Reply to ‘Letter to the editor: is maximal diaphragm tissue velocity suited for the assessment of diaphragm contractility?’ Article Swipe

MAKE IMPACT

Thomas Poulard , Martin Dres , Marie‐Cécile Nierat , Isabelle Rivals , Jean‐Yves Hogrel , Thomas Similowski , Jean‐Luc Gennisson , Damien Bachasson · YOU? · · 2021 · Open Access · · DOI: https://doi.org/10.1113/jp281433

We appreciate the opportunity given to respond to Dr Beltrami's Letter (Beltrami 2021) regarding our recent publication (Poulard et al. 2020) in The Journal of Physiology. Below, we have paraphrased the main criticisms, each followed by our response. ‘Participants underwent a single trial, where a minimum of three stimulations were performed at every intensity, and two to three of these were used for subsequent analysis. In my opinion, this does not constitute an adequate setting to evaluate inter-session reliability’ One must note that inter-session reliability was not assessed in the current work. However, we assessed intra-session reliability of Pdi,tw, Vdi,max and TFdi,tw by comparing values obtained at any given stimulation intensity. As clearly mentioned in the Discussion: ‘between-day reliability of Vdi,max remains to be investigated.’ A different design will be chosen to investigate this question. ‘The design notwithstanding, Poulard et al. (2020) misinterpreted the message given by their standard error of measurement (SEM) […] the relative SEM for Vdi,max is 2.5 bigger than that of Pditw […] the ICC of both measures are also not comparable because that of Vdi,max (0.86, 95%; CI = 0.81–0.90) is clearly outside the boundaries of that for Pdi,tw (95%; CI = 0.96–0.98) […] SEM [For TFdi,tw] was not 10%; but rather ∼50%; if expressed in relative terms’ As previously discussed with Dr Beltrami in several email exchanges, the message we tried to convey was that SEM and ICC for Vdi,max appeared to be acceptable, and importantly, much more reliable than TFdi,tw. It must also be noted that one must be careful when inferring SEM values expressed as a percentage from SEM and average values. To achieve this, analysis of log-transformed variables is required to estimate errors when the standard deviations expressed as percentages (coefficients of variation) apply more accurately to a full population given the between-individual variability (Hopkins, 2000). Moreover, one should not lose sight of the fact that these analyses involve all responses at all intensities and are probably less reliable than what can be expected when considering supramaximal stimulation only. ICC for Vdi,max was >0.85, which can be considered sufficiently high to be considered as a reproducible index (Koo & Li, 2016). Dr Beltrami's statement regarding the SEM of TFdi,tw is also incorrect because percentage and percentage points were not confused. All SEM values were presented in the corresponding unit for each variable (i.e. cmH2O for Pdi,tw, mm‧s− 1 for Vdi,max and % for TFdi,tw). Although the ICC for Vdi,max was outside the 95% confidence interval of that for Pdi,tw, the ICC was >0.85, which supports Vdi,max as a reproducible index when it comes to relative reliability (Koo & Li, 2016). ‘Pdi,tw values are typically only used from supramaximal stimulation, so although it is interesting that there was a correlation between the two variables when low stimulation intensities are considered, it is unclear from the data whether there is a relationship between Pdi,tw and Vdi,max.’ In the current work, change in cervical magnetic stimulation intensity was used as an experimental paradigm for eliciting various levels of Pdi,tw. Therefore, obtaining supramaximal Pdi,tw was not essential for our work. We used a broad range of cervical magnetic stimulation intensities to investigate the sensitivity of indices derived from ultrafast ultrasound imaging to changes in diaphragm contraction levels. Within our publication, Fig. 7A clearly illustrates the relationship between Pdi,tw and Vdi,max. In the text, we clearly state that a significant relationship was found between Pdi,tw and Vdi,max within all participants (ρ = 0.64 – 1.00, all P < 0.05), as well as at the group level, as assessed using repeated measures correlation coefficient [R = 0.75 (95% CI = 0.65–0.83), P < 0.0001]. ‘We recalculated the data from Poulard et al. (2020), and found no relationship between Pdi,tw and Vdi,max (R2 = 0.01, F1,11 = 0.157, P = 0.699)’ Dr Beltrami did not find the same results as we did because he did not perform the same analysis as us (although it remains unclear which exact analysis Dr Beltrami performed because we found a coefficient of determination of 0.39 when considering all data points sent to Dr Beltrami). In the publication, we did not calculate a coefficient of determination (R2) but, instead, a repeated measure correlation coefficient (R) was calculated (Bakdash & Marusich, 2017). As mentioned in the ‘Statistical analysis’ section of our paper (Poulard et al. 2020), this technique ‘considers the independence of repeated measures between individuals, so that potential confounding factors, such as between-participant variability, do not interfere.’ Dr Beltrami (2021) also raises some concerns about the fact that Vdi,max may be independent of Pdi,tw because several participants reached supramaximality for Pdi,tw and not for Vdi,max, or vice versa. The key concept behind this part of the Discussion is that the recruitment of accessory inspiratory muscles may also contribute to Pditw, particularly at high stimulation intensities (Wragg et al. 1994; Laghi et al. 1996; Attali et al. 1997). In some participants, Vdi,max reached a plateau at ∼90% of maximal stimulation intensity, whereas Pditw increased up to 100%. Because we hypothesized Vdi,max to be highly specific of diaphragm contraction and, unlike Pdi,tw, potentially unaffected by the recruitment of neck inspiratory muscles, we suggested that Vdi,max may be an interesting index to detect supramaximality of evoked diaphragm contraction. ‘Although Vdi,max might prove itself better at identifying contractility/weakness of the diaphragm, inferences in this direction cannot be made from the available data.’ We consider that Dr Beltrami's concerns regarding the potential power of Vdi,max for the detection of diaphragm dysfunction is beyond the scope of this paper for which we must recall the main objective: capturing a 300 ms lasting evoked contraction of the diaphragm using ultrafast ultrasound and investigating potential relationships between metrics derived from ultrafast ultrasound with the reference method (i.e. Pdi,tw). Accordingly, we can only concur with this statement because it was never an objective of the current work to determine whether Vdi,max could identify diaphragm weakness as the participants recruited were all healthy and with normal respiratory function. For these reasons, we explicitly stated that ‘Further studies will focus on this specific point [i.e. the ability of Vdi,max to determine diaphragm dysfunction], with the perspective that Vdi,max may be one parameter, among others, guiding clinicians through the assessment of diaphragm contractility.’ (Poulard et al 2020) We are currently investigating the diagnostic power of our approach in patients with suspicion of diaphragm dysfunction. In time, we will be pleased to share our findings with the scientific community. No competing interests declared. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. The PhD fellowship of TP is funded by the Fondation EDF that is supporting the RespiMyo project, which includes the current work. This study was also supported by the Association Française contre les Myopathies (AFM).

Type

letter

Language

en

Landing Page

https://doi.org/10.1113/jp281433

PDF

https://onlinelibrary.wiley.com/doi/pdfdirect/10.1113/JP281433

OA Status

bronze

References

9

Related Works

10

OpenAlex ID

https://openalex.org/W3135662678

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W3135662678

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.1113/jp281433

Digital Object Identifier

Title

Reply to ‘Letter to the editor: is maximal diaphragm tissue velocity suited for the assessment of diaphragm contractility?’

Work title

Type

letter

OpenAlex work type

Language

en

Primary language

Publication year

2021

Year of publication

Publication date

2021-03-04

Full publication date if available

Authors

Thomas Poulard, Martin Dres, Marie‐Cécile Nierat, Isabelle Rivals, Jean‐Yves Hogrel, Thomas Similowski, Jean‐Luc Gennisson, Damien Bachasson

List of authors in order

Landing page

https://doi.org/10.1113/jp281433

Publisher landing page

PDF URL

https://onlinelibrary.wiley.com/doi/pdfdirect/10.1113/JP281433

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

bronze

Open access status per OpenAlex

OA URL

https://onlinelibrary.wiley.com/doi/pdfdirect/10.1113/JP281433

Direct OA link when available

Concepts

Reliability (semiconductor), Session (web analytics), Diaphragm (acoustics), Contractility, Psychology, Computer science, Medicine, Mathematics, Physics, Internal medicine, Acoustics, Loudspeaker, Quantum mechanics, Power (physics), World Wide Web

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

References (count)

9

Number of works referenced by this work

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload