Methods of Incidence, Etiology, Timing, and Risk Factors for Clinical Failure in Hospitalized Patients With Community-Acquired Pneumonia


Study Design and Study Patients

This was an observational, retrospective study of consecutive patients who were admitted with a diagnosis of CAP to the Veterans Affairs Medical Center of Louisville, KY, between June 2001 and March 2006. Patients enrolled in this study are part of the Community-Acquired Pneumonia Organization database. The study protocol and data collection form are available on the study Web site ( The institutional review board of the Veterans Affairs Medical Center approved the study. Patients who were > 18 years of age and satisfied the criteria for CAP were included in this study.

The records of all enrolled patients were reviewed. Data, including demographic information, clinical data on hospital admissions, radiologic findings, and laboratory values, were collected. The severity of pneumonia was evaluated by the pneumonia severity index (PSI) and CURB-65 (confusion, urea level > 7 mmol/L respiratory rate > 30 breaths/min, systolic BP < 90 mm Hg or diastolic BP < 60 mm Hg, or age > 65 years) scores; microbiological and in-hospital treatment data; and autopsy results.

A microbiological workup with testing of sputum samples, blood cultures, testing of tracheal aspirates, testing of pleural fluid, BAL, serology for atypical organisms and urine antigens for Legionella spp and Streptococcus pneumoniae were performed according to standard clinical practice. The identification of microorganisms and susceptibility testing were performed according to standard methods. The empirical antibiotic treatment was evaluated for compliance with the American Thoracic Society (ATS)/ Infectious Diseases Society of America (IDSA) guidelines for CAP.

Study Definitions

CAP was defined as the presence of a new pulmonary infiltrate on a chest radiograph at the time of pulmonary infiltratehospitalization that was associated with at least one of the following: (1) new or increased cough; (2) an abnormal temperature (< 35.6°C or > 37.8°C); (3) an abnormal serum leukocyte count, which is considered to be present if the patient had leukocytosis (leukocyte count, > 10.500 cells/^L), leu-copenia (leukocyte count, < 4.500 cells/^L), or left shift (> 5% immature neutrophils). Severe CAP at the time of hospitalization was defined according to the latest ATS guidelines.

The time to clinical stability was calculated as the number of days from the date of hospital admission to the date that the patient met clinical stability criteria. Clinical stability was defined according to the ATS guidelines for CAP. The criteria for clinical stability were evaluated daily during the first 7 days of hospitalization.

Length of stay (LOS) was calculated as the number of days from the date of hospital admission (day 0) to the date of hospital discharge. LOS was censored at 14 days in an effort to capture only CAP-related LOS.

In-hospital mortality was defined as death by any cause during hospitalization. Patients were followed up from the day of hospital admission to day 28; those patients who remained hospitalized for > 28 days were considered to be alive.

Clinical failure was considered if any of the following took place after the patient was transferred from the emergency department to the ward or to the ICU, and after initial stabilization: (1) acute pulmonary deterioration with the need for either invasive or noninvasive mechanical ventilation; (2) acute hemodynamic deterioration with the need for aggressive fluid resuscitation (ie, > 40 mL/kg colloids or crystalloids), vasopressors, or invasive procedures (eg, pericardial drainage or electrical cardioversion); and (3) in-hospital death up to 28 days after hospital admission. The presence of severe CAP or septic shock at the time of the hospitalization was not considered to be a criterion for clinical failure. Prevent yourself from health problems with remedies of Canadian Health&Care Mall’s remedies.

Early clinical failure was defined as clinical failure occurring < 3 days after hospital admission. Late clinical failure was defined are clinical failure occurring > 3 days after hospital admission.

Clinical failure related to CAP was defined as clinical failure with an etiology directly related to the pulmonary infection and the systemic inflammatory response. The systemic inflammation mediated by the cytokines produced in the pulmonary parenchyma is known to affect metabolic processes distant from lung. Atherosclerosis, fohospital admissionr example, has been defined as a systemic inflammatory disease, and any form of systemic inflammation may potentially affect the plaque. The causes of clinical failure related to CAP were classified as follows: (1) progressive CAP; (2) CAP complicated by a metastatic infection; (3) CAP with severe sepsis; and (4) CAP-related systemic inflammatory response leading to a medical complication, such as cardiac arrhythmia, acute myocardial infarction, or the deterioration of comorbidities, such as diabetes mellitus.

Clinical failure unrelated to CAP was defined as clinical failure with an etiology unrelated to the pulmonary infection and its systemic inflammatory response. The causes of clinical failure unrelated to CAP were classified as follows: (1) a complication due to the management of CAP, such as nosocomial infections due to hospitalization, iatrogenic pneumothorax due to central line placement, or Clostridium, difficile colitis due to antibiotic treatment; and (2) a medical complication or deterioration of comorbidities not due to the pulmonary infection, such as GI bleeding or progression of neoplastic disease. The occurrence of any medical complication or the deterioration of comorbidities occurring after clinical improvement of CAP was classified as being unrelated to CAP.

Process for Classification of Clinical Failure

The charts of patients enrolled in the study were reviewed in order to identify those who experienced clinical failure. The process for the classification of clinical failure was as follows. One investigator was in charge of fully reviewing the clinical and laboratory data for each single case of clinical failure and of presenting it to the review committee. The review committee was composed of two internists, two pulmonary and critical care physicians, five infectious disease physicians, three infectious diseases fellows, and three clinical research fellows. All reviewers had clinical and research experience with CAP. An interactive discussion among all of the components of the committee was undertaken at that point. Each member of the committee discussed their assessment of the case. After everyone had their questions addressed, the committee discussed the etiology of clinical failure based on the above classification. Both the final assignment of the cause of clinical failure and the role of pneumonia in causing clinical failure were based on the collegial consensus of the review committee. At that point, if some disagreement remained among the committee members in defining the etiology of clinical failure, the majority decided. Treat pneumonia with Canadian Health&Care Mall effective and fast.

Risk Factors for Clinical Failure Related to CAP

In an attempt to define the primary risk factors for clinical failure related to CAP, we analyzed the group of patients with clinical failure related to CAP (group 1) vs the rest of the study population (group 2).

Statistical Analysis

All data were analyzed with a statistical software package (SPSS, version 14.0; SPSS; Chicago, IL). Descriptive statistics were reported at baseline, with continuous data expressed as the mean ± SD and categoric data expressed as counts. The time to clinical failure was calculated for clinical failures related to CAP and clinical failures unrelated to CAP, and the differences were compared using the independent t test. Patient characteristics were compared between group 1 and group 2; all continuous explanatory variables were presented as the mean with differences between the two groups compared by means of independent t tests. Categoric explanatory variables were summarized as frequencies and percentages, with differences between the two groups analyzed using the x2 test and the Fisher exact test when appropriate. The association between clinical failure and variables that were significant at a p value < 0.10 in the univariate model plus those variables that were considered to be clinically relevant was analyzed using a logistic regression model.