HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Pierre Magdeleinat Permission Requests Citing Articles Citing Articles via HighWire Google Scholar Articles by Schussler, O. Articles by Regnard, J. F. PubMed Articles by Schussler, O. Articles by Regnard, J. F. Related Collections Lung - other Related Article

---

Original Articles: General Thoracic

Should We Change Antibiotic Prophylaxis for Lung Surgery? Postoperative Pneumonia Is the Critical Issue

^a Department of Thoracic Surgery, Hotel-Dieu Hospital, APHP, Paris V University, Paris, France
^b Department of Anaesthesia and Surgical Intensive Care, Hotel-Dieu Hospital, APHP, Paris V University, Paris, France
^c Department of Respiratory and Critical Care Medicine, Hotel-Dieu Hospital, APHP, Paris V University, Paris, France
^d Department of Microbiology, Hotel-Dieu Hospital, APHP, Paris V University, Paris, France
^e Nosocomial Infectious Surveillance Committee, Hotel-Dieu Hospital, APHP, Paris V University, Paris, France

Accepted for publication August 4, 2008.

^_* Address correspondence to Dr Schussler, Department of Thoracic Surgery, Hôpital Hôtel Dieu, 1 place Parvis de Notre Dame, Paris, 75004, France (Email: schussler.olivier{at}neuf.fr).

General thoracic surgery: The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.

 

	Abstract

Top Abstract Introduction Patients and Methods Results Comment Footnotes Acknowledgments References

 
Background: The recommended antibiotic prophylaxis by second-generation cephalosporins reduces the incidence of wound infection and empyema, but its effectiveness on postoperative pneumonias (POPs) after major lung resection lacks demonstration. We investigated risk factors and characteristics of POPs occurring when antibiotic prophylaxis by second-generation cephalosporin or an alternative prophylaxis targeting organisms responsible for bronchial colonization was used.

Methods: An 18-month prospective study on all patients undergoing lung resections for noninfectious disease was performed. Prophylaxis by cefamandole (3 g/24 h, over 48 hours) was used during the first 6 months, whereas amoxicillin-clavulanate (6 g/24 h, over 24 hours) was used during the subsequent 12 months. Intraoperative bronchial aspirates were systematically cultured. Patients with suspicion of pneumonia underwent bronchoscopic sampling for culture.

Results: Included were 168 patients in the first period and 277 patients in the second period. The incidence of POP decreased by 45% during the second period (P = 0.0027). A significant reduction in antibiotic therapy requirement for postoperative infections (P = 0.0044) was also observed. Thirty-day mortality decreased from 6.5% to 2.9% (P = 0.06). Multivariate analysis showed that type of resection, intraoperative colonization, chronic obstructive pulmonary disease, gender, body mass index, and type of prophylaxis were independent risk factors of POP. A case control-study that matched patients of the two periods according to these risk factors (except for antibiotic prophylaxis) confirmed that the incidence of POP was lowered during the second period.

Conclusions: Targeted antibiotic prophylaxis may decrease the rate of POPs after lung resection and improve outcome.

Postoperative pneumonia (POP) is a frequent and severe complication of major lung resection, with a mortality rate of 20% to 30% [1, 2]. Antibiotic prophylaxis is recommended in pulmonary resection [3, 4] because it is a "clean-contaminated" operation in which bacterial contamination may occur during the opening of the bronchial tree. The antibiotic prophylaxis recommended in France is a first- or second-generation cephalosporin [4]. Antibiotic prophylaxis has been shown to decrease the incidence of wound infections and to be effective in the prevention of postoperative empyema. Several trials of different cephalosporins [5] or ampicillin-sulbactam [6, 7] reported the possible impact of prophylaxis in reducing the incidence of POP.

We recently published a prospective observational study [1] of patients undergoing major lung resections during a 6-month period with cefamandole, a second-generation cephalosporin, being used for antibiotic prophylaxis. The conclusions of the study were:

1 POP was frequent (25%) and severe (mortality of 19%) and occurred early in the postoperative course.

2 Responsible bacteria were mostly community-acquired bacteria (Streptococcus pneumoniae or Haemophilus spp).

3 The intraoperative rate of bronchial colonization with these potentially pathogenic bacteria was high.

4 Intraoperative bronchial colonization increased the risk of developing early POP, which was often sustained by bacteria recovered from intraoperative bronchial colonization,

5 These POP occurred frequently while the patients were still receiving cefamandole antibiotic prophylaxis.

On the basis of these results and with the agreement of our Hospital Nosocomial Infectious Surveillance Committee, we decided to change the antibiotic prophylaxis with a drug that was active not only against bacteria mostly involved in wound infections (ie, Staphylococcus) but also against bacteria responsible for initial colonization. Antibiotic prophylaxis by cefamandole was thus switched to a short-time (24 hours) high-dose (6 g) course of amoxicillin-clavulanate.

After the change of antibiotic prophylaxis, as also suggested by Nosocomial Infectious Surveillance Committee of the hospital, we continued our prospective study, with monthly in-hospital audits of incidence, severity, and microbiology of both POP and wound infections. A 1-year observation period of amoxicillin-clavulanate prophylaxis was decided to account for possible seasonal effects. This article reports the data comparing the two different antibiotic prophylaxis regimens. We also evaluated the entire cohort of patients to study possible risk factors for POP. Finally, a case-control study was also performed by matching patients of either period according to factors predictive of POP.

	Patients and Methods

Top Abstract Introduction Patients and Methods Results Comment Footnotes Acknowledgments References

 
Patients
Patients undergoing thoracotomy for scheduled lobectomy or pneumonectomy for noninfectious disease and without signs of acute respiratory infections were eligible for this study. Inclusion criteria remained the same during the first 6-month and the second 12-month periods and have been previously reported [1]. The initial 6-month period with cefamandole (CF) prophylaxis lasted from June 2001 to December 2001. The second period, with amoxicillin-clavulanate (AC) prophylaxis began in January 2002 and lasted 12 months. The study was conducted according to French laws on biomedical research. Informed consent was obtained from all the patients. The research was conducted according to recommendations outlined in the Declaration of Helsinki.

Study Design
All data concerning patient's characteristics, results of microbiologic studies, treatment procedures, and outcome were prospectively collected by a standardized questionnaire, as previously reported [1]. In particular, information was collected about age, sex, lung function, indication for lung resection, Karnofsky index, and C-reactive protein level (CRP). White blood count (WBC), chest roentgenogram, and clinical examination were systematically performed to eliminate an undergoing pneumonia or bronchitis. Nutritional status was assessed by determination of body mass index and evaluation of possible weight loss in the previous 6 months. Lung function was evaluated by spirometry and in almost all cases by calculation of predictive postoperative function with perfusion lung scanning.

Procedures
Antibiotic prophylaxis was administered during the induction of anesthesia. Patients were intubated with a double-lumen endobronchial tube. Within 30 minutes after induction of anesthesia, bilateral quantitative endobronchial aspirate (QEBA) was performed. A patient was considered colonized if the QEBA culture at 48 hours showed a predominant bacterium over a cutoff value of 104 CFU/mL. Lung resections were done according to standard techniques. In particular, bronchial section represented the last step of the operative procedure.

In both periods of the study, closure of the bronchial stump was achieved by a mechanical stapling device, whenever possible; otherwise interrupted sutures with absorbable monofilament were used. Side, type of resection, possible associated sleeve bronchial resection or chest-wall resection, previous thoracotomy, and total procedure time were recorded.

During the first period of the study, patients received an antibiotic prophylaxis by CF 1.5 g at the induction of anesthesia and 3 g/24 hours for 48 hours postoperatively. During the second period, AC (2 g) was administered at induction and at postoperative hours 8 and 16. In case of contraindication for β-lactams, a prophylaxis by levofloxacin was administered.

Postoperative analgesia was achieved by either intravenous patient-controlled analgesia with morphine or a single dose of intrathecal morphine switched to patient-controlled analgesia. Patients were kept in the semirecumbent position. A regular program of physiotherapy was started on the day of the operation. Oral alimentation was started on the first postoperative day after lobectomy and on the second day after pneumonectomy. In cases of previous head and neck operations or if recurrent nerve paralysis was observed, a special program for realimentation was started.

Outcome Assessment
As previously reported [1], our general policy was to maintain a very high index of clinical suspicion for POP and to try to identify the bacteria involved by a combination of quantitative fiberoptic bronchoscopy aspiration, plugged telescopic catheter (PTC), or protected specimen brush (PSB) sampling in case of (1) abnormal radiographic findings (new or changing radiographic infiltrates that persisted after physiotherapy or fiberoptic bronchial aspiration), (2) fever exceeding 38°C, and (3) one of the following criteria: purulent secretions or an increase of more than 30% of the CRP or WBC count during the last 24 hours (with WBC 12 x 10⁹/L).

POP was considered documented if bacteria were identified in blood culture or at the 48-hour culture of the fiberoptic sample with the following thresholds: PTC or PSB at 10³ CFU/mL or more, or QEBA at 10⁶ CFU/mL or more. If no bacteria were cultured or if the significant cutoff value was not reached, pneumonia was considered as probable if clinical and radiologic improvement occurred after the administration of antibiotics.

Acute bronchitis was defined by an increase and modification of the sputum (purulent) with a laboratory criterion of predominant bacteria at 10⁷ CFU/mL or higher, at sputum culture or 10⁵ CFU/mL or higher, or at fiberoptic bronchoscopy aspiration without radiological abnormality.

All postoperative pulmonary complications were reviewed secondarily by a pneumologist, a surgeon, and an intensive care physician. Wound sepsis was defined by a reddened, painful, and indurated wound not necessarilassociated with bacteria isolation. Empyema was defined by the presence of purulent fluid in the pleural drainage or by the isolation of pathogens from the pleural cavity. Other nosocomial infections were defined according to standard definitions from the Centers for Disease Control and Prevention (Atlanta, GA). Need for antibiotics other than antibiotic prophylaxis was also recorded.

Several outcome variables were recorded, including duration of stay in the intensive care unit (ICU), total hospital stay, and reintubation. Operative mortality was calculated by considering deaths occurring within 30 postoperative days or during the postoperative hospitalization.

Statistical Analysis
The results are expressed as percentages and mean ± standard deviation. Patients of the two periods were compared with respect to demographic, surgical, and postoperative management data, as well as to risk factors for POP as reported in the literature and, specifically, those identified in the first period of our study [1]. Continuous variables were compared by a nonparametric test (Mann-Whitney) and categoric variables, by the 2 or the Fisher exact test, as appropriate. The potential risk factors were also tested in the entire cohort of patients, including those receiving a prophylaxis different from that planned. Variables with p 0.1 were entered into a multivariate regression analysis. Statistical significance was accepted at values of p < 0.05.

To further explore the effect of antibiotic prophylaxis (CF vs AC), a case-control study was performed. Patients of the two groups were matched according to factors predictive of POP at either univariate or multivariate analysis with p < 0.05, with the obvious exception of type of antibiotic prophylaxis. All data processing and analysis were performed by using the SEM statistical software (SILEX Development, Mireffleurs, France).

	Results

Top Abstract Introduction Patients and Methods Results Comment Footnotes Acknowledgments References

 
Patients
Between June 2001 and January 2003, 507 patients underwent major lung resections in our department, and 478 were included in the study: 175 during the first 6-month period and 303 in the second 12-month period. The study excluded 29 patients (11 of 186 in the first period, 18 of 321 in the second period) because of a preexisting infection. Prophylaxis by either CF or AC was not given to 20 patients (7 of 186 [3.7%] in the first period and 13 of 161 [4%] in the second period) because of allergy to β-lactams. In addition 13 of 321 patients (4.0%) in the second period received CF because lung resection was done immediately after mediastinoscopy with frozen sections and CF had been used for prophylaxis in mediastinoscopy. This last group of patients is analyzed separately. Finally, 168 patients received CF and 277 patients received AC.

The main indications for major lung resection were non-small cell lung cancer (NSCLC). Almost all patients' characteristics, risks factors for POP, and surgical procedures were similar during the two periods (Table 1) [8]. Postoperative pain management remained the same.

The number of patients with sputum retention who required fiberoptic aspiration was not significantly different during the two periods. Pathogens responsible for POP are summarized in Table 2.

Risk Factors for POP
For the whole period of the study, 12 risk factors of developing POP were identified at univariate analysis and are listed in Table 1. Multivariate analysis showed that extent of resection, intraoperative bronchial colonization, chronic obstructive pulmonary disease (COPD), type of antibiotic prophylaxis, male sex, and body mass index 25 kg/m² were independent risk factors for POP.

Case-Control Study: CF vs AC
Patients of the first and second periods were matched according to all factors identified at either univariate or multivariate analysis (with the exception of antibiotic prophylaxis). Matching on the basis of the factors identified at univariate analyses resulted in of 79 matched pairs. The incidence of POP in matched patients of the CF and AC periods was 21 of 79 (26.6%) and 11 of 79 (13.9%, p = 0.048), respectively. Matching was performed on the basis of the five factors identified at multivariate analysis resulted in 122 matched pairs. The incidence of POP among these matched subjects was 30 of 122 (24.5%) in the CF period vs 13 of 122 (10.7%, p = 0.0043) in the AC period.

	Comment

Top Abstract Introduction Patients and Methods Results Comment Footnotes Acknowledgments References

 
During the second period of the study, we observed a significant decrease in POP incidence, possibly related to the change of antibiotic prophylaxis. This drop was predominantly caused by the decrease in the incidence of pneumonia sustained by S pneumoniae and Haemophilus spp, two pathogens that were likely to be highly susceptible to the new antibiotic prophylaxis by high-dose AC [9]. On the other hand, the incidence of POP sustained by either S aureus or gram-negative bacteria other than Haemophilus spp remained stable, without an increase in strains with resistance to antibiotics. This decrease was not accompanied by an increase in other site infections, such as wound infection or empyema.

Interestingly, the decrease in the incidence of postoperative events was achieved while the duration of the initial prophylaxis administration was reduced from 48 to 24 hours. The decreased incidence of POP was concomitant with a global diminution of antibiotic requirement and duration of ICU stay. As expected, among pathogens cultured in intraoperative bronchial aspirates, 25.6% were responsible for a subsequent respiratory infection in the first period, whereas this occurred in only 10% in the second period (p = 0.05). Globally, bacteria of the initial colonization with an increased susceptibility to AC were more likely not to become pathogenic compared with the first period.

This study has a before-and-after design that was imposed by the reasons explained in the introductory section. We are conscious that this kind of study is theoretically susceptible to unmeasured temporal confounders and regression to the mean; however we think this is unlikely for several reasons:

First, patient characteristics, indication for lung resection, type of operation, and risk factors for POP were very similar between the two groups, with the exception of more frequent alcohol intake during the second period, which obviously cannot explain the decreased incidence of POP (Table 1). The criteria for inclusion were very strict and no patients were clinically infected or receiving antibiotic therapy at the time of operation.

Second, patient management was standardized, and every effort was made not to change any aspect from one period to the other: strict clinical rules were applied for POP suspicion, and bacteriologic sampling, including FOB, was always performed before antibiotic administration. Also identical was the rest of the perioperative care, including preoperative hospital stay, surgical approaches, and pain management.

Third, the percentage of bacterial documentation of POP was in the range of what has been reported [10] for nosocomial pneumonia (40% to 70%) and was stable during the two periods (57.2% and 63 % under CF and AC, respectively), as was the incidence of acute bronchitis, wound infection, and empyema. These results are not in favor of a drift classification possible between nondocumented pneumonia, documented pneumonia, or acute bronchitis.

Fourth, a time-related effect is unlikely: the monthly incidence of POP was stable during each 6-month period but dropped when prophylaxis was changed. With a second-generation cephalosporin, the percentage of POP was in the range of what has been reported in other prospective studies (12% to 30%) [1] and very close to the 24% reported in a very recent retrospective study done in another French center [11]. However, POP developed in 4 of 13 patients in the second period of our study who had CF prophylaxis, a percentage similar to that observed in the first period but much higher than that of the second period.

Fifth, a seasonal effect can be excluded because the second period included a whole year and the monthly incidence of POP was always lower.

Finally, a sudden change in the percentage of bronchial colonization at the time of operation could partly explain the decrease in POP occurrence. In agreement with this hypothesis, the intraoperative bronchial colonization of our patients was lower during the second period. However, we hypothesize that the real colonization of our patients did not change and that the observed decrease is due to intraoperative samplings that were performed after administration of antibiotic prophylaxis.

AC, with its good and rapid penetration in bronchial secretion as described by Jehl and colleagues [12], could have influenced the incidence of the observed colonization in the second period. Other studies have similarly shown that antibiotic prophylaxis could have an effect on the incidence of observed intraoperative colonization [13]. Of note, Bold and colleagues [7] reported that targeting the bacteria isolated from the intraoperative bronchial aspiration by ampicillin-sulbactam, which has an antibacterial spectrum similar to AC, had a favorable impact on the incidence of very early postoperative pulmonary infection compared with cefazolin.

The decrease in intraoperative colonization might also be due to a sudden change in population characteristics; however, this is unlikely. Microbiology of observed colonization was very similar in both periods and similar to that described in the literature in either patients with COPD [14, 15] or those undergoing major lung resection [7, 16–19]. Furthermore, the incidence of both COPD and lung cancer, which are the principal risk factor of colonization, was the same during the two periods.

The bronchial colonization is accepted now as a determinant of the evolution of COPD [20, 21]. It is also associated with the occurrence of postoperative pulmonary complications [1, 13, 19]. A new stress, such as an operation and mechanical ventilation, could lead to development of a new infection in the presence of pathogens otherwise responsible for simple colonization.

This study suggests that a short-time targeted antibiotic prophylaxis against bacteria that colonize the bronchi of patients at the time of operation for major lung resection may be crucial in the prevention of POP. The results of this study have to be confirmed by a randomized trial.

	Acknowledgments

Top Abstract Introduction Patients and Methods Results Comment Footnotes Acknowledgments References

 
We are indebted to the different members of the Anesthesia, Intensive Care, Microbiology, and Thoracic Surgery departments for their indispensable contribution and their goodwill throughout the course of this long study, and to Professor Jean Pierre Haberer for valuable advice and encouragement.

	Footnotes

Top Abstract Introduction Patients and Methods Results Comment Footnotes Acknowledgments References

 
^_* These two authors contributed equally to the work.

	References

Top Abstract Introduction Patients and Methods Results Comment Footnotes Acknowledgments References

 

1. Schussler O, Alifano M, Dermine H, et al. Postoperative pneumonia after major lung resection Am J Respir Crit Care Med 2006;173:1161-1169.[Abstract/Free Full Text]
2. Bernard A, Ferrand L, Hagry O, et al. Identification of prognostic factors determining risk groups for lung resection Ann Thorac Surg 2000;70:1161-1167.[Abstract/Free Full Text]
3. Mangram AJ, Horan TC, et al. Guideline for prevention of Surgical site infection Infect Control Hosp Epidemiol 1999;20:250-278.[Medline]
4. Martin C, Andrivon F, Botto M, et al. Recommandations pour la pratique de l'antibioprophylaxie en chirurgie. Actualisation 1999. Société francaise d'Anesthésie et Réanimation. http://www.sfar.org/antibiofr.htlm 1999Accessed: Dec 28 2006.
5. Turna A, Kutlu CA, Ozalp T, et al. Antibiotic prophylaxis in elective thoracic surgery: cefuroxime versus cefepime Thorac Cardiovasc Surg 2003;51:84-88.[Medline]
6. Wertzel H, Swobola L, Frank U, et al. Operative antibiotic prophylaxis in general thoracic surgery Thorac Cardiovasc Surg 1992;40:326-329.[Medline]
7. Boldt J, Piper S, Uphus D, et al. Preoperative microbiologic screening and antibiotic prophylaxis in pulmonary resection operations Ann Thorac Surg 1999;68:208-211.[Abstract/Free Full Text]
8. Arozullah AM, Khuri SF, Henderson WG, et al. Development and validation of a multifactorial risk index for predicting postoperative pneumonia after major noncardiac surgery Ann Intern Med 2001;135:847-857.[Abstract/Free Full Text]
9. White A, Kaye C, Poupard J, Pypstra R, Woodnutt G, Wynne B. Augmentin (amoxicillin/clavulanate) in the treatment of community-acquired respiratory tract infection: a review of the continuing development of an innovative antimicrobial agent J Antimicrob Chemother 2004;53(Suppl 1):i3-i20.[Medline]
10. Arya SC, Agarwal N. Nosocomial infections in adult intensive-care units Lancet 2003;362:493ii-494ii.
11. Radu DM, Jaureguy F, Seguin A, et al. Postoperative pneumonia after major pulmonary resections: an unsolved problem in thoracic surgery Ann Thorac Surg 2007;84:1669-1673.[Abstract/Free Full Text]
12. Jehl F, Peter JD, Engler JM, et al. Pharmacokinetics of amoxycillin-clavulanate in bronchial secretions after multiple doses to intensive care Clin Microbiol Infect 1999;293(suppl 3):777.
13. Frey DJ, Reichmann AK, Mauch H, Kaiser D. ["Single-shot" antibiotic prophylaxis in thoracic surgery; reduction of the postoperative infection rate] Infection 1993;21(suppl 1):S35-S44.[Medline]
14. Monso E, Ruiz J, Rosell A, et al. Bacterial infection in chronic obstructive pulmonary disease. A study of stable and exacerbated outpatients using the protected specimen brush. Am J Respir Crit Care Med 1995;152:1316-1320.[Abstract]
15. Cabello H, Torres A, Celis R, et al. Bacterial colonization of distal airways in healthy subjects and chronic lung disease: a bronchoscopic study Eur Respir J 1997;10:1137-1144.[Abstract]
16. Sok M, Dragas AZ, Erzen J, et al. Sources of pathogens causing pleuropulmonary infections after lung cancer resection Eur J Cardiothorac Surg 2002;22:23-27discussion 27–9.[Abstract/Free Full Text]
17. Ioanas M, Angrill J, Baldo X, et al. Bronchial bacterial colonization in patients with resectable lung carcinoma Eur Respir J 2002;19:326-332.[Abstract/Free Full Text]
18. Wansbrough-Jones MH, Nelson A, New L, et al. Bronchoalveolar lavage in the prediction of post-thoracotomy chest infection Eur J Cardiothorac Surg 1991;5:433-434discussion 435.[Abstract]
19. Belda J, Cavalcanti M, Ferrer M, et al. Bronchial colonization and postoperative respiratory infections in patients undergoing lung cancer surgery Chest 2005;128:1571-1579.[Medline]
20. Sethi S, Maloney J, Grove L. Airway inflammation and bronchial colonization in chronic obstructive Pulmonary Disease Am J Resp Crit Care Med 2006;173:991-999.[Abstract/Free Full Text]
21. Tumkaya M, Atis S, Ozgge C. Relationship between airway colonization, inflammation and exacerbation frequency in COPD Resp Med 2007;101:729-737.[Medline]

This article has been cited by other articles:

				R. Engelman and R. Brown Invited Commentary. Ann. Thorac. Surg., December 1, 2008; 86(6): 1733 - 1734. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Pierre Magdeleinat Permission Requests Citing Articles Citing Articles via HighWire Google Scholar Articles by Schussler, O. Articles by Regnard, J. F. PubMed Articles by Schussler, O. Articles by Regnard, J. F. Related Collections Lung - other Related Article