The Coronavirus Disease 2019 (COVID-19) pandemic severely impacted patient care in Intensive Care Units (ICU) over the world as the disease can develop into an Acute Respiratory Distress Syndrome (ARDS) (1, 2). ARDS causes pulmonary edema, severely impaired gas exchange, and reduced lung compliance (3–5). As a result, invasive mechanical ventilation is often necessary to maintain sufficient gas exchange, which can in turn lead to aggravation of lung injury through excessive transpulmonary pressure or high tidal volumes resulting in Ventilator Induced Lung Injury (VILI) (6, 7). Despite current lung-protective ventilation strategies aiming to reduce VILI, mortality in ARDS remains high (8, 9).

“Time-Controlled Adaptive Ventilation” (TCAV) is a relatively new approach that utilizes Airway Pressure Release Ventilation (APRV) mode. TCAV can be conceptualized as continuous positive airway pressure (CPAP), that is intermittently interrupted by a release phase during which gas is expelled. The CPAP phase, which constitutes approximately 90% of the total respiratory cycle, allows for time-dependent lung recruitment. The release phase terminates when 75% of the expiratory peak flow is achieved (10, 11). Experimental studies have demonstrated that this approach restores lung volume while preventing VILI (12, 13). To date, there is a lack of studies that have tested the efficacy of TCAV in relation to patient-centered endpoints in a clinical setting (14).

Due to inflammation-induced microthrombi leading to microcirculatory disturbances and disruptions in hypoxic pulmonary vasoconstriction, ARDS can result in an increase in physiological dead space volume (15, 16). Numerous studies have identified increased dead space ventilation as an independent predictor of higher mortality in ARDS (15, 17, 18). The Ventilatory Ratio (VR) is an easy to obtain bedside index of ventilatory impairment. It exhibits a near-exponential relationship with physiological dead space fraction in subjects with ARDS and is associated with outcome (19, 20). VR is calculated by dividing the product of measured minute ventilation and measured PaCO₂ by the product of ideal minute ventilation and ideal PaCO₂ for the patient (19, 21, 22).

Under physiological conditions, VR is approximately 1. If pulmonary gas exchange is impaired, as in ARDS, a given minute ventilation may no longer be sufficient to achieve physiological PaCO₂ values. Therefore, elevated VR values indicate ineffective ventilation with an inability to achieve adequate decarboxylation at given minute ventilation levels.

It was hypothesized that ventilation with APRV mode could achieve continuous, progressive lung tissue recruitment in ARDS patients over hours to days, while preventing subsequent de-recruitment, leading to corresponding reductions in VR and increases in P/F ratio over time. The aim of this dual-center retrospective cohort study was therefore to evaluate the effects of APRV on pulmonary gas exchange utilizing the VR to develop hypotheses for improving patient care and reducing morbidity.

A total of 107 patients were included in the study. In 48 patients, APRV was used at least once, whereas in 59 patients, only LTVV was used. Overall patient characteristics are presented in Table 1.

In this retrospective cohort study of patients with moderate to severe ARDS, we found that overall in-hospital-mortality did not differ between ventilation with APRV and ventilation with LTVV. However, we found improved oxygenation and indications of reduced dead space ventilation and higher ventilation efficiency when patients were transitioned to APRV after prior ventilation with LTVV.

To capture the effects of APRV over an extended period, a duration of 72 h of continuous APRV following at least 12 h of observed LTVV was chosen. There was a statistically significant effect on the P/F ratio over time. After transition to APRV, mean P/F ratio at 24 h was 43% higher than before the transition.

Numerous studies have reported a positive effect of APRV on oxygenation (23–26). The observed increase is likely attributable to enhanced alveolar recruitment and stabilization. Gradual opening and stabilization of opened alveoli are key elements of TCAV. Repetitive alveolar collapse and expansion (RACE) is effectively prevented (27–29). Prolonged CPAP-Phases and terminating P_low at 75% of expiratory peak flow has shown to be the most effective way to achieve this stability when using APRV (30). Strict protocol adherence when ventilating patients using TCAV is therefore paramount. Although the average P/F ratio increased, this effect was not equally strong in every patient. Individual patient trajectories demonstrated a heterogeneous response. In 4 of 8 patients an increase in P/F-Ratio of ≥50 mmHg within 24 h could be seen, whereas the remaining patients showed minimal or no improvement. This responder distribution suggests that mean effects were driven by clinically meaningful improvements in a subset of patients rather than uniform small changes across the cohort. A common characteristic among patients who improved was that the transition to APRV occurred within the first 72 h following endotracheal intubation. In contrast, all 4 patients with little to no improvement in oxygenation were transitioned to APRV on the third day of mechanical ventilation or later. Particularly in this early phase of ARDS, inappropriate ventilation patterns and mechanical stress can damage the vulnerable lung (31). In these 4 patients, VILI may have already developed to some extent as a result of atelectrauma through RACE caused by LTVV. Due to viscoelastic properties of the lung, the short release phase used in TCAV can prevent de-recruitment of opened lung areas and achieve alveolar stabilization (32). Therefore, early application of TCAV has the potential to prevent the progression of ARDS and VILI, thereby facilitating enhanced lung healing (33, 34).

After transition to APRV, there was a significant reduction in VR. This reduction was already statistically significant after 6 h. The mean VR approached a value of 1.0 after the transition to APRV. This implies that, with a constant minute ventilation, more CO₂ could be eliminated, or alternatively, a lower minute ventilation was sufficient to achieve a defined CO₂ elimination. Consequently, a lower VR could be interpreted as an indicator of more efficient ventilation under APRV compared to LTVV. One potential reason for this effect could be improved lung tissue recruitment and prevention of de-recruitment, resulting in an increase in physiologically active alveolar surface area. An additional explanation is the diffusion of CO₂-containing respiratory gas from the alveolar compartment toward the central airways during the prolonged CPAP-phase (35). This effectively reduces the physiological dead space and achieves a VR that is otherwise only observed in patients with healthy lungs. Similar to the present study, some retrospective studies have investigated VR after initiation of APRV. Comparable to our results, these studies reported a decrease in mean PaCO₂ despite a lower minute ventilation and, consequently, a reduction of the VR (36, 37). It should be noted that in studies demonstrating a correlation between physiological dead space ventilation and VR, mechanical ventilation was exclusively performed using LTVV (19, 22). Therefore, it remains questionable to what extent VR can also estimate dead space ventilation in APRV. Nevertheless, a VR closer to 1.0 suggests higher efficiency in the physiological utilization of the applied minute ventilation. In our study, this assumption is further supported by the fact, that VR continued to show a decreasing trend even after 6 h, reaching its lowest value after 24 h. Similar to the decreasing VR, the P/F ratio showed a steady increase in the hours following the transition. This dynamic development, with continuous improvement in both oxygenation and CO₂ elimination under APRV, could imply ongoing recruitment of collapsed alveolar surface area. Our data indicate that maximum lung recruitment is achieved within a period of approximately 24 h. While strategies aimed at alveolar recruitment over seconds or minutes could not achieve significant success, a continuous recruitment over hours or even days might be the optimal method to reopen collapsed lung areas (38, 39). This gradual reopening of the de-recruited lung might be less damaging to the injured lung than previously published approaches (27).

Some studies have demonstrated lower mortality with APRV compared to LTVV (24–26). However, this association could not be reproduced in the present study. A potential explanation for this could be the retrospective study design, which lacked a standardized experimental protocol and randomization of the therapeutic strategies. This could have led to many difficult-to-measure covariates, which might have significantly influenced mortality. Furthermore, the sample size was small.

When APRV was used as a rescue therapy, only 2 out of 15 patients survived the remainder of their ICU stay. In these patients, pulmonary function had already been severely impaired, immensely worsening the clinical prognosis. In such cases, it is reasonable to conclude that APRV was unlikely to facilitate a significant recovery.

It has to be noted that there was no standard operating procedure regarding the use of APRV during the observed time period in the included ICUs. This resulted in APRV being used in heterogeneous situations and settings. Although TCAV was used as a template by the practitioners, in some cases, P_low was not set to 0 cmH₂O. In TCAV, the end-expiratory lung volume is not determined by the extrinsically applied PEEP, but by the short expiratory time (15). A P_low set higher than 0 cmH₂O can significantly affect the expiratory airflow and undermine the underlying principles of TCAV (40).

Our study highlights the potential of APRV to enhance gas exchange in patients with moderate to severe ARDS. Based on the observed changes in blood gasses over time, it can be concluded that APRV may potentially recruit alveolar surface area, reduce dead space ventilation, improve oxygenation, and enhance ventilation efficiency. With TCAV, there is an established, easy-to-use protocol which incorporates fundamental physiological advantages and has shown promising results in recent studies. Taken together, these findings suggest that APRV, particularly when applied early and according to TCAV principles, may improve oxygenation and ventilation efficiency in selected ARDS patients. However, given the small sample size, these results should be viewed as hypothesis-generating and warrant confirmation in adequately powered prospective trials.