Since the first description of ARDS, PEEP has remained an essential component in its management, aiming to improve gas exchange and keep the airways and alveoli open. However, the response to PEEP varies among patients, depending on pulmonary recruitability. The assessment of this potential can be performed in different ways, but there is still no consensus on the best clinical method for its measurement.

The aim of this study was to compare four common clinical methods of assessing response to PEEP, using mechanical ventilator data and correlating them with clinical and mechanical characteristics of patients with ARDS.

This is a secondary analysis study of a prospective cohort of patients with ARDS. Inclusion criteria were age > 18 years, diagnosis of severe ARDS (PaO2/FiO2 < 100 sustained for at least 2 hours), and mechanical ventilation (MV) time = 48 hours. Patients with contraindications for alveolar recruitment maneuver (ARM), such as intracranial hypertension (ICH), undrained pneumothorax, and severe hemodynamic instability, were excluded. Patients were assessed by four methods before ARM: Stress Index (SI), morphology of the PxV curve, hysteresis, and Recruitment to Inflation ratio (RI-ratio). They were classified as "potentially recruitable" or "poorly recruitable" and underwent ARM as a rescue strategy. After the maneuver, the recruitment response was assessed.

A total of 86 individuals were evaluated, with a mean age of 60.9 ± 12.1 years. The average length of stay was 17.4 ± 12 days, and the average MV time was 13.5 ± 7.8 days, with a mortality rate of 44.2% (n = 38). The morphology of the PxV curve (AUC = 0.827 [95% CI 0.73–0.92]; p = 0.047), hysteresis (AUC = 0.896 [95% CI 0.83–0.96]; p = 0.035), and RI-ratio (AUC = 0.878 [95% CI 0.80–0.95]; p = 0.037) were equivalent predictors, with no statistically significant difference between them (SMD: 0.126, 95% CI -0.113 - 0.129; p = 0.867). SI had an AUC of 0.64 (p = 0.061), not being a good predictor of response to PEEP. The "cutoff points" found were lower than those established in previous studies, both in hysteresis (VCe – VCi = 378 ml; GINI index = 0.792) and in RI-ratio (0.335; GINI index = 0.756).

The clinical methods used for evaluating recruitment potential were accurate and can be clinically applied to estimate recruitment potential in patients with severe ARDS, although the cutoff points found were lower than those described in previous studies.

Patients who could benefit from ARM may be incorrectly classified as non-recruitable, which could lead to the omission of the maneuver, even when rescue of severe refractory hypoxemia is necessary. This could hypothetically contribute to a reduction in morbidity and mortality and promote better clinical outcomes in this heterogeneous syndrome.