QUESTION: Does the instillation of normal saline before suctioning reduce the incidence of ventilator-associated pneumonia in intubated and ventilated adults?

DESIGN: Randomised, controlled trial with blinded outcome assessment.

SETTING: The medical/surgical intensive care unit of a tertiary oncology hospital in Brazil.

PARTICIPANTS: Adults expected to require at least 72 hours of mechanical ventilation via an endotracheal or tracheostomy tube. Previous ventilation within the past month and contraindications to bronchoscopy were exclusion criteria. Randomisation of 262 participants allotted 130 to the intervention group and 132 to a control group.

INTERVENTIONS: Closed tracheal suction systems with heat and moisture exchangers were used with both groups and were changed regularly. All patients were nursed with backrest elevation to 45 degrees. medical or nursing staff, who were blinded to group allocations, requested suctioning when any of the following occurred: visible or audible secretions, ventilator-patient asynchrony, and increased peak inspiratory pressures or decreased tidal volumes attributed to secretions. Respiratory therapists performed the suctioning according to a standardised procedure that included preoxygenation. The therapists instilled 8 mL of normal saline prior to suctioning in the intervention group only.

OUTCOME MEASURES: The primary outcome was the incidence of ventilator-associated pneumonia (VAP). If VAP was suspected because of radiographic evidence plus either fever, leukocytosis, or purulent secretions, a bronchoscopy with standardised lavage was performed. VAP was considered confirmed if the bacterial density of the lavage fluid exceeded 1000 colony-forming units/mL. Secondary outcome measures included time to VAP, duration of mechanical ventilation, length of stay and mortality in the intensive care unit, unscheduled ventilation circuitry changes due to secretions, and number of suctions per day.

RESULTS: all participants completed the study. Significantly fewer participants in the saline group developed VAP (14/130) than in the control group (31/132), relative risk reduction 0.54 (95% CI 0.18 to 0.74). This indicates that one patient will avoid developing VAP for every 8 patients in which saline instillation is used. Significant benefits of saline instillation were also seen in the incidence of VAP (9 vs 21 per 1000 days of mechanical ventilation, p = 0.01) and in the time to first VAP (p = 0.02). The groups did not differ significantly on the remaining secondary outcomes.

CONCLUSION: Instillation of normal saline before tracheal suctioning decreases the incidence of VAP in mechanically ventilated adults.