laboratory tests. In CAP patients unresponsive to apparently appropriate antibiotic therapy,
transbronchial or open lung biopsy may be necessary. Compromised hosts respond more
slowly than normal hosts to effective therapy. Normal hosts with severe CAP usually show
some improvement in three to five days, but in compromised hosts,*7 to 10 days may be
needed before clinical improvement is noted. The duration of antibiotic therapy, IV/PO, for
CAP in normal hosts is one to two weeks, whereas in compromised hosts two to three weeks
are often necessary because of impaired host defenses (1,8,10).
Obviously, the prognosis in severe CAP is also a function of host factors, i.e.,
cardiopulmonary reserve/impaired HI/CMI. Inappropriate/delayed empiric therapy length-
ens LOS and is associated with a worse prognosis (1,10,59).
Clinical and Therapeutic Approach to Severe CAP
Patients presenting with severe CAP often require ventilatory, volume, or pressor support. The
clinician’s first task is to support vital functions and rapidly consider the treatable/reversible
causes of severe CAP mimics. The cause of CAP mimics is usually fairly straightforward based
on history, physical findings, and routine laboratory tests. The CXR/chest CT scan is helpful in
eliminating diagnostic possibilities, limiting diagnostic possibilities, and sometimes in making
a specific diagnosis.
If the mimics of severe CAP can be reasonably ruled out, the clinician’s next task is to
determine the likely pathogen based on history, physical findings, routine laboratory tests, and
aspects of the clinical presentation, including assessment of cardiopulmonary function, HI/
CMI status, and the degree of hypoxemia (Table 11).
The most common cause of severe CAP in normal hosts is viral pneumonias. The classic
severe viral pneumonia in adults is influenza A pneumonia. As mentioned previously,
influenza A pneumonia most commonly occurs alone. Alternately, it may be complicated by
bacterial CAP, either simultaneously initially (with MSSA/CA-MRSA) or sequentially after a
5–7 day interval of improvement with subsequent CAP due toS. pneumoniaeorH. influenzae.In
cases without bacterial superinfection, prognosis is related to degree and duration of
hypoxemia. In pandemic influenza A, as in 1918–1919, the majority of the deaths occurred in
young, healthy adults without comorbidities and were due to severe hypoxemia uncompli-
cated by bacterial pneumonia. During the past decade, avian influenza (H5N1) strains have
circulated in Asia and Europe. Unlike influenza A, avian influenza (H5N1) is not efficiently
transmitted from person-to-person, and for this reason does not, as yet have pandemic
potential. However, in contrast to human influenza A, avian influenza (H5N1) is fatal in the
majority of cases and affects primarily young healthy adults. Deaths from avian influenza
(H5N1) occurs from severe hypoxemia uncomplicated by bacterial pneumonia.
In the spring of 2009, the swine influenza (H1N1) pandemic began in Mexico and quickly
spread throughout the world. Although large numbers of the population were affected by
swine influenza (H1N1), there were relatively few mortalities. In the fatal cases of swine
influenza (H1N1) pneumonia, like avian influenza (H5N1) pneumonia, fatalities died from
severe hypoxemia also uncomplicated by bacterial pneumonia. The majority of fatalities with
swine influenza (H1N1) pneumonia were young healthy adults without comorbidities (60–65).
Optimal empiric therapy is based on correlating epidemiologic and clinical findings to
arrive at a presumptive clinical diagnosis directed at the most likely pulmonary pathogen.
Empiric therapy is continued until diagnostic possibilities are eliminated, and if possible, a
specific etiologic diagnosis is made. Empiric therapy should be continued if clinically effective.
REFERENCES
- Cunha BA. Severe community-acquired pneumonia. Crit Care Clin 1998; 14:105–118.
- Cunha BA. Severe community-acquired pneumonia: determinants of severity and approach to therapy.
Infect Med 2005; 22:53–58. - Wilson PA, Ferguson J. Severe community-acquired pneumonia: an Australian perspective. Intern
Med J 2005; 35:699–705. - Moine P, Vercken J-P, Chevret S, et al. Severe community-acquired pneumonia. Chest 1994; 105:
1487–1495.
Severe Community-Acquired Pneumonia in Critical Care 175