Perioperative cetylpyridinium chloride spray to the tongue dorsum and bacteria recovered from tracheal aspirates: a randomised feasibility study

To assess the feasibility of a randomised trial of perioperative cetylpyridinium chloride (CPC) spray to the tongue dorsum in patients undergoing oral cancer resection and reconstruction, and to obtain preliminary, hypothesis-generating estimates of its effect on oral hygiene and on the bacterial profile of tracheal aspirates. This single-centre, randomised, parallel-group, open-label feasibility trial (UMIN000044932) allocated 17 patients to a 0.3% CPC tongue-dorsum spray or to a no-spray control arm. Feasibility outcomes were enrolment rate, adherence to the spray regimen, completeness of oral and tracheal-aspirate sampling, and safety. Preliminary efficacy outcomes were the Simplified Oral Hygiene Index (OHI-S) and tongue-coating score, and the bacterial load and amplicon sequence variants (ASVs, a measure of sequence-variant richness derived from 16S rRNA V3–V4 sequencing) recovered from oral samples and tracheal aspirates. Among the microbiological outcomes, the between-group difference in tracheal-aspirate ASV counts at extubation was treated as a key exploratory comparison. No imputed numerical ITT estimates were generated for postoperative oral-hygiene, oral microbiological, or tracheal-aspirate outcomes because some participants had structurally missing post-treatment values and no pre-specified imputation rule was available; these outcomes are therefore reported as outcome-specific evaluable-sample analyses. Of 17 randomised patients (9 CPC, 8 control), 14 were included in the per-protocol analysis (CPC n = 8; control n = 6); three patients were excluded after randomisation (two for reconstructive-flap necrosis, one for poor postoperative oral hygiene). Adherence in the per-protocol CPC group was complete (184/184 scheduled administrations across 8 participants). OHI-S and tongue-coating scores decreased significantly within the CPC group from baseline to postoperative day 7 (Wilcoxon signed-rank test, p = 0.042 and p = 0.018; unadjusted), but not within the control group; no significant between-group differences were detected at any time point. ASVs in tracheal aspirates were numerically lower in the CPC group, but the between-group difference did not reach statistical significance (Mann–Whitney U test, p = 0.073). Post-extubation tracheal-aspirate outcomes are reported for the corresponding evaluable-sample sets. Postoperative pneumonia occurred in one control patient and no surgical-site infections were observed; the trial was not designed or powered for these clinical endpoints. In this small, exploratory feasibility trial, perioperative CPC tongue-dorsum spray was deliverable within routine perioperative care, with complete recorded administration in the per-protocol CPC cohort and no recorded intervention-related adverse events. The findings provide preliminary, hypothesis-generating signals regarding oral hygiene and the bacterial profile of tracheal aspirates, but do not support any inference of clinical efficacy for the prevention of postoperative pneumonia. An adequately powered multicentre randomised trial is required.