Use of Bronchoalveolar Lavage Semiquantitative Cultures in Cystic Fibrosis

Abstract

Section:

To assess bronchoalveolar lavage (BAL) in adult CF patients with respiratory symptoms, we studied BAL fluid (BALF) culture results from 28 bronchoscopies in 11 patients. Patients were asked to provide sputum for culture. All but two patients were receiving antibiotics at the time of bronchoscopy, with 13 bronchoscopies done on patients who had been receiving antibiotics for more than 10 d. Gram stain of the BALF was positive in 18 cases. In all but one BALF, > 10,000 colony-forming units per milliliter (cfu/ml) BALF of one or more pathogens was identified. The final case grew Burkholderia cepacia, which was not grown in the sputum. In only six cases (21%) were the sputum and BALF culture results the same. Prior to 11 bronchoscopies, the sputum was not adequate. The remaining 11 cases either had different pathogens in the BAL (six cases), or had some but not all of the BALF pathogens in the sputum. BALF cultures changed therapy in 13 (48%) of cases. Semiquantitative culture of BALF was a useful diagnostic tool in CF in patients in whom empiric therapy failed.

Cystic fibrosis (CF) is a genetic disease that is associated with significant pulmonary morbidity. The cause of the morbidity is multifactorial, including basic underlying defects of the chloride channel. However, the clinical manifestations of the disease are most evidenced by patients' inability to handle infections. This is especially apparent when patients develop pulmonary infection with Pseudomonas aeruginosa. A major aspect of the management of CF is management of the pulmonary infections that occur in the disease (1, 2).

Proper identification of the organisms causing pulmonary infections, and determination of their sensitivity to antimicrobial agents, is part of the basic management of patients with CF. Identification of organisms in sputum, using either induced or spontaneously produced samples, has been the standard method for assessing respiratory infections in CF. In children or adults unable to produce sputum, throat swabs may be obtained (3, 4). The identification of Pseudomonas in sputum or in a throat swab is felt to be diagnostic of lower respiratory infection with Pseudomonas, and antibiotic sensitivities of the organism are determined. Patients are treated on the basis of these sensitivities (3, 5). Other pathogens, such as Staphylococcus aureus, have shown discordance between the throat and lower-respiratory cultures (4, 6).

In the case of other pulmonary infections, the utility of sputum for diagnosing the cause of lower respiratory tract infections has been recognized as limited (7-9). In patients receiving antibiotics, there is a high rate of colonization of the upper respiratory tract that may lead to the overgrowth of one or another pathogen, making it more abundant in the upper than in the lower respiratory tract. Also, the patient may be unable to provide adequate lower-respiratory-tract samples. For these reasons, more invasive procedures have been evaluated. These include bronchoscopy with the protected-brush sampling (10, 11) or semiquantitative bronchoalveolar lavage (BAL) sampling (12, 13). The latter is a technique that has been used for many years in immunosuppressed patients and patients requiring mechanical ventilation. We and others have shown that the value of BAL in identifying lower-respiratory-tract infections relies on the use of a semiquantitative culture technique. In these two patient populations, cutoff values of 10,000 to 100,000 colony-forming units per milliliter (cfu/ml) of BAL sample have allowed specific diagnosis of lower-respiratory-tract infections (12, 13). The results of semiquantitative culture techniques are affected by previously administered antibiotics. For patients receiving antibiotics at the time of BAL, the overall diagnostic yield is much lower for identifying a pathogen in the lower respiratory tract.

In the care of adults with CF, it has become clear that even though sputum production is part of the disease, not all patients are able to provide an adequate sputum sample for culture (1, 5). In addition, some patients do not seem to respond to antibiotic choices made on the basis of their sputum cultures. The widespread use of antibiotics in the treatment of CF patients has led to their improved survival. However, the use of empiric antibiotic therapy eventually leads to the selection of resistant organisms. This is usually appreciated clinically when an empiric trial of antibiotics fails.

The purpose of the current study was to determine the utility of bronchoscopy and semiquantitative BAL in trying to identify pathogens in the lower respiratory tract of patients with CF. Using previously established techniques of semiquantitative culture of BAL fluid (BALF) samples, we found that most patients had more than 10,000 cfu of bacteria/ml BALF. This was despite the use of antibiotics at the time of bronchoscopy. In many cases, the organism identified in the BALF was not found in a concurrent sputum sample.

METHODS

Section:

Adults with CF undergoing bronchoscopy were eligible for the study. During a 20-mo period of (November 1993 to July 1995), all patients with known CF followed at the CF Clinic at the University of Cincinnati Medical Center or Children's Hospital Medical Center in Cincinnati were considered for bronchoscopy for one or more of the following reasons: (1) Diagnostic reasons. This would include patients who either had no adequate sputum or sputum that grew only normal pharyngeal flora, and who were failing to respond to antibiotics. (2) Patients with apparently adequate sputum but who were failing to respond to therapy in terms of fever, chest roentgenographic findings, or dyspnea, and who were also considered to have experienced therapeutic failure. (3) Therapeutic reasons, for persistent atelectasis and mucous plugging, and therefore to enhance pulmonary toilet. (4) As part of an initial evaluation for subsequent gene therapy (two patients). In both of these cases the patients had completed 2 wk of intravenous antibiotic therapy but still had cough, sputum production, and symptoms. Both were stable enough to undergo subsequent gene therapy. All patients gave informed consent for bronchoscopy. A control group of 11 healthy subjects (10 nonsmokers) underwent bronchoscopy under a research protocol. The protocols used in the study were approved by the Human Research Committee at the University of Cincinnati Medical Center.

Patients' maximal temperature, chest roentgenographic findings, and antibiotics were noted at the time of bronchoscopy. Sputum for culture and sensitivity was obtained within the 48 h prior to bronchoscopy. Every effort was made to obtain an adequate sputum specimen; however, many patients were unable to provide a sample.

Bronchoscopy was performed in a standard fashion. The technique of BAL for assessment followed a protocol established by one of us (R.P.B.) in 1985, and has been used for all immunocompromised patients undergoing diagnostic bronchoscopy at University Hospital (12). During the time of the study, the same technique for lavage and the same semiquantitative method for bacterial culture were used on patients at Children's Hospital. Prior to bronchoscopy, all patients were sedated with a combination of intravenous narcotics and midazolam. Topical anesthesia was achieved with 2% lidocaine in the nose and upper respiratory tract. Topical lidocaine was also applied as anesthesia in the lower respiratory tract. Lavages were performed in the areas of disease in all patients. Most patients had diffuse disease, and lavage was performed in the upper lobes, lingula, or middle lobe. The bronchoscope was advanced to the most distant airway in the subsegment leading to the area of interest. A 20-ml aliquot of normal saline was instilled, aspirated, and held separately. Next, 120 ml of normal saline were instilled, immediately aspirated with a hand-held syringe, and pooled. An aliquot of this fluid was sent for semiquantitative bacterial cultures done according to previously described techniques (12). Using a quantitative loop, we plated 0.01 ml of the BALF directly onto each of four culture media: sheep blood agar, MacConkey's agar, blood agar containing colistin and nalidixic acid (CNA agar), and chocolate blood agar. In addition, 0.01 ml of BALF from all but one bronchoscopy was plated onto selective media for growing B. cepacia. The one exceptional bronchoscopy was performed when selective media were not available in our laboratory. Another BALF aliquot was used for a total cell count and differential cell count with slides prepared with a cytocentrifuge as previously described (15). An additional slide was prepared for Gram stain and was interpreted according to standard microbiologic techniques. Gram-stained smears were considered positive if at least one organism was seen per ×1,000 oil- immersion field (12).

Patients were observed in the postoperative bronchoscopy area for at least 2 h. Antibiotic therapy was continued in most patients, and the antibiotics for some were changed as a result of the bronchoscopy. Patients were followed up clinically for at least 2 wk after bronchoscopy. Chest roentgenograms were done after bronchoscopy in those patients who had persistent pulmonary symptoms of cough, fever, or chills, or new symptoms of cough, fever, or chills.

RESULTS

Section:

Eleven patients underwent 28 bronchoscopies with BAL. Seven were male, and the median age was 23 yr, with a range from 18 to 41 yr. Patients 1 and 2 underwent gene therapy and had baseline FEV₁ values of 45% and 50% predicted, respectively. All of the remaining patients had severe obstructive disease, with the most recent pulmonary function studies showing an FEV₁ below 40% predicted. Other details of the patients at the time of bronchoscopy are summarized in Table 1. Many of the patients were afebrile at the time of bronchoscopy. All patients had chronic changes on their chest roentgenogram. However, five of the patients had no acute infiltrates noted. Antibiotics are also noted in Table 1. The number of days of antipseudomonal antibiotics prior to bronchoscopy were noted, and we elected to see whether prolonged use of antibiotics would affect results. Prolonged use of antibiotics was considered as more than 10 d of antibiotic therapy. Two patients were receiving no prior antibiotics, 13 lavages were done after 1 to 9 d, and 13 lavages were performed after 10 or more days of antibiotics. The most common organism identified was Pseudomonas aeruginosa (Table 2). Also identified were B. cepacia, and Stenotrophomonas maltophilia (formerly Pseudomonas xanthomonas), and in one case each, Streptococcus pneumoniae and Branhamella catarrhalis. All but one patient had > 10,000 organisms identified. This one patient initially had only 1–10,000 hemolytic streptococci identified. However B. cepacia was subsequently detected in this case. This was the only BALF sample not specifically cultured for B. cepacia which was identified in mycobacterial and fungal cultures. The patient in this case had had no positive sputum cultures of B. cepacia for the preceding 2 yr, but subsequent bronchoscopy and sputum samples were positive for B. cepacia.

Table 1. CLINICAL CHARACTERISTICS AT TIME OF BRONCHOSCOPY
Procedure No.	Patient No.	Age/Sex	Chronic X-ray Changes	Acute X-ray Changes	Temperature	Indication for Bronchoscopy^*	Segment Lavaged	Antibiotics^*
1	1	23/Male	Diffuse	None	98.6	Gene therapy protocol	RLL	Ceftazidime/Tobramycin
2	2	27/Male	Upper	None	98.6	Gene therapy protocol	RLL	Ceftazidime/Tobramycin
3	2	27/Male	Upper	Local	100.4	Gene therapy protocol	RLL	None
4	2	27/Male	Upper	None	98	Gene therapy protocol	RLL	Ceftazidime/Tobramycin
5	3	39/Female	Upper	None	98	Infection	RUL	Aztreonam/Tobramycin
6	4	39/Male	Diffuse	Local	100.6	Infection	RML	Ceftazidime/Tobramycin
7	5	24/Male	Diffuse	Local	98.6	Collapse^*	RML	Ceftazidime/Tobramycin
8	5	24/Male	Diffuse	Local	98.4	Collapse^*	RUL	Ciprofloxacin/Tobramycin
9	5	24/Male	Diffuse	Diffuse	101.8	Infection	RUL	Ciprofloxacin/Tobramycin
10	5	24/Male	Diffuse	Local	101.2	Infection	RUL	Ceftazidime/Tobramycin
11	5	24/Male	Diffuse	Local	98.6	Collapse	RUL	Imipenem/Tobramycin
12	5	24/Male	Diffuse	Local	100	Infection	RUL	Imipenem/Tobramycin
13	5	24/Male	Diffuse	Local	98.6	Collapse^*	RUL	Aztreonam/Amikacin
14	5	24/Male	Diffuse	Local	98.6	Collapse^*	RUL	Ciprofloxacin/Aztreonam/Amikacin
15	5	24/Male	Diffuse	Local	98	Collapse^*	RUL	Aztreonam/Amikacin
16	5	24/Male	Diffuse	Local	98	Infection	RUL	Ciprofloxacin/Aztreonam/Amikacin
17	5	24/Male	Diffuse	Local	100.6	Infection	RUL	Ciprofloxacin/Tobramycin
18	6	21/Male	Diffuse	Local	102	Infection	RML	Ceftazidime/Tobramycin
19	7	23/Male	Upper	Diffuse	100.2	Infection	RML	Ceftazidime/Tobramycin
20	7	23/Male	Upper	Local	98.6	Collapse^*	RUL	Ciprofloxacin/Aztreonam
21	7	23/Male	Upper	None	101	Infection	RUL	Ciprofloxacin/Ticarcillin/Tobramycin
22	7	23/Male	Upper	Diffuse	100	Infection	RML	Aztreonam/Tobramycin
23	7	23/Male	Upper	Local	98.6	Collapse^*	RUL	Doxycycline Aztreonam/Tobramycin
24	8	26/Female	Upper	Local	98	Infection	RUL	Ciprofloxacin
25	9	34/Female	Upper	Local	101	Infection	RUL	Ceftazidime/Tobramycin
26	9	34/Female	Upper	Local	98.6	Infection	RUL	Pipercillin-tazobactam/Tobramycin
27	10	12/Female	Diffuse	Local	101	Infection	RUL	Ceftazidime/Tobramycin
28	11	35/Male	Upper	Local	98	Infection	RUL	Ciprofloxacin

^*Collapse = segmental collapse on chest roentgenogram.

Table 2. BRONCHOALVEOLAR LAVAGE SEMIQUANTITATIVE BACTERIAL CULTURE RESULTS^*
Culture Results	Pseudomonas aeruginosa	Burkholderia cepacia	Alcaligenes	Streptococcus pneumoniae	Moraxella catarrhalis
> 100,000 cfu/ml BAL
Two organisms	7
One organism	11	1	2	1
10–100,000 cfu/ml BAL
Two oranisms	3
One organism		1			1

^*All but one patient had at least one organism > 10,000 cfu/ml BAL.

Gram stains of BAL specimens were done in 27 of the 28 patients who had BAL. The Gram stains were positive in 20 of the 27 cases. Differential cell counts were also performed on the BAL samples. The differential counts showed > 75% neutrophils in all but one sample. In that case, there were 33% neutrophils. The normal result in our laboratory is < 5% neutrophils in BAL samples (15). Thus, examination of the cellular component of the BAL samples revealed that they were always purulent, and usually showed a positive Gram stain for microorganisms.

We compared the results of BAL examination with the duration of therapy prior to bronchoscopy (Table 3). There was no significant difference in the diagnostic yield for a pathogen for patients who had recently been started on antibiotic therapy and those who had completed at least 10 d of treatment. The culture results were not affected by the duration of therapy. There was also no significant effect of duration of therapy on the Gram-stain results.

Table 3. COMPARISON OF RESULTS VERSUS NUMBER OF DAYS OF ANTIBIOTICS PRIOR TO UNDERGOING BRONCHOSCOPY
	No Therapy	< 10 d	> 10 d
Number of studies	2	13	13
Highest culture results:
> 100,000 cfu/ml BAL	2	8	10
10 to 100,000 cfu/ml BAL	0	4	3
< 10,000 cfu/ml BAL	0	1	0
Gram stain:
Positive	1	8	9
Negative	0	5	4
Not done	1	0	0

A comparison was made between sputum results and BAL results. Table 4 summarizes the individual bronchoscopy results compared with sputum results, if available. In only six (21%) instances were the sputum-culture and BAL results identical. For 11 bronchoscopies, no adequate sputum sample was obtained within 48 h of bronchoscopy. In four instances, the sputum grew only normal pharyngeal flora. For seven bronchoscopies, some but not all of the BAL pathogens were grown in the sputum.

Table 4. COMPARISON OF SPUTUM VERSUS BAL RESULTS
Procedure No.	Sputum Results	BAL Results^*	Sputum versus BAL Culture
1	None	Pseudomonas aeruginosa	Only BAL positive
2	Pseudomonas aeruginosa	Pseudomonas aeruginosa (2)^†	Partial
3	Pseudomonas aeruginosa	Pseudomonas aeruginosa (2)	Partial
4	Pseudomonas aeruginosa	Pseudomonas aeruginosa (2)	Partial
5	Normal pharyngeal flora	Pseudomonas xanthomonas	Different
6	Normal pharyngeal flora	Pseudomonas aeruginosa	Different
7	Pseudomonas aeruginosa	Pseudomonas aeruginosa (2)	Partial
8	Pseudomonas aeruginosa	Pseudomonas aeruginosa	Same
9	None	Pseudomonas aeruginosa (2)	Only BAL positive
10	Pseudomonas aeruginosa (2)	Pseudomonas aeruginosa (2)	Same
11	None	Pseudomonas aeruginosa (2)	Only BAL positive
12	Pseudomonas aeruginosa (2)	Pseudomonas aeruginosa (2)	Same
13	Pseudomonas aeruginosa (1)	Pseudomonas aeruginosa (2)	Partial
14	None	Pseudomonas aeruginosa (2)	Only BAL positive
15	Pseudomonas aeruginosa (2)	Pseudomonas aeruginosa (2)	Same
16	Pseudomonas aeruginosa (1)	Pseudomonas aeruginosa (1)	Same
17	Pseudomonas aeruginosa (1)	Streptococcus pneumoniae, Pseudomonas aeruginosa (1)	Partial
18	Pseudomonas aeruginosa (1)	Pseudomonas aeruginosa (1)	Same
19	None	Pseudomonas aeruginosa (2)	Only BAL positive
20	Pseudomonas aeruginosa (2)	Pseudomonas aeruginosa (2)^§	Partial
21	None	Pseudomonas aeruginosa (1)	Only BAL positive
22	None	Pseudomonas aeruginosa (1)	Only BAL positive
23	None	Pseudomonas aeruginosa (2)	Only BAL positive
24	None	Branhamella catarrhalis	Only BAL positive
25	Normal pharyngeal flora	No organism > 10,000 cfu^¶	Partial
26	Normal pharyngeal flora	Burkholderia cepacia	Partial
27	None	Alcaligenes	Only BAL positive
28	None	Burkholderia cepacia (2)	Only BAL positive

^* Results of BAL at > 10,000 cfu/ml BALF.

^†Organism (number of isolates).

^§Different sensitivity patterns from sputum isolates.

^¶Mycobacterial culture grew B. cepacia.

Since patients were undergoing bronchoscopy for clinical failure of current antibiotics, we examined the effect of BAL results on therapy. Overall, 13 of 28 (46%) bronchoscopies led to a change in therapy (Table 5). In six cases a new pathogen was identified in the BAL sample, and in seven cases antipseudomonal therapy was changed as a result of sensitivity testing of the organisms recovered. The remaining 15 lavage results did not lead to an alteration in therapy. The reason for not changing therapy was that the organism identified in the BAL sample was sensitive to the antimicrobial therapy that the patient was receiving prior to bronchoscopy. It is interesting to note that patients bronchoscoped in the first 10 d after the beginning of antibiotic therapy were more likely to have a therapeutic change based on BAL findings (nine of 15 instances) than were those bronchoscoped after a longer duration of therapy (four of 13 cases), although this difference was not statistically significant. We do not know whether patients improved more rapidly because of antibiotic changes.

Table 5. CHANGE IN THERAPY AS RESULT OF BAL, EFFECT OF ANTIBIOTIC THERAPY PRIOR TO BRONCHOSCOPY
	No Therapy	< 10 d	> 10 d
Number of studies	2	13	13
Added new anti-Pseudomonas	1	2	4
Discovered new pathogen	1	5	0
No change in therapy	0	6	9

One patient had a marked worsening of her chest roentgenogram within 5 d after bronchoscopy. Figure 1 shows chest roentgenograms done prior to bronchoscopy in this case, and a repeat study done 5 d later. The initial film shows marked upper-lobe disease, which is characteristic for CF. The patient's roentgenogram done 5 d later demonstrated a large abscess with an air–fluid level. This was in the area of a previous bulla in the right upper lobe. Its position was confirmed by subsequent, serial chest roentgenograms. As the lesion grew smaller, it was clear that it was above the minor fissure. The abscess was presumably due to the lavage having been performed in the anterior segment of the right upper lobe. The patient was subsequently proved to have an Alcaligenes organism. Her antibiotics were changed, which was followed by gradual clinical improvement, and by 4 wk, her chest roentgenogram had returned to its baseline appearance.

Fig. 1. Chest roentgenogram prior to bronchoscopy (A) and 5 d after bronchoscopy (B) in a patient who developed a lung abscess in the right upper lobe. The causative agent was eventually cultured from the BALF and the patient responded to a new antibiotic regimen with subsequent complete resolution of the abscess.

Download Figure (Part 1) | Download Figure (Part 2) | Download Powerpoint

[More] [Minimize]

DISCUSSION

Section:

The proper identification of a pulmonary pathogen is a central tenet of the treatment of pulmonary infection. In the case of lower respiratory infection, the identification of pathogens has been the subject of multiple studies (10, 12, 16, 17), and sputum analysis has been shown to be inadequate for a significant percentage of cases. The use of invasive techniques has led to better diagnosis. The current study examines patients with advanced CF who continued to have symptoms despite empiric antipseudomonal therapy. In this select group, bronchoscopy with lavage added new information and changed therapy in half of the cases.

Two studies have compared the use of a protected catheter brush and BAL to immediate autopsy findings in ventilated patients (18, 19). These studies demonstrated that by using a semiquantitative technique, one could accurately identify which patients had pneumonia versus other causes of infiltrate. In addition, the studies showed a good correlation between the organism identified by the bronchoscopic technique that subsequently cultured from lung tissue.

In the nonventilated patient, semiquantitative cultures of BALF have been shown to have significant predictive value for pneumonia (12, 13, 20). Several studies have compared the BALF culture results for patients with pneumonia with those of other patients who underwent bronchoscopy but had no clinical evidence of pneumonia. We previously reported findings in a group of 92 unselected patients undergoing bronchoscopy, which included 15 patients with acute bacterial pneumonia (12). In that study we found that 13 of 15 patients with pneumonia had bacteria at > 100,000 cfu/ml. Of 19 patients who had bacteria at > 10,000 cfu/ml, 15 were felt to have acute bacterial pneumonia (12). In another prospective study of 357 lavages in 225 patients, Cantral and colleagues found similar sensitivity and specificity for semiquantitative culture results. They did note that some normal volunteers had > 10,000 cfu/ ml of normal oral flora (e.g., alpha-hemolytic streptococci), which they considered a contaminant from the upper airways. Cantral and colleagues suggested that identification of pulmonary pathogens at > 10,000 cfu/ml BALF should be considered diagnostic for lower-respiratory-tract pneumonia (20). However, these studies did not involve patients with CF.

Pulmonary infections in patients with CF are recognized as a major cause of morbidity and mortality (1). Adult patients often have chronic infection with P. aeruginosa, although other bacteria can be identified (21, 22). The use of sputum to identify organisms as well as to choose appropriate antibiotic therapy has been part of standard therapy in CF. In the present study, we identified new pathogens in the BALF samples from many of our patients, and the BALF findings changed therapy in nearly 50% of the patients. BAL has been used to identify new pathogens in many other conditions, such as human immunodeficiency virus (HIV) infection and patients following transplantation (17, 23). In some patients, empiric therapy correctly anticipated the results of BAL. In several instances, a new pathogen or sensitivity pattern was identified and the patient's therapy was changed. In two patients there was some delay in correctly identifying the pathogen. In one patient, whose roentgenograms are shown in Figure 1, there was a 5-d delay in correctly identifying Alcaligenes in the BAL sample, a problem also noted by others (24). During this time, the patient developed worsening symptoms. Once appropriate antibiotics were started, she did improve. The other patient's BAL sample was initially not specifically tested for B. cepacia. There was therefore a delay of several weeks in recognizing its presence. The patient remained colonized with B. cepacia over the next 6 mo, with positive sputum and bronchoscopy specimens. Both of these cases point to the need for the microbiology laboratory to be aware that a BAL sample is from a patient with CF in order to maximize the utility of this technique.

CF patients are usually treated with antibiotics for increased cough, fever, worsening findings in pulmonary function studies, weight loss, or a combination of these factors. All but two of our bronchoscopies were done in symptomatic patients. Even the two patients who were studied at the end of a 2-wk period of antibiotic treatment prior to gene therapy had significant sputum production and abnormal chest roentgenograms. Therefore, all of the CF patients in our study appeared to have significant lower respiratory infection, and the finding of bacteria in the lower respiratory tract should not be considered benign colonization.

Konstan and associates (25) used bronchoscopy and BAL to study CF patients with minimal symptoms. They found significant colonization with bacteria, usually P. aeruginosa (25). In addition, both Konstan and colleagues and others have found increased neutrophil numbers and other cytokine responses in cases of bacterial pulmonary infection (25-27). The finding of pathogens at more than 10,000 cfu/ml BAL was associated with inflammation, although not necessarily pneumonia.

We were initially concerned that previous treatment with antibiotics would affect the yield of semiquantitative culture results with BALF. This is a significant problem in other groups of patients, such as immunocompromised patients and those with ventilator-associated pneumonia (14). In these groups, semiquantitative culture results are often below the 10,000 cfu/ml cutoff if the patient has been receiving antibiotics for more than 12 h. Sputum studies done with semiquantitative techniques have demonstrated a large burden of organism in patients with CF (22). Although therapy reduces the number of pathogens found in sputum, the concentration is often greater than 100,000 cfu/ml. In our study, we found that duration of antimicrobial therapy had little effect on the previously established cutoff for semiquantitative culture results. We did not quantitate for higher concentrations of organisms, and it is probable that bacteria are much more concentrated in the BALF and that therapy does reduce the absolute concentration. The purpose of this study was to assess the 10,000 cfu/ ml cutoff for identifying potential pathogens. Examining individual results demonstrates how BAL may be useful in identifying a new pathogen. Bronchoscopy 17 was performed in a patient who had undergone extensive therapy and was receiving ciprofloxacin and tobramycin on the basis of previous sputum findings. The patient underwent bronchoscopy because of a worsening state. He was found to have S. pneumoniae at a significant concentration in his BALF. The addition of antibiotics directed at this pathogen led to resolution of the patient's fever, and his symptoms returned to their baseline levels.

We conclude that the use of BAL to identify potential pathogens in the lower respiratory tract of adult CF patients is usually well tolerated and provides useful information. The patients in the present study represent a subgroup of those treated at our center. BAL was useful in patients who were unable to provide an adequate sputum specimen or in whom symptoms persisted despite apparently adequate therapy. Proper processing of the BAL sample is important, including the use of semiquantitative culture techniques and culturing for all potential pathogens, including B. cepacia. The results of BAL led to a change of therapy in nearly half of the patients in the present study.

Supported in part by Grant 51832 from the National Institutes of Health-Heart and Lung.

References

Section:

1.	Konstan, M. W., and M. Berger. 1993. Infection and inflammation of the lung in cystic fibrosis. In P. B. Davis, editor. Cystic Fibrosis. Marcel Dekker, New York. 219–276. Google Scholar
2.	Wood R. E., Boat T. F., Doershuk C. F.State of the art: cystic fibrosis. Am. Rev. Respir. Dis.1131976833878 Abstract, Medline, Google Scholar
3.	Ramsey B. W., Wentz K. R., Smith A. L., Richardson M., Williams-Warren J., Hedges D. L., Gibson R., Redding G. J., Lent K., Harris K.Predictive value of oropharyngeal cultures for identifying lower airway bacteria in cystic fibrosis patients. Am. Rev. Respir. Dis.1441991331337 Abstract, Medline, Google Scholar
4.	Hoiby N.Microbiology of lung infections in cystic fibrosis patients. Acta Paediatr. Scand.301(Suppl.)19823354 Crossref, Google Scholar
5.	Thomassen M. J., Klinger J. D., Badger S. J., van Heekeren D. W., Stern R. C.Cultures of thoracotomy specimens confirm usefulness of sputum cultures in cystic fibrosis. J. Pediatr.1041984352356 Crossref, Medline, Google Scholar
6.	Konstan M. W., Hilliard K. A.Comparison of throat with bronchoalveolar lavage cultures in determining lower airway bacterial colonization in cystic fibrosis. Pediatr. Pulmonol.6(Suppl.)1991281 Google Scholar
7.	Meduri G. U.Diagnosis and differential diagnosis of ventilator-associated pneumonia. Clin. Chest Med.1619956193 Crossref, Medline, Google Scholar
8.	Hayner C. E., Baughman R. P.Nosocomial pneumonia: a review of diagnostic approaches. Infect. Med.121995322330 Google Scholar
9.	Barrett-Conner E.The non-value of sputum culture in the diagnosis of pneumococcal pneumonia. Am. Rev. Respir. Dis.1031971843848 Google Scholar
10.	Wimberley N., Faling L. J., Bartlett J. G.A fiberoptic bronchoscopic technique to obtain uncontaminated lower airway secretions for bacterial culture. Am. Rev. Respir. Dis.1191980337343 Google Scholar
11.	Baughman R. P., Thorpe J. E., Staneck J., Rashkin M., Frame P. T.Use of protected specimen brush in patients with endotracheal or tracheostomy tubes. Chest911987233235 Crossref, Medline, Google Scholar
12.	Thorpe J. E., Baughman R. P., Frame P. T., Wesseler T. A., Staneck J. L.Bronchoalveolar lavage for diagnosing acute bacterial pneumonia. J. Infect. Dis.1551987855861 Crossref, Medline, Google Scholar
13.	Kahn F. W., Jones J. M.Diagnosing bacterial respiratory infection by bronchoalveolar lavage. J. Infect. Dis.1551987862869 Crossref, Medline, Google Scholar
14.	Meduri G. U., Chastre J.The standardization of bronchoscopic techniques for ventilator associated pneumonia. Chest1021992557S564S Crossref, Medline, Google Scholar
15.	Baughman R. P., Strohofer S., Kim C. K.Variation of differential cell counts of bronchoalveolar lavage fluid. Arch. Pathol. Lab. Med.1101986341343 Medline, Google Scholar
16.	Fagon J. Y., Chastre J., Hance A. J., Guige M., Trouillet J. L., Domart Y., Pierre J., Gibert C.Detection of nosocomial lung infection in ventilated patients: use of a protected specimen brush and quantitative culture techniques in 147 patients. Am. Rev. Respir. Dis.1381988110116 Abstract, Medline, Google Scholar
17.	Baughman R. P., Dohn M. N., Frame P. T.The continuing utility of bronchoalveolar lavage to diagnose opportunistic infection in AIDS patients. Am. J. Med.971994515522 Crossref, Medline, Google Scholar
18.	Chastre, J., J. Y. Fagon, M. Bornet-Lecso, S. Calvat, M. C. Dombret, R. al Khani, F. Basset, and C. Gibert. 1995. Evaluation of bronchoscopic techniques for the diagnosis of nosocomial pneumonia. Am. J. Respir. Crit. Care Med. 152:231–240. Google Scholar
19.	Torres A., el-Ebiary M., Padro L., Gonzalez J., de la Bellacasa J. P., Ramirez J., Xaubet A., Ferrer M., Rodriguez-Roisin R.Validation of different techniques for the diagnosis of ventilator-associated pneumonia. Comparison with immediate postmortem pulmonary biopsy. Am. J. Respir. Crit. Care Med.1491994324331 Abstract, Medline, Google Scholar
20.	Cantral D. E., Tape T. G., Reed E. C., Spurzem J. R., Rennard S. I., Thompson A. B.Quantitative culture of bronchoalveolar lavage fluid for the diagnosis of bacterial pneumonia. Am. J. Med.951996601607 Crossref, Google Scholar
21.	Branger C., Fournier J. M., Loulergue J., Bouvet A., Goullet P., Boutonnier A., de Gialluly C., Couetdic G., Chomarat M., Jaffar-Banjee M. C., Mariani P.Epidemiology of Staphylococcus aureus in patients with cystic fibrosis. Epidemiol. Infect.1121994489500 Crossref, Medline, Google Scholar
22.	Bilton D., Pye A., Johnson M. M., Mitchell J. L., Dodd M., Webb A. K., Stockley R. A., Hill S. L.The isolation and characterization of non-typeable Haemophilus influenzae from the sputum of adult cystic fibrosis patients. Eur. Respir. J.81995948953 Medline, Google Scholar
23.	Sternberg R. I., Baughman R. P., Dohn M. N., First M. R.Utility of bronchoalveolar lavage in assessing pneumonia in immunosuppressed renal transplant recipients. Am. J. Med.951993358364 Crossref, Medline, Google Scholar
24.	Burge D. R., Noble M. A., Campbell M. E., Krell V. L., Speert D. P.Xanthomonas maltophilia misidentified as Pseudomonas cepacia in cultures of sputum from patients with cystic fibrosis: a diagnostic pitfall with major clinical implications. Clin. Infect. Dis.201995445448 Crossref, Medline, Google Scholar
25.	Konstan M. W., Hilliard K. W., Norvell T. M., Berger M.Bronchoalveolar lavage findings in cystic fibrosis patients with stable, clinically mild lung disease suggest ongoing infection and inflammation. Am. J. Respir. Crit. Care Med.1501994448454 Abstract, Medline, Google Scholar
26.	Bonfield T. L., Panuska J. R., Konstan M. W., Hilliard K. A., Hilliard J. B., Ghnaim H., Berger M.Inflammatory cytokines in cystic fibrosis lungs. Am. J. Respir. Crit. Care Med.152199521112118 Abstract, Medline, Google Scholar
27.	Wilmott R. W., Kassab J. T., Kilian P. L., Benjamin W. R., Douglas S. D., Wood R. E.Increased levels of interleukin-1 in bronchoalveolar washings from children with bacterial pulmonary infections. Am. Rev. Respir. Dis.1421990365368 Abstract, Medline, Google Scholar

Correspondence and requests for reprints should be addressed to Dr. Robert P. Baughman, 231 Bethesda Avenue, Rm. 6004, P.O. Box 670564, Cincinnati, OH 45267-0564.

American Journal of Respiratory and Critical Care Medicine

Abstract

Abstract

Related