Is maximal diaphragm tissue velocity suited for assessment of diaphragm contractility?

The recent work by Poulard et al. (2020) offers a novel, non-invasive technique to assess diaphragm contractility, which has long been target of respiratory physiologists, and the importance this topic could not be over-emphasized during current pandemic. In addition discomfort balloon's placement, magnetic stimulation phrenic nerves is also pleasant, avenues such as ultrasound-based techniques warrant further investigation. should therefore commended for their efforts. While out two methods tested thickening fraction (TFdi,tw) showed little promise, maximal axial velocity (Vdi,max) was concluded promising because it ‘is reliable, sensitive change in cervical intensity, correlates with twitch transdiaphragmatic pressure (Pdi,tw).’ However, analysis data (kindly provided upon request), some instances were noticed where different interpretation derived from data. first issue regards study design. Participants underwent single trial, minimum three stimulations performed at every or these used subsequent analysis. my opinion, does constitute an adequate setting evaluate intra-session reliability: studies using Pdi,tw normally perform several (replicates) are averaged together generate point (Wüthrich 2014), aiming minimize variability. By replicates separate observations, probably stacked odds against themselves finding reasonable A much-preferred set-up would have repeat whole protocol twice short break in-between. design notwithstanding, misinterpreted message given standard error measurement (SEM), claiming that SEM Vdi,max ‘were comparable those observed Pdi,tw’ (Poulard 2020). Indeed, 1.55 cmH2O 1.89 mm s−1 Vdi,max, 95% CI narrowly matching (upper bound 1.75, lower 1.70). direct comparison numbers, however, unwarranted: measurements units and, more crucially, average value 11.6, whereas, 5.6. other words, relative 2.5 bigger than Pdi,tw. Furthermore, ICC both measures (0.86, = 0.81–0.90) clearly outside boundaries (95% 0.96–0.98). Finally, TFdi,tw SEM, apparently mistook percentage points: 18.8% 10.41%, respectively. Therefore, 10%, but rather ∼50% if expressed terms. manuscript changes intensity Although interpretations correct, obscures relationship between actual clinical use Pdi,tw, relies on values elicited supramaximal only. European Respiratory Society/American Thoracic Society guidelines recommend fixed criteria continuous assessing potential diaphragmatic weakness, (maximal) below 18 (Laveneziana 2019) 20 (Gibson 2002) considered suspicious. context, we recalculated (2020), found no (R2 0.01, F1,11 0.157, P 0.699), possibly results reproducibility Vdi,max. participants highest scores, rendering any attempt draw quadrants true/false negative/positive findings based established untenable, although arguably purpose investigation (i.e. all healthy). Importantly, mean inadequate method: possible itself properly discriminative healthy/abnormal responses. when five did reach supramaximality, note ‘interestingly, four supramaximality either, suggesting absence directly related supramaximality’. This observation implies mechanistic link assessments. 11 13 (or 85% participants). expected 80% (four five) particular subgroup fall into category. Looking side coin, fact eight seven show plateau seen stronger evidence phenomena reaching stimulation) independent. higher prevalence even repeatable method pre-requisite 2002). summary, I contend settings used, repeatability compared bring caution about conclusion practical application method. might prove better identifying contractility/weakness diaphragm, inferences direction cannot made available No competing interests declared. Sole author. None. author indebted Professor Will Hopkins discussion leading up letter.