Depression and all-cause mortality risk in HIV-infected and HIV-uninfected US veterans: a cohort study.

OBJECTIVES: The contribution of depression to mortality in adults with and without HIV infection is unclear. We hypothesized that depression increases mortality risk and that this association is stronger among those with HIV infection. METHODS: Veterans Aging Cohort Study (VACS) data were analysed from the first clinic visit on or after 1 April 2003 (baseline) to 30 September 2015. Depression definitions were: (1) major depressive disorder defined using International Classification of Diseases, Ninth Revision (ICD-9) codes; (2) depressive symptoms defined as Patient Health Questionnaire (PHQ)-9 scores ≥ 10. The outcome was all-cause mortality. Covariates were demographics, comorbid conditions and health behaviours. RESULTS: Among 129 140 eligible participants, 30% had HIV infection, 16% had a major depressive disorder diagnosis, and 24% died over a median follow-up time of 11 years. The death rate was 25.3 [95% confidence interval (CI) 25.0-25.6] deaths per 1000 person-years. Major depressive disorder was associated with mortality [hazard ratio (HR) 1.04; 95% CI 1.01, 1.07]. This association was modified by HIV status (interaction P-value = 0.02). In HIV-stratified analyses, depression was significantly associated with mortality among HIV-uninfected veterans but not among those with HIV infection. Among those with PHQ-9 data (n = 7372), 50% had HIV infection, 22% had PHQ-9 scores ≥ 10, and 28% died over a median follow-up time of 12 years. The death rate was 27.3 (95% CI 26.1-28.5) per 1000 person-years. Depressive symptoms were associated with mortality (HR 1.16; 95% CI 1.04, 1.28). This association was modified by HIV status (interaction P-value = 0.05). In HIV-stratified analyses, depressive symptoms were significantly associated with mortality among veterans with HIV infection but not among those without HIV infection. CONCLUSIONS: Depression was associated with all-cause mortality. This association was modified by HIV status and method of depression ascertainment.

The impact of prescribed opioids on CD4 cell count recovery among HIV-infected patients newly initiating antiretroviral therapy.

OBJECTIVES: Certain prescribed opioids have immunosuppressive properties, yet their impact on clinically relevant outcomes, including antiretroviral therapy (ART) response among HIV-infected patients, remains understudied. METHODS: Using the Veterans Aging Cohort Study data, we conducted a longitudinal analysis of 4358 HIV-infected patients initiating ART between 2002 and 2010 and then followed them for 24 months. The primary independent variable was prescribed opioid duration, categorized using pharmacy data as none prescribed, short-term (< 90 days) and long-term (≥ 90 days). Outcomes included CD4 cell count over time. Analyses adjusted for demographics, comorbid conditions, ART type and year of initiation, and overall disease severity [ascertained with the Veterans Aging Cohort Study (VACS) Index]. Sensitivity analyses examined whether effects varied according to baseline CD4 cell count, achievement of viral load suppression, and opioid properties (i.e. dose and known immunosuppressive properties). RESULTS: Compared to those with none, patients with short-term opioids had a similar increase in CD4 cell count (mean rise per year: 74 vs. 68 cells/µL; P = 0.11), as did those with long-term prescribed opioids (mean rise per year: 74 vs. 75 cells/µL; P = 0.98). In sensitivity analysis, compared with no opioids, the effects of short-term prescribed opioids were statistically significant among those with a baseline CD4 cell count ≥ 500 cells/µL (mean rise per year: 52 cells/µL for no opioids vs. 20 cells/µL for short-term opioids; P = 0.04); findings were otherwise unchanged. CONCLUSIONS: Despite immunosuppressive properties intrinsic to opioids, prescribed opioids appeared to have no effect on CD4 cell counts over 24 months among HIV-infected patients initiating ART.

Prevalence and correlates of obstructive sleep apnoea among patients with and without HIV infection.

OBJECTIVES: In HIV-uninfected populations, obstructive sleep apnoea (OSA) is commonly associated with cardiovascular disease, metabolic syndrome, and cognitive impairment. These comorbidities are common in HIV-infected patients, but there are scarce data regarding OSA in HIV-infected patients. Therefore, we examined the prevalence and correlates of OSA in a cohort of HIV-infected and uninfected patients. METHODS: An observational cohort study was carried out. Electronic medical record and self-report data were examined in patients enrolled in the Veterans Aging Cohort Study (VACS) between 2002 and 2008 and followed until 2010. The primary outcome was OSA diagnosis, determined using International Classification of Diseases, 9th edition (ICD-9) codes, in HIV-infected compared with uninfected individuals. We used regression analyses to determine the association between OSA diagnosis, symptoms and comorbidities in adjusted models. RESULTS: Of 3683 HIV-infected and 3641 uninfected patients, 143 (3.9%) and 453 (12.4%) had a diagnosis of OSA (p<0.0001), respectively. HIV-infected patients were more likely to report symptoms associated with OSA such as tiredness and fatigue. Compared with uninfected patients with OSA, HIV-infected patients with OSA were younger, had lower body mass indexes (BMIs), and were less likely to have hypertension. In models adjusting for these traditional OSA risk factors, HIV infection was associated with markedly reduced odds of OSA diagnosis (odds ratio 0.48; 95% confidence interval 0.39-0.60). CONCLUSIONS: HIV-infected patients are less likely to receive a diagnosis of OSA. Future studies are needed to determine whether the lower prevalence of OSA diagnoses in HIV-infected patients is attributable to decreased screening and detection or to a truly decreased likelihood of OSA in the setting of HIV infection.

Impact of defined clinical population and missing data on temporal trends in HIV viral load estimation within a health care system.

OBJECTIVES: Community viral load (CVL) estimates vary based on analytic methods. We extended the CVL concept and used data from the Veterans Health Administration (VA) to determine trends in the health care system viral load (HSVL) and its sensitivity to varying definitions of the clinical population and assumptions regarding missing data. METHODS: We included HIV-infected patients in the Veterans Aging Cohort Study, 2000-2010, with at least one documented CD4 count, HIV-1 RNA or antiretroviral prescription (n = 37 318). We created 6-month intervals including patients with at least one visit in the past 2 years. We assessed temporal trends in clinical population size, patient clinical status and mean HSVL and explored the impact of varying definitions of the clinical population and assumptions about missing viral load. RESULTS: The clinical population size varied by definition, increasing from 16 000-19 000 patients in 2000 to 23 000-26 000 in 2010. The proportion of patients with suppressed HIV-1 RNA increased over time. Over 20% of patients had no viral load measured in a given interval or the past 2 years. Among patients with a current HIV-1 RNA, mean HSVL decreased from 97 800 HIV-1 RNA copies/mL in 2000 to 2000 copies/mL in 2010. When current HIV-1 RNA data were unavailable and the HSVL was recalculated using the last available HIV-1 RNA, HSVL decreased from 322 300 to 9900 copies/mL. HSVL was underestimated when using only current data in each interval. CONCLUSIONS: The CVL concept can be applied to a health care system, providing a measure of health care quality. Like CVL, HSVL estimates depend on definitions of the clinical population and assumptions about missing data.

Towards a combined prognostic index for survival in HIV infection: the role of 'non-HIV' biomarkers.

BACKGROUND: As those with HIV infection live longer, 'non-AIDS' condition associated with immunodeficiency and chronic inflammation are more common. We ask whether 'non-HIV' biomarkers improve differentiation of mortality risk among individuals initiating combination antiretroviral therapy (cART). METHODS: Using Poisson models, we analysed data from the Veterans Aging Cohort Study (VACS) on HIV-infected veterans initiating cART between 1 January 1997 and 1 August 2002. Measurements included: HIV biomarkers (CD4 cell count, HIV RNA and AIDS-defining conditions); 'non-HIV' biomarkers (haemoglobin, transaminases, platelets, creatinine, and hepatitis B and C serology); substance abuse or dependence (alcohol or drug); and age. Outcome was all cause mortality. We tested the discrimination (C statistics) of each biomarker group alone and in combination in development and validation data sets, over a range of survival intervals, and adjusting for missing data. RESULTS: Of veterans initiating cART, 9784 (72%) had complete data. Of these, 2566 died. Subjects were middle-aged (median age 45 years), mainly male (98%) and predominantly black (51%). HIV and 'non-HIV' markers were associated with each other (P < 0.0001) and discriminated mortality (C statistics 0.68-0.73); when combined, discrimination improved (P < 0.0001). Discrimination for the VACS Index was greater for shorter survival intervals [30-day C statistic 0.86, 95% confidence interval (CI) 0.80-0.91], but good for intervals of up to 8 years (C statistic 0.73, 95% CI 0.72-0.74). Results were robust to adjustment for missing data. CONCLUSIONS: When added to HIV biomarkers, 'non-HIV' biomarkers improve differentiation of mortality. When evaluated over similar intervals, the VACS Index discriminates as well as other established indices. After further validation, the VACS Index may provide a useful, integrated risk assessment for management and research.

Visceral leishmaniasis infection in a rheumatoid arthritis patient treated with infliximab.

Anti-TNFalpha strategies can result in significant clinical benefits in rheumatoid arthritis (RA), but with an increased rate of opportunistic infections. Visceral leishmaniasis (VL) is a severe disease that can develop in immunocompromised hosts, principally in HIV patients. VL in RA patients treated with TNFalpha antagonists is an extremely rare event, and only one case has been described. Here we report a case of VL, occurring after 9 infusions of infliximab in association with azathioprine, in a patient who developed blood cytopenia, fluctuant fever, and splenomegaly.

Pseudomonas aeruginosa ventilator-associated pneumonia: comparison of episodes due to piperacillin-resistant versus piperacillin-susceptible organisms.

We sought to determine the epidemiological characteristics of patients in an intensive care unit (ICU) who developed ventilator-associated pneumonia (VAP) caused by piperacillin-resistant Pseudomonas aeruginosa (PRPA; n=34) or piperacillin-susceptible P. aeruginosa (PSPA; n=101). According to univariate analysis, the factors associated with the development of PRPA VAP were presence of an underlying fatal medical condition, immunocompromised status, longer previous hospital stay, less-severe illness at the time of ICU admission, duration of mechanical ventilation before onset of VAP, number of classes of antibiotic received, and previous exposure to imipenem or fluoroquinolone. Multivariate logistic regression analysis identified the following significant independent factors: presence of an underlying fatal medical condition (odds ratio [OR], 5.6), previous fluoroquinolone use (OR, 4.6), and initial disease severity (OR, 0.8). We concluded that the clinical characteristics of patients who develop PRPA VAP differ from those of patients who develop PSPA VAP. Restricted fluoroquinolone use is the sole independent risk factor for PRPA VAP that is open to medical intervention.

Nosocomial pneumonia in ventilated patients: a cohort study evaluating attributable mortality and hospital stay.

PURPOSE: Although nosocomial pneumonia is a common problem in intubated and ventilated patients, previous studies have not clearly demonstrated that nosocomial pneumonia actually results in increased mortality or prolongs hospitalization of these patients. In an attempt to answer these questions, we have performed a cohort study in which patients who developed nosocomial pneumonia and control subjects were carefully matched for the severity of underlying illness and other important variables. PATIENTS AND METHODS: Case patients were 48 ventilated patients with nosocomial pneumonia identified on the basis of results of protected specimen brush quantitative culture and identification of intracellular organisms in cells recovered by bronchoalveolar lavage. For matching cases and their respective controls, the following variables were used: age (+/- 5 years), Simplified Acute Physiologic Score (+/- 3 points), indication for ventilatory support, date of admission, and duration of exposure to risk. RESULTS: Successful matching was achieved for 222 of 240 (92.5%) variables. The mortality rate in cases was 26 of 48 (54.2%) compared with 13 of 48 (27.1%) in controls. The attributable mortality was 27.1% (95% confidence interval [CI], 8.3% to 45.9%; p < 0.01) and the risk ratio for death was 2.0 (95% CI, 1.61 to 2.49). The mean length of stay was 34 days for cases and 21 days for controls (p < 0.02). In the case of pneumonia due to Pseudomonas or Acinetobacter species, the mortality rate was 71.4%, the attributable mortality was 42.8% (95% CI, 14.5% to 69.0%), and the risk ratio was 2.50 (95% CI, 1.31 to 4.61). CONCLUSION: Pneumonias occurring in ventilated patients, especially those due to Pseudomonas or Acinetobacter species, are associated with considerable mortality in excess of that resulting from the underlying disease alone, and significantly prolong the length of stay in the intensive care unit.

Diagnosis of nosocomial bacterial pneumonia in intubated patients undergoing ventilation: comparison of the usefulness of bronchoalveolar lavage and the protected specimen brush.

PURPOSE: To compare the usefulness of specimens recovered using a protected specimen brush and those recovered by bronchoalveolar lavage in the diagnosis of nosocomial pneumonia occurring in intubated patients undergoing ventilation, we performed both procedures in patients suspected of having pneumonia because of the presence of a new pulmonary infiltrate and purulent tracheal secretions. PATIENTS AND METHODS: Twenty-one patients (16 men and five women) with an average age of 57 +/- 12 years were studied. They had been receiving mechanical ventilation for 8 +/- 6 days before inclusion in the trial. The clinical suspicion for nosocomial bacterial pneumonia was high in these patients. Fiberoptic bronchoscopy was performed in each patient. Bronchoscopy specimens were obtained by a protected specimen brush and by bronchoalveolar lavage, and were then processed for quantitative bacterial and fungal culture using standard methods. Total cell counts were performed on an aliquot of resuspended original lavage fluid. Differential cell counts were made on at least 500 cells. In addition, 300 cells were examined at high-power magnification and the percentage of cells containing intracellular microorganisms and the average number of extracellular organisms per oil-immersion field were determined. RESULTS: Quantitative culture of specimens recovered using the protected specimen brush were positive (more than 10(3) colony-forming units [cfu]/ml) in five of five patients with subsequently confirmed pneumonia, and negative (less than 10(3) cfu/ml) in 13 of 13 patients without bacterial pneumonia, but results were not available until 24 to 48 hours after the procedure. Quantification of intracellular organisms in cells recovered by lavage was also useful in distinguishing patients with pneumonia (more than 25 percent of cells with intracellular organisms in five of five patients) from those without pneumonia (less than 15 percent of cells with intracellular organisms in all cases), and results were available immediately. In contrast, quantitative culture of lavage fluid and differential cell counts were of little value in identifying infected patients. CONCLUSION: The protected specimen brush and microscopic identification of intracellular organisms in cells recovered by lavage yield useful and complementary information, and together permit rapid and specific treatment of most patients with nosocomial pneumonia.

Nosocomial legionellosis, Paris, France. Evidence for transmission by potable water.

During a five-week period in 1981, six cases of legionellosis due to Legionella pneumophila serogroup 1 were recognized in a hospital in Paris, France. Four cases were clearly nosocomial in origin. There was a direct association between development of disease and exposure to potable hot water (p = 0.003). The entire hot water system was contaminated with L. pneumophila serogroup 1; monoclonal antibody testing demonstrated that the case isolate and the potable water isolates belonged to the same subgroup. Although serogroup 1 was isolated from both the cooling tower and its drift, the cooling tower isolate was antigenically distant from the case isolate. In other nosocomial outbreaks of legionellosis, multiple sources have been found within the hospital environment, but an epidemiologic association of disease with potable water had not been shown. The significant association of cases with exposure to the potable hot water supply, and the identification of case and potable water isolates of the same subtype, suggest that the potable hot water was responsible for transmission of disease in this outbreak.

Clinical and echocardiographic observations in pulmonary valve endocarditis.

Clinical and echocardiographic data from 12 patients with pulmonary valve endocarditis are described. Seven patients had isolated pulmonary endocarditis and in 5 patients other valves were infected (aortic, tricuspid, mitral or all 3). Two patients were heroin addicts and 4 had underlying heart disease (congenital heart disease in 3 and aortic regurgitation in 1 patient). The organisms involved were alpha streptococci in 3 patients (all with underlying heart disease), Staphylococcus aureus in 4, Streptococcus D bovis in 1 patient and Candida guillermondii in 1. M-mode and 2-dimensional echocardiography was performed in 10 patients and revealed vegetations in 8. Pulsed Doppler echocardiography was performed in 6 patients and revealed pulmonary regurgitation in all 6. Seven patients had pulmonary emboli. Four patients underwent surgery. Four patients died, including 1 after cardiac surgery. Five patients, including the patient infected with Candida guillermondii, recovered with antibiotic treatment.

Prosthetic valve endocarditis. The case for prompt surgical management.

Clinical and morphologic features are described in 27 patients with prosthetic valve endocarditis. The interval from valve replacement to onset of symptoms of prosthetic valve endocarditis was less than 2 months in 10 patients, longer than 2 months but less than 6 months in seven patients, and longer than 6 months in 10 patients. The most frequent infecting organism was Staphylococcus (11 patients). In nearly all patients, infection spread behind the site of attachment of the valve prosthesis and resulted in valve ring abscesses. Twenty-three of the 28 infected prostheses were partially or almost completely detached, and in 15 patients the infection destroyed the entire valve anulus, burrowing to adjacent structures in six. Despite prolonged bactericidal antibiotic therapy, bacterial cultures of prosthetic valves removed at operation or autopsy were positive in 14 patients. Standard valve replacement was attempted in nine patients. All were hospital survivors, but two of these patients evidenced rapid postoperative valve dehiscence and required a complex surgical procedure at reoperation. The 14 other surgically treated patients had almost complete destruction of the annular root, and surgical repair was achieved by complex surgical techniques. There were five postoperative deaths, but nine patients survived with no further evidence of infection (mean follow-up 34 months). All patients with early prosthetic valve endocarditis who recovered underwent this type of operative technique. Total exclusion of the infected annular root, as described, may offer in patients with extensive endocarditic lesions the only possibility to eradicate the infection and to reduce the mortality.

Acute pneumonitis with bilateral pleural effusion after talc pleurodesis.

We describe a patient who developed acute pulmonary distress with bilateral interstitial infiltrates and pleural effusion following talc pleurodesis. Talc particles, obtained by bronchoalveolar lavage, were identified by transmission electron microscopy and chemical analysis. The patient improved with corticosteroid therapy. Acute respiratory failure can be a potential hazard of talc pleurodesis.

Incidence and morbidity of cytomegaloviral infection in patients with mediastinitis following cardiac surgery.

To determine the incidence and morbidity of infections with CMV associated with mediastinitis after conventional cardiac surgery, 115 consecutive adult patients with mediastinitis were evaluated with viral cultures of blood and urine. Shedding of CMV was seen in 29 patients (25 percent) within a mean period of 37 +/- 22 days after cardiopulmonary bypass. Viremia was documented in 79 percent (23) of these 29 patients. Acute renal failure and enzymatic abnormalities (AST and LDH) were significantly more common in patients with virologically proven infection with CMV (p less than 0.05). In patients who survived the initial period of bacterial infection, major differences in their clinical course were observed according to their virologic status. After the 15th day of hospitalization following the débridement, the persistence of local infection was more frequent (p less than 0.05) and the mortality was higher (p less than 0.01) in CMV-infected patients. Moreover, the mean duration of hospitalization in the ICU for survivors was 69 +/- 36 days in viral shedders, compared with 48 +/- 27 days in nonshedders (p less than 0.05). Infection with CMV in mediastinitis occurs frequently and is associated with persistence of local infection, prolonged hospitalization, and increased late mortality.

Intragastric pH profile during acute respiratory failure in patients with chronic obstructive pulmonary disease. Effect of ranitidine and enteral feeding.

The ability of H2 receptor antagonists and continuous enteral alimentation to maintain high intragastric pH in patients with chronic obstructive pulmonary disease (COPD) requiring mechanical ventilation was evaluated by continuously monitoring intragastric pH prior to and following sequential addition of ranitidine or continuous enteral alimentation (or both) to their therapeutic regimen. Prior to therapy, intragastric pH was less than 4.0 for 75 +/- 10 percent of the time, but never less than 1.0. Nevertheless, this moderate gastric acidity was associated with evidence of mucosal injury. Ranitidine failed to continuously maintain a high intragastric pH (pH less than 4.0 for 35 +/- 11 percent of the time; p greater than 0.2 compared to patients treated with placebo). Following administration of continuous enteral alimentation, intragastric pH fell, and ranitidine therapy only partially blocked this increase in gastric acidity induced by continuous enteral alimentation. We conclude that without treatment, patients with COPD who have acute respiratory failure may develop gastric mucosal injury despite the presence of only moderate intragastric acidity; however, ranitidine and continuous enteral alimentation are not effective in maintaining a high intragastric pH.

Pulmonary fibrosis following pneumonia due to acute Legionnaires' disease. Clinical, ultrastructural, and immunofluorescent study.

During a recent nosocomial outbreak, 20 critically ill patients with acute Legionnaires' disease were admitted to the intensive care unit of Hopital Bichat, Paris. Pulmonary specimens were obtained at surgery or immediately after death in 12 patients and were examined by light, immunofluorescent, and electron microscopy. Five of these 12 patients showed evidence of pulmonary fibrosis. In all of these five patients, infection with Legionella pneumophila was evidenced by bacteriologic methods, and other diseases known to cause fibrosis were excluded. The condition of four patients deteriorated rapidly with respiratory failure, and they died with pulmonary fibrosis. Only one patient finally recovered but was left with pulmonary sequelae. Two distinctive morphologic patterns were observed, one in which interstitial fibrosis was predominant and one in which intra-alveolar organization and fibrosis were also present. The alveolar epithelial lining and the basement membranes were disrupted in all patients, as evidenced by ultrastructural observations and by immunofluorescent studies showing gaps in the distribution of type 4 collagen and laminin. Types 1 and 3 collagen accumulated in areas corresponding to thickened interstitium and intra-alveolar fibrosis. Thus, some patients who survive the acute pneumonia of Legionnaires' disease may develop pulmonary fibrosis, and this process may lead to functional impairment or death despite prompt and appropriate treatment.

Fiberoptic bronchoscopy in ventilated patients. Evaluation of cardiopulmonary risk under midazolam sedation.

One hundred seven acutely ill ventilated patients were prospectively studied to ascertain the severity and frequency of alterations in gas exchange and hemodynamic parameters during brief bronchoscopy. Sedation was performed using midazolam (0.1 mg/kg IV) without topical anesthesia. An average decline in PaO2 of 26 percent was observed at the end of the procedure, compared to the baseline value, and this was associated with a mild increase in PaCO2 in spite of the use of a special adapter. Alterations in mean systolic blood pressure appeared to be modest, consisting of a 10 percent decrease from the control level, related to sedation, and a 10 percent rise from baseline during the procedure, associated with a concomitant mild tachycardia. At that time, central hemodynamic measurements performed in a subset of 31 patients showed a significant increase in cardiac output associated with higher pulmonary wedge pressure. Fourteen patients developed hypoxemia of less than 60 mm Hg on FIO2 adjusted to 0.8. Of the ten risk factors univariately associated with hypoxemia, only the presence of ARDS (p less than 0.001) and "fighting" the ventilator during the procedure (p less than 0.05) remained significant after stepwise logistic regression. Attempts to prevent hypoxemia in critically ill patients should focus on inducing complete sedation, with careful attention to hemodynamic status, or providing maximal levels of oxygen to the ventilator (or both).

Long-term outcome and quality of life of patients requiring multidisciplinary intensive care unit admission after cardiac operations.

Patients with organ failure or severe infection after cardiac operations may require prolonged stays in the intensive care unit. This study examined long-term mortality and determined quality of life for surviving patients in this group. This observational cohort study was conducted at Bichat Hospital, Paris, an academic tertiary care center. The study group consisted of 116 consecutive patients who underwent cardiac operations and were transferred to the multidisciplinary intensive care unit between January 1986 and December 1987. Patients referred for mediastinitis were automatically excluded. Respiratory failure (88.8%) and hemodynamic instability (81.9%) were the main causes of transfer; an infection was present in 23.3% of patients at entry into the intensive care unit. Twenty-seven patients (23.3%) died in the intensive care unit. Presurgical New York Heart Association functional class, postoperative bacteremia before admission to the intensive care unit, and severity of illness on admission to the intensive care unit were independent predictors of death in the intensive care unit. After an average follow-up of 81 months (range 70 to 93 months), 69% of the patients alive at transfer from the intensive care unit were still alive. Preoperative New York Heart Association functional class was the only long-term independent prognostic factor. Quality of life, as evaluated by the Nottingham Health Profile, was good for more than 70% of the survivors and was not influenced by any recorded variables, with the exception of age.

Evaluation of clinical judgment in the identification and treatment of nosocomial pneumonia in ventilated patients.

To evaluate the accuracy of clinical judgment in the diagnosis and treatment of nosocomial pneumonia in ventilated patients, we studied 84 patients suspected of having nosocomial pneumonia because of the presence of a new pulmonary infiltrate and purulent tracheal secretions. We prospectively evaluated the accuracy of diagnostic predictions and therapeutic plans independently formulated by a team of physicians aware of all clinical, radiologic and laboratory data, including the results of Gram-stained bronchial aspirates. Definite (n = 51) or probable (n = 33) diagnoses could be established in all patients by strict histopathologic and/or bacteriologic criteria. Only 27/84 patients were diagnosed as having pneumonia. Organisms responsible for pneumonias were identified by quantitative cultures of samples obtained using a protected specimen brush or pleural fluid cultures. Four hundred eight predictions were made for the 84 studied patients. Clinical diagnoses for patients subsequently diagnosed as having pneumonia were accurate in 81/131 cases (62 percent). Furthermore, only 43/131 (33 percent) therapeutic plans proposed for these patients represented effective therapy. Common causes of inappropriate treatment included failure to diagnose pneumonia (50 plans), failure to effectively treat highly resistant organisms (21 plans), and failure to treat all organisms in cases of polymicrobial pneumonia (14 plans). Therapeutic plans formulated for patients without pneumonia included the unnecessary use of antibiotics in 45/277 cases (16 percent). These findings indicate that the use of clinical criteria alone does not permit the accurate diagnosis of nosocomial pneumonia in ventilated patients, and commonly results in inappropriate or inadequate antibiotic therapy for these patients.