Critical Care Medicine
Approach to the patient with PNA
- 1. Description of the problem
2. Emergency Management
- 3. Diagnosis
- 4. Specific Treatment
- 5. Disease monitoring, follow-up and disposition
Special considerations for nursing and allied health professionals.
What's the evidence?
Types and subtypes: Community-acquired pneumonia, hospital-acquired pneumonia, nosocomial pneumonia, ventilator-associated pneumonia, aspiration pneumonia, healthcare-associated pneumonia
1. Description of the problem
What every clinician needs to know
Pneumonia refers to an acute infection of the lung. Although the term "pneumonia" may be used to describe non-infectious pulmonary processes (eg, the idiopathic interstitial pneumonias), in these chapters the term "pneumonia" is used to describe an acute infectious process. Traditionally, pneumonia has been categorized by the setting in which the host resides -- i.e., community-acquired vs. hospital-acquired, or nosocomial, infection (CAP vs. HAP). A subset of HAP that develops in the ICU in a patient who has received at least 48 hrs of mechanical ventilation is called ventilator-associated pneumonia (VAP). The distinction between CAP and HAP historically evolved based on the type/virulence of the pathogens involved. The pathogens seen in CAP include Streptococcus pneumoniae, Haemophilus influenzae, and Legionella sp. In HAP the pathogens include more resistant organisms such as Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus (MRSA).
With the diffusion of healthcare delivery beyond the confines of the acute inpatient setting, this distinction between these two syndromes has become obscured, such that patients in the community may now present to the emergency department with infections due to bacteria generally considered limited to HAP. This change in microbiology led to a change in nomenclature with the new concept of a healthcare-associated pneumonia (HCAP) -- no longer defined by one's location but by one's risk for acquiring multi-resistant pathogens -- a function more of one's comorbidities and by exposure to either healthcare venues or treatments (most notably antibiotics). The concept of HCAP is more broadly described in the chapter on nosocomial pneumonia.
Not all infectious pneumonias are caused by bacteria. Other potential pathogens include viruses, mycobacteria, fungi, and parasites, although the last few play a much more important role in immunosuppressed subjects. Aspiration pneumonia, on the other hand, represents a chemical pneumonitis where gastric contents are aspirated into the lung. This syndrome may mimic infectious pneumonia but in many cases - particularly when community acquired - is not specifically caused by an infectious agent. However, it is not uncommon for these cases of bland aspiration pneumonia to become superinfected.
Irrespective of the causative agent or subtype, pneumonia tends to be associated with a general constellation of manifestations. One set of findings relate to the systemic signs and symptoms of infection. Other findings are those confined to the lung that result from the localized inflammatory process that ensues from the specific infection. More specifically, the most common presenting findings include fever, an elevated white blood count, cough, and dyspnea. However, in the elderly and immunocompromised, fever may be absent and the WBC count may remain in the normal range. Occasionally the only objective evidence of infection is a shift of the WBCs toward immature forms. In severe cases, hypothermia occurs rather than fever. Given the potential range in disease severity associated with pneumonia (again irrespective of the type) some patients may seem only mildly ill, while others progress rapidly to shock and acute respiratory failure.
Patients also usually have tachypnea and hypoxemia, reflecting the direct effects of inflammation in the lungs. Although tachypnea is a very sensitive sign in pneumonia, pulse oximetry may fail to demonstrate the disturbance in oxygenation that is a hallmark of pneumonia. This relates to the inherent inaccuracies of this test and the fact that secondary hyperventilation may improve the oxygen saturation. Thus, arterial blood gases are much more sensitive, as reflected by the alveolar-arterial PO2 gradient. Sputum production is variable, with some patients presenting with copious sputum production while others (again mainly the elderly) may only have a nonproductive cough. Rarely and depending on the pathogen, the patient may describe hemoptysis.
In the setting of a rigorous clinical trial testing investigational agents for the treatment of pneumonia, the diagnosis of pneumonia often requires the presence of at least 2 or 3 signs/symptoms. The presence of an infiltrate on a chest radiograph (CXR) is central to the diagnosis of pneumonia. Without evidence of direct parenchymal involvement on CXR, it is unlikely that an infectious pneumonia explains a patient's syndrome. However, standard CXR has limitations. First, portable anterio-posterior films without an accompanying lateral image may miss up to 20% of acute infiltrates. Similarly, even a posterio-anterior film with a lateral is less sensitive then a dedicated CT scan of the chest (especially in immunosuppressed subjects).
Unfortunately, many other disease processes besides pneumonia can cause the same signs and symptoms and radiographic findings often attributed to pneumonia. At present, there is no objective gold standard for the diagnosis of pneumonia and it remains a clinical diagnosis. Historically, specific patterns of either signs and symptoms and/or CXR findings were thought to implicate a particular pathogen. This observation has certainly not panned out. No one finding or combination of findings is pathognomonic for a particular pathogen, and clinicians should avoid using such a deterministic approach to this disease.
Key management points
The key management points are to focus on 1) establishing the diagnosis, 2) assessing disease severity and determining patient triage, and 3) initiating therapy.
- For the diagnosis, all patients require a detailed history. The physical exam itself is often unhelpful for the diagnosis of pneumonia as reliance on just physical exam findings has been shown to both over- and under-diagnose pneumonia. Specific aspects in the history should be elicited, including the duration of symptoms, their progression, occupational/environmental exposures, and the presence of risk factors for HCAP (see relevant chapter for more detail). A careful assessment of the patient's immune status is also essential (eg, HIV disease, recent chemotherapy, etc.).
-To evaluate disease severity, the vital signs should be reviewed. Evidence of tachycardia, hypotension, and/or shock require urgent attention while assessing the patient. Hypoxemia or airway instability should be corrected immediately and the need for mechanical ventilation assessed.
Since the timeliness and appropriateness of antibiotics improves outcome in virtually all forms of pneumonia, clinicians should strive to ensure timely antibiotic treatment (preferably within 6 hrs of symptom onset or presentation). These antibiotics must also be active against the likely pathogen to optimize outcomes.
2. Emergency Management
Emergency management of the patient with pneumonia includes establishing an airway, if needed, and maintaining adequate oxygenation and ventilation, and for those patients in shock rapid fluid resuscitation is crucial. In ~50% of patients who present with septic shock the infection is in the lung. In all patients, if there is any concern for a significant or growing pleural effusion it should be tapped to determine whether there is an accompanying empyema, since this must be drained urgently.
Diagnostic criteria and tests
Chest imaging represents the most important diagnostic test for establishing the presence of pneumonia. A CXR is preferred but at times the infiltrate is seen only on chest CT. Other diagnostic tests focus on identifying the etiologic pathogen. This generally involves cultures of sputum, blood and pleural fluid (if present). In patients who cannot produce sputum, inducing sputum with the help of respiratory therapy should be considered. Blood cultures are recommended for patients with HCAP. Although positive in only 10-15% of patients, in some cases the blood culture may be the only way to identify the etiologic pathogen. In CAP, the evidence supporting routine blood cultures is more limited. Except for critically ill patients with CAP, blood cultures are more likely to grow a contaminant than a true pathogen. In mechanically ventilated patients, clinicians should consider obtaining lower airway cultures rather than simply tracheal aspirates.
Lower airway cultures are more reliable than tracheal aspirates and are less likely to be confounded by upper airway colonization. Examples of lower airway cultures include both bronchoscopic and non-bronchoscopic techniques. Bronchoscopic approaches include traditional broncheal alveolar lavage (BAL) and bronchial brush. Mini-BAL and blind-brush are options for non-bronchoscopic alternatives. Both bronchoscopic and non-bronchoscopic means for obtaining culture material are considered to be equivalent for diagnostic purposes in the non-immunocompromised host.
Other tools for identifying the pathogen include urinary antigen testing and serum antibody studies. Urinary antigen tests are commercially available for Streptococcus pneumoniae and selected Legionella species. The urinary antigen tests have limited sensitivity and specificity since they are positive only in the presence of select serotypes/species. Measuring specific antibody titers and then re-evaluating the subject weeks later to determine the convalescent titer has no value outside of clinical research.
In immunosuppressed patients, special stains should be ordered to evaluate sputum and BAL specimens for the presence of Pneumocystis jiroveci (PJP) and selected fungi. In otherwise normal hosts, these tests are of little value. When clinically suspected based on the clinical and epidemiologic scenario, acid fast stains and subsequent cultures are appropriate to rule out mycobacterial disease. Genetic probing of sputum and lower airway material can also prove helpful in the proper setting.
In the end, the diagnosis of pneumonia is a clinical one. The physician must assemble and integrate the information and patient presentation along with results from objective testing to determine if the patient has pneumonia. Again, each aspect of diagnostic testing undertaken in pneumonia has limited sensitivity. In other words, the differential diagnosis for an elevation in the WBC count is huge, as is the differential for an abnormal chest film. Decision aids exist to help make the diagnosis more objective.
For example, the clinical pulmonary infection score (CPIS) integrates various aspects of the clinical presentation and the objective testing into a score. The higher the score, the more likely the patient is to have infectious pneumonia rather than some alternate process. The CPIS, though, has very limited utility. It has not been examined outside of use in VAP. Even within the arena of VAP, the diagnostic value of the CPIS appears quite limited.
A recently developed assay to measure serum procalcitonin appears to identify subjects likely to have a bacterial infection (as opposed to either some other syndrome) and may be a useful guide for initiating antimicrobial treatment. However, this assay is not widely used in the United States at present and the data are not definitive enough to use for clinical decisions at present.
The differential diagnosis for patients who present with an acute respiratory syndrome along with radiographic evidence of an infiltrate is quite broad. Whether the patient is being evaluated for HAP or CAP, other considerations include:
-Aspiration or inhalation injury
-Congestive heart failure
-Connective tissue diseases with pulmonary manifestation
-Idiopathic interstitial pneumonia (eg, UIP, NISP, etc.)
-Trauma (pulmonary contusion)
-Transfusion-related acute lung injury
-Nearly any cause of acute lung injury/acute respiratory distress syndrome
-Drug reaction/radiation pneumonitis
In most instances, the clinical scenario excludes many of these possibilities (eg, no history of transfusion, no trauma).
In some cases, several of these syndromes may coexist with an acute pneumonia. For example, pneumonia may often lead to an exacerbation of heart failure.
Confirmatory testing in pneumonia includes interventions to better evaluate the pulmonary parenchyma and to obtain more sensitive material.
If a question exists about the presence of an infiltrate on a chest radiograph, formal PA and lateral films should be ordered if there has been only an AP portable film. Physicians should not simply jump to CT scanning. However, if there is a question about the pattern of infiltrates after reviewing a PA and lateral, or if there is a need to better evaluate the mediastinum, then a CT is appropriate. Selected aspects of an infiltrate may also be better appreciated only with the resolution of a CT. For example, a CT can confirm the presence of a cavity, which can substantially shift the differential diagnosis in terms of microbiology. Similarly, the absence of an infiltrate nearly always excludes an acute pneumonic process. Note that ~10% of patients with PJP pneumonia will have normal chest films and the presence of the infiltrate is detected only on a CT scan. Similarly, in immunosuppressed subjects, the CT clearly is more sensitive for the diagnosis of pneumonia as the burden of organisms needed to lead to an infection may be lower than what is required in a more normal host.
For material for culture, bronchoscopy represents a compromise between sputum cultures and tracheal aspirates and an open lung biopsy. Bronchoscopy is often indicated when there is/are: accompanying hemoptysis, an abnormality on imagining suggesting malignancy, nonresolving pneumonias, atypical clinical features and/or exposures, and in immunosuppressed subjects. In patients requiring mechanical ventilation, lower airway cultures should be obtained.
4. Specific Treatment
First-line therapy includes antibiotics. For all forms of pneumonia, multiple studies confirm that initial appropriate antibiotic therapy improves outcomes. For therapy to be "appropriate" it must be given in a timely manner (within several hours of presentation) and be active in vitro against the culprit pathogen. Because of the need to ensure in vitro activity, and given that one may not know culture results for several days, it is important to design empiric and protocolized regimens based on the nature of the pneumonia and local susceptibility data. This is true both for patients who present to the ED and those with CAP. Failure to administer initially appropriate antibiotic therapy can increase the risk of death up to four-fold.
Thus, in patients with CAP, it remains crucial to use regimens against S. pneumoniae, the most common pathogen in this syndrome. However, many strains of S. pneumoniae are penicillin and/or macrolide resistant. The clinical implications of this resistance are unclear at best. In the US it remains common to treat "atypical" pathogens such as C. pneumoniae and Mycoplasma species. The importance of these pathogens is difficult to establish, and in some nations, treatment for these organisms is not considered routine. Multiple guidelines exists to aid in antibiotic decision making and include those from the American Thoracic Society (ATS)/Infectious Disease Society of America (IDSA), the British Thoracic Society, and others. Note that compliance with national guidelines with respect to antibiotic decision making has also been shown to lead to improved outcomes.
For patients at risk for bacteria such as P. aeruginosa, MRSA, and other multidrug-resistant organisms, broader initial therapy is recommended. These pathogens are of particular concern in persons with HCAP, HAP, and VAP, and those who are immunosuppressed. There is a formal ATS/IDSA guideline covering these syndromes as well. Often multiple agents are needed as part of the initial treatment regimen to ensure that at least one of them is active in vitro against what is eventually found to be the primary organism. It is crucial to narrow therapy when cultures return. This helps limit the development of resistance and contains cost.
Selected patients may be at risk for pathogens such as (but not limited to) cytomegalovirus, filamentous molds, and PJP. In patients with risk factors for these organisms, empiric treatment is recommended pending cultures. In those with AIDS empiric PJP treatment (eg, TMP-SMZ) should be given along with traditional antibiotics. Corticosteroids should also be given if there is significant hypoxemia. The role for corticosteroids in other settings where pneumonia is a concern is controversial, and routine corticosteroid administration is not recommended for CAP, HCAP, and HAP.
In patients with non-resolving pneumonia, clinicians should consider:
1) Is there a resistant pathogen not being treated?
2) Is there an alternate diagnosis causing the syndrome (eg, a non-infectious process)?
3) Is there a complication of the infection requiring treatment?
On average, patients require 3-4 days to achieve clinical stability after initial antibiotic treatment for pneumonia. In the patient progressing/failing to respond, a careful review of all cultures is needed to ensure that a resistant organism has not been missed. In addition, there may be an undrained empyema that requires drainage. If the patient has developed diarrhea there is the possibility that he or she is becoming more ill because of antibiotic-associated colitis. If the presentation is confusing, it may be that with time the situation will become more clear and an alternate diagnosis will become evident, such as pulmonary embolism or heart failure. Given the diagnostic limitations of the approach to pneumonia, it is crucial to constantly re-evaluate the clinical situation and to have a low threshold for questioning the original diagnosis. Often part of the re-evaluation requires new chest imaging. For example, infiltrates that resolve in a few days are not likely to have been due to infection.
5. Disease monitoring, follow-up and disposition
Expected response to treatment
Prognosis is best determined based on traditional measures of outcome prediction. Specifically, the presence or absence of key organ failures (eg, respiratory, shock) is a major determinant of outcome. For both CAP and HAP, severity of illness scoring tools perform moderately well at predicting outcomes but have certain limitations. For CAP, the CURB-65 scoring tool and the Pneumonia Severity Index correlate with mortality. For HAP and VAP there are no well-validated scores uniquely developed for outcome prediction for those syndromes.
Patients require several days to improve. For both HAP syndromes and CAP, most patients who will improve will do so by day 3. In VAP, the PaO2/FiO2 ratio by day 3 is the best predictor of outcome. Multiple biomarkers correlate with outcomes; however, these tend to be nonspecific.
If there is a suspicion of an incorrect diagnosis, the clinician must focus on determining how the patient's clinical evolution points to an alternate diagnosis. For example, if despite broad-spectrum antibiotics the patient is not improving, one must consider infections outside the chest and/or non-infectious syndromes. First steps include repeat culturing along with re-imaging.
The recommended follow-up varies based on the pneumonia type. For CAP, guidelines recommend a follow-up radiograph at 6 to 8 weeks after onset to ensure that the infiltrate has resolved and that there is not a confounding malignancy. For VAP, no specific follow-up is recommended.
Special considerations for nursing and allied health professionals.
What's the evidence?
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.