Hospital-acquired pneumonia (HAP), or nosocomial pneumonia, is a serious complication of hospitalized patients. HAP is associated with increased mortality, increased morbidity including prolongation of hospital stay, and increased economic costs. Many of the same risk factors that predispose elderly people to community-acquired pneumonia (CAP) also put them at risk for HAP. Nevertheless, it is not clear whether increasing age itself is an important independent risk factor for HAP. Studies have found mixed results on this question. In fact, one of the largest cohort studies, using a computerized system database available in several hospitals for studying ventilator-associated pneumonia (VAP), a major component of HAP, found that patients developing VAP were statistically significantly younger than those who did not develop HAP, although the actual difference in mean age, 61.7 versus 64.6 years respectively, was not very great. This result is somewhat surprising because the other risk factors associated with HAP in most studies are conditions that might be expected to increase with age and are similar to the risk conditions for CAP.
Although there are several studies focusing on CAP in the elderly, this is not the case for HAP. Until better information is available, therefore, the approach to and the management of HAP in the elderly are the same as for HAP in other patients, with a few exceptions that will be described.
Some people have divided HAP into early (usually occurring within five days of admission to the hospital) and late (occurring after five days). Opinions differ as to whether the organisms involved in early versus late pneumonia are different. Authors who emphasize a difference in the pathogenic flora argue that HAP should be defined as occurring after some arbitrary time from admission to the hospital, often after five to seven days. These authors claim that any pneumonia occurring before that time is likely secondary to community-acquired organisms colonizing the patient before hospital admission. Therefore, they argue that patients with pneumonia developing during the first several days of hospitalization should be considered as having CAP and managed according to the guidelines for that condition.
As in CAP, HAP in elderly patients may present in unusual ways, especially with a change in mental status, and may not have other typical findings, such as fever, that readily indicate pneumonia. Therefore, a high degree of suspicion for HAP in the elderly must be maintained.
Diagnosis of HAP can be difficult. Although, obviously, a new infiltrate on chest x-ray, fever, purulent sputum production and increased white blood cell count should suggest the development of HAP, unfortunately most of these findings are not specific and occur in many other conditions. This is particularly true of fever and leukocytosis. Sputum production, although somewhat more specific, certainly is not limited to pneumonia. Furthermore, the more critically ill the patient, the more difficult the diagnosis. Many hospitalized patients may already have pulmonary infiltrates on chest x-rays, making the appreciation of new infiltrates difficult.
Making a diagnosis of VAP is especially problematic in patients with acute lung injury or the acute respiratory distress syndrome. These patients have preexisting diffuse infiltrates, making it very difficult to appreciate changes in a chest x-ray that may represent a new pneumonia. At the same time, it is well established that the inflammatory condition of acute lung injury/acute respiratory distress syndrome can cause fever, leukocytosis and purulent pulmonary secretions. These issues make the study of HAP and VAP difficult because there is no gold standard for diagnosis. The situation has been made clearer in VAP through multiple studies evaluating the use of quantitative cultures obtained by bronchoalveolar lavage and protective specimen brush through a fiberoptic bronchoscope. These have led to the conclusion that quantitative cultures can be useful in identifying patients with pneumonia.
Although Streptococcus pneumoniae is one of the causative organisms of HAP, it is a less common etiology in HAP than in CAP. Staphylococcus aureus, including methicillin-resistant S. aureus (MRSA), and gram-negative bacilli, including Pseudomonas and Acinetobacter, which tend to be resistant to some antibiotics, are greatly increased in the etiology of HAP compared with CAP. However, much more important than this general information is knowledge about the incidence and antibiotic resistance pattern of microorganisms causing HAP in each institution. Considerable variability exists among hospitals and geographic regions in both the likely organisms for HAP and their antibiotic resistance. Therefore, although general guidelines can be used as a starting point, it is necessary to adapt these guidelines to the patterns found in a particular institution.
The general treatment guidelines essentially are based on whether or not resistant Pseudomonas and Acinetobacter or MRSA is likely to be causing the HAP in the individual institution and patient. If these organisms are not likely to be responsible, one can treat the other likely organisms with any of several options: ceftriaxone; a quinolone such as levofloxacin, moxifloxacin, or ciprofloxacin; ampicillin-sulbactam; or ertapenem. On the other hand, in the more likely case of multi-drug resistant organisms such as Pseudomonas or Acinetobacter (which are of particular concern), double coverage is recommended with a beta-lactam plus either an antipseudomonal fluoroquinolone (ciprofloxacin or levofloxacin) or an aminoglycoside. Suggested options for the beta-lactam include an antipseudomonal cephalosporin such as cefepime or ceftazidime; an antipseudomonal carbapenem such as imipenem or meropenem; or a beta-lactam/beta-lactamase inhibitor such as piperacillin-tazobactam.
If MRSA is a consideration, vancomycin can be added. One of the described antibiotic choices should be started as empirical therapy, but if quantitative cultures have been obtained, the treatment can be tailored to the predominant organism grown once the cultures are available. Again, if the quantitative cultures are negative (i.e., if they fall below the accepted cutoffs for the diagnosis of pneumonia), antibiotics can be discontinued, although this obviously requires some clinical judgment. If pneumonia is considered unlikely by a low quantitative culture growth, infections in other sites should be considered and appropriate diagnostic studies performed.
Several strategies have been suggested to reduce the incidence of HAP. These strategies have usually been tested regarding VAP, but the data may be extrapolated in some cases to patients at risk for HAP but not receiving mechanical ventilation. Most of these strategies have been tested in relatively small trials and do not reach the level of IA evidence, the highest level of evidence basis. However, some of the strategies have been found to have consistent positive results in smaller studies and because they are not associated with toxicity and are inexpensive, they can be recommended for implementation.
The most important strategy for reducing VAP appears to be keeping the patient in the semi-recumbent rather than recumbent position. Three trials have evaluated the efficacy of this approach. It has been previously shown that supine positioning is independently associated with the development of VAP, presumably because of increased risk of esophageal reflux and aspiration. The trials have tested elevating the head of the bed to 45 degrees. Two trials measured gastroesophageal reflux and aspiration events as surrogate outcomes and showed a decreased frequency in both of these outcomes. The third trial found a statistically significant reduction in VAP in patients nursed with the head of the bed elevated to 45 degrees compared with those maintained in the supine position. There was no difference in mortality in this relatively small study. Because there were no adverse effects observed in patients randomly assigned to the semi-recumbent position and because this is a low-cost, low-risk strategy, it is recommended that all patients in the critical care unit be kept in a semi-recumbent position unless there is a specific contraindication. Contraindications might include particular types of recent surgery or shock refractory to vasoactive therapy. Although this is a simple strategy, it can be difficult to implement in all patients in whom it is indicated. It requires cooperation, especially by the nurses who need to understand both the rationale for the strategy and its potential importance. This involves education of the physicians, nurses and respiratory therapists, but also requires leadership by all three of these groups.
Patients who are mechanically ventilated are known to be at risk for stress gastric ulceration. Data from the 1980s suggested that antacids and possibly H2 blockers, which reduce the gastric acidity in an attempt to decrease stress ulceration, were associated with increased risk for VAP. Sucralfate, which protects the mucosa from stress ulceration but does not affect the gastric pH, was found to have lower rates of VAP. Several other studies also found that sucralfate was superior to H2 blockers in reducing the incidence of VAP. However, a more recent large well-designed trial of 1,200 mechanically ventilated patients found that H2 blockers were more effective than sucralfate in preventing clinically important gastrointestinal bleeding and failed to find a significant difference in VAP between these two agents. H2 blockers are more expensive than sucralfate, but they are easier to administer. Therefore, with our current state of knowledge, one can choose between H2 blockers or sucralfate to prevent stress ulceration in mechanically ventilated patients.
Continuous or intermittent aspiration of (i.e., removal of) pooled secretions above the endotracheal tube cuff has been suggested to reduce VAP development. This requires the use of specially designed endotracheal tubes with a suction port above the cuff. This method has been studied in small trials with results that are mixed but generally favorable. However, because of the mixed results and the small numbers of studies, this approach needs to be studied further before it can be recommended for routine use.
Several studies of beds that rotate the position of the patient (so-called oscillating beds) have been performed. A meta-analysis of six of these randomized, controlled trials found a statistically significant reduction in the risk for pneumonia, but five of these studies were limited to surgical patients or patients with neurologic impairment. The sixth study, which was primarily in medical ICU patients, found no significant benefit. The cost of these beds is relatively expensive. However, there is reasonable evidence that this practice may be effective in selected surgical patients or patients with neurological problems. Further studies in medical ICU patients should be done before use of oscillating beds can be recommended.
Several studies have looked at the frequency of changing ventilator circuits. In general, more frequent changes are associated with a higher incidence of colonization, although there was no difference in the rate of VAP. Based on these findings, ventilator circuits can be changed less frequently based on the monitoring of colonization, with cost savings in addition to the lower incidence of colonization.
No clear evidence favors any specific form of enteral feeding in reducing VAP incidence.
Excerpted from Buckner JK, Duke JR Jr., Good JT Jr., et al. Venous thromboembolism. In: Petty TL, Seebass JS, eds. Pulmonary Disorders of the Elderly. Philadelphia: American College of Physicians; 2007.