Hospital-acquired pneumonia has a characteristic etiological pattern reflecting the oropharyngeal colonization by specific flora suffered by patients after hospital admittance. Several factors, such as coexisting diseases, the length of hospital stay, some therapeutic procedures, and the occurrence of certain complications, modulate the timing and characteristics of airway colonization and the probability of developing pneumonia. As shown in Table 1, the bacterial pathogens most frequently associated with nosocomial pneumonia are enteric aerobic Gram-negative bacilli and Staphylococcus aureus. Other organisms such as Streptococcus pneumoniae or Hemophilus influenzae are less common, but a variety of other bacteria, such as streptococci and anaerobes, may be present. Opportunistic bacteria, viruses, and fungi are rarely seen, except in immunosuppressed patients. Legionella pneumophila is an endemic cause of pneumonia in some institutions. It is generally accepted that between a third and a half of cases present a polymicrobial etiology.
Mig e lYhLTJifi* w h m irMhia lactora oral r*f] I h r J r ■ jjvV-1'1 ftpl flv+ft inifpmwH.tilfmrt f mr-ii Qiin-ntqra-» tuctt iwnsmy^Ti. Lhta tw
Unho i' res-ran Sng* tvwi
The accumulated experience of the circumstances that predict the etiology of hospital-acquired pneumonia has identified the key importance of four specific factors: the severity of the episode, the presence of coexisting illness, previous antibiotic therapy, and the length of hospital stay.
Severe presentation of nosocomial pneumonia may be due to the presence of specific circumstances or to particularly virulent causative organisms such as methicillin-resistant Staph. aureus or highly resistant Gram-negative bacteria (Pseudomonas aeruginosa or Acinetobacter species). Certain comorbidities or preceding antibiotic treatments may predispose to certain organisms (Table2). Previous antibiotic treatments during hospital admittance clearly predispose to infection by more resistant organisms such as Ps. aeruginosa, Acinetobacter species, and methicillin-resistant Staph. aureus. Finally, the etiological pattern is also influenced by the length of hospital stay: in 'early- onset' pneumonias (< 5 days) organisms such as Staph. aureus, H. influenzae, and Strep. pneumoniae are predominant, whereas in 'late-onset' pneumonias (> 5 days) Gram-negative bacilli are much more common.
Table 2 Conditions that predispose to specific causes of nosocomial pneumonia
According to the official statement recently published by the American Thoracic Society, patients can be categorized in three groups.
1. Patients without risk factors who present with either mild to moderate nosocomial pneumonia occurring at any time during their stay in hospital or early-onset severe nosocomial pneumonia.
2. Patients with mild to moderate nosocomial pneumonia and risk factors, who develop pneumonia at any time.
3. Patients with severe early-onset nosocomial pneumonia but without risk factors, or patients with severe pneumonia occurring at any time and risk factors. Diagnostic methods
The goals of microbiological studies in the diagnosis of nosocomial pneumonia are as follows.
1. To confirm the infectious origin of the clinical and radiographic features. The efficacy of diagnosis based on clinical and radiological features only is poor in terms of both sensitivity and specificity. Consequently, it is difficult to differentiate pneumonia from other non-infectious causes of febrile pneumonitis.
2. To identify the causative organisms. The etiological spectrum of nosocomial pneumonia is very broad and the diagnostic efficacy of routine diagnostic methods is poor. Although some predisposing factors may provide a guide to which pathogens are most likely to be found, reliable etiological diagnosis is usually obtained by means of specific diagnostic tests.
3. To establish the severity of the episode. Prognosis is partially dependent on the specific infecting organism.
Examination and culture of proximal airways secretions
Sputum or tracheal aspirate microbiological processing (Gram stain and cultures) is the simplest method of studying the etiology of nosocomial pneumonia. This procedure is considered to be very sensitive because the causative organisms involved in pneumonia are usually present in tracheobronchuial secretions. In contrast, its specificity is poor and it has a high rate of false-positive results (i.e. positive cultures in patients without pneumonia) owing to the difficulty of differentiating pneumonia from airways infection or colonization. Nevertheless, although this method is unsatisfactory for separating pneumonia from other causes of febrile lung infiltrates, it can be used to exclude some potential pathogens and, occasionally, can provide conclusive evidence about obligate pathogens (e.g. L. pneumophila) or some opportunistic agents.
Several recent studies have suggested that quantitative cultures of endotracheal aspirates could provide reliable information by substantially increasing the specificity of the procedure. According to these studies, true pathogenic organisms are usually cultured in concentrations higher than 10 6 colony-forming units.
The sensitivity of blood cultures for diagnosing nosocomial pneumonia is low. Additionally, it is not uncommon for blood infection in critical patients to arise from a simultaneous extrapulmonary source. Owing to this lack of specificity, blood-cultured organisms should also be found in respiratory secretions before they can be considered as the causative organisms of pneumonia.
In the event of significant parapneumonic pleural effusion, a pleural fluid sample should be stained and cultured to exclude empyema. If positive, these cultures provide extremely reliable information on the etiology of pneumonia. In practice, however, this procedure is of little help for demonstrating the causative organisms.
Serological studies are of little use in the diagnosis of nosocomial pneumonia. However, they can provide valuable epidemiological information for the identification of hospital-acquired L. pneumophila infections.
Urine cultures are used to exclude an extrapulmonary source of infection. Additionally, detection of L. pneumophila antigen could be useful in institutions where this etiology is common.
The poor performance of non-invasive methods, based on the examination and culture of samples which are easily obtainable by various microbiological techniques, has prompted the recovery of more representative samples by a variety of instrumental procedures. These invasive techniques ( Table..3) enable samples to be obtained directly from, or close to, the site of infection, so that the yield of microbiological processing is substantially increased.
V-tHflctanpc ttWdißJ amfif anprç TrarsJtracic ne«*a Espiraicr ftaflSWhfil Kfilfcfl
Was this article helpful?