Ventilator-associatedpneumonia (VAP) is amongthemost frequently occurring infections among critically ill patients, and its development is associatedwith poor patient prognosis. As such, VAP has been an important topic of clinical research in intensivecaremedicine,pulmonology, infectiousdiseases, and clinicalmicrobiology. Yet, aftermore than 30 years of clinical studies on diagnosis, treatment, and prevention, a myriad of uncertainties remain. Among the many challenges that confront our efforts to effectively treat VAP is that there is currently no reliable gold standard fordiagnosis.Diagnosis is essentiallybasedonacombinationof clinical symptoms,microbiological test results, and radiological interpretations. These criteria, even in combination, are highly subjective and have suboptimal specificity. Subjective diagnosis of VAP hampers unbiased evaluation of the efficacy of preventive measures in unblindedstudies and increases the riskof assessmentbiaswhen VAP rates are used as a quality metric. Poor specificity of the diagnosis leads to unnecessary antibiotic use and its associatedharms.Ofnote, thediagnostic approachwithbetter specificity, ie, bronchoscopic procedures combinedwith quantitative microbiological cultures, is hardly used because it is invasive and expensive. The other sizable challenge in VAP management is that the evidence for preventive strategies is less robust than it appears on first examination. For instance, most guidelines recommend semirecumbent patient positioning. However, this recommendation is based on a prematurely interrupted controlled trial of 87 patients, in which the beneficial effects of the semirecumbent position could not be separated from the detrimental effects of continuous gastric feeding in control patients randomized to a complete supine position.1 The only other randomized trial of this intervention in intensive care unit (ICU) patients questioned the feasibility of the 45° position and failed to demonstrate a benefit from the 30° position.2 Some guidelines also recommend topical antimicrobial prophylaxis, such as selective decontamination of the digestive tract andselectiveoropharyngealdecontamination.These treatments have been associatedwith statistically significant reductions in the incidence of VAP inmany randomized trials and with improved patient outcome in settings with low levels of antibiotic resistance.3 Yet, absence of evidence of benefit in settings with higher levels of antibiotic resistance and fear of selection for antibiotic resistancehavepreventedwidespread adoption of these measures. There is greater consensus for theuseof chlorhexidinegluconate oropharyngeal care as a strategy for VAP prevention, in part because of its perceived safety, both for individual patients andwith respect to avoiding antibiotic resistance.However, althoughchlorhexidine iswidelyusedandhasbeen identified as an important process of high-quality care by the Institute forHealthcare Improvement, theevidence that it prevents VAP or improves patient outcome isweaker than for selective decontamination. In this issue of JAMA Internal Medicine, Klompas and coworkers4 question the effectiveness and safetyof chlorhexidineoral care inpatients receivingmechanicalventilation inthe ICU. This conclusion is based on the results of a new metaanalysis of 16 randomized trials evaluating this intervention in 1868 cardiosurgical patients (3 studies) and 1762 intensive care patients (13 studies).This stratifiedmeta-analysis suggests that chlorhexidine oral care is effective in cardiosurgery patients, whogenerallyundergo intubationandventilationonly forshort periodsaftersurgery,butnotamongintensivecarepatients,who generally undergo intubation for a longer period. In the intensive care patients, the authors found only statistically nonsignificanteffectsonthepreventionofVAP; inopen-label trials the summary risk ratio (RR) was 0.61 (95% CI, 0.35-1.04) and in double-blind trials the RR was 0.88 (95% CI, 0.66-1.16). Furthermore, there was a statistically nonsignificant result of increasedmortalityamong intensivecarepatients randomized to receivechlorhexidineoralcare(RR,1.13[95%CI,0.99-1.29]),with some evidence of a dose-response effect. Related article page 751 Chlorhexidine Oral Care andMechanical Ventilation Original Investigation Research
Read full abstract