JF TIMSIT (MD)
Réanimation médicale et infectieuse
Hôpital Bichat-Claude Bernard
46 rue Henri Huchard
75018 Paris
33 1 40 25 77 03
33 1 42 26 64 38
| Table of contents |
Abstract
In the management of nosocomial pneumonia, the issue of pretreatment for suspected pneumonia vs non specific antibiotic therapy must be considered as two distinct entities. Many techniques, either blind or directed by a bronchoscopy, are available, and must be performed before any new antimicrobial treatment for suspected pneumonia is initiated.
If bacteriological analysis are not available in emergency, a bacteriologic specimen should be sampled and frozen.
Plan
It has been repeatedly demonstrated in experimental animals not receiving antibiotics that confluent bronchopneumonia and lung abscess are associated with bacterial lung tissue concentrations exceeding 103 cfu/g (1,2) ,
However, Marquette and colleagues (3) , in a pig model free of antibiotic therapy, have shown a clear relationship between histologic evidence of pneumonia and lung bacterial count, although they were not able to find a clear cut threshold for quantitative cultures to discriminate the presence or absence of pneumonia. The relationship disappeared after 4 days of ceftriaxone treatment (4). Similarly, the administration of topical and parenteral antimicrobial agents has been experimentaly shown to markedly decrease the lobar bacterial burden of lungs with bronchopneumonia, and even to increase the number of sterile lungs (5).
Similar results have been shown in humans when antimicrobial have been instituted recently to treat VAP (6,7).
Finally, after ceftriaxone therapy, the sensitivity of protected specimen brush (PSB), bronchoalveolar lavage culture (BAL-C) and quantitative endotracheal aspirates (EA) decreased dramatically as compared to lung histology (4).
On the opposite, the PSB technique appears to work well when pneumonia develop as superinfection in patients who have been receiving systemic antimicrobials for several days before the appearance of the new pulmonary infiltrates.
Evidence level 6: In experimental model, recent antimicrobials (topical or parenteral) decrease the accuracy of lung culture , PSB, BAL-C and EA in diagnosing histological pneumonia.
When nosocomial pneumonia occurred in patients receiving a course of antibiotics for a previous infections, however, the etiologic organisms are likely to be resistant to the antibiotics and their growth unaffected by them. Current antibiotic therapy could be then defined as antimicrobials instituted before any signs related to VAP. Current antimicrobials is usually defined as no change during the last 72 hours.
Recent antimicrobials have been defined as new antibiotic class started within the last 24 hours(8). In this situation, the impact of antimicrobial on the accuracy of techniques should largely vary according to the susceptibility of the causative micro-organism(s) responsible for VAP was susceptible or not.
The impact of antimicrobials on the accuracy of bacteriological samplings depends on the MICs of antimicrobial used, and the pharmacokinetic dynamic properties of antimicrobials. The bronchial concentration of molecules such as vancomycin is low and could explain partly why protected BAL remained positive after 3 to 5 days of vancomycin in MRSA-VAP (9). Similarly, aminoglycosides poorly penetrates in the lung and could be inactivated by low ph environment . Then, it might not influenced largely the accuracy of samplings.
On the opposite, antimicrobial such as fluoroquinolones could concentrate in the bronchial mucosa and could inhibit the growth of some, even non susceptible, bacteria(10). Finally antimicrobials such as aminoglycosides or fluoroquinolones or imipenem (on P. aeruginosa) processed an important post-antibiotic effect and might influenced the accuracy of the exams for a longer period than other antimicrobials.(evidence level:8).
In a pharmacokinetic modelisation of the bacterial clearance of infected lungs, Forrest and coworkers have shown that time to eradication of bacterias susceptible to ciprofloxacin in tracheal aspirates is much more rapid when the 24-h area under the concentration time curve/ MIC ratio is greater than 250 11. In general, for infections caused by extremely susceptible bacteria, even small dose are able to eradicate very rapidly bacterias from bronchial samples and to rapidly increase the rate of false positive exams.
To avoid the effects of antibiotics, on culture results, antibiotics are frequently discontinued before the sampling procedure. Essentially no information is available to validate this practice or to determine the minimum number of hours required without antibiotic therapy before samples can be taken, with maximum cultural yield12 .
In one study in which follow-up cultures of PSB specimen were obtained in 43 cases, 40% and 65% of the PSB were sterile after 24 and 48 hours of effective treatment respectively(14). The same teams have shown in the ATS meeting in 1995 that after 12 hours of antimicrobial therapy, the total number of colonies/ml recovered by the PSB decrease from 1.3 ± 0.8 X 105 cfu/ml to 3 ± 4 X 104 cfu/ml and that 25 % of the PSB obtained after 12 hours of treatment are below the 103 cfu/ml threshold(29).
Similarly, Baughman et al have shown the results of repeated non bronchoscopic P-BAL of patients with VAP after 3-5 day of adequate antibiotic treatment. They found that follow up P-BAL were sterile (7 cases) or <= 10 2 cfu/ml (4 cases) in 11/13 cases of VAP not related to MRSA. On the opposite, the clearance of MRSA under vancomycin (1g b.i.d) was much lower. Follow up pBAL was sterile in 2/12 cases and >= 104 cfu/ml in 5/12 cases (9).
It should then be made clear that performing bacteriologic samplings for diagnostic purposes after the initiation of a new antibiotic therapy in patient suspected of having developped VAP can lead to a high number of false negative results.(evidence level 3). However, the effect of recent antimicrobials in decreasing sensitivity is probably depending on bacteria and seems to be lower for MRSA.
Rouby et al compared (30) histologically proven pneumonia and 29 patients free from pneumonia. All patients with suspected VAP received antimicrobials before they died. The sensitivity of protected non bronchoscopic BAL (P-BAL) as compared to post-mortem lung histology and culture was 80% whereas the specificity was 66%. In 73% cases, the micro-organisms were partially or completely in accordance with those recovered from the lung biopsy.
Similarly, Torres et al (7) compared guided lung biopsies histology and cultures (>=103 cfu/g) to conventional BAL, Protected BAL, PSB and EA in 25 patients. The previous duration of antibiotic therapy (17/25 patients) was 9.5 +- 7.9 days and the causes of previous antimicrobials were not detailed. Previous antimicrobials decrease both sensitivity and specificity of EA (>=105 cfu/ml: ON AB: Se 50%, SP: 71%, OFF AB: Se 75%, Sp 100%), PSB (>=103 cfu/ml: ON AB: Se 50%, SP: 86%, OFF AB: Se 87%, Sp 100%). On the contrary, the accuracy of conventional BAL (>=104 cfu/ml:ON AB: Se 83%, SP: 75%, OFF AB: Se 62%, Sp 80%) and Protected BAL (>=104 cfu/ml ON AB: Se 67%, SP: 89%, OFF AB: Se 25%, Sp 100%) remained unchanged (7,21).
Antibiotic therapy is also associated with a significant rate of false positive PSB (41 to 60%) and BAL (35%) culture results. PSB growths >= 103 cfu/ml were found in patients without pneumonia by histologic and tissue culture criteria. This is believed to occur because antibiotics predispose to high levels of colonization of the upper and lower airways(15).
We compared the accuracy of PSB, BAL-C and direct examination of BAL fluid between patients not receiving antimicrobials in the last 72 hours (OFF AB, n=96) and patients who received antimicrobials for an earlier septic episode not modified during the last 72 hours (ON AB n=65). The sensitivity and the specificity of each test were not different between the ON AB and the OFF AB group as well as the percentage of complete agreement between the 3 procedures(22). All but two (MRSA twice) microorganisms were resistant to previous antimicrobial treatment.
Recently, Souweine and al reported 63 episodes of suspected VAP (8). Three groups were defined: no antibiotic treatment, current antimicrobials (initiated >72 earlier) and recent antibiotic group (new antibiotic class started within the last 24 hours). The sensitivity of PSB and BAL culture was not affected by current antimicrobials. On the contrary, recent antibiotic induced a decrease in sensitivity of both PSB (40%) and BAL (38%). The specificity of both exams remained greater than 90%. The accuracy of the percentage of intracellular organisms of BAL fluid (ICO count: threshold >= 5%) was not affected by previous antimicrobials (NO AB: 71%, Current AB 50%, Recent AB: 67%). Finally, the discrimination of PSB, BAL culture as measured by area under the ROC curve remained unchanged and greater than 0.85 for PSB, BAL and ICO when new antimicrobials have been introduced in the last 24 hours. The authors proposed to decreased the threshold of PSB (>= 102 cfu/ml) and culture of BAL (>= 103 cfu/ml) in that condition. However, recent antibiotics were totally ineffective in 5/12 patients.
If unchanged within the last 3 days, previous intravenous antimicrobials did not influence the percentage of discordant result between PSB and EA (21 vs 16%, p>0.99) and between PSB and BAL C (15 vs 18%, p=0.8)(22,23).
On the opposite, the rate of discordance between ICB (cut-off: 5%) and PSB (cut-off: 103 cfu/ml) seems to be lower in patients not receiving antibiotics (0/31) than in patient receiving antibiotics either currently (2/5) or who received antibiotics > 48 hours but < 72 hours (4/13) (24)
However, topical antimicrobials (SDD) seems to increase the rate of discordant results between PSB and EA (38 vs 12%, p=0.07) (25)
Finally, regarding mechanically ventilated patients, nosocomial pneumonia occurring in patients already treated with antimicrobials are due to resistant microorganisms and that previous antibiotic therapy, if not related to septic signs associated to nosocomial pneumonia does not influenced the accuracy of EA, PSB, ICO and BAL-C (evidence level 3).
On the opposite recent antimicrobial (i.v. or topical) decreased the accuracy and the concordance between bacteriological samplings (evidence level 3). In that condition, broncho-alveolar lavage could possibly be less affected than other quantitative techniques (7)
When antimicrobials have been recently modified, a negative finding indicates either that the patient has been successfully treated for pneumonia and the bacteria are eradicated, or that the lung infection was not present to begin with.
In the management of nosocomial pneumonia, the issue of pretreatment for suspected pneumonia vs non specific antibiotic therapy must be considered as two distinct entities. Many techniques, either blind or directed by a bronchoscopy, are available, and must be performed before any new antimicrobial treatment for suspected pneumonia is initiated. If bacteriological analysis are not available in emergency, a bacteriologic specimen should be sampled and frozen. Specimen from PSB could be frozen at -80°C with good reliability enabling PSB to be performed around the clock (26). Good reliability is also obtained at +4°C for non bronchoscopic plugged telescoping catheter (27) and is suggested for other techniques(28).
1-Johanson WG, Higuchi JH, Woods DE et al - Dissemination of pseudomonas aeruginosa during lung infection in hamsters. Role of oxygen induced lung injury. Am Rev Respir Dis 1985; 132:358-361 (evidence level:6)
2-Seidenfield JJ, Mullis RC, Fowler SR et al- Bacterial infection and acute lung injury in hamsters- Am Rev Respir Dis 1986; 134:22-6 (evidence level:6)
3-Marquette CH, Wallet F, Chopin MC et al - Relationship between microbiologic and histologic features in bacterial pneumonia - Am J Respir Crit Care Med 1996; 154:1784-1787 (evidence level 6).
4-Wermert D, Marquette CH, Copin MC et al - Influence of pulmonary bacteriology and histology on the yield of diagnostic procedures in ventilator-associated pneumonia - Am J Respir Crit Care Med 1998; 158:1390-147 (evidence level 6)
5- Johanson WG, Seidenfield JJ, De Los Santos R et al - Prevention of nosocomial pneumonia using topical and parenteral antimicrobial agents. Am Rev Respir Dis 1988; 137:265-272 (evidence level 6)
6- Rouby JJ, Martin de Lassale E, Poete P et al- Nosocomial bronchopneumonia in the critically ill : histologic and bacteriologic aspects - Am Rev Respir Dis 1992; 146:1059-1066 -( Evidence level: 3)
7- Torres A, Fabregas N, Ewig S et al - Sampling methods for ventilator-associated pneumonia: validation using different histologic and microbiologic references - Crit Care Med 2000; 28:2799-2804 (evidence level 3)
8- Souweine B, Veber B, Bedos JP et al- Diagnostic accuracy of protected specimen brush and bronchoalveoalr lavage in nosocomial pneumonia: impact of previous antimicrobial treatments. Crit Care Med 1998; 26:236-244
9- Baughman RP, Kerr MA - Lack of effectiveness of vancomycin in clearing S. aureus from the lung of patients with ventilator associated pneumonia - 2001 ATS international conference, May 18-23 2001. Am J respir Crit Care Med ; 163:5:A928 (evidence level 3)
10- Fabre D, Bressole F, Gomeni R et al - Steady state pharmacokinetics of ciprofloxacin in plasma from patients with nosocomial pneumonia: penetration of the bronchial mucosa - Antimicrobial Agents chemother 1991;2521-2525 (evidence level 5)
11- Forrest a, Nix DE, Ballow CH et al - Pharmacodynamics of intravenous ciprofloxacin in seriously ill patients - Antimicrob Agents Chemother 1993; 1073-81
12-Pingleton S, Fagon JY, Leeper KV - Patient selection for clinical investigation of ventilator associated pneumonia. Criteria for evaluating diagnostic techniques - Chest 1992; 102: 553S-556S (evidence level 7)
13- Montravers Ph, Fagon JY, Chastre J et al - Follow-up protected specimen brushes to assess treatment in nosocomial pneumonia - Am Rev Respir Dis 1993:147:38-44 (level of evidence: 3)
14-Blavia R, Dorca J, Verdaguer R et al - Bacteriological follow-up of nosocomial pneumonia by successive protected specimen brushes (Abstract) Eur Respir J 1991;A823 (evidence level 3)
15-Chastre J, Viau F, Brun P et al - Prospective evaluation of protected specimen brush for the diagnosis of pulmonary infections in ventilated patients - Am Rev Respir Dis 1984;130:924-929 (evidence level 3)
16- Torres A, El-Ebiary M, Padro J et al - Validation of different techniques for the diagnosis of ventilator associated pneumonia. Comparison with immediate post-mortem pulmonary biopsy - Am Rev Respir Dis 1994; 149:324-331 (evidence level 3)
17- Marquette CH, Copin MC, Wallet R et al - Ventilator associated pneumonia . prospective evaluation of the diagnostic yield of the protected specimen brush, endotracheal aspirates and broncho-alveolar lavage using the histopathological diagnosis as gold standard - Am J Respir Crit Care Med 1995; 151:1878-1888
18- Chastre J, Fagon JY, Bornet-Lesco S et al - Evaluation of bronchoscopic techniques for the diagnosis of nosocomial pneumonia - Am J respir Crit Care Med 1995; 152:231-240
19- Papazian L, Thomas P, Garbe L et al - Bronchoscopic or blind sampling techniques for the diagnosis of ventilator-associated pneumonia - Am J Respir Crit Care Med 1995; 152:1982-91 (evidence level 3)
20- Kirtland SH, Corley DE, Winterbauer RH et al - The diagnosis of ventilator-associated pneumonia - a comparison of histologic, microbiologic and clinical criteria - Chest 1997; 112:445-457 (evidence level 3)
21- Fabregas N, Ewig S, Torres A et al - Clinical diagnosis of ventilator-associated pneumonia revisited : comparative validation using immediate post-mortem lung biopsies- Thorax 1999; 54:867-873 (evidence level 3)
22- Timsit JF, Misset B, Renaud B et al - Effect of previous antimicrobial therapy on the accuracy of the main procedures used to diagnose nosocomial pneumonia in patients who are using ventilation - Chest 1995; 108:1036-1040 (evidence level 3)
23-Jourdain B, Joly-Guillou ML, Dombret MC et al- Usefulness of Quantitative cultures of BAL fluid for diagnosing nosocomial pneumonia in ventilated patients - Chest 1997; 111:411-418 (evidence level 3)
24-Dotson RG, Pingleton SK - The effect of antibiotic therapy on recovery of intracellular bacteria from broncho-alveolar lavage in suspected ventilator associated nosocomial pneumonia. Chest 1993 103:541-546 (evidence level 3)
25- Jourdain B, Novara A, Joly-Guillou ML et al - Role of the quantitative cultures of endotracheal aspirates in the diagnosis of nosocomial pneumonia - Am J Respir Crit Care Med 1995; 152:241-6 (evidence level 3)
26- Georges H, Santre C, Leroy O et al - Reliability of quantitative cultures of protected specimen brush after freezing - Am J Respir Crit Care Med 1996;153:855-857 (evidence level 3)
27- Delassence A, Joly-Guillou ML, Martin-Lefevre L et al - Accuracy of delayed cultures of plugged telescoping catheter samples for diagnosing bacterial pneumonia - Crit Care Med 2001;29:1311-7.
28-Middleton RM, Eyre J, Kirkpatrick MB - In vitro evaluation of the effect of overnight refrigeration on bacterial growth from protected specimen brush cultures. Abstr. Am Rev Respir Dis 1993; 147:A652
29- Prats E, Dorca J, Irigaray R et al - Bacteriological follow-up of ventilator-associated pneumonia during the first 24 hours of antibiotic treatment. - 1995 international conference Am J Respir Crit Care Med 1995;151:A791 (evidence level 3)