Nosocomial pneumonia accounts for approximately 10 to 15 per cent of all hospital-acquired infections. It is currently the second leading cause of nosocomial infections, after urinary tract infections. The overall risk of acquiring nosocomial pneumonia is approximately 6.0 to 8.6 infections per 1000 admissions. The incidence of nosocomial pneumonia is greatly increased for all patients in ICUs, where respiratory infections have been reported to be the most frequent type of nosocomial infection. The reported incidence in ICUs ranges from 12 to 29 per cent. The risk to patients receiving mechanical ventilation is as much as 20 times greater, with rates as high as 25 to 70 per cent.
The spectrum of potential pathogens associated with nosocomial pneumonia differs from that of community-acquired pneumonia. The bacterial pathogens that are most frequently associated with nosocomial pneumonia are enteric Gram-negative bacilli and Staph. aureus, but accumulating data suggest that the etiology is polymicrobial in up to 50 per cent of mechanically ventilated patients.
The spectrum of potential pathogens can be identified by assessing several factors such as the severity of the episode, the presence of coexisting illness, the prior antibiotic therapy, and the length of time in hospital.
Knowledge of these factors allows patients to be separated into easily identifiable groups which form the basis for therapeutic decisions. In order to classify a patient appropriately, three questions must be asked.
1. Is the pneumonic process mild to moderate, or is it severe (Table.?)?
Table 2 Definition of severe nosocomial pneumonia
2. Are specific host or therapeutic factors predisposing to specific pathogens present?
3. Is the pneumonia of early onset (occurring within less than 5 days of admission) or of late onset (occurring 5 days or more after admission)? According to the American Thoracic Society official statement published recently, patients can be categorized into three groups.
1. Patients without risk factors who present with mild to moderate nosocomial pneumonia occurring at any time during their stay in hospital or with severe nosocomial pneumonia of early onset.
2. Patients with mild to moderate nosocomial pneumonia and risk factors who develop pneumonia at any time.
3. Patients with severe nosocomial pneumonia of early onset and no risk factors, or patients with severe pneumonia occurring at any time and risk factors.
The core organisms of group 1 are the enteric Gram-negative bacilli, H. influenzae, Strep. pneumoniae, and methicillin-sensitive Staph. aureus. These micro-organisms can be treated with second-generation or non-pseudomonal third-generation cephalosporins ( Ta.b.!§...3.). Combinations of b-lactams and b-lactamase inhibitors can be also used. In the case of penicillin allergy, fluoroquinolones, or clindamycin plus aztreonam can be given. Monotherapy is usually appropriate in this setting.
Table 3 Core organisms and core antibiotics in immunocompetent patients
Patients from group 2 are susceptible to certain additional bacteria beyond the core organisms, such as anaerobes, Staph. aureus, Legionella species, and Ps. aeruginosa. These should be considered according to the presence or absence of specific risk factors for these organisms ( Iable 4). All of these patients should be treated for the core organisms as above, but usually require the addition of other antimicrobial agents to provide cover for other likely pathogens. For example, clindamycin or metronidazole are active against anaerobes and can be added to the core antibiotics in witnessed or suspected cases of gross aspiration, although a b-lactam plus b-lactamase inhibitor combination may be sufficient. Vancomycin should be added in cases with coma or head trauma until methicillin-resistant Staph. aureus (MRSA) is excluded. If Legionella infection is suspected (e.g. patients on high doses of steroids), erythromycin (with or without rifampin) should be administered.
Treatment of patients from group 3 should be directed against the core organisms as well as against more resistant and virulent Gram-negative bacilli such as Acinetobacter species and Ps. aeruginosa. These micro-organisms require the use of combination antimicrobial therapy. This can be achieved by an aminoglycoside or ciprofloxacin plus an anti-pseudomonal penicillin, a b-lactam plus b-lactamase inhibitor combination, ceftazidime or cefoperazone, imipenem, or aztreonam. In some cases it may be possible to complete therapy using a single agent. The decision to continue combination therapy can be made after treatment for 2 or 3 days based on the clinical response and microbial cultures. For instance, if Ps. aeruginosa, Acinetobacter species, Enterobacter species, or MRSA are not isolated, monotherapy can be used with confidence since there are data showing good efficacy with monotherapy in severe nosocomial pneumonia not caused by these micro-organisms. However, if one of the micro-organisms mentioned above is present, antibiotic combination therapy has to be continued to reduce mortality in bacteremic cases and to avoid the emergence of resistant mutants.
There are several approaches to anti-pseudomonal therapy. The first is a combination of a b-lactam with an aminoglycoside, which in theory increases the synergism against Ps. aeruginosa. However, the existence of this synergism in vivc is not clear since aminoglycosides do not penetrate the pulmonary tissue well and are inactivated by the acidic environment. A second possibility is a combination of two b-lactam antibiotics, which may act by antagonism in some cases or even by induction of b-lactamases for one or both antibiotics. Finally, combination of ciprofloxacin with a b-lactam antibiotic may provide synergy, good parenchymal penetration, and lower toxicity compared with aminoglycosides.
Particular antibiotic issues when choosing empirical treatment of nosocomial pneumonia
In addition to the factors discussed above, the choice of antibiotic regimens in a specific hospital depends on the following.
1. The sensitivity pattern of the predominant flora in the unit: it is necessary to adjust antibiotic strategies according to the specific bacterial flora and sensitivity patterns of the ICU.
2. The type and degree of immunosuppression, and the type of antibiotics received either for prophylaxis or treatment, since these factors can modify the spectrum of pathogens.
3. Antibiotic-related aspects such as the ability to produce endotoxins and to select mutant strains. This issue is of particular importance and deserves careful study by the physicians.
Duration of therapy should be individualized depending upon the severity of illness, the rapidity of clinical response, and the infecting agent. No carefully controlled studies documenting duration of therapy have been reported, but antibiotics should be continued for a minimum of 14 to 21 days to reduce the chances of relapse when treating enteric Gram-negative bacilli or MRSA. In contrast, cure rates exceeding 95 per cent have been noted for nosocomial pneumonia caused by methicillin-sensitive Staph. aureus, H. influenzae, and Strep. pneumoniae. A 7- to 10-day course of treatment may be adequate for these pathogens.
There are several possible explanations for a rapid deterioration or failure to improve. These include the possibility that the process is not pneumonia, or that certain host, bacterial, and therapeutic factors have not been considered. The non-infectious processes that may be mistakenly labeled as nosocomial pneumonia include atelectasis, congestive heart failure, pulmonary embolism with infarction, lung contusion in trauma patients, and chemical pneumonitis from aspiration. Patients with acute respiratory distress syndrome can have fibroproliferative diffuse alveolar damage, and mechanically ventilated patients can have pulmonary hemorrhage.
Host factors associated with non-response to treatment include underlying fatal conditions, superinfection, chronic obstructive pulmonary disease, and immunosuppression. Again, bacterial variables may also be associated with an adverse outcome of initial therapy. The infecting pathogen can be resistant at the outset to the chosen therapy or may acquire resistance during therapy, particularly if the agent is Ps. aeruginosa. Finally, pneumonia can be due to other pathogens (i.e. fungi).
Certain complications during therapy can also lead to failure to respond. Some patients with nosocomial pneumonia can have other sources of fever such as sinusitis, vascular-catheter-related infection, urinary tract infections, or pseudomembranous enterocolitis. Complications of the original pneumonia, including the development of lung abscess or empyema, can lead to failure.
Was this article helpful?