Effect of nasal high flow therapy on work of breathing and CO2 tension: using an in vitro and mathematical modelling approach (2017)
Type of ContentTheses / Dissertations
Thesis DisciplineMechanical Engineering
Degree NameDoctor of Philosophy
PublisherUniversity of Canterbury
AuthorsAdams, Cletus Fiifishow all
A breathing therapy is a technique (invasive or non-invasive) that improves respiration. Nasal high flow (NHF) is a relatively new breathing therapy which involves the delivery of humidified and warmed air via a nasal cannula at a constant flow rate (up to 60 L/min). Conditions alleviated with NHF include sleep apnea, hypercapnic respiratory failure and hypoxemic respiratory failure. The proposed mechanisms of action of NHF include improvement in oxygenation; attenuation of inspiratory resistance; generation of positive airway pressure; and reduction in work of breathing (WOB) (energy expended by the respiratory muscles during breathing). This work was motivated by the scarce data on the effect of NHF on work of breathing and gas exchange. Models of the upper airways of an adult and neonate were 3D printed. A physiologically realistic breathing flow was driven through the models (with a piston pump) along with the recording of tracheal pressure (with and without NHF). Airway resistance (with and without NHF) was defined by the coefficients of a quadratic fit to the tracheal pressure versus breathing flow. This work is the first to estimate airway resistance under NHF conditions in this way. In the adult model, with and without NHF, tracheal pressure during mouth closed breathing was 3 times that of mouth open breathing. Airway resistance increased in an NHF-dependent manner in both breathing directions (inspiration and expiration) and both mouth states of breathing (mouth open and mouth closed). At peak breathing flows of 30 L/min (for the adult model) and 5 L/min (for the neonate model), the neonatal airway resistance was found to be 4 times that of the adult model. The increased pressures generated by mouth closed breathing suggests that the efficacy of NHF at preventing lung collapse (atelectasis) may be improved via closed mouth breathing. A mathematical model for WOB (based on the Otis equation, which uses the experimentally measured airway resistance) was developed by taking into account reported NHF-induced physiological changes in minute volume and I:E ratio. WOB per minute (rWOB/min) during NHF was greater than without NHF, if minute volume did not change with NHF. A fall in minute volume by 40 % reduced rWOB/min by 90 % (at NHF of 60 L/min) compared to a no NHF condition. NHF can drive part or all of the inspiratory breathing flow across the upper airway and produce a fall in inspiratory rWOB/min. The rWOB/min for mouth closed breathing was 3 times that of mouth open breathing assuming equal fall in minute volume. The adult rWOB/min was found to be 16 times that of the neonate. The effect of NHF on alveolar CO2 tension was estimated by metering CO2 at a specific rate into the piston pump chamber and recording tracheal CO2 tension during breathing, with and without NHF. Apnea was simulated by intermittently stopping the breathing flow for a definite time. End-tidal CO2 concentration (EtCO2) reduced with NHF and reached a plateau at NHF of 40 L/min. Compared to breathing without NHF, 40 L/min of NHF reduced the maximum end-tidal CO2 by 18 %, which suggests a lowering of alveolar CO2 tension and rWOB/min (via attenuation of hyperventilation). It is speculated that NHF may ameliorate the frequency of apnea per hour. A two compartment (lung and dead space) mathematical model of the respiratory system was developed to predict EtCO2 tension during apnea under NHF conditions. The model assumed a sinusoidal breathing flow, constant metabolic CO2 production, instantaneous gas mixing and flushing of the nasopharynx by 3 % of NHF. Apnea was modelled by making the breathing flow zero over a definite time. The model capnogram and EtCO2 for apnea, spontaneous breathing and NHF matched those of the bench-top experiment. The EtCO2 at NHF of 30 L/min differed from that of experiment by 0.9 % (V/V).