Ventilator-associated pneumonia in patients with respiratory failure. A diagnostic approach

From bedside to bedside: how iterative clinical research influenced the diagnosis and management of pneumonia at the Elvis Presley Trauma Center

Ventilator-associated pneumonia is a well-acknowledged complication after hospitalization for injury or surgical emergency. The contribution to the literature on this topic by Dr Timothy Fabian and the Memphis group at the Elvis Presley Trauma Center resulted in the contemporary recognition that the diagnosis and management of pneumonia is an essential component of surgical critical care. During three decades, the Memphis group, under Dr Fabian's leadership, performed numerous clinical studies that led to the publication of over 40 articles concerning the epidemiology, diagnosis, and treatment of pneumonia after injury. The purpose of this review is to survey the consecutive studies from Memphis specifically that led to the development of a clinical pathway that has stood the test of time. Examination of the research output during this period provides a case study in how bedside clinical research can inform clinical practice and is a model for applied science in the intensive care unit.

The Effect of Lung Lavage Volume Return on the Diagnosis of Ventilator-Associated Pneumonia

BACKGROUND

Bronchoalveolar lavage (BAL) is a commonly used tool in the diagnosis of ventilator-associated pneumonia (VAP). Previous protocols recommend 30% lavage return, though no studies have investigated this relationship. This study aims to assess the influence of BAL volume return on VAP diagnosis.

MATERIALS AND METHODS

A retrospective review was performed of a prospectively maintained database for BAL performed from January 2015 to January 2016 in the trauma and surgical ICU at a level 1 trauma center. In total, 147 ventilated patients with clinical suspicion for pneumonia underwent 264 BALs. A protocol was used with five aliquots of 20 cc of saline instilled. Quantitative cultures were performed with 10ˆ5 colony-forming organisms as the threshold for VAP diagnosis. BAL was repeated at 6-8 d on 50 patients. Univariate and multivariate regression analyses were performed to investigate the predictors of VAP diagnosis.

RESULTS

Patients with >40% lavage return had increased rates of VAP diagnosis (odds ratio [OR] 2.86, P = 0.002). Increasing volume return also trended toward a lower false-negative rate. Temperature, leukocytosis, and X-ray infiltrate were not associated with increased VAP diagnosis. Concurrent antibiotic therapy at the time of BAL predicted decreased VAP diagnosis (OR 0.58, P = 0.04). On multivariable analysis, only >40% return remained associated with increased rate of VAP diagnosis (OR 4.00, P = 0.004).

CONCLUSIONS

This study found that >40% lavage volume return was associated with increased VAP diagnosis. Clinicians should consider the reliability of a negative BAL if clinical suspicion of VAP is high and lavage return is <40%. Additional investigation is needed to further elucidate this association.

[Nosocomial pneumonia from a radiological perspective]

Due to the high morbidity and mortality, nosocomial pneumonia represents a serious risk in hospitalized patients. The increased risk of infections with multidrug-resistant (MDR) pathogens makes a timely diagnosis and prompt therapy indispensable. A newly occurring or progressive infiltrate in any patient who has been hospitalized for more than 48 h should be viewed with suspicion. In contrast to community acquired pneumonia (CAP), radiography plays a limited role in the diagnosis of hospital-acquired pneumonia (HAP). This is partly due to the technical challenges in imaging of patients who are in a lying position as well as the numerous other possible differential diagnoses. Careful analysis of the various radiological features, such as temporal progression, distribution and appearance can help to narrow down the differential diagnoses. In the absence of a single gold standard, clinical features and appropriate radiological features in addition to cultures obtained from respiratory secretions can help to maximize the diagnostic efficacy and expedite the treatment with appropriate antibiotic therapy.

Risk stratification for the development of a subsequent pneumonia after a nondiagnostic bronchoalveolar lavage

BACKGROUND

Broncho-alveolar lavage (BAL) is an invasive bedside procedure to define type and concentration of pathologic organisms causing ventilator associated pneumonia (VAP). We evaluated if the absence of pathogens on final results represented a lavage aspect of the BAL as a therapeutic procedure to eliminate organisms.

METHODS

BAL results collected from 2008 to 2009 were stratified as positive (POS) ≥ 100,000 cfu), indeterminate (INT) ≤ 100,000 cfu pathologic organisms, or negative defined as mixed flora (MF) or sterile (STR). The INT, MF, and STR results were assessed by incidence of a subsequent POS sample.

RESULTS

Nine-hundred forty-nine BALs performed on 490 SICU patients were interpreted as POS in 227 patients (46%). 237 non- POS patients needed a subsequent BAL. Any pathogen on the first BAL (INT group) indicates a high likelihood for subsequent BAL which will be POS. Monthly cumulative sum analysis (CUSUM) of yield was unable to identify any specific period in which BAL performance varied from trend.

CONCLUSION

MF and STR represent adequate sampling of secretions that are clinically benign. Any pathogen, regardless of concentration, should be considered a biomarker for future pneumonia. CUSUM analysis suggest better training in timing and indication may decrease unnecessary procedures yielding negative results.

Mixed Flora: Indication for Therapy or Early Warning Sign?

“Mixed flora” is a commonly returned result yielding not in either indication for therapy or identification of potential causative organisms. We sought to determine whether mixed flora (MF) was in fact a harbinger of impending pneumonia or a benign result that could be therapeutically ignored. Bronchoalveolar lavage (BAL) results of injured adults undergoing mechanical ventilation in a trauma intensive care unit were stratified by identified organisms and by colony counts. The incidence of mixed flora as a component of the specimen report was compared for diagnostic (greater than 105 colony forming units/mL) versus nondiagnostic results using χ2 accepting P < 0.05 as significant. Nondiagnostic specimens were then stratified as MF only or MF and other identified pathogenic organisms. This group was further evaluated to determine the use of antibiotic therapy and development of pneumonia. Finally, patients with nondiagnostic reports and subsequent BAL were analyzed to determine specific species if subsequent BAL were required or if later pneumonia occurred. During 2007, 159 BALs were performed on injured patients of which 93 were diagnostic for pneumonia, whereas 66 were nondiagnostic. Of the diagnostic specimens, 15 (16%) included mixed flora. Of the 66 nondiagnostic specimens, 39 (59%) contained mixed flora. Nine (60%) of the 15 with diagnostic mixed flora were started on antibiotic therapy for an average of 6.2 days. The remaining 39 (82%) patients with mixed flora received no antibiotic therapy and never developed pneumonia. These data demonstrate that in the absence of diagnostic threshold of an identifiable pathogenic organism, therapy for pneumonia should not be instituted or continued.

Predictive value of broncho-alveolar lavage fluid Gram's stain in the diagnosis of ventilator-associated pneumonia: a prospective study

BACKGROUND

Quantitative broncho-alveolar lavage (qBAL) is increasingly being used for diagnosing ventilator-associated pneumonia (VAP). The current study prospectively evaluates the accuracy of broncho-alveolar lavage fluid Gram's stain (GS) in predicting both the presence of VAP and the class of causative microorganism in patients suspected of VAP.

METHODS

Patients suspected of VAP in a trauma or surgical intensive care unit underwent bronchoscopic qBAL with GS. Presence and class of organisms seen on GS were correlated respectively with the presence of VAP, as diagnosed by qBAL, and class of causative microorganism. VAP was defined as qBAL >10(5) colony forming units/mL. All data were gathered prospectively.

RESULTS

During a 28-month study period, 229 patients underwent 309 qBALs for suspected VAP. Seventy-one (23%) specimens were positive for VAP (qBAL>10(5) CFU/mL). Fifty-four specimens (77%) had one causative microorganism, 13 (18%) had two, 3 (4%) had three, and 1 (1%) demonstrated four microorganisms giving a total of 93 VAPs. Forty-one (62%) of 66 specimens showing moderate or many microorganisms on GS were positive for VAP. However, 7 (4%) of 167 specimens showing none and 23 (30%) of 76 showing few microorganisms on GS were also positive for VAP. Of the 64 qBAL specimens positive for VAP and where the GS showed microorganisms, 6 (23%) of 26 showing only G+ microorganisms on GS had G- VAP (G- alone, 4; G+ and G-, 2), and 1 (8%) of 12 showing G- microorganisms only had G+ and G- VAP. Of the seven qBAL specimens positive for VAP where the GS did not show microorganisms, one had G+ and six had G- VAP. With the threshold of positivity of GS at more than none, the sensitivity, specificity, positive, and negative predictive values of GS for the presence of VAP were 90%, 67%, 45%, and 96% respectively.

CONCLUSIONS

Broncho-alveolar lavage fluid GS is poor in predicting the presence of VAP and predicting the class of causative microorganism. Using GS to determine necessity of and to select class of antimicrobial therapy will result in delayed or inappropriate VAP therapy or both.

Intensive-care unit lung infections: The role of imaging with special emphasis on multi-detector row computed tomography

Nosocomial pneumonia is the most frequent hospital-acquired infection. In mechanically ventilated patients admitted to an intensive-care unit as many as 7-41% may develop pneumonia. The role of imaging is to identify the presence, location and extent of pulmonary infection and the presence of complications. However, the poor resolution of bedside plain film frequently limits the value of radiography as an accurate diagnostic tool. To date, multi-detector row computed tomography with its excellent contrast resolution is the most sensitive modality for evaluating lung parenchyma infections.

Ventilator-associated pneumonia in injured patients: do you trust your Gram's stain?

BACKGROUND

The results of sputum or bronchoalveolar lavage (BAL) fluid Gram's stain have been used to guide presumptive antibiotic therapy for ventilator-associated pneumonia (VAP) in injured patients, despite reported variability in sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Our aim was to evaluate the utility of Gram's stain of BAL fluid in the diagnosis of VAP.

METHODS

We conducted a retrospective chart review of all mechanically ventilated trauma patients who developed pneumonia over a 5-year period in whom Gram's stain and final culture data were available.

RESULTS

One hundred fifty-five records with complete data sets were reviewed. VAP was diagnosed by Centers for Disease Control and Prevention criteria and confirmed by BAL and quantitative culture in all patients. Overall accuracy of Gram's stain in diagnosing VAP for any organism was 88% (137 true-positives). When assessed for the ability to predict pneumonia caused by a specific organism, the accuracy decreased significantly, with only 63% of Gram-negative VAPs and 72% of Gram-positive VAPs accurately identified by Gram's stain. However, the absence of Gram-positive organism of Gram's stain excludes Gram-positive VAP in 80% of patients.

CONCLUSION

All trauma patients should be covered presumptively for gram-negative organisms, as they encompass 70% of infections, but are not reliably identified by Gram's stain. As 88% of VAP can be identified by the presence of any organism on Gram's stain, it may be useful in the early diagnosis of VAP but cannot reliably be used to guide presumptive therapy.

Antibiotic-treated infections in intensive care patients in the UK

The purpose of this audit was to study reasons for starting antibiotic therapy, duration of antibiotic treatment, reasons for changing antibiotics and the agreement between clinical suspicion and microbiological results in intensive care practice. We conducted a multicentre observational audit of 316 patients. Data on demographic details, site, treatment and nature of infection were collected. The median duration of antibiotic therapy was 7 days. Infections were community-acquired in 160 patients (55%). Antibiotics were started on clinical suspicion of infection in 237 patients (75%). Pulmonary infections were the most common, representing 52% of all proven infections. Gram-negative organisms were the most common cause of proven infections (n = 90 (50%)). The antibiotic spectrum was narrowed in light of microbiology results in 78 patients (43%) and changed due to antibiotic resistance in 38 patients (21%). We conclude that the mean duration of treatment contrasts with existing published guidelines, highlighting the need for further studies on duration and efficacy of treatment in intensive care.

A formula for prediction of posttraumatic pneumonia based on early anatomic and physiologic parameters

BACKGROUND

Identification of risks for development of ventilator-associated pneumonia (VAP), which might be identified early after injury, would allow for prognostic estimates and targeting of high-risk cohorts for clinical trials of preventive strategies. This study was performed to develop an equation that can be applied to estimate the probability of pneumonia based on parameters collected in the early postinjury interval.

METHODS

Over a 28-month period, patient admissions were reviewed for mechanism and severity of injury, patterns of injury, shock, and need for emergent intubation. Early deaths (<48 hours) were excluded. VAP diagnosis required > or = 10(5) colony-forming units/mL organisms in the bronchoalveolar lavage effluent. Multiple logistic regression analysis was used to develop the prediction equation and estimate odds ratios. The equation was then tested on consecutive patients admitted over a 2-month period.

RESULTS

We reviewed 9,721 admissions (77% blunt, 23% penetrating). VAP incidence was 5.6%. Overall mortality was 2% (21% for patients with VAP vs. 1% for no VAP; p < 0.0001). Multiple logistic regression analysis for all patients produced the following equation: f(x) = -3.08 - 1.56 (MOI) - 0.12 (GCS) + 1.37 (SCI) + 0.30 (chest AIS) + 1.87 (lap) + 0.67 (tx) + 0.05 (ISS) + 0.66 (int), where MOI is mechanism of injury (penetrating = 1, blunt = 0), GCS is Glasgow Coma Scale score, SCI is spinal cord injury (yes = 1, no = 0), lap is emergent laparotomy (yes = 1, no = 0), ISS is Injury Severity Score, tx is units of blood transfused in the resuscitation room, and int is intubation in either the field or the resuscitation room (yes = 1, no = 0). The probability of VAP was calculated as follows: P(VAP) = e(f)(x)/1 + e(f)(x). This formula was concordant in 95% and discordant in 5%.

CONCLUSION

It is possible to accurately predict risk for VAP in trauma patients based on data available early after injury. This calculation could be useful for counseling families relative to prognosis and research protocols, and addressing hospitalization issues with third-party payors.

Physiologic factors contributing to a transition in oral immunity among mechanically ventilated adults

Ventilator-associated pneumonia (VAP), a specific type of nosocomial pneumonia, occurs in approximately 21% of patients in intensive care, and the mortality can be as high as 71%. VAP causes considerable mortality and morbidity, and it exponentially increases health care costs. The incidence of VAP is associated with oropharyngeal colonization of gram-negative bacteria. Within 48 h of hospital admission, the composition of the oropharyngeal flora of critically ill patients undergoes a change from the usual gram-positive streptococci and dental pathogens to a predominant gram-negative flora that includes more virulent organisms, which predispose patients to VAP. Identification and understanding of this oral transition from gram-positive to predominantly gram-negative flora may assist health care professionals in differentiating among oral immune markers that suggest compromised immunity. The purpose of this article is to provide a review of the literature that promotes an understanding of current knowledge about the transition of oral immunity in mechanically ventilated patients.

Role of bronchoscopy in modern medical intensive care unit

This article gives a broad overview of the increasingly important applications of bronchoscopy, flexible (FOB) and rigid (RB), in a modern medical intensive care unit. Special emphasis is made to bronchoscopy use in mechanically ventilated patients. Therapies such as endobronchial stenting and Nd:YAG laser are being used to improve respiratory failure and facilitate weaning from mechanical ventilation. Practical applications of recent advancements in technology (endobronchial stenting, laser therapy, and so forth), the increasing use of rigid bronchoscopy, and the new generation of flexible bronchoscopes like battery bronchoscopes, and ultra-thin bronchoscopes, are also discussed. The risks, potential benefits, complications, and suggested technique of performing bronchoscopy in mechanically ventilated patients are reviewed.

Acute respiratory distress syndrome caused by pulmonary and extrapulmonary injury: a comparative CT study

PURPOSE

To determine computed tomographic (CT) differences between acute respiratory distress syndrome (ARDS) due to pulmonary injury (ARDS(p)) and extrapulmonary injury (ARDS(ex)).

MATERIALS AND METHODS

CT appearances in 41 patients (27 male, 14 female; mean age, 47.1 years +/- 17.1 [SD]; age range, 17-79 years; those with ARDS(p), n = 16; those with ARDS(ex), n = 25) were categorized as typical or atypical of ARDS by two observers. The extent of individual CT patterns was also quantified.

RESULTS

Typical CT appearances were more frequent in ARDS(ex) than ARDS(p) (18 [72%] of 25 vs five [31%] of 16 patients, respectively; P <.01). Sensitivity, specificity, and accuracy of a typical CT pattern for the diagnosis of ARDS(ex) were 72%, 69%, and 71%, respectively. Atypical appearances were characterized by more extensive nondependent intense parenchymal opacification (IPO) (P =.03) and cysts (P =.05), whereas typical CT appearances had more extensive dependent IPO (P =.01). Typical appearances at CT were independently related to the cause of ARDS (odds ratio, 8.9; 95% CI: 1.8, 44.2; P <.01) but were independent of the time from intubation. Foci of nondependent IPO were more extensive in ARDS(p) (P =.05) than ARDS(ex), but this finding was ascribable to differences in time to CT (after intubation) between ARDS(p) and ARDS(ex).

CONCLUSION

The differentiation between ARDS(p) and ARDS(ex) can, with some caveats, be based on whether the CT appearances are typical or atypical of ARDS but not on any individual CT pattern in isolation.

Identification of Early Predictors for Post-Traumatic Pneumonia

We demonstrated that the standard clinical criteria of fever, leukocytosis, purulent sputum, and infiltrate on chest radiograph are nonspecific for the diagnosis of post-traumatic pneumonia, and only ∼50 per cent of patients with these conditions have pneumonia. Quantitative cultures of bronchoalveolar lavage effluent will differentiate pneumonia (requiring antibiotic therapy) from systemic inflammatory response syndrome (not requiring antibiotics). Early identification of patients at risk for pneumonia can target populations for clinical research. Because risk factors for pneumonia when diagnosed by quantitative cultures have not been defined we reviewed our recent experience to identify variables predictive of pneumonia. Patients over a 22-month period who survived >48 hours were identified from the trauma registry. Pneumonia was defined as growth of ≥105 organisms per milliliter in the bronchoalveolar lavage effluent. Risk factors evaluated included injury severity and severity of shock. There were 7503 patients (75% with blunt and 25% with penetrating injuries). The incidence of pneumonia was 6 per cent (7% of patients with blunt and 2% of patients with penetrating injuries). Logistic regression analysis identified age; Glasgow Coma Scale score; Injury Severity Score; transfusion requirements during resuscitation; spinal cord injury; chest injury severity; and emergent femur fixation, craniotomy, and laparotomy as being independent predictors of pneumonia. We conclude that multiple risk factors, which are all able to be determined early after injury, are predictive of post-traumatic pneumonia. Prompt identification of this high-risk group of patients allows prognostic considerations relative to patient management schemes and targets populations for prophylactic measures or immunomodulation.

New Epidemiology for Postoperative Nosocomial Infections

Changes in health care delivery systems over the last decade have resulted in a major increase in outpatient surgery and a higher severity of illness for inpatients. We sought to determine the effects of this change on the epidemiology of postoperative surgical infections. Historical data on incidence and epidemiology of infection were obtained from peer-reviewed articles published between 1960 and 1999 (MEDLINE). All nosocomial infections in 5035 patients admitted to a tertiary-care university hospital surgical intensive care unit between January 1994 and December 1997 were prospectively identified and classified as wound, urinary tract, bloodstream, or pneumonia. Incidence of bacterial isolates at each site was also recorded. From these data we determined infection rates per 100 admissions. We also identified all device-related nosocomial infections and calculated infection rates. Comparisons between time periods were made. In the 1960s wound infections constituted the predominant postoperative infection at 46 per cent. This was replaced by urinary tract infection in the 1970s (44%) and 1980s (32%) and closely followed by bloodstream infections (25%). In the 1990s nosocomial pneumonia became the most common postoperative infection, comprising 43 per cent of surgical intensive care unit infections. Analysis of the bacteriology also revealed changing trends with primarily Gram-positive organisms in the 1960s followed by an increase in methicillin-resistant Staphylococcus in the 1970 to 1980s, and currently resistant Gram-negative bacteria predominate. The incidence of fungal infections has steadily increased. This survey identified a new epidemiology for postoperative surgical infections. Over the last several decades the reported wound infections have been markedly decreased and there is little change in urinary tract infection. Nosocomial pneumonia with resistant Gram-negative bacteria now predominates along with increased incidence of fungal infections. Currently, postoperative infections are now more severe, involve critical organs, and require close monitoring of the changing patterns of pathogens.

A comparative analysis of patients with early-onset vs late-onset nosocomial pneumonia in the ICU setting

STUDY OBJECTIVE

To compare the clinical outcomes of critically ill patients developing early-onset nosocomial pneumonia (NP; ie, within 96 h of ICU admission) and late-onset NP (ie, occurring after 96 h of ICU admission).

DESIGN

Prospective cohort study.

SETTING

A medical ICU and a surgical ICU from a university-affiliated urban teaching hospital.

PATIENTS

Between July 1997 and November 1998, 3, 668 patients were prospectively evaluated.

INTERVENTION

Prospective patient surveillance and data collection.

RESULTS

Four hundred twenty patients (11.5%) developed NP. Early-onset NP was observed in 235 patients (56.0%), whereas 185 patients (44.0%) developed late-onset NP. Among patients with early onset NP, 114 patients (48. 5%) spent at least 24 h in the hospital prior to ICU admission, compared to 57 patients (30.8%) with late-onset NP (p = 0.001). One hundred eighty-three patients (77.9%) with early-onset NP received antibiotics prior to the development of NP, as compared to 162 patients (87.6%) with late-onset NP (p = 0.010). The most common pathogens associated with early-onset NP were Pseudomonas aeruginosa (25.1%), oxacillin-sensitive Staphylococcus aureus (OSSA; 17.9%), oxacillin-resistant S aureus (ORSA; 17.9%), and Enterobacter species (10.2%). P aeruginosa (38.4%), ORSA (21.1%), Stenotrophomonas maltophilia (11.4%), OSSA (10.8%), and Enterobacter species (10.3%) were the most common pathogens associated with late-onset NP. The ICU length of stay was significantly longer for patients with early-onset NP (10.3 +/- 8.3 days; p < 0.001) and late-onset NP (21. 0 +/- 13.7 days; p < 0.001), as compared to patients without NP (3.5 +/- 3.2 days). Hospital mortality was significantly greater for patients with early-onset NP (37.9%; p = 0.001) and late-onset NP (41.1%; p = 0.001) compared to patients without NP (13.1%).

CONCLUSIONS

Both early-onset and late-onset NP are associated with increased hospital mortality rates and prolonged lengths of stay. The pathogens associated with NP were similar for both groups. This may be due, in part, to the prior hospitalization and use of antibiotics in many patients developing early-onset NP. These data suggest that P aeruginosa and ORSA can be important pathogens associated with early-onset NP in the ICU setting. Additionally, clinicians should be aware of the common microorganisms associated with both early-onset NP and late-onset NP in their hospitals in order to avoid the administration of inadequate antimicrobial treatment.

Lower respiratory tract infections in the intensive care unit: consequences of antibiotic resistance for choice of antibiotic

Pneumonia in the intensive care unit (ICU) has been associated with highly virulent pathogens that often exhibit resistance to multiple antibiotics and mortality rates of 30-70%. Pseudomonas aeruginosa and Enterobacteriaceae are the leading pathogens, followed by Staphylococcus aureus and polymicrobial etiologies. Recent clinical studies using monotherapy for nosocomial pneumonias resulted in low eradication rates for P. aeruginosa and staphylococci. An additional problem of these studies was the development of resistance by P. aeruginosa during the antibiotic treatment; also the selection of highly resistant strains like Xanthomonas maltophilia and Acinetobacter species was a major concern. However, several prospective studies comparing monotherapy versus combination therapy in nosocomial pneumonia of ICU patients have shown that a response rate of 60% is achievable, which is comparable to historic rates for combination therapy regimens. Only infections induced by P. aeruginosa, S. aureus, or other highly resistant pathogens (Acinetobacter, X. maltophilia, etc.) should be treated with well-defined antibiotic combinations.

Postoperative Pneumonia

Despite the advances made in surgical critical care, the diagnosis of one of the most common infections seen in critically ill patients remains a challenge. Ventilator-associated pneumonia is associated with a 20 to 25 per cent mortality rate. There are numerous risk factors for ventilator-associated pneumonia, including underlying disease, prolonged mechanical ventilation, direct lung injury, and shock. The standard clinical criteria for pneumonia are inaccurate. Quantitative cultures of bronchoalveolar lavage effluent are accurate for the diagnosis, and it is safe to base antibiotic therapy on the results of the quantitative cultures.

Prevention of infection in multiple trauma patients by high-dose intravenous immunoglobulins

OBJECTIVE

To investigate the activity of intravenous immunoglobulin (IVIG) as a prophylactic agent against infection in trauma victims.

DESIGN

Prospective, randomized, double-blind, placebo-controlled study.

SETTING

A 20-bed university intensive care unit.

PATIENTS

Thirty-nine trauma patients with injury severity scores (ISSs) of 16-50.

INTERVENTIONS

Penicillin was given at the time of admission and continued at least until day 4. Twenty-one patients received IVIG and 18 patients received human albumin at 1 g/kg in four divided doses (days 1, 2, 3, and 6). The two groups had similarities in age, gender, Acute Physiology and Chronic Health Evaluation II score, risk of death, and Glasgow Coma Scale score, but differing ISSs (p = .02), at the time of admission. Blood was collected on days 1, 4, and 7.

MEASUREMENTS AND MAIN RESULTS

Clinical variables related to infection were recorded. The complement components C3c, C4 and CH50, IgG, and the fractions of IgG were measured. The serum bactericidal activity (SBA) was assessed at 37 degrees C (98.6 degrees F) and 40 degrees C (104.0 degrees F) at the time of admission and during the course of IVIG administration. Controlling for ISS, IVIG-treated patients had fewer pneumonias (p = .003) and total non-catheter-related infections (p = .04). Catheter-related infections (p = .76), length of stay in the intensive care unit, antibiotic days, and infection-related mortality did not differ between the two groups. A significantly increased trend in IgG and its subclasses was shown on days 4 and 7 in the IVIG group but not in the control group (p<.000001). No important differences were noted in complement fractions. The SBA of the groups was similar on day 1, but significantly higher on days 4 and 7 (p<.000001) in the IVIG group, remaining so controlling for complement and ISS. SBA was higher at 40 degrees C (104.0 degrees F) compared with 37 degrees C (98.6 degrees F) (p<.0001) under all three conditions. In both groups, low SBA (on days 1, 4, and 7) was associated with increased risk of pneumonia (p<.01) and non-catheter-related infections (p = .06 for day 1; p<.01 for days 4 and 7).

CONCLUSIONS

Trauma patients receiving high doses of IVIG exhibit a reduction of septic complications and an improvement of SBA. Early SBA measurement may represent an index of susceptibility to infection.

Radiology of pneumonia

Chest radiography is the imaging technique of choice in evaluating patients with suspected pneumonia because of its low radiation dose, low cost, and wide accessibility. In daily practice, radiographs are used to confirm the clinical diagnosis of pneumonia, characterize the extent and severity of disease, search for complications such as empyema, monitor the response to therapy, and examine for possible alternative or additional diagnoses. Although CT scan has no defined role in the routine assessment of patients with either community-acquired or nosocomial pneumonias, its advantages of superior contrast resolution and cross-sectional display can often be helpful in the analysis of complex cases, particularly when radiographic evidence of associated central obstruction, cavitation, lymphadenopathy, or empyema is equivocal. In the immunocompromised patient population, high-resolution CT has been shown to be more sensitive than plain film radiography in the early detection of pulmonary infections.

Utility of Gram's stain and efficacy of quantitative cultures for posttraumatic pneumonia: a prospective study

OBJECTIVE

This prospective trial examined the efficacy of using bronchoalveolar lavage (BAL) for the diagnosis of pneumonia (PN) and the utility of Gram's stain (GS) for dictating empiric therapy.

SUMMARY BACKGROUND DATA

Posttraumatic nosocomial PN remains a significant cause of morbidity and mortality. However, its diagnosis is elusive, especially in multiply injured patients. The systemic inflammatory response syndrome of fever, leukocytosis, and a hyperdynamic state is common in trauma patients, especially patients with pulmonary contusion. Bronchoscopy with BAL with quantitative cultures of the lavage effluent may distinguish between PN and systemic inflammatory response syndrome, and GS of the lavage effluent may guide empiric therapy before quantitative culture results.

METHODS

Mechanically ventilated trauma patients with a clinical diagnosis of PN (fever, leukocytosis, purulent sputum, and new or changing infiltrate on chest radiograph) underwent bronchoscopy with BAL. Effluent was sent for GS and quantitative cultures. The diagnostic threshold for PN was > or =10(5) colony-forming units (CFU)/mL, and antibiotics were continued. Antibiotics were stopped for < 10(5) CFU/mL and the diagnosis of systemic inflammatory response syndrome was made. Causative organisms for PN were compared to GS.

RESULTS

Over a 2-year period, 232 patients underwent 443 bronchoscopies with BAL (71% men, 29% women; mean age, 41). The mean injury severity score was 30. Sixty percent of the patients had pulmonary contusion, and 59% were cigarette smokers. The overall incidence of PN was 39% and was no different regardless of the number of BALs a patient had. The false-negative rate of BAL was 7%. GS identified gram-positive organisms in 80% of patients with gram-positive PN and 40% of patients with gram-negative PN. GS identified gram-negative organisms in 52% of patients with gram-positive PN and 77% with gram-negative PN. The duration of the intensive care unit stay relative to the timing of BAL was beneficial for guiding empiric therapy. BAL in week 1 primarily identified Haemophilus influenzae and gram-positive organisms; Acinetobacter sp. and Pseudomonas sp. were more common after week 1.

CONCLUSIONS

Bronchoscopy with BAL is an effective method to diagnose PN and avoids prolonged, unnecessary antibiotic therapy. Empiric therapy should be adjusted to the duration of the intensive care unit stay because the causative bacteria flora changes over time. GS of BAL effluent correlates poorly with quantitative cultures and is not reliable for dictating empiric therapy.

Nosocomial pneumonia in the critical care unit

Nosocomial pneumonia poses a major threat to the recovery of patients receiving mechanical ventilation. In addition, nosocomial pneumonia is often difficult to diagnose. This article examines the extent of the threat and some of the difficulties facing the critical care physician when diagnosing nosocomial pneumonia.

Diagnosis of nosocomial pneumonia in mechanically ventilated patients: repeatability of the bronchoalveolar lavage

The repeatability of the bronchoalveolar lavage (BAL) was assessed prospectively in 44 mechanically ventilated patients with suspected nosocomial pneumonia. Two BAL were performed in the same lung area (contiguous segment) during two fibroscopic procedures performed with a thirty minute interval. All the bronchoscopies were performed by the same operator. The statistical analysis looked out for bias (MacNemar test), agreement, and repeatability (kappa test). In the 44 patients studied, the qualitative repeatability (i.e., presence or absence of bacteria) was excellent (95.4%). However, in the 16 patients having at least one positive culture, these results were more controversial. The quantitative repeatability for bacteria (same log10 for both BAL of the same patient) was the lowest of all the results (26.7%). The distinction between presence and absence of bacterial pneumonia (based on the 10[4] cfu/ml threshold) showed a repeatability of 75% with no bias, an agreement of 47% and a just-significant kappa test (test = 1.97; p = 1.96 for a 5% risk error). BAL seems to have excellent repeatability when sterile. Its repeatability when positive needs further studies to be assessed.

Nosocomial pneumonia. Diagnosis and prevention

Ventilator-associated pneumonia (VAP) is an important complication in patients with respiratory failure who undergo endotracheal intubation and mechanical ventilation. VAP cannot be accurately diagnosed by clinical or radiographic criteria or culture of endotracheal aspirates; however, it can be accurately diagnosed by histopathologic examination of lung tissue, rapid cavitation of a pulmonary infiltrate, culture of empyema fluid, percutaneous lung needle aspiration, simultaneous recovery of the same microorganism from cultures of respiratory secretions, and blood and quantitative culture of lower respiratory tract secretions obtained by bronchoscopy. VAP can be prevented by proper decontamination and use of ventilatory support equipment, practice of proper nursing techniques during care of the mechanically ventilated patient, and use of face mask ventilation in selected patients.

Rapid identification of Staphylococcus aureus in bronchoalveolar lavage fluid using a DNA probe (Accuprobe)

OBJECTIVE

Staphylococcus aureus is one of the prominent causative agents of ventilator-associated pneumonia (VAP). Gram staining of bronchoalveolar lavage (BAL) fluid is not always reliable. A nonisotopid probe (Accuprobe) has been developed by Gen-Probe for the specific identification of S. aureus isolated from cultures. This study was undertaken to assess the reliability of this probe for the early diagnosis of S. aureus VAP.

DESIGN

A prospective study in 120 consecutive patients.

SETTING

Department of intensive care medicine at a university hospital.

PATIENTS

120 ventilated patients (70 males and 50 females; mean age 52 +/- 12 years; mean simplified acute physiologic score = 13 +/- 4) were studied.

INTERVENTIONS

164 bronchoalveolar lavages were performed (none of the patients received prior antibiotic therapy).

MEASUREMENTS AND RESULTS

S. aureus was identified 29 times at significant concentrations (> or = 10(4) cfu/ml) and 7 times at < 10(4) cfu/ml. The sensitivity and specificity of the Accuprobe system were 100 and 96%, respectively. We found agreement between quantitative cultures and probes in 96.3% of cases.

CONCLUSIONS

We conclude that this probe provides a rapid (< or = 7 h) and accurate diagnosis of S. aureus pulmonary infection.

Comparison of quantitative cultures to semiquantitative loop cultures of bronchoscopic protected specimen brush samples

STUDY OBJECTIVE

To compare quantitative growth from a calibrated loop to growth from the standard serial dilution technique of culturing bronchoscopic protected specimen brush (PSB) samples and to determine the effect of refrigeration of the PSB sample on subsequent quantitative growth.

DESIGN

Laboratory stock cultures were sampled with a PSB and cultured by both standard 100-fold serial dilution as well as 1:100 mL and 1:1,000 mL calibrated loops. Stock cultures were also sampled with a PSB and growth before and after refrigeration for 24 h at 4 degrees C (both serial dilution and calibrated loops) was compared. Clinical PSB samples from seven patients suspected of having lower respiratory tract infections were cultured by both techniques as well.

SETTING

Clinical research laboratory and teaching hospital.

PATIENTS AND INTERVENTIONS

PSB samples from inpatients and outpatients who had clinically indicated bronchoscopy. No interventions.

MEASUREMENTS AND RESULTS

Quantitative growth from the 1:1,000 mL calibrated loop was within 1 log10 of growth from the serial dilution technique for 20 of 21 organisms, including 2 yeasts. Except for Haemophilus influenzae (known to be cold intolerant), there were no important differences in growth of bacteria before and after 24 h at 4 degrees C. For quantitative bacterial growth, there was a significant correlation between serial dilution and the 1:1,000 mL loop cultures (r=0.86, p < 0.0001).

CONCLUSION

In this study, quantitative growth from a single 1:1,000 mL loop culture of PSB samples was comparable to growth from the standard serial dilution technique. Our results also suggest that overnight refrigeration of PSB samples may be possible in certain clinical situations.

Prophylaxis for stress-related gastrointestinal hemorrhage: a cost effectiveness analysis

OBJECTIVE

To assess the cost-effectiveness of prophylaxis for stress-related gastrointestinal hemorrhage in patients admitted to the intensive care unit.

DESIGN

Decision model of the cost and efficacy of sucralfate and cimetidine, two commonly used drugs for prophylaxis of stress-related hemorrhage. Outcome estimates were based on data from published studies. Cost data were based on cost of medications and costs of treatment protocols at our institutions.

MEASUREMENTS AND MAIN RESULTS

The marginal cost-effectiveness of prophylaxis, as compare with no prophylaxis, was calculated separately for sucralfate and cimetidine and expressed as cost per bleeding episode averted. An incremental cost-effectiveness analysis was subsequently employed to compare the two agents. Sensitivity analyses of the effects of the major clinical outcomes on the cost per bleeding episode averted were performed. At the base-case assumptions of 6% risk of developing stress-related hemorrhage and 50% risk-reduction due to prophylaxis, the cost of sucralfate was $1,144 per bleeding episode averted. The cost per bleeding episode averted was highly dependent on the risk of hemorrhage and, to a lesser degree, on the efficacy of sucralfate prophylaxis, ranging from a cost per bleeding episode averted of $103,725 for low-risk patients to cost savings for very high-risk patients. The cost per bleeding episode averted increased significantly if the risk of nosocomial pneumonia was included in the analysis. The effect of pneumonia was greater for populations at low risk of hemorrhage. Assuming equal efficacy, the cost per bleeding episode averted of cimetidine was 6.5-fold greater than the cost per bleeding episode averted of sucralfate.

CONCLUSIONS

The cost of prophylaxis in patients at low risk of stress-related hemorrhage is substantial, and may be prohibitive. Further research is needed to identify patient populations that are at high risk of developing stress-related hemorrhage, and to determine whether prophylaxis increases the risk of nosocomial pneumonia.

Initiation of mechanical ventilation in the emergency department

Mechanical ventilation is frequently initiated by emergency physicians. Further, the physician on duty in the emergency department is frequently responsible for evaluating ventilated patients who decompensate in the intensive care unit when other physicians are not present in the hospital. A bewildering array of features on new mechanical ventilators has made their appropriate and effective use increasingly complex. Knowledge of the pathophysiology of acute respiratory failure and changes in lung physiology during positive pressure ventilation will aid the emergency physician in choosing an appropriate ventilator modality and initial settings to maximally benefit patients with respiratory insufficiency due to various causes. An appreciation of the adverse effects of mechanical ventilation and problems commonly encountered in patients on ventilators will prepare the emergency physician to rapidly assess and effectively manage the patient who deteriorates in this setting.

Using bronchoalveolar lavage to distinguish nosocomial pneumonia from systemic inflammatory response syndrome: a prospective analysis

OBJECTIVE

Ventilator-associated pneumonia (PN) is difficult to distinguish from trauma-induced systemic inflammatory response syndrome (SIRS), especially in patients with multiple injuries. Previous work using bronchoscopy and quantitative cultures demonstrated significant bacterial growth in about one-third of patients with clinical evidence of PN. In this prospective study, antibiotic therapy for PN was based solely on quantitative bronchoalveolar lavage (BAL) cultures.

METHODS

Mechanically ventilated trauma patients underwent bronchoscopy with BAL when they developed clinical evidence of PN: fever (temperature > 100.5 degrees F), white blood cells > 10,000 or > 10% immature forms, purulent sputum, and new or changing infiltrate on chest roentgenogram. Patients with other infections or those receiving antibiotics for any other reason were excluded. Empiric antibiotic therapy for PN was started at the time of BAL. If the quantitative cultures revealed > or = 10(5) colony-forming units (CFU)/mL, that patient was defined as having PN and was treated. If the cultures revealed < 10(5) CFU/mL, that patient was defined as having SIRS, and empiric therapy was stopped.

RESULTS

Forty-three patients (88% blunt, 12% penetrating) underwent bronchoscopy with BAL 55 times. Mean age was 40 and Injury Severity Score was 25. Twenty patients had > or = 10(5) CFU/mL (47%) and 23 had < 10(5) CFU/mL (53%). There were no differences in age, Injury Severity Score, temperature, white blood cell count, or ventilator days before BAL between groups. Sixty-five percent of those with SIRS improved after empiric therapy was stopped (average 3.3 days), and 35% underwent repeat BAL. Three patients with the initial diagnosis of SIRS developed PN (13% of SIRS). Mortality for PN was 15%, compared with 17% for SIRS; no deaths were related to antibiotic therapy.

CONCLUSIONS

SIRS, which can mimic PN, is common in trauma patients. These entities can be distinguished by bronchoscopy with BAL. Basing antibiotic therapy solely on quantitative BAL cultures is efficacious in trauma patients.

The effect of late-onset ventilator-associated pneumonia in determining patient mortality

STUDY OBJECTIVE

To determine whether the development of late-onset ventilator-associated pneumonia (VAP) is associated with an increased risk of hospital mortality.

DESIGN

Prospective cohort study.

SETTING

ICUs of two university-affiliated teaching hospitals.

PATIENTS

Three hundred fourteen patients admitted to an ICU who required mechanical ventilation for greater than 5 days.

INTERVENTIONS

Prospective patient surveillance and data collection.

MEASUREMENTS

The primary outcome measures were the development of late-onset VAP (ie, occurring > 96 h after intubation) and hospital mortality.

RESULTS

Late-onset VAP was observed in 87 patients (27.7%). Thirty-four (39.1%) patients with late-onset VAP died during hospitalization compared with 85 patients (37.4%) without late-onset VAP (relative risk, 1.04; 95% confidence interval [CI], 0.76 to 1.43). Twenty patients (6.4%) developed late-onset VAP due to a "high-risk" pathogen (ie, Pseudomonas aeruginosa, Acinetobacter sp, Xanthomonas maltophilia) with an associated mortality rate of 65%. Stepwise logistic regression analysis identified five variables as independent risk factors for hospital mortality (p < 0.05): an organ system failure index of 3 or greater (adjusted odds ratio [AOR], 3.4; 95% CI, 2.0 to 5.8; p < 0.001), having a nonsurgical diagnosis (AOR, 2.1; 95% CI, 1.3 to 3.6; p = 0.002), a premorbid lifestyle score of 2 or greater (AOR, 1.8; 95% CI, 1.1 to 2.9; p = 0.015), acquiring late-onset VAP due to a "high-risk" pathogen (AOR, 3.4; 95% CI, 1.2 to 10.0; p = 0.025), and having received antacids or histamine type-2 receptor antagonists (AOR, 1.7; 95% CI, 1.0 to 2.9; p = 0.034). Additionally, we found the occurrence of late-onset VAP due to high-risk pathogens to be the most important predictor of hospital mortality among patients developing VAP (AOR, 5.4; 95% CI, 2.8 to 10.3; p = 0.009).

CONCLUSIONS

Nosocomial pneumonia due to certain high-risk microorganisms is an independent risk factor for hospital mortality among patients requiring prolonged mechanical ventilation. We suggest that future investigations of late-onset VAP stratify patient outcomes according to the distribution of high-risk pathogens when reporting their results.

Elastin fibers and the diagnosis of bacterial pneumonia in the adult respiratory distress syndrome

OBJECTIVE

It has been inferred from previous work that 40% potassium hydroxide preparations of lower respiratory tract secretions that demonstrate elastin fibers have a 100% specificity and positive predictive value in diagnosing bacterial pneumonia in intubated, mechanically ventilated patients without the adult respiratory distress syndrome (ARDS). Our aim was to assess the specificity of 40% potassium hydroxide preparations in diagnosing bacterial pneumonia in patients with ARDS and suspected pneumonia.

DESIGN

Prospective, case-referral clinical study.

SETTING

Referral hospital.

PATIENTS

Of 24 patients with ARDS who were intubated and mechanically ventilated with suspected bacterial pneumonia, 22 were assessable and evaluated for this report.

INTERVENTIONS

Tracheo-bronchial aspirates were obtained from all patients and analyzed for elastin fibers using 40% potassium hydroxide.

MEASUREMENTS AND MAIN RESULTS

Of the 22 assessable patients, ten patients did not have a complicating bacterial pneumonia. Six of these ten patients had potassium hydroxide preparations that demonstrated elastin fibers (false positives). The other four patients had preparations that did not demonstrate elastin fibers (true negatives). Specificity was 40%.

CONCLUSION

Elastin fiber preparations are not specific for diagnosing bacterial pneumonia in patients with ARDS.

Bacteremic nosocomial pneumonia. A 7-year experience in one institution

STUDY OBJECT

To describe the epidemiology, microbiology, and outcome of nosocomial pneumonia with secondary bloodstream infection.

DESIGN

Prospective cohort study.

SETTING

Tertiary care Canadian teaching hospital.

PATIENTS

Inpatients.

MEASUREMENT

All inpatient blood cultures were concurrently monitored over an 89 month period. Following chart review, patients experiencing nosocomial bloodstream infection due to pneumonia were identified. A standardized definition of pneumonia was used.

RESULTS

One hundred forty-nine episodes occurred in 145 patients, 0.66/1,000 hospital admissions, 8.4% of all nosocomial bloodstream infections. No change in rate occurred in the study period. Fifty-four percent of episodes developed in one of seven ICUs. Staphylococcus aureus was the most frequently identified etiologic organism (27%). The ICU and non-ICU cases did not differ in etiology. No organism became more prevalent during the study period. Twenty percent of patients died within 1 week of first positive culture; episodes associated with Pseudomonas species had a much higher mortality rate (45%) than other infections (14%) (p = 0.002). The ICU and non-ICU infections had a similar mortality rate.

CONCLUSION

Pneumonia is an important cause of nosocomial bloodstream infection, but it is not increasing in frequency or changing in etiology in our institution. The ICUs are a major contributor to this problem but have the same case short-term mortality rate and microbial etiology as non-ICU cases. Cases associated with Pseudomonas have a much higher mortality rate.

Evaluation of the invasive techniques for diagnosing bacterial respiratory infections

Bacterial community-acquired respiratory infections are usually sustained by strains highly responsive to antibiotic therapy. Thus, the clinical approach is based on an empirical treatment and does not require the isolation of the causative pathogen and the determination of the bacterial susceptibility to antibiotics. On the other hand, Gram-negative bacteria, most commonly multidrug resistant, frequently affect immunocompromised and nosocomial patients and their identification in cultures is absolutely necessary for proper antibacterial treatment. To this aim, two conventional methods are used, i.e. the blood culture, which is positive only in 20% of pneumonia cases, and the sputum culture, which is not invasive but easily contaminated by oropharyngeal flora. Consequently, invasive techniques for sampling the pathologic specimen, such as the BAL and the PSB, performed with the help of fiberoptic bronchoscope, are needed. The diagnostic power and the limits of both these techniques are analyzed. Moreover, the opportunity to obtain quantitative cultures, which may discriminate between contamination and infection is considered.

A comparison of bronchoscopic vs blind protected specimen brush sampling in patients with suspected ventilator-associated pneumonia

BACKGROUND

Pneumonia is a common complication in patients undergoing mechanical ventilation and increases ICU mortality. The clinical diagnosis of ventilator-associated, however, pneumonia is unreliable, and many consider bronchoscopic-directed protected specimen brush sampling and quantitative culture the diagnostic method of choice. Bronchoscopy, however, is expensive and not readily available in many ICUs.

OBJECTIVE

To test the hypothesis that "blind" protected specimen brush (PSB) sampling may produce results similar to that of bronchoscopic-directed sampling.

SETTING

The medical ICU of a university-affiliated teaching hospital.

INTERVENTION

Patients with suspected ventilator-associated pneumonia (VAP) who had not received antibiotics for at least 48 h underwent "blind" and bronchoscopic-directed PSB sampling with quantitative culture.

RESULTS

Fifty-five paired PSB specimens were obtained from 53 patients. There was an 85% quantitative agreement between the blind and bronchoscopic-directed specimens. The agreement was independent of the bronchopulmonary segment from which the bronchoscopic sampling was directed.

CONCLUSION

The results of this study are consistent with the notion that blind PSB sampling and quantitative culture may prove to be a useful, cost-effective, and minimally invasive method of diagnosing VAP.

Significant reduction in methicillin-resistant Staphylococcus aureus ventilator-associated pneumonia associated with the institution of a prevention protocol

OBJECTIVE

To determine whether the institution of a methicillin-resistant Staphylococcus aureus prevention protocol was associated with a decrease in methicillin-resistant S. aureus ventilator-associated pneumonia in long-term, acute care ventilator patients.

DESIGN

A retrospective chart review comparing the number of episodes of clinical pneumonia per patient ventilator day in the 12 months preceding and 24 months following the introduction of the protocol.

SETTING

University affiliated, long-term, acute care ventilator hospital.

PATIENTS

Long-term, acute care ventilated patients who presented with clinical pneumonia.

INTERVENTIONS

Addition of a methicillin-resistant S. aureus prevention protocol. In addition to universal precautions, the protocol consisted of mupirocin 2% ointment applied to the anterior nares, and whole body washing with chlorhexidine. All patients were given mupirocin and chlorhexidine twice weekly. Patients were cohorted in the same room if they were, or had been, infected or colonized with methicillin-resistant S. aureus in any anatomical location or at any time. This procedure replaced strict isolation of methicillin-resistant S. aureus-infected or colonized individuals.

MEASUREMENTS AND MAIN RESULTS

Clinical pneumonia was diagnosed when a patient developed fever, bronchorrhea, increased white blood cell count, methicillin-resistant S. aureus isolated from the tracheal aspirate, and new or increasing infiltrate on chest roentgenograph. During the 12 months preceding the protocol, there were 0.2% episodes of methicillin-resistant S. aureus ventilator-associated pneumonia per ventilated patient day compared with 0.026% in the 24 months after the protocol (p < .001). The relative and absolute risk reductions associated with the introduction of the protocol were 87% and 6, respectively.

CONCLUSIONS

The period following the institution of the protocol showed a significant reduction in episodes of clinical pneumonia compared with the 12-month period preceding the use of the protocol (p < .001). Thus, we conclude that the introduction of this protocol is associated with a significant decrease in methicillin-resistant S. aureus ventilator-associated pneumonia.

Elastin fiber analysis in acute diffuse lung injury caused by smoke inhalation

OBJECTIVE

The evaluation of various techniques to diagnose or exclude ventilator-associated bacterial pneumonia has been a focus of much research. One such technique involves elastin fiber detection. It has been inferred from previous work that 40% potassium hydroxide preparations of respiratory secretions that demonstrate elastin fibers have a 100% specificity in diagnosing bacterial pneumonia in intubated, mechanically ventilated patients without acute diffuse lung injury. The purpose of this investigation was to ascertain if elastin fibers might be detected in respiratory secretions in acute, diffuse lung injury in the absence of pneumonia (i.e., assess specificity).

DESIGN

An animal model using a standardized smoke inhalation protocol to cause acute, diffuse lung injury was used.

MATERIALS AND METHODS

Respiratory secretions collected from the endotracheal tubes from eight sheep that underwent the standardized smoke inhalation protocol and were examined with 40% potassium hydroxide. Histologic data were obtained from autopsy to diagnose or exclude lung injury and pneumonia.

MEASUREMENT AND MAIN RESULTS

We found six (false) positive elastin fiber preparations in the absence of histologic pneumonia. Specificity was 0.25.

CONCLUSIONS

We concluded that seeing these results, given a true specificity of 0.99 inferred from previous work, is highly improbable with a probability of 2.74 x 10(-7). Thus, elastin fiber analysis is likely to be highly nonspecific for diagnosing pneumonia in the setting of acute diffuse lung injury.

Use of elastin fibre detection in the diagnosis of ventilator associated pneumonia

BACKGROUND

Elastin fibre detection could be a simple and reliable marker of ventilator associated pneumonia. To confirm this, a prospective study was undertaken to evaluate the diagnostic yield of elastin fibre detection in the diagnosis of ventilator associated pneumonia.

METHODS

Seventy eight mechanically ventilated patients were evaluated by examining endotracheal aspirates for the presence of elastin fibres. All patients were previously treated with antibiotics. Quantitative bacterial cultures of endotracheal aspirates and protected specimen brush samples were also performed. Patients were classified into three diagnostic categories: group 1, definite pneumonia (n = 25); group 2, probable pneumonia (n = 35); and group 3, controls (n = 18).

RESULTS

Patients with definite and probable pneumonia were grouped together. The presence of elastin fibres in endotracheal aspirate samples was more frequent in groups 1 and 2, being found in 19 of the 60 patients compared with five of the control group. Although the presence of elastin fibres had a low sensitivity (32%), it was a reasonably specific marker (72%) of pneumonia. This specificity increased to 86% and 81% respectively when only Gram negative bacilli and Pseudomonas aeruginosa pneumonia were considered. Again, calculated sensitivity was 43% and 44% when analysing cases infected by Gram negative bacilli and Ps aeruginosa, respectively. The negative predictive value of the detection of elastin fibres in pneumonia caused by Ps aeruginosa was 81%. Detection was more frequent with infection by Gram negative bacilli (14/19), particularly with Ps aeruginosa (8/14). By contrast, pneumonia due to Gram positive cocci or non-bacterial agents uncommonly resulted in positive elastin fibre preparations (4/19, 21%). When analysing patients with and without chronic obstructive pulmonary disease, the diagnostic value of elastin fibre detection did not change.

CONCLUSIONS

Potassium hydroxide preparation of elastin fibres is a rapid and simple specific marker of ventilator associated pneumonia and may be a useful technique to help diagnose pulmonary infections in mechanically ventilated patients, although this assessment is at present limited to patients without adult respiratory distress syndrome.