Ventilator-associated pneumonia (VAP) has been a known complication in the intensive care unit (ICU) since the late 1950s. Originally VAP was recognized as a cause of rising rates of Gram-negative, necrotizing pneumonia, which was uncommon at the time, and was attributed to ventilator and respiratory therapy equipment contaminating patients [1]. Subsequently, a number of studies demonstrated that critically ill patients had respiratory tract colonization, by their own Gram-negative flora, and these organisms often proliferated in endotracheal tube biofilm, and condensated in ventilator circuits, where they were often re-inoculated into patients during endotracheal suctioning and tubing circuit changes [2].

VAP was commonly reported in the 1980s where it occurred in up to 28% of mechanically ventilated patients, with the highest rates early in the course of intubation (3% per day risk up to day 5) [3]. These high rates were reported, in spite of controversies about overdiagnosis using clinical definitions, and whether bronchoscopic sampling was needed. VAP was not only the most common ICU-acquired infection, but had a mortality rate as high as 50%, with at least 25% of these deaths directly attributable to the infection, and not the underlying diseases [2, 3]. More recent studies have estimated a much lower attributable mortality for VAP [4].

Early and appropriate therapy has been consistently demonstrated to reduce mortality, and the efficacy of therapy has been challenged by the presence of multidrug-resistant Gram-negative and Gram-positive pathogens. In addition to endotracheal intubation itself, other risk factors for VAP include underlying serious illness (coma, acute lung injury, aspiration gastric colonization) and a variety of interventions (e.g., H2 blockers, reintubation, supine head position, low endotracheal tube cuff pressure). This information was used in the early part of this century to develop “ventilator bundles”, which dramatically reduced the reported rates of VAP. In fact, at one point, it was presumed possible to have “zero VAP”, and there was a belief that VAP was a medical error, fully preventable with simple interventions such as head of the bed elevation, daily awakening and weaning, and provision of oral care [5].

New classification

In the last several years, a new classification of pneumonias acquired during ICU stay has emerged and reflects the development of non-invasive ventilation and more commonly elderly and frail patients being admitted to the ICU. The new classification expanded hospital-acquired pneumonia (HAP) into ventilated and non-ventilated ICU-acquired pneumonias, while a new diagnosis emerged for ventilator-associated tracheobronchitis (VAT) [6,7,8]. Unlike VAP, patients with ventilated HAP were usually intubated after the onset of infection, and not with a preceding period of 48 h of ventilation. The definition of VAT shares the same criteria as VAP, except without the presence of new pulmonary infiltrates on portable chest radiograph [8]. The absence of lung infiltrates does not exclude the possibility that a percentage of VAT could be actual VAP, if a computed tomography scan is performed. It is thus possible that some reports of “zero VAP” were created artificially by reporting possible VAP as “ventilator-associated tracheobronchitis”, or identifying intubated HAP patients that fulfil VAP criteria, as “ventilated HAP” and not VAP, among other potential explanations.

Old and new challenges

A recurrent issue in VAP is making an accurate diagnosis in patients with a clinical suspicion of pneumonia [9]. In daily ICU practice, clinicians still use the presence of new radiographic infiltrates plus at least two of the classical clinical criteria for VAP diagnosis. Overall these criteria had 70% sensitivity and specificity in a postmortem study [9]. Recent developments in VAP diagnosis include the use of bedside lung ultrasound to detect pulmonary infiltrates compatible with pneumonia, and molecular point-of-care tests of respiratory secretions to identify potential pathogens. In skilled hands lung ultrasound has an important complementary role in VAP diagnosis [10]. The advantage of lung ultrasound is its non-invasive use at the bedside as both a diagnostic tool, and as a method to follow the response of VAP to treatment. The appropriate use of lung ultrasound in the diagnosis and management of VAP is still being defined in terms of patient-centered outcomes. Other unsolved topic in VAP is the use of invasive or non-invasive respiratory sampling for microbiological diagnosis. Potential new randomised controlled trials (RCTs) focusing on comparing each strategy associated with protocols for antibiotic stewardship, or applying molecular diagnostic methods could add to the field [11]. A main challenge still remains defining a gold-standard for VAP diagnosis.

Rapid and accurate microbial diagnosis of VAP is still a matter of debate. Recent advances in molecular tests provide promising tools for identifying pathogens and resistance profiles. A pilot RCT using the polymerase chain reaction (PCR) to detect methicillin-resistant S. aureus (MRSA) in the bronchoalveolar lavage (BAL) of mechanically ventilated patients has demonstrated better diagnostic performance and antibiotic management than with traditional methods [12]. Other multicenter studies show a very good sensitivity and good concordance of rapid molecular tests for both MRSA and Gram-negative bacilli with conventional cultures [13, 14].

In the last decade, the emergence of multi-drug and extensively drug-resistant (MDR and XDR) Gram-negative bacilli has presented a tremendous challenge for clinicians. Experts commissioned by the World Health Organisation (WHO) prioritized carbapenem-resistant Acinetobacter, ESBL-producing Enterobacterales, and carbapenem-resistant Pseudomonas aeruginosa as the major challenges for the future [15]. Importantly, there is a worldwide variability of the prevalence of these microorganisms and their different mechanism of resistance. In the last 5 years, several new antibiotics have been studied and approved for use in VAP [16]. Most of them include the combination of a beta-lactam or a carbapenem with a beta-lactamase inhibitor (ceftazidime–avibactam, ceftolozane-tazobactam, imipenem–relebactam, meropenem–vaborbactam) or beta-lactams with new mechanisms of action (e.g., cefiderocol), some of them with broad activity against almost all MDR/XDR microorganism [16]. Although timely and accurate treatment is fundamental for better outcome, empiric overtreatment is also frequent in VAP and necessitates an organized approach to antibiotic stewardship. Importantly, risk factors and scores for MDR have not been accurate enough so far to better target initial empiric treatment.

“VAP is back”

With the advent of the pandemic caused by the coronavirus disease 2019 (COVID-19) in 2020, much has changed and VAP returned, or always has been, as a main issue in ICUs worldwide [17]. Many series report high rates of VAP, in spite of modern prevention efforts, with reported rates over 40%, using bronchoscopic diagnosis [18]. Clearly VAP has not gone away, and this resurgence may be explained by the realization in high-income countries that it is now “politically acceptable” to accurately report this illness. Other factors that can explain the resurgence of high incidence rates of VAP during COVID-19 include the severity of COVID-19 illness per se and its associated treatments (e.g., deep sedation, prolonged mechanical ventilation, corticosteroid and anti-IL 6 treatments), along with a decrease in nurse-to-patient ratios, and less compliance with preventive measures [17].

Based on lessons from the past, we have learned the key management issues in VAP, a disease that is not going away. In the future, we need to develop new approaches and future investigations should focus on epidemiology, prevention, diagnosis and treatment of VAP (Table 1).

Table 1 Main challenges and lessons learned from the VAP legacy
Full size table