Leak Compensation Algorithms: The Key Remedy to Noninvasive Ventilation Failure?

Noninvasive ventilation (NIV) was first used in the early 1980s for the management of chronic respiratory failure and nocturnal hypoventilation. The rapid development of new ventilatory technologies able to synchronize with the patient's demand has made NIV accessible for a large number of indications, such as chronic obstructive diseases, neuromuscular disorders and restrictive pulmonary diseases.^1–3 In this population, the use of NIV has proved to reduce the length of hospital stay, morbidity, and mortality compared with invasive ventilation.^4–6 Noninvasive ventilation is also associated with a significant reduction in the incidence of nosocomial infections, which is 3–5 times lower than that with invasive ventilation.^6–8 Ventilator-associated pneumonia is now described as tube-associated pneumonia, especially in immunosuppressed patients and those with comorbidities.⁹ Because NIV can be provided without the need for deep sedation, it is becoming a major therapeutic tool even outside the ICU. However, it should be applied by a trained and experienced team, because it is not totally exempt from serious adverse effects and complications.

One of the main factors leading to NIV failure is leak. This decreases arterial oxygenation by reducing tidal volumes and patient synchrony, and also causes mouth and throat dryness, conjunctivitis, and sleep disturbances.¹⁰ In an attempt to reduce air leak by tightening the straps of the mask, most caregivers are increasing patient discomfort by inducing nasal skin lesions, which potentially occurs in 50% of patients.^11,12 Minimizing leak is a major clinical concern that can only be achieved by adjusting the patient interface and pressurization levels. Specific algorithms for leak compensation have thus been incorporated into last-generation ventilators to reduce these adverse effects.

Two types of technologies for leak compensation are available: pressure control compensation and volume control compensation. It is important to understand that we are not talking about ventilatory modes stricto sensu, but about algorithms used in the ventilator to compensate air leak. While using volume control compensation, leak volume is estimated from the difference between inspiratory and expiratory volumes. Given the fact that leak is also occurring during exhalation, the ventilator cannot accurately assess the proportion of leak occurring during the inspiratory phase. Thus, tidal volume is often underestimated, leading to inaccurate compensations of the pressure support.¹³ The operating mode of pressure control compensation is more basic, because it simply measures the inspiratory pressure and increases the inspiratory flow in the case of significant leak to reach the pressure set by the operator.¹ Storre et al¹⁰ showed that pressure control compensation has higher compensation capabilities than volume control compensation. Nevertheless, the marked increase in inspiratory flow during compensation might have non-negligible side effects, such as impairing face mask seal and inducing more leak and patient discomfort, or causing gastric insufflation. During NIV, aerophagia can occur in 30–40% of patients, depending on respiratory system characteristics, lower esophageal sphincter pressure, head position, and inspiratory flow.^14,15 Gastric distention compresses the lungs and induces high intrathoracic pressures that therefore require high inspiratory pressure to ventilate the lung. This generates more air leak, and the ventilator compensates by increasing inspiratory flow and thereby creating a vicious circle.^15,16

Another adverse effect of air leak is patient asynchrony. A relevant indicator of a ventilator's efficiency is the inspiratory delay, which is the time needed for the ventilator to reach positive pressure in the patient's airway from the start of inspiratory effort.¹⁷ A long inspiratory delay significantly increases the patient's work of breathing, because it reduces the amount of assistance delivered to the patient in the early stages of the breathing effort.^18–20 Therefore, the ventilator must have efficient triggering and pressurization capacity to promptly meet the patient's demand and reduce the work of breathing and discomfort. The sensitivity of the inspiratory trigger is a key element that depends both on the sensing technology and the sealing level of the inspiratory circuit. Significant leak can induce a pressure drop or generate flows that are perceived as patient inspiratory efforts by the ventilator. This inevitably results in auto-triggering that is a highly undesirable phenomenon that can only be resolved by reducing triggering sensitivity, which in turn creates late cycling or failed triggering, occurring in 12–23% of patients.²¹

In the past 2 decades, NIV has become a new standard of care for adults and children. Improving patient tolerance to NIV by optimizing mask sealing and ventilator synchrony is a new challenge that requires close monitoring of the patient's vital signs and a strong knowledge of the triggering technologies. In a qualitative review of randomized trials published recently, Carron et al¹⁶ concluded that NIV efficiency was mainly determined by the choice of an adequate device, even when performed by experienced teams.

In this issue, Itagaki et al²³ evaluate the leak compensation algorithms of various ICU ventilators and their effects on patient-ventilator synchrony. The authors make extensive investigations and find interesting results that are essential to help clinicians decide which ventilators best meet their expectations and allow them to minimize the risk of NIV failure. This paper also represents a major effort to evaluate and discuss the technical origins of asynchrony, including missed triggering, auto-triggering double-triggering, and cycling problems. Finally, Itagaki et al²³ explain in an easy understandable way the clinical consequences of asynchrony and find some interesting correlations with patient physiological characteristics.

• Copyright © 2017 by Daedalus Enterprises