Background: Respiratory secretions are a common complication in patients receiving invasive mechanical ventilation, yet no validated tool is available to classify secretion type and amount in this setting. Methods: We conducted a prospective study in 2 mixed ICUs in southern Brazil (January 2022–January 2024). Adult patients intubated for >24 h and requiring suctioning were eligible. Secretions were collected during routine suctioning, photographed, and classified using the newly developed Respiratory Secretion Scale (RESES). The scale categorizes secretions into 7 types (mucoid, mucopurulent-1, mucopurulent-2, purulent, bloody-mucoid, bloody, bloody-purulent) and 3 amounts (small, medium, large). Inter-observer reliability was evaluated using Cohen’s kappa. Associations between secretion characteristics and clinical variables were analyzed using chi-square, t test, analysis of variance (ANOVA) test, and multinomial logistic regression. Results: Among 70 subjects included in the validation phase, inter-observer agreement was excellent (kappa type = 0.87; amount = 0.85; combined = 0.82). In 398 subjects analyzed for clinical associations, purulent and bloody-mucoid secretions were more frequent among nonsurvivors, whereas mucoid secretions predominated among survivors (P = .03). Large secretion amounts were associated with respiratory disease, pulmonary infection focus, and longer duration of mechanical ventilation (all P < .05). In the multinomial logistic regression, the diagnosis and the pulmonary infection focus remained significantly associated with both secretion type (Relative Risks RR = 1.28, 95% CI: 1.05–1.56, P = .02; RR = 1.55, 95% CI: 1.12–2.14, P = .008, respectively) and amount (RR = 1.62, 95% CI: 1.08–2.44, P < .02; RR = 1.71, 95% CI: 1.09–2.68, P = .02, respectively), although chest imaging with consolidation was associated only with secretion amount (RR = 1.05, 95% CI: 1.01–1.09, P < .007). Conclusions: The RESES is a simple, visual tool that standardized secretion assessment in mechanically ventilated subjects, improving clinical communication and providing a foundation for future research.
Ferrari et al. (Thu,) studied this question.