Respiratory infections in patients undergoing mechanical ventilation Jordi Rello, Thiago Lisboa, Despoina Koulenti Lancet Respir Med 2014; 2: 764–74 Published Online August 21, 2014 http://dx.doi.org/10.1016/ S2213-2600(14)70171-7 Critical Care Department, Hospital Universitari Vall d´Hebron, Barcelona, Spain, Centro de Investigación Biomédica en Red Enfermedades Respiratorias, Barcelona, Spain, and Universitat Autonoma de Barcelona, Barcelona, Spain (Prof J Rello PhD); Critical Care Department and Infection Control Committee, Programa de Pós-Graduação Pneumologia, Hospital de Clinicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil, and Rede Institucional de Pesquisa e Inovação em Medicina Intensiva, Complexo Hospitalar Santa Casa, Porto Alegre, Brazil (T Lisboa MD); and 2nd Critical Care Department, Attikon University Hospital, Athens, Greece, and Burns Trauma and Critical Care Research Centre, The University of Queensland, Brisbane, QLD, Australia (D Koulenti PhD) Correspondence to: Dr Jordi Rello, Critical Care Department, Hospital Universitari Vall d´Hebron, PS Vall d´Hebron 119, Anexe A - 5a planta, 08035 Barcelona, Spain [email protected]
Lower respiratory tract infections in mechanically ventilated patients are a frequent cause of antibiotic treatment in intensive-care units. These infections present as severe sepsis or septic shock with respiratory dysfunction in intubated patients. Purulent respiratory secretions are needed for diagnosis, but distinguishing between pneumonia and tracheobronchitis is not easy. Both presentations are associated with longlasting mechanical ventilation and extended intensive-care unit stay, providing a rationale for antibiotic treatment initiation. Diﬀerentiation of colonisers from true pathogens is diﬃcult, and microbiological data show Staphylococcus aureus and Pseudomonas aeruginosa to be of great concern because of clinical outcomes and therapeutic challenges. Key management issues include identiﬁcation of the pathogen, choice of initial empirical antibiotic, and decisions with regard to the resolution pattern.
Introduction In the Extended Prevalence of Infection in Intensive Care II (EPIC II) study,1 which looked at 1265 intensivecare units (ICUs) in 75 countries, 51% of adults admitted to ICUs were infected, and the respiratory tract was the focus of infection in 64% of cases. Airway infection in intubated patients is the main reason for antibiotic prescription in medical ICUs. Because no gold standard exists for the diagnosis of respiratory tract infections in intubated patients,2 prescription of antibiotics for patients with purulent respiratory secretions is common clinical practice in the ICU setting. The panel shows the clinical challenges in management of respiratory tract infections in ventilated patients with a hypothetical clinical scenario. In this Review we discuss ventilator-associated respiratory infections (VARIs) in adult patients, with an emphasis on diagnosis, microbiological causes, and management.
showing the limitations of present deﬁnitions and diagnostic criteria. The clinical pulmonary infection score was created to predict the pretest probability of pneumonia.7 It combines information on body temperature, volume and appearance of tracheal secretions, chest radiograph ﬁndings, white blood cell count, oxygenation, and tracheal aspirate culture.7 Many studies have used the clinical pulmonary infection score to identify patients with pneumonia because it allows an objective assessment of clinical variables for pneumonia diagnosis.8,9 Unfortunately, despite the use of objective data, such as white blood cell count and oxygenation, the clinical pulmonary infection score also includes variables that are either subjective or retrospective, such as chest radiograph ﬁndings, secretion appearance, and microbiological data, which might
Patients with lower respiratory tract infection usually present with progressive hypoxaemia and fever, which is unlike the sudden onset of rigor and temperature rise associated with bloodstream infections. In our view, the low sensitivity and speciﬁcity of present diagnostic criteria is the most important diﬃculty in the assessment and diagnosis of mechanically ventilated patients with suspected lower respiratory tract infections.3 Deﬁnition of a clinical syndrome on the basis of the clinical presentation of VARI is thus a challenge for clinicians. Because the diagnostic criteria include many subjective components (eg, chest radiographs, assessment of respiratory secretions, or auscultation), interobserver variability in the identiﬁcation of ventilator-associated pneumonia (VAP) is high.4,5 In a prospective survey6 of a nationally representative group of US hospitals, participants were asked to classify standardised vignettes of possible cases of VAP as having pneumonia or not having pneumonia. Agreement at the hospital level on classiﬁcation of cases as VAP or not was nearly random,
• Diﬀerentiation of ventilator-associated pneumonia and tracheobronchitis on the sole basis of clinical signs can be a diﬃcult task at the bedside in view of chest radiograph interpretation variability and other subjective clinical variables • High rates of pathogen resistance have emerged in patients with ventilator-associated pneumonia, making therapeutic choice a challenge, especially for Gram-negative pathogens • Improvements in pathogen detection, use of biomarkers to guide treatment, and new antibiotic delivery devices are promising strategies to optimise ventilator-associated respiratory infection treatment • Clinical response assessed 3 days after ventilatorassociated respiratory infection onset is a key issue in optimisation of therapy • Severity of illness and underlying conditions are the most important variables inﬂuencing survival. • Immunocompromise and delay in adequate antibiotic therapy, such as management of hypoxaemia and shock, results in increased use of health-care resources
www.thelancet.com/respiratory Vol 2 September 2014
Panel: Hypothetical clinical scenario of a mechanically ventilated patient developing pneumonia A 52-year-old man with underlying chronic obstructive pulmonary disease underwent colectomy for colon cancer. He presented a complex postoperative course, receiving ceftriaxone and metronidazole. After 1 week on mechanical ventilation, the patient was not successfully weaned and a tracheostomy was done. Respiratory secretions became purulent 2 days after tracheostomy and the chest radiograph was consistent with basal atelectasis. The patient developed worsening hypoxaemia the following day, with a temperature increase to 38·5°C. The white blood cell count was 14 000 cells per μL and the C-reactive protein concentration was 181·63 μmol/L (normal values