HRC 2012-2017
Description
The hipóteses was that dental treatment can contribute to reducing mortality in ICU. The objectives was to evaluate dental treatment outcomes and safety in the Intensive Care Unit (ICU) through a 6-year retrospective analysis of objective metrics related to mortality, hospitalization and number of dental consultations. Methods: Data collected included the frequency of dental consultations, incidence of ventilator-associated pneumonia (VAP), patient demographics, length of ICU stay, and mortality. In the statistical analysis, logistic regression models were utilized to explore associations between dental care and patient outcomes, calculating odds ratios for the outcome of mortality, with adjustments for potential confounders. Results: More than three consultations were associated with a lower risk of mortality in both crude (OR: 3.65, p < 0.0001) and adjusted analyses (OR: 4.24, p < 0.0001), suggesting a protective effect. Patients who received more frequent dental care, which involved the treatment of potential infection sources and the removal of bacterial plaque, demonstrated improved survival outcomes. VAP did not significantly increase mortality risk in this cohort. Dental procedures were not conclusively linked to mortality reduction; however, they were not associated with any significant adverse effects, indicating that they are safe for ICU patients. Conclusions: The findings indicate that regular dental treatment in the ICU may be beneficial to patient survival and do not pose additional safety risks. While VAP did not independently predict mortality, comprehensive dental care was a protective factor. Clinical relevance: These results underscore the potential value of integrating dental treatment into ICU patient care teams and highlight the need for further prospective, multicenter studies to confirm these benefits and inform clinical practices.
Files
Steps to reproduce
For statistical analysis, contingency tables were established between the analyzed variables and mortality. This was followed by the estimation of simple logistic regression models to calculate the crude odds ratios and their respective 95% confidence intervals. Subsequent analysis involved estimating multilevel multiple logistic regression models that accounted for variables at the individual level (first level) and the contextual level (the year as the second level), with the year treated as a contextual variable due to its association with specific patient cohorts. In the multivariable analysis, all variables with a p-value of ≤0.20 from the simple analyses were included, with those maintaining a p-value of ≤0.05 being retained in the final models. The adjusted odds ratios and their 95% confidence intervals were then derived from these multiple models. All analyses were conducted using the R software (R Core Team, 2018. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria), with a significance level set at 5%.