key: cord-0006779-limy65sf
authors: Nieto, J. M. Sanchez; Alcaraz, A. Carillo
title: The role of bronchoalveolar lavage in the diagnosis of bacterial pneumonia
date: 1995
journal: Eur J Clin Microbiol Infect Dis
DOI: 10.1007/bf01691489
sha: 06b002d20b2a9096f4b2961f00d0603b78fd79f9
doc_id: 6779
cord_uid: limy65sf

Bronchoalveolar lavage (BAL) has become an invaluable diagnostic tool with important clinical implications in both opportunistic infections and the pulmonary pathology of immunologic disease. Until recently, the use of BAL was limited primarily to two areas: the study of interstitial lung diseases and the diagnosis of lung infections by opportunistic microorganisms in severely immunocompromised patients with lung infiltrates. Over the past decade, the use of BAL has been expanded to include the conventional diagnosis of bacterial pneumonia in non-immunocompromised patients. In the past, different clinical studies proposed using BAL to quantify cultures in the sample obtained as a means of increasing the tool's effectiveness. Recent developments have led to a number of newer applications of BAL, such as bronchoscopic BAL, non-bronchoscopic BAL and protected BAL. The most important use of BAL in the non-immunocompromised patient is the diagnosis of pneumonia in the mechanically ventilated patient.

J.M. Sanchez Nieto 1., A. Carillo Alcaraz 2 Bronchoalveolar lavage (BAL) has become an invaluable diagnostic tool with important clinical implications in both opportunistic infections and the pulmonary pathology of immunologic disease. Until recently, the use of BAL was limited primarily to two areas: the study of interstitial lung diseases and the diagnosis of lung infections by opportunistic microorganisms in severely immunocompromised patients with lung infiltrates. Over the past decade, the use of BAL has been expanded to include the conventional diagnosis of bacterial pneumonia in non-immunocompromised patients. In the past, different clinical studies proposed using BAL to quantify cultures in the sample obtained as a means of increasing the tool's effectiveness. Recent developments have led to a number of newer applications of BAL, such as bronchoscopic BAL, non-bronchoscopic BAL and protected BAL. The most important use of BAL in the non-immunocompromised patient is the diagnosis of pneumonia in the mechanically ventilated patient.

Bronchoalveolar lavage (BAL), first introduced in the 1970s and used initially as an experimental procedure to study the cytologic and humoral components present in the alveolar surface, has become an invaluable diagnostic tool with important clinical implications in both opportunistic infections and the immunologic pulmonary pathologies. Over the past decade, the use of BAL has been expanded to include the conventional diagnosis of bacterial pneumonia in nonimmunocompromised patients. This review briefly summarises the main findings on the use of BAL in the diagnosis of bacterial pneumonia and describes several new, recently developed uses for BAL.

The use of BAL allows us to explore in vivo the epithelial surface of the lung under normal and pathologic conditions, something which cannot be achieved with other internal organs. The alveolar surface through a lavage precedes the appearance of fiberoptic bronchoscopy at the beginning of this century. Jackson (1) introduced a series of modifications to the rigid bronchoscope, creating an inhalation channel that permitted bronchial lavages to be used. For 20 years bronchial lavage was used for therapeutic ends, particularly in patients with bronchiectasis (2) . Later, newer designs were developed, which brought better results from the lavage (3) . With the appearance of M6tras's catheter, it was possible to enlarge the area in which the lavage was used, since it allowed segmentary bronchi to be channelled (4) . This catheter enabled the first studies of the immunologic function of alveolar macrophages, which were carried out in human volunteers (5) (6) (7) (8) (9) . As a result of the introduction of fiberoptic bronchoscopy in the 1960s by Ikeda et al. (10) , various experiments were performed using healthy volunteers who were also smokers (11) . The findings of Reynolds and Newball (12) served as the basis for numerous studies that analyzed the different cells and substances involved in the inflammatory damage and immunopathogenesis of many lung diseases (13) (14) (15) (16) (17) (18) (19) .

Bronchoalveolar lavage became popular immediately, mainly due to its easy and repeatable use, its safeness and its ability to explore a large portion of lung tissue. Its value as an exploratory tool for the lung is largely derived from the fact that the fluid obtained by means of BAL reproduces exactly the inflammatory changes present in the lung tissue (20) . Two international symposiums have approved and recommended the use of BAL in clinical practice (21, 22) .

Bronchoalveolar lavage is performed by injecting and re-aspirating sterile saline solution through a suction channel in the fiberoptic bronchoscope, placed in a subsegmentary bronchi. The amount of liquid used for the diagnosis of infectious lung diseases has not been standardised. The BAL Cooperative Group Steering Committee has recommended using 240 ml for evaluation of patients with interstitial lung disease (23) . In studies carried out using BAL to obtain bacterial cultures, including from patients under mechanical ventilation, the amount of liquid used varied between 100 and 240 ml (24) . According to some studies, at least 120 ml is needed to obtain secretions from the most distal portions of lung subsegments (25) . A 5 ml sample of fluid is the minimum recommended for correct microscopic and microbiological processing (26) . The effect of dilution on the results of bacterial cultures has not been studied.

Until a few years ago, the use of BAL was limited primarily to two main areas: the study of interstitial lung diseases and the diagnosis of lung infections by opportunistic microorganisms or by obligate pathogens, where the contamination of fluid by the oropharynx flora does not represent a diagnostic problem (27) . In severely immunocompromised patients with lung infiltrates, BAL is considered a principal diagnostic tool (28) . In animal studies BAL was as sensitive as lung biopsy in the detection of Pneumocystis carinii (29) . Bronchoalveolar lavage is also the most advanced diagnostic tool when used in AIDS patients and in those suspected to be infected with Pneumocystis carinii (30) . The use of BAL for the diagnosis of bacterial pneumonia, however, has been questioned, since it is believed that samples obtained by BAL will be invariably contaminated by oropharynx bacteria after their passage through the channel of the fiberoptic bronchoscope in the same way that bronchoscopy samples are contaminated, as shown by the work of Bartlett et al. immunocompromised patients with infiltrated lungs were studied along with 18 controls by means of fiberoptic bronchoscopy with BAL. The samples taken were processed quantitatively. In the 18 controls, none of whom had evidence of respiratory infection, the presence of more than 1% squamous epithelial cells suggested contamination of the sample by oropharynx flora. Of the 13 patients with bacterial pneumonia confirmed by other methods, each had less than 1% squamous cells and at least one microorganism that was isolated at a concentration of 105 cfu/ml. One organism cultured from one BAL at a concentration of 105 cfu/ml and the presence of i % squamous cells resulted in a sensitivity of 88 % and a specificity of 100 %. However, contamination of BAR fluid occurred in 26 % of the cases.

In a more recent study, Kirkpatrick and Bass (34) showed that the BAL fluid could be contaminated by the bacterial oropharynx flora. Samples for quantitative culture were obtained from eight healthy volunteers using a protected specimen brush (PSB), in accordance with the Wimberly technique (35), and BAL. Cultures of samples obtained by BAL were positive in seven of eight cases, while only one sample obtained by PSB was positive. Although the concentration of microorganisms was not significant (< 105 cfu/ml), the study showed that contamination by the upper airway is common in BAL cultures; the authors attributed the contamination in part to the introduction of lidocaine into the fiberoptic broncho-scope channel, which had been done prior to the use of BAL.

The majority of studies investigating the use of BAL in bacterial pneumonia were carried out in mechanically ventilated patients. Table 1 shows the most representative studies of this type. Until very recently, it was widely accepted that the PSB described by Wimberly et al. (35, 36) was the best way to obtain microorganisms for identification in patients with pneumonia associated with mechanical ventilation and to differentiate between colonisation of the main respiratory tract and distal lung infection. The initial studies by Higuchi et al. (37) and Chastre et al. (38) , which correlated PSB cultures with the histological findings and quantitative cultures of lung tissue, stimulated clinical and experimental research in this area.

Since then, numerous studies have proven the diagnostic efficiency of the PSB in pneumonia associated with mechanical ventilation (39) (40) (41) (42) (43) (44) . In a total of 18 studies evaluating the PSB in 524 patients with mechanical ventilation, the overall sensitivity of the tool was 90 %, and the specificity was 94 % (45). The use of the PSB, however, is not without disadvantages. The percentage of false-positive and false-negative results varies between 10 and 30% (46) , particularly in patients with underlying pathologies that favour colonisation of the distal airway (47, 48) and those with mechanical ventilation and prolonged antibiotic therapy (38, 41, 44) . The small volume of secretions obtained using the PSB also contributes to these rates (37, 49) . In addition, the results of culture tend to be prolonged (requiring > 24 h), and fewer cases of polymicrobial pneumonia are detected (50) . Finally, complications of PSB use, such as pneumothorax and endobronchial haemorrhage, have been described (44) . In four recent studies, the efficiency of BAL versus PSB was compared in animals (51) and in patients on mechanical ventilation (52) (53) (54) . Despite good results achieved by BAL when used in non-intubated patients, its use in mechanically ventilated patients has been more controversial, mainly due to the heterogeneous populations that have been studied and the different methods used.

In the study by Johanson et al. (51) , carried out in 35 monkeys with prolonged mechanical ventilation, cultures of endotracheal aspirates, BAL fluid, PSB samples and needle aspirates were compared with the histological findings of lung biopsy. The 'bacterial charge' of the lung was expressed as the 'bacterial index' (calculated by the sum of the logarithmic concentration of individual bacterial species). BAL recovered 74 % of all species present in the lung tissue, compared with 41% obtained by PSB and 56 % by needle aspiration. The specificity of BAL, however, was slightly inferior to that of the PSB: 8 % of the samples obtained using BAL in animals without pneumonia had significant growth. The authors of the study concluded that the quantitative analysis of samples obtained by BAL reflected the microbiological spectrum of the lung, and thus BAL was judged the most efficient of the three tools studied. BAL was the most sensitive of all three tools, although the specificity of the three methods was similar.

Chastre et al. (52) were the first to compare BAL and the PSB in mechanically ventilated humans. In a group of 21 patients suspected to have pneumonia, all of whom had received antibiotic treatment for more than ten days without the use of the bacterial index or the establishment of different cut-off points, those patients with pneumonia were readily differentiated from those without pneumonia. These authors' study concluded that Torres et al. (54) also compared the diagnostic value of quantitative cultures of both PSB specimens and BAL fluid from 25 mechanically ventilated patients with nosocomial pneumonia of short evolution who had also been treated with empiric antibiotics for fewer than 12 hours. The BAL sample was processed by different microbiological methods than those used by other authors (32, 33, 51, 52) . The cut-off point for cultures of both PSB specimens and BAL fluid was established at 104 cfu/ml. The diagnostic correlation between both tools was excellent. When both tools were combined, sensitivity was 84 %. The sensitivity and specificity of BAL were 59 % and 71%, respectively, while values for the PSB were 59 % and 86 %. However, although the sensitivity of both tools was identical, the amount of organisms collected was different. In only 14 of 25 patients was there concordance between the type and amount of organisms by both tools. Lastly, Guerra and Baughman (55) studied the benefits of BAL in 60 patients with mechanical ventilation, obtaining a sensitivity of 60 %. Although organism counts in BAL cultures from a group of control patients did not exceed 104 cfu/ml, this study lacked a viable reference test and did not allow an exaggeratedly high specificity in comparison with studies done by other authors. Table   1 shows the more significant results achieved using BAL in ventilator-associated pneumonia.

The specificity of BAL cultures in diagnosing bacterial pneumonia is limited in part by the contamination by colonising bacteria of the upper airway. Contaminants can be found in significant quantities in up to one-third of the samples obtained (33, 52, 54) . Recommendations for avoiding contamination include avoiding suction by the fiberoptic bronchoscope channel before performing BAL, placing the patient in the Trendelenburg position and discarding the first aliquot from the fluid obtained (23) . These problems have stimulated the search for new solutions in improving BAL's efficiency while retaining its maximum benefits in diagnostics.

Initially, the use of BAL without the introduction of a fiberoptic bronchoscope in the distal airway was intended for the diagnosis of infection by

Pneumocystis carinii in AIDS patients (56, 57) .

Other authors (58-60) later applied non-bronchoscopic BAL (NB-BAL) in patients with ventilator-associated pneumonia. Pugin et al. (60) compared the quantitative cultures obtained by bronchoscopic BAL and non-bronchoscopic BAL in 28 patients with prolonged mechanical ventilation. The results showed a strong correlation between both BAL tools for the type of microorganism isolated. The predominant species recovered was the same in 93 % of the cases, and there were no significant differences in either bacterial index or cellularity. The authors concluded that the good sensitivity and specificity obtained with NB-BAL lies in the fact that the bacterial spectrum of nosocomial pneumonia in mechanically ventilated patients is similar in all lung lobules and in both lungs, even when a more radiographically localised infiltrate exists.

Despite this study and others such as that by Rouby et al. (61) , who defined ventilatorassociated pneumonia as a diffuse and multifocal process with varying grades of severity, limited data suggest the usefulness of this technique, which is less expensive than fiberoptic bronchoscopy, in clinically unstable patients, particularly when an experienced bronchoscopist is not available. On the other hand, in the event of a local pulmonary process in the upper lung lobules, it is preferable to obtain samples by fiberoptic bronchoscopy. Table 2 .

Meduri et al. (63) have described a use of the P-BAL in which a double-light, protected, balloontipped catheter is employed. Placement of this catheter requires the aid of a fiberoptic bronchoscope. Furthermore, due to its thickness, the catheter needs a work channel of 2.6 mm. Once the fiberoptic bronchoscope is installed in the zone chosen for P-BAL, the distal balloon must be inflated with 1.5 to 2 ml of air, and the distal diaphragm that seals the whole system must be expelled using 3 ml of sterile saline solution. Protected BAL is performed through a catheter with 5 aliquots of 30 ml of saline solution. The control group comprised 18 patients with neither Pneumonia nor mechanical ventilation. The study group included 28 patients with suspected pneumonia, most of whom were intubated and mechanically ventilated. Ninety-one percent of the samples obtained by P-BAL contained < 1% squamous epithelial cells, and 59 % of the samples from patients without pneumonia showed no growth. Establishing a cut-off point of 10 4 cfu/ml, 1 of every 33 patients without pneumonia had a false-positive result, and 1 of every 13 patients with pneumonia had a false-negative result (positive predictive value and sensitivity, 97 %; negative predictive value, 92 %). Microscopic analysis using Gram stain had a sensitivity and specificity of 92 % and 97 %, respectively. Fluid was not recovered from 2 of the 49 patients included in the study.

Following the development of P-BAL, a curious variation of P-BAL was proposed by Castella et al. (64) . These authors used a protected specimen brush, such as that designed by Wimberly et al. (35) , in a heterogenous group of 38 patients. The brush was extracted from the catheter, and in the proximal extreme an intramuscular needle was inserted, through which two aliquots of 20 ml of saline solution were introduced. Three endoscopic tools were introduced in sequence: classic or conventional BAL, P-BAL and bronchial smear by means of a PSB. The sensitivity of P-BAL (95 %) was greater than that of the PSB (55 %), while specificity for both tools was similar (94 and 89 %, respectively). The specificity obtained using the classic BAL was far inferior to the values obtained with the other two techniques (42 %).

Recently, we described another variation of BAL, which we refer to as 'protected alveolar lavage' (PAL) (65) . For this, we used a Combicath catheter with an external diameter of 1.7 mm, such as that used by Rouby et al. (62) , which we introduced through the work channel of a standard fiberoptic bronchoscope. Once wedged, the fiberoptic bronchoscope is placed into a subsegmented bronchi, in accordance with the radiographic location of the infiltrate and/or the direct observation of endobronchial secretions. Next, a Combicath catheter is introduced, and once the distal plug is expelled, the internal catheter is moved inside approximately 3 cm beyond the end of the external catheter. Eight 5 ml aliquots of isotonic sterile saline are then instilled, and suctioning is performed manually after every instillation. Fifty percent of the fluid obtained is processed for cytologic study and the other 50 % for bacteriologic study, including quantitative culture. We studied 20 patients suspected to have pneumonia and compared the bacteriological results obtained with the results of two consecutive uses of PSB and PAL, using a cut-off point of 103 cfu/ml and 104 cfu/ml, respectively. Table 3 .

Recently, Rouby et al. (61) described meticulously the histologic and bacteriologic aspects of pneumonia in patients with mechanical ventilation. At the same time, they evaluated the diagnostic benefit of a protected mini-BAL technique carried out 'blindly', or nonbronchoscopically (non-B-mini-P-BAL), in a clinical model using a unilateral post-mortem lung test. Eighty-three critically ill patients on mechanical ventilation were studied. In 69 of these patients, non-B-mini-P-BAL was performed shortly before death. The good results obtained with this method, known as 'blind lavage', were attributed to two aspects. First, the catheters are introduced by means of an artificial airway without bronchoscopic help and descend spontaneously to the lower lung lobules. Second, the lesions occurring in nosocomial bronchopneumonia are usually distributed in the lower lobules. In ten patients with positive lung biopsy cultures and without evidence of alveolar infection, histologic evidence of bronchiolitis was found. Lesions were associated with bacterial concentrations of 105 cfu/g. These authors con-cluded that a cut-off point of 104 cfu/g provides an inaccurate indication of the rate of active bronchopulmonary infection in patients receiving antibiotics. False-negative results obtained with the mini-BAL were not linked to pneumonia 'sterilised' by antibiotic treatment, and false-positive results often were not related to bronchiolitis. The sensitivity and specificity of the non-B-mini-P-BAL was 70 %. Until a few years ago, three tools had been widely used for the diagnosis of conventional bacterial pneumonia: analysis of sputum or endotracheal aspirates (43, (66) (67) (68) , analysis of samples obtained by percutaneous ultrathin needle aspiration, perfected by Zavala and Schoell (69) and Dorca et al. (70) , and examination of bronchial smears obtained by the PSB, described by Wimberly et al. (35, 36) . The first method is inaccurate, with high sensitivity and low specificity (43, 66) . In recent years a number of studies have evaluated the efficacy of quantitative bacterial culture in intubated patients with suspected pneumonia. An increase in the efficacy of this method was shown using a cut-off value of 106 cfu/ml (67) or 105 cfu/ml (68) . Further studies are indicated in order to validate this diagnostic tool.

The second method offers excellent specificity, approaching 100 %, but is less sensitive than other invasive tools (71) . In addition, most authors reject its use in mechanically ventilated patients due to the risk of pneumothorax. Only two studies of this method have been performed in humans on mechanical ventilation (66, 72) . The last tool mentioned, examination of bronchial smears, also presents the problem of false-negative results, which, for some authors, renders it unacceptable, especially in critically ill patients (73, 74) . These problems have stimulated research for alternative diagnostic tools such as BAL. Bronchoalveolar lavage achieves a greater and more representative volume of secretions from the lower respiratory tract than the other tools mentioned. Its role in the detection of opportunistic and obligated pathogens in immunosuppressed patients has been substantially established, but its value in the rapid identification and diagnosis of bacterial pneumonia is still under investigation. Figure 1 shows the main diagnostic indications for the variations of BAL in patients suspected to have pneumonia.

The employment of culture quantification techniques has improved the specificity of the BAL in the diagnosis of bacterial lung infection (23, 33) .

Nevertheless, it is difficult to reach a consensus on when to fix a cut-off point that establishes a safe diagnosis of infection, above all in patients receiving antibiotics. In these patients a decrease in the bacterial count relative to the pre-established cutoff point may indicate either an absence of pneumonia or a diminishing of the bacterial charge due to the administered antibiotic. On the other hand, an increase may represent active pneumonia due to resistant bacteria or cotonisation of the lower airway without lung tissue infection.

The difference in the cut-off point of bacterial growth, established by various authors as 103 to 105 cfu/ml, depends mainly on the type of bacteriologic processing and on three variables (32, 34, 52, 54) : the concentration of pathogenic microorganisms in the lung area reached by BAL, the dilution attained with the lavage liquid and the final amount recovered (24) . The dilution of lung secretions in the BAL fluid varies from 10 to 100 times; therefore, a count of 104 colonies is equivalent to 105 or 106 bacteria per millilitre, the figure considered indicative of lung infection (75, 76) . However, in the presence of antibiotic treatment, it is impossible to attain 100 % sensitivity with BAL and other invasive tools based on the identification of bacteria in distal respiratory samples, since some lungs with histologic indications of pneumonia in patients receiving antibiotic treatment are 'sterile' (61) . In patients without mechanical ventilation, the manner of administering the topic anaesthesia also influences the sensitivity of BAL. On one hand, the anaesthetic inoculates the bronchial tree against the microorganisms that accumulate in the channel of the fiberoptic bronchoscope, and on the other hand, lidocaine has bacteriostatic properties. However, the concentrations of the anaesthetic upon dilution by the BAL fluid do not seem sufficient to inhibit bacterial growth (34, 77, 78) .

The specificity of BAL in excluding bacterial pneumonia depends on whether colonising bacteria of the oropharynx and tracheobronchial tree are recovered. Despite measures taken to diminish contamination of the interior of the fiberoptic bronchoscope channel, most authors are inclined to look for technical solutions that minimise the risk of contamination of the BAL fluid. The protected BAL was designed to increase specificity over that of conventional BAL through the use of special catheters that impede contact between colonising bacteria and the instilled liquid (62) (63) (64) (65) . In our work the term 'protected alveolar lavage' (65) seems most adequate to differentiate protected lavage from classic or non-protected BAL. The introduction of catheters distally, to the end of the fiberoptic bronchoscope, washes or clearly shows the small airway and the alveolus. It is clear that the first results obtained using the various types of protected lavages improved the specificity over that of other methods. Until recently, the use of the PSB was considered optimal in mechanically ventilated patients with bacterial pneumonia (64 The cytologic processing of BAL fluid, used systematically, does not seem to offer diagnostic advantages. Different studies have shown an increase in the percentage of neutrophils in patients with active pneumonia (59, 60, 65), but these neutrophils are not diagnostic and can be associated with pulmonary processes that are inflammatory but non-infectious (80) . In those patients with pneumonia suspected to be caused by a virus or opportunistic microorganisms or presumed to be due to non-infectious pathology such as alveolar haemorrhage, it is suggested that a cytologic test with special staining be performed (63) (64) (65) .

Other research tests, such as detection of pneumonia markers, have been described, including the visualisation of elastin fibres in tracheal aspirates (81) , applied more recently to BAL analysis (60) . Although the presence of elastin fibres is very specific in the existence of necrotising bacterial pneumonia, sensitivity is low and false-positive results may occur in patients with adult respiratory distress syndrome (82) . Figure 2 summarises the main applications of BAL and PAL for the diagnosis of pneumonia.

Most studies have reported rates of false-negative results of 30 to 40 % with BAL (83, 84) . Several factors account for this. First, there is the lack of standardisation of sampling techniques used by different investigators. Second, the sensitivity of BAL will be reduced if the patient is receiving antibiotics at the time of sampling. It is possible that the frequency with which false-negative findings occur depends not so much on the sampling technique as on whether and in how many doses appropriate antibiotics were administered prior to specimen collection and whether the pathogens concerned are sensitive to ongoing therapy. Likewise, false-positive results are caused by a number of factors (83, 84) . First, upper airway organisms can contaminate the BAL fluid during the procedure. The P-BAL was designed to avoid the contamination of this fluid. Second, prior antibiotic therapy can lead to overgrowth of antibiotic-resistant organisms, causing a false-positive result (86) . Finally, false-positive results can occur in the setting of heavy colonisation of the airway. In one recent study in ventilated patients without pneumonia, Rodriguez de Castro et al. (87) found a significant relationship between the ability of the PSB and/or BAL to recover microorganisms and the duration of mechanical ventilation.

Bronchoalveolar lavage is a tool that presents very few complications. In fact, most of the undesirable effects are related to the use of fiberoptic bronchoscopy. It is rare that severe complications, such as pneumothorax, arise (88) . Hypoxemia and arrhythmias can occur during use of the fiberoptic bronchoscope, but this is usually foreseen by an increase in the FIO2 (89) (90) (91) (92) . Arterial hypoxemia is more pronounced and lasting in patients with pneumonia than in those without lung infection, possibly due to the alveoli being filled by the instilled liquid (93) . Recently, it has been proposed that BAL procedures, bronchoscopic or other, could be responsible for a translocation of endotoxins from the infected or colonised airways to the interior of the lung capillaries or lymphatic channels through an alveoli capillary wall damaged by the infection. This passage ofendotoxins or bacterial products in circulation could be responsible for a systemic sepsis-like response, with fever and hypertension after the use of BAL (93) . The type and severity of these changes have yet to be clarified. The use of variations of BAL requiring a smaller amount of saline solution and/or the use of lavages without bronchoscopy should diminish the frequency and intensity of these adverse effects, although these hypotheses must still be confirmed.

Important advances in the use of BAL in the diagnosis of bacterial pneumonia have been made, but important questions also have yet to be answered. The first international conference on the clinical investigation of pneumonia associated with mechanical ventilation (26) recommended the use of the quantitative cultures of BAL specimens and established a cut-off point of 104 cfu/ml.

It was also recognised that contradictory results will continue to be reported, especially when treatments such as antibiotics and prolonged mechanical ventilation co-exist. The mini-protected BAL and the non-bronchoscopic BAL are other important diagnostic options, above all in pneumonia linked to mechanical ventilation. As these methods become further refined, they may even replace the PSB or conventional BAL, methods previously considered state of the art. It remains to be seen which of the variations of protected lavage will become of greatest diagnostic value and under which clinical circumstances.

It is hoped that there will be a prompt standardisation of the type of material used and of the technical procedures to be followed. It is still not known whether the systemic application of these and other invasive tools improve the prognosis of patients with nosocomial pneumonia. Although the use of BAL in the diagnosis of bacterial pneumonia in immunocompromised patients is still under research, the introduction of new variations such as the mini-protected-BAL has contributed to an increasing number of indications for bronchoalveolar lavage.

Bronchoscopy: past, present and future

Bronchial lavage for disinfection and immunization of the bronchial tree

A new flexible double-lumen catheter for bronchospirometry

Le cath6risme bronchique. Vigot-Freres

Human alveolar macrophages: comparison of phagocytic, glucose utilization, and ultrastructure in smokers and nonsmokers

The human alveolar macrophage: isolation, cultivation in vitro and studies of morphologic functional characteristics

Alveolar macrophages: structural and functional differences between nonsmokers and smokers of marijuana and tobacco

Low yield of pulmonary surfactant in cigarette smokers

Bronchopulmonary lavage in man

Flexible bronchofiberscope

Induction of aryl hydrocarbon hydroxylase in human pulmonary alveolar macrophages by cigarette smoking

Analysis of proteins and respiratory cells obtained from human lungs by bronchial lavage

Immunocompetent cells from the lower respiratory tract of normal human lungs

Normal human bronchial immunoglobulin and proteins: effects of cigarette smoking. American Review of Respiratory Disease

Biochemical analyses of bronchoalveolar lavage fluids of normal healthy volunteers

Immunoglobulin secreting cells in normal human bronchial lavage

Inflammatory and immune processes in the human lung in health and disease: evaluation by bronchoalveolar lavage

Free secretory component and other proteins in human lung lavage. American Review of Respiratory Disease

Reaginic antibody in the lung lining fluid: analysis of normal human bronchoalveolar lavage IgE and comparison to immunoglobulins G and A

Bronchoalveolar lavage in pulmonary fibrosis: comparison of cells obtained with lung biopsy and clinical features

Le lavage broncho-alv6olaire chez I'homme-colloque

Summary of the proceedings of the 1984 International Conference on Bronchoalveolar Lavage

Group Steering Committee: Bronchoalveolar constituents in healthy individuals, idiopathic pulmonary fibrosis, and selected comparison groups

The role of bronchoalveolar lavage in diagnosing non-opportunistic bacterial pneumonia

Anatomical distribution of bronchoalveolar lavage fluid as assessed by digital subtraction radiography

The standarization of bronchoscopic techniques for ventilator-associated pneumonia

State of the art. Bronchoalveolar lavage. American Review of Respiratory Disease

Bronchoalveolar lavage in the diagnosis of diffuse pulmonary infiltrates in the immunosuppressed

Diagnosis of pneumonia due to pneumocystis by subsegmental pulmonary lavage via the fiberoptic bronchoscope

Diagnosis of pulmonary disease in acquired immune deficiency syndrome (AIDS)

Should fiberoptic bronchoscopy aspirates be cultured? American Review of Respiratory Disease

Bronchoalveolar lavage for diagnosing acute bacterial pneumonia

Diagnosing bacterial respiratory infection by bronchoalveolar lavage

Quantitative bacterial cultures of bronchoalveolar lavage fluids and protected brush catheter specimens from normal subjects

A fiberoptic bronchoscopy technique to obtain uncontaminated lower airway secretion for bacterial culture

Use of a bronchoscopic protected brush for the diagnosis of pulmonary infections

Bacteriologic diagnosis of nosocomial pneumonia in primates

Prospective evaluation of the protected specimen brush for the diagnosis of pulmonary infections in ventilated patients. American Review of Respiratory Disease

Reliability of the bronchoscopy protected catheter brush in intubated patients

Protected transbronchial needle aspiration and protected specimen brush in the diagnosis of pneumonia

Detection of nosocomial lung infections in ventilated patients. Use of protected specimen brush and quantitative culture techniques in 147 patients

Agusti Vidal A: Diagnostic value of telescoping plugged catheter in mechanically ventilated patients with bacterial pneumonia using the Metras catheter

Comparison of tracheal aspirates and protected brush catheter specimens for identifying pathogenic bacteria in mechanically ventilated patients

An&lisis de la informaci6n bacteri-ol6gica aportada mediante cat~ter telescopado de doble luz y oclusi6n distal en el diagn6stico de infiltrados pulmonares en pacientes bajo ventilaci6n mec~.nica. Revista Clinica EspaSola

Evaluation of the protected brush catheter and bronchoalveolar lavage in the diagnosis of pneumonia

Evaluation of the available invasive and non-invasive techniques for diagnosing nosocomial pneumonias in mechanically ventilated patients

Diagnosis of bacterial pulmonary infections with quantitative protected catheter cultures obtained during bronchoscopy

Pulmonary infiltrates in immunocompromised patients. Diagnostic value of telescoping plugged catheter and bronchoalveolar lavage

Diagnosis of nosocomial pneumonia in intensive care unit patients

Nosocomial pneumonia in patients receiving continues mechanical ventilation. Prospective analysis of 52 episodes with use of a protected specimen brush and quantitative culture techniques. American Review of Respiratory Disease

Bacteriologic diagnosis of nosocomial pneumonia following prolonged mechanical ventilation

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

Quantification of BAL cells containing intracellular bacteria rapidly identifies ventilated patients with nosocomial pneumonia

Agusti Vidal A: Diagnostic value of quantitative culture of bronchoalveolar lavage and telescoping plugged catheter in mechanically ventilated patients with bacterial pneumonia

Use of bronchoalveolar lavage to diagnose bacterial pneumonia in mechanically ventilated patients

Nonbronchoscopic bronchoalveolar lavage for the diagnosis of Pneumocystis carinii pneumonia in the AIDS

Nonbronchoscopic lung lavage for diagnosis of opportunistic infection in AIDS

Diagnostic value of nonbronchoscopic bronchoalveolar lavage during mechanical ventilation

Comparison of nonbronchoscopic bronchoalveolar lavage to open lung biopsy for the bacteriologic diagnosis of pulmonary infections in mechanically ventilated patients

Diagnosis of ventilator associated pneumonia by bacteriologic analysis of bronchoscopic and nonbronchoscopic "blind" bronchoalveolar lavage fluid

Nosocomial bronchopneumonia in the critically ill. Histologic and bacteriologic aspects

A prospective study of protected bronchoalveolar lavage in the diagnosis of nosocomial pneumonia

Protected bronchoalveolar lavage. A new bronchoscopic technique to retrieve uncontaminated distal airway secretions

Lavado broncoalveolar protegido: nuevo m6todo endosc6pico en el diagn6stico de las neumoNas bacterianas

Estudio comparativo del lavado alveolar protegido frente al cat~ter telesc6pico ocluido en pacientes con sospecha de neumonia yen ventilaci6n mec~.nica

Etiologic diagnosis of bacterial nosocomial pneumonia in seriously ill patients

Diagnostic efficiency of endotracheal aspirates with quantitative bacterial cultures in intubated patients with suspected pneumonia. Comparison with the protected specimen brush

Rodriguez-Roisin R: Validation of different techniques for the diagnosis of ventilator-associated pneumonia. Comparison with immediate postmortem pulmonary biopsy

Ultrathin needle aspiration of the lung in infections and malignant disease

Efficacy, safety, and therapeutic relevance of transthoracic aspiration with ultrathin needle in nonventilated nosocomial pneumonia

Tecnicas invasivas en el diagn6stico de las neumonfas. Archives de Bronconeumologia

Diagnostic value of nonfluoroscopic percutaneous lung needle aspiration in patients with pneumonia

New advances in diagnosing nosocomial pneumonia in intubated patients

Lamer CH: Procedures for the diagnosis of pneumonia in ICU patients

Bacteriology of expectorated sputum with quantitative culture and wash technique compared to transtracheal aspirates

Diagnostic accuracy of transtracheal aspiration bacteriologic studies

Role of bronchoalveolar lavage in the diagnosis of pneumonia

Safety and efficacy of using high-dose topical and nebulized anesthesia to obtain endobronchial cultures

The effect of antibiotic therapy on recovery of intracellular bacteria from bronchoalveolar lavage in suspected ventilator-associated nosocomial pneumonia

Fibroproliferative phase of ARDS. Clinical findings and effects of corticosteroids

Diagnosis of nosocomial pneumonia in intubated intensive care unit patients

Diagnosis of pneumonia in patients with adult respiratory distress syndrome

Invasive diagnostic testing is not needed routinely to manage suspected ventilator-associated pneumonia

Invasive diagnostic testing should be routinely used to manage ventilated patients with suspected pneumonia

Clinical significance of borderline quantitative protected brush specimen culture results

Rodriguez-Roisin R: Specificity of endotracheal aspiration, protected specimen brush and bronchoalveolar lavage in mechanically ventilated patients. American Review of Respiratory Disease

Quantitative cultures of protected brush specimens and bronchoalveolar lavage in ventilated patients without suspected pneumonia

Bronchoalveolar lavage induced pneumothorax

The effect of fiberoptic bronchoscopy on cardiac rhythm

Arrhythmias from fiberoptic bronchoscopy

Management of arterial hypoxemia induced by fiberoptic bronchoscopy

The effect of bronchofiberoptic examination on oxygenation status

Diagnostic bronchoalveolar lavage in patients with pneumonia produces sepsis like systemic effects