Introduction: Inspiratory muscle training (IMT) is an intervention used in various clinical settings, including in critical care to ameliorate the effects of ventilator-induced diaphragm dysfunction. Metabolic oxygen consumption during IMT has not previously been explored. A recent study by our group explored the metabolic cost of IMT in mechanically ventilated individuals. We sought to supplement this with an understanding of the response of the healthy respiratory system to IMT.

Methods: The study conformed to the principles of the Declaration of Helsinki. Maximum inspiratory pressure (PImax) was measured with a differential pressure transducer during a sustained maximal inspiratory effort against an occlusion and calculated as the greatest one-second mean pressure.

IMT was applied at 4cmH₂O (“sham”) and at 30%, 50% and 80% of each participant’s PImax using an inspiratory threshold loading device (POWERbreathe Plus IMT). Sham training was delivered using an inverted Philips Threshold PEP device. Breath-by-breath oxygen consumption (VO₂) was measured using the Beacon Caresystem (Mermaid care A/S, Noerresundby, Denmark). Pressure at the airway opening was measured continuously using a port in the bacterioviral filter attached between the facemask and the IMT device.

IMT was applied for twelve breaths at each load with five minutes of tidal breathing between loads; the order of loads was randomised. VO₂ was recorded as a mean of the two minutes immediately following each IMT dose (recording during IMT bouts was found to be unfeasible due to the substantial negative pressures generated). Tension-time index of the respiratory muscles (TTmus) was calculated as mean airway pressure divided by PImax, multiplied by the respiratory duty cycle (PI/PImax x Ti/Ttot). Friedman’s ANOVA was used to examine whether VO₂ differed with IMT dose, with Dunn’s post hoc testing (using Bonferroni correction for multiple comparisons) for differences in VO₂ at individual doses. Linear mixed effects modelling (LMM) was used to quantify the relationship between IMT dose and VO₂ and between TTmus and VO₂.

Results: Thirty healthy adults (eighteen female) were studied (median (IQR) age 32.0 (24.3 – 44.5) years, mean (SD) PImax 119 (48)cmH₂O. Distribution of VO₂ differed significantly with IMT dose (p<0.001). Baseline median (IQR) VO₂ (4.42 (4.81 – 6.50) ml/min/kg) was not significantly different to sham (4.90 (4.11 – 5.03), p=0.305) or 30% (4.38 (3.69 – 5.23), p=1.000) but 50% (4.64 (4.09 – 5.28)) and 80% (5.09 (4.81 – 6.50)) IMT doses were significantly higher (p=0.043 and p<0.001 respectively). VO₂ at 30% and 80%, sham and 80%, and 50% and 80% doses were significantly different (p<0.001, p=0.004 and p=0.043 respectively). LMM showed a significant dose-response relationship between IMT dose and VO₂ (slope (95% confidence interval): 0.013 (0.009 – 0.018)ml/min/kg per %PImax increase in IMT dose, p<0.001). VO₂ was also significantly related to TTmus: slope (95% CI) 3.74 (2.67 – 4.81)ml/min/kg per unit increase in TTmus (p<0.001).

Conclusion: Oxygen consumption during inspiratory muscle training exhibits a significant positive dose-response relationship. There is also a significant positive relationship between VO₂ and respiratory muscle effort relative to capacity. Examining the metabolic cost of breathing may offer an option to guide prescription of IMT.