the other hand, engineering groups far too often develop advanced signal processing techniques without a clear vision of clinical problems to which the new concepts might be applied. Naturally, this calls for a more active collaboration between clinical and engineering teams and we can only hope that our position statement would facilitate such research partnerships.

References