Michael B. Kirkpatrick

Diagnosis and Treatment of Ventilator-Associated Pneumonia—Impact on Survival: A Decision Analysis

STUDY OBJECTIVE To determine the impact of antibiotic treatment of ventilator-associated pneumonia (VAP) on survival. DESIGN Decision analysis. PATIENTS A hypothetical cohort of immunocompetent patients receiving mechanical ventilation who have suspected bacterial pneumonia. The analysis was performed separately for the following diagnostic techniques: clinical criteria, bronchoscopic protected specimen brush (PSB), and nonbronchoscopic protected BAL (pBAL). Additional factors accounted for in the analysis included the presence or absence of prior antibiotic use, mortality of antibiotic-treated and untreated pneumonia, mortality attributable to VAP, development of antibiotic resistance, and mortality due to adverse drug reactions. MEASUREMENTS AND RESULTS The overall survival of patients who receive antibiotic therapy was compared to survival if antibiotic therapy had been withheld. Antibiotic treatment of clinically diagnosed VAP was associated with lower overall survival than withholding treatment. Antibiotic treatment of VAP diagnosed by invasive (PSB) or semi-invasive (pBAL) techniques was associated with better survival than withholding treatment, although withholding antibiotic therapy was favored as the mortality rate of antibiotic-treated VAP approached 70%. CONCLUSIONS Invasive or semi-invasive diagnostic techniques should be used to diagnose VAP, guide therapy, and thereby potentially improve survival. A prospective, randomized trial assessing outcome according to diagnostic technique is needed.

Clinical Investigations In Critical Care: ArticlesDiagnosis and Treatment of Ventilator-Associated Pneumonia—Impact on Survival: A Decision Analysis

STUDY OBJECTIVE To determine the impact of antibiotic treatment of ventilator-associated pneumonia (VAP) on survival. DESIGN Decision analysis. PATIENTS A hypothetical cohort of immunocompetent patients receiving mechanical ventilation who have suspected bacterial pneumonia. The analysis was performed separately for the following diagnostic techniques: clinical criteria, bronchoscopic protected specimen brush (PSB), and nonbronchoscopic protected BAL (pBAL). Additional factors accounted for in the analysis included the presence or absence of prior antibiotic use, mortality of antibiotic-treated and untreated pneumonia, mortality attributable to VAP, development of antibiotic resistance, and mortality due to adverse drug reactions. MEASUREMENTS AND RESULTS The overall survival of patients who receive antibiotic therapy was compared to survival if antibiotic therapy had been withheld. Antibiotic treatment of clinically diagnosed VAP was associated with lower overall survival than withholding treatment. Antibiotic treatment of VAP diagnosed by invasive (PSB) or semi-invasive (pBAL) techniques was associated with better survival than withholding treatment, although withholding antibiotic therapy was favored as the mortality rate of antibiotic-treated VAP approached 70%. CONCLUSIONS Invasive or semi-invasive diagnostic techniques should be used to diagnose VAP, guide therapy, and thereby potentially improve survival. A prospective, randomized trial assessing outcome according to diagnostic technique is needed.

The Acute Chest Syndrome of Sickle Cell Disease

The acute chest syndrome (ACS), characterized by fever, chest pain, leukocytosis and a new infiltrate on chest roentgenogram, is a common complication of sickle hemoglobinopathies. The major differential diagnoses of ACS are pneumonia and pulmonary vaso-occlusive disease, which may occur simultaneously. Bacterial pulmonary infections are documented infrequently in ACS with the exception being in the pediatric population under 5 years of age. Because there are no clinical or laboratory parameters that clearly allow for distinction between pneumonia and vaso-occlusive disease, empiric use of antibiotics directed against S. pneumoniae and other pathogens commonly seen in community-acquired pneumonias remain a mainstay of therapy.

Comparison of Four Methods for Assessing Airway Bacteriology in Intubated, Mechanically Ventilated Patients

A prospective evaluation of lower airway bacteriology from intubated, mechanically ventilated patients was performed by comparing the qualitative and quantitative recovery of bacteria using four different techniques. Twelve intubated, mechanically ventilated patients who satisfied accepted clinical criteria for the suspicion of ventilator-associated pneumonia were studied. Airway secretions were obtained from each patient by: (1) blind endotracheal aspiration (ET); (2) Accu-cath pulmonary culture catheter (Accu); (3) bronchoscopic protected specimen brush (BPSB); and (4) bronchoalveolar lavage (BAL). ET specimens were cultured semi-quantitatively (1+ to 4+) aerobically, and all other specimens were cultured quantitatively both aerobically and anaerobically. The BPSB recovered 9 organisms in > or = 10(3) colony forming units/ml, a standard number often used to indicate significant growth. Of these 9 organisms, 7 were recovered at > or = 10(3) cfu/ml by Accu, and 6 were recovered at > or = 10(4) cfu/ml by BAL. All 8 aerobic isolates recovered in > or = 10(3) cfu/ml by BPSB also were recovered by ET aspirate. Five of these were recovered in > or = 3+ semi-quantitative growth by ET aspirate. Of 30 organisms recovered in < 3+ semi-quantitative growth by ET aspirate, 28 were recovered in < 10(3) cfu/ml by BPSB, indicating a negative predictive value of 93%. Thus, it appears that these four methods provide reasonably similar qualitative and quantitative recovery of bacteria from the lower airways of intubated, mechanically ventilated patients. In addition, routine Grams stain and semi-quantitative aerobic culture of endotracheal aspirate may provide useful information in patients with suspected ventilator-associated pneumonia.

Clinical Use of Cocaine: A Review of the Risks and Benefits

SummaryFollowing the isolation of cocaine from the extract of coca leaves in the late 1700s by Albert Neiman, the local anaesthetic properties of the drug have been evaluated. The anaesthetic effect of cocaine is believed to be the result of reversible blockade of nervous impulse conduction by the prevention of sodium ion movement within the cell membrane. The many undesired effects, however, are the result of adrenergic stimulation by means of prevention of noradrenaline (norepinephrine) uptake. The clinical use of cocaine in the modern era is associated primarily with surgical procedures involving the manipulation of mucous membranes, particularly those of the upper respiratory tract. Recently, based on problems with the potential for illicit use and adverse effects, alternative regimens have been investigated. In some settings these have been superior to cocaine. Further studies comparing various anaesthetic regimens are warranted.

The role of bacterial super infection in extensive pulmonary tuberculosis: Data from protected brush cultures in untreated patients from the University of South Alabama Medical Center, Mobile, Alabama, USA

Pulmonary tuberculosis (TB) and bacterial pneumonia are both characterized by fever, cough, and purulent sputum production. Although TB alone can cause these symptoms, the possibility of a concomitant bacterial pneumonia has led some clinicians to treat these patients empirically with antibacterial agents. Neither the benefit of such empiric antibiotic therapy nor transtracheal aspirate cultures from patients with pulmonary TB have yielded consistent results. Consequently, we performed a prospective study to obtain lower airway secretions via a bronchoscopic protected specimen brush (PSB) technique for quantitative aerobic and anaerobic cultures from untreated patients with extensive pulmonary TB (defined as cavitary disease or involvement of > or = 3 segments). We obtained bronchoscopic samples from 3 untreated men aged 21, 61, and 60 years with extensive pulmonary TB. There was no significant bacterial growth (aerobic or anaerobic) from the specimens obtained. These results, therefore, do not support the hypothesis that bacterial pneumonia is a common concomitant of extensive pulmonary TB.