Treatment of Postpneumonectomy Empyema with Debridement Followed by Continuous Antibiotic Irrigation

TEbT

Edhaapanbit

C

R2F(OCd

©P

reatment of Postpneumonectomympyema with Debridement Followedy Continuous Antibiotic Irrigationhomas Ng, MD, FACS, Beth A Ryder, MD, Donna E Maziak, MDCM, FACS, Farid M Shamji, MD, FACS

BACKGROUND: The goal of this study was to determine the efficacy of treating postpneumonectomy empyema(PPE) with debridement followed by continuous antibiotic irrigation for pneumonectomyspace sterilization.

STUDY DESIGN: All patients presenting with PPE were evaluated. Patients with bronchopleural fistula (BPF)underwent thoracotomy for fistula closure and debridement. Patients without BPF underwentvideo-assisted thoracic surgery debridement. All patients then underwent intraoperative place-ment of an 8-F irrigation catheter and a 36-F drainage catheter. Two weeks of continuousantibiotic irrigation, as determined by cultures, were followed by collection of chest cultures on3 consecutive days. If cultures returned negative, antibiotic was instilled into the chest and allcatheters were removed. If cultures were positive, another 2 weeks of irrigation were reinitiated,adjusting the antimicrobial agent based on culture results. This regimen was repeated until threeconsecutive negative cultures were obtained.

RESULTS: Over a 5-year period, 8 consecutive patients with PPE were evaluated. Two had BPF. Mean agewas 56 years. Median time to empyema after pneumonectomy was 20 days (range 12 to 497days). Mean irrigation duration was 40 days (range 18 to 72 days) and mean followup was 580days (range 75 to 1,666 days). There was no treatment-associated morbidity or mortality. Nopatients experienced empyema recurrence during followup.

CONCLUSIONS: PPE can be successfully treated with thoracic debridement followed by continuous antibioticirrigation. This method avoids the morbidity of rib resection or thoracic cavity reductionprocedures. Closure of BPF, if present, is a prerequisite. Debridement can be performed byvideo-assisted thoracic surgery in patients without fistula. ( J Am Coll Surg 2008;206:
1178–1183. © 2008 by the American College of Surgeons)
puircf

MOaBitBswas

mpyema after pneumonectomy has a reported inci-ence of 2% to 16%.1,2 The true incidence may beigher, because this complication can occur many dec-des after surgery.3 Up to 80% of these cases are associ-ted with bronchopleural fistula (BPF).1 Historically,ostpneumonectomy empyema (PPE) has been man-ged through sequential chest wall procedures, begin-ing with rib resection for open drainage, and followedy procedures to obliterate the thoracic space. If thenfected postpneumonectomy space could be sterilized,hese procedures could be avoided. We present our ex-

ompeting Interests Declared: None.

eceived October 30, 2007; Revised December 8, 2007; Accepted January 9,008.rom the Department of Surgery, Brown Medical School, Providence, RINg, Ryder) and the Division of Thoracic Surgery, University of Ottawa,ttawa, Ontario, Canada (Maziak, Shamji).orrespondence address: Thomas Ng, MD, 2 Dudley St, Ste 470, Provi-

tence, RI 02905.

11782008 by the American College of Surgeons

ublished by Elsevier Inc.

erience with thoracic debridement followed by contin-ous antibiotic irrigation for thoracic space sterilization

n the treatment of PPE. This algorithm aims to avoidib resection (open window drainage) or any thoracicavity reduction procedures (ie, thoracoplasty or musclelap).

ETHODSn presentation, intravenous antibiotics were initiated andchest tube was inserted for patients suspected of havingPF. Patients were then taken to the operating room. Lung

solation was achieved with a double-lumen endotrachealube, and bronchoscopy was performed to evaluate forPF. Patients with BPF underwent thoracotomy for clo-

ure of BPF and thoracic cavity debridement. Patientsithout BPF underwent thoracic debridement by video-

ssisted thoracic surgery (VATS) using two or three portites. The absence of BPF was reconfirmed by intraopera-
ive air leak evaluation. After debridement, all patients un-
ISSN 1072-7515/08/$34.00doi:10.1016/j.jamcollsurg.2008.01.005

d(d

dbsrtlilfctmtaptacrics

ucaRff

ROtphcEwto

sopewpn1

tmct(btaoebct7mr(dTTatwi

Fa

1179Vol. 206, No. 6, June 2008 Ng et al Treatment of Postpneumonectomy Empyema

erwent placement of the chest tube irrigation systemFig. 1), comprised of an 8-F irrigation catheter and a 36-Frainage catheter.Postoperatively, continuous antibiotic irrigation was un-

ertaken for 2 weeks. The antimicrobial agent was selectedased on culture and sensitivity results. A continuous infu-ion rate of 200 mL/hour was empirically chosen. At thisate, 5 times the daily intravenous dose was delivered intohe chest in a 24-hour period. The drainage catheter waseft on water seal at all times. After 2 weeks of antibioticrrigation, saline irrigation was initiated for 24 hours, fol-owed by 3 consecutive daily cultures of the chest tubeluid. Fluid was obtained by temporarily disconnecting thehest tube from the drainage system tubing (care was takeno avoid contamination of both open ends), allowing theost recent saline effluent to be cultured. If all three cul-

ures were negative, antibiotic was instilled into the chest,nd both chest tubes were removed. If the cultures wereositive, another 2 weeks of antibiotic irrigation were reini-iated, adjusting the antimicrobial agent based on culturend sensitivity results. This regimen was repeated untilhest tube cultures were negative. In addition, all patientseceived 6 weeks of intravenous antibiotic. Failure to ster-lize the thoracic cavity was defined as persistent positiveultures after 90 days of treatment, at which point rib re-ection for open thoracic window drainage was considered.

Outcomes measures of this approach included the fail-re rate of thoracic cavity sterilization (persistent positiveultures), rate of empyema recurrence, treatment duration,nd treatment morbidity and mortality. Our Institutionaleview Board approved this study and also waived in-

ormed consent. The study was a retrospective analysisrom a prospective database.

ESULTSver a 5-year period, 8 consecutive patients with PPE were

reated with the algorithm described earlier. All patientsresented with fever and leukocytosis.Two of eight patientsad BPF. These two patients presented with productiveough and a decreasing fluid level seen on chest radiograph.mpyema was diagnosed in all patients by thoracentesis,hich returned culture-positive purulent fluid. Both pa-

ients with BPF underwent initial pneumonectomy at

Abbreviations and Acronyms

BPF � bronchopleural fistulaPPE � postpneumonectomy empyemaVATS � video-assisted thoracic surgery

ther institutions: one for blunt trauma and the other for s

quamous cell carcinoma. One patient had empyema with-ut BPF at the time of pneumonectomy for necrotizingneumonia. Five patients had empyema without BPF afterlective pneumonectomy for squamous cell carcinoma; allere from our own thoracic surgical service. During thiseriod, we performed 44 elective pneumonectomies foron-small cell lung carcinoma, for an empyema rate of1.4% (5 of 44), all without BPF.The two patients with empyema and BPF underwent

horacotomy for fistula closure, thoracic cavity debride-ent, and placement of chest irrigation catheters. Staple

losure of the BPF was possible in both patients, becausehe bronchial stump was left long after pneumonectomywhich is also the likely cause of bronchial dehiscence inoth patients). The single patient with empyema at theime of pneumonectomy also had thoracic debridementnd placement of irrigation catheters in addition to theriginal pneumonectomy procedure.The five patients withmpyema but without BPF underwent VATS thoracic de-ridement and placement of chest irrigation catheters. Spe-ific details of each patient are shown in Table 1. All pa-ients were men, with a mean age of 56 years (range 16 to4 years). The mean time to empyema was 90 days, with aedian of 20 days (range 12 to 497 days). The mean du-

ation of irrigation was 40 days, with a median of 34 daysrange 18 to 72 days). Mean followup duration was 580ays, with a median of 314 days (range 75 to 1,666 days).he 2 patients with N2 nodal metastasis (patients 4 and 5,able 1) received preoperative concurrent chemotherapynd radiation. We successfully sterilized the thoracic cavi-ies of all patients (achieved negative chest tube cultures)ithin 90 days. There was no morbidity or mortality dur-

ng the treatment period. Five of the six patients with non-

igure 1. Irrigation system. (A) 8-F irrigating catheter. (B) 36-F drain-ge catheter.

mall cell carcinoma died during followup, all without ev-

Table 1. Patient CharacteristicsPtNo. Age, y Initial procedure Pathology

Time toempyema, d Second procedure Microbiology

Antibioticirrigation

Irrigationduration, d* Followup, d

1 50 Right pneumonectomy T3N1, Squamouscell carcinoma

12 VATS, debridement,irrigation system

MRSA Vancomycin 18 457 died ofrecurrentcancer

2 44 Right pneumonectomy,debridement,irrigation system

Necrotizingpneumonia

— — Enterococcusserratia VRE

Vancomycin,imipenem,linezolid,quinupristin

69 1,666 alive

3 16 Right pneumonectomy Blunt trauma 34 with BPF Thoracotomy, stapleclosure andintercostals musclebuttress of BPF,debridement,irrigation system

Enterococcus,enterobacterstreptococcus,candida, MSSA

Vancomycin,aztreonam,amphotericin

49 (24 d in-hospital,25 d at home)

1,460 alive

4 69 Left pneumonectomy T2N2, Squamouscell carcinoma


MSSA Vancomycin 29 244 died ofrecurrentcancer



MRSA, enterobactercandida, CNS

Vancomycin,imipenem,amphotericin

72 384 died ofrecurrentcancer



Streptococcus Cefazolin 27 75 died ofrespiratoryfailure



Bacteroides, CNS Metronidazole,vancomycin

39 127 died ofrecurrentcancer


20 with BPF Thoracotomy, stapleclosure andserratus musclebuttress of BPF,debridement,irrigation system

CNS, MSSA,candida

Vancomycin,amphotericin

21 224 alive

*All irrigation days in-hospital except patient number 3.BPF, bronchopleural fistula; CNS, coagulase negative staphylococcus; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus; VATS, video-assisted thoracicsurgery; VRE, vancomycin-resistant enterococcus.

11

80

Ng

etal

Treatment

ofPostpneum

onectomy

Empyem

aJ

AmC

ollSurg

ipdc

DFnsmntmmaaatrtcpnippa

cCttGowsbrpiwBstit

mtiV

StacbCes

camtbwotdhtcs

maicoftptWibTccpaase

twtitm


dence of recurrent empyema at the time of death. Oneatient with carcinoma and the two patients with benigniagnosis are alive in followup, all without evidence of re-urrent empyema.

ISCUSSIONactors associated with empyema and BPF after pneumo-ectomy include poor vascular supply to the bronchialtump, long bronchial stump, malignancy at the bronchialargin, preoperative chemotherapy and radiation, poor

utritional status, diabetes, steroid use, active infection athe time of resection, prolonged chest tube duration, andechanical ventilation.1,2 Operative techniques to mini-ize PPE include careful bronchial stump handling, avoid-

nce of a long bronchial stump, use of tissue buttress,voidance of spillage, perioperative intravenous antibiotics,nd intraoperative antibiotic irrigation.1,2 As reported inwo large series of PPE by Zaheer and colleagues3 and Ee-ola and associates,4 we found Staphylococcus aureus to behe most common organism isolated. So, intraoperativeontamination is likely the major cause of empyema afterneumonectomy in patients without BPF. Despite recog-ition of these risk factors and careful operative technique,

t is unlikely that PPE can be entirely eliminated, becauseneumonectomy, by definition, is a clean contaminatedrocedure resulting in a large residual space that providesn environment for potential infection.

Historically, treatment of PPE has involved multiplehest wall procedures. Clagett and Geraci5 and Stafford andlagett6 described creating a thoracic window by rib resec-

ion for open drainage, then closing the window and fillinghe thoracic space with antibiotic solution. Clagett anderaci5 reported an initial success rate of 60%, with unrec-

gnized BPF as the main cause of failure. Zaheer and co-orkers3 reported results from the Mayo Clinic using the

ame technique as that described by Clagett and colleagues,ut with the addition of BPF closure by muscle flap. Thisesulted in an 81% success rate but required a median of 7rocedures, with a mortality of 7.1%. Obliteration of thenfected pneumonectomy space can also be accomplishedith use of multiple muscle flaps7-9 or by thoracoplasty.10,11

ut all of these extensive chest wall procedures can haveignificant morbidity and mortality in elderly and frail pa-ients with PPE. In addition, many patients who undergonitial rib resection for open drainage will never have theirhoracic window closed.4,9,12

For these reasons, more recent series have describedethods of treating PPE without the need for rib resec-

ion or thoracic reduction procedures. These methodsnvolve thoracic debridement by rethoracotomy13,14 or by
ATS15-17 and thoracic cavity antibiotic irrigation.13,15,17-19 t
chneiter and coauthors14 described success with repeatedhoracotomy for radical debridement, but this required anverage of 3.5 procedures per patient. Successful outcomesan also be obtained by combining a single thoracic de-ridement procedure with antibiotic chest irrigation.13,15,17

ritics point out that these series suffer from selection bias,xcluding patients with BPF, late presentation, or both, andhort-term followup.3,8

In our series, all patients presenting with PPE were in-luded, regardless of timing of presentation and presence orbsence of BPF. All patients underwent thoracic debride-ent followed by postoperative antibiotic irrigation for

horacic space sterilization. Debridement was performedy VATS except in the presence of BPF, when debridementas performed during thoracotomy for BPF closure. Webserved no treatment morbidity and mortality, no failureo achieve thoracic sterilization, and no recurrent empyemauring followup. The 2 patients with benign diagnosesave had longterm followup (greater than 4 years). All pa-ients avoided rib resection for open drainage and thoracicavity reduction procedures. Our proposed algorithm ishown in Figure 2.

Despite the success of our treatment strategy for PPE,any questions remain unanswered, such as the optimal

ntibiotic dose (concentration, rate of delivery, duration ofrrigation before obtaining chest fluid culture), the finan-ial cost, and the impact of therapy on the patient’s qualityf life. The daily chest irrigation dose was empirically set ative times the daily intravenous dose. Because of the con-inuous chest drainage and the thickened nature of theleura in an empyema cavity, we estimated that very little ofhe irrigated antibiotic would be absorbed into the serum.

e found that the patients, when receiving vancomycinrrigation alone, had minimal detectable levels of the anti-iotic in their serum when tested (patients 3, 4, and 7,able 1). It is likely that an even higher intrathoracic dosean be used without adverse effects. Whether this will de-rease treatment time is unknown. Our preference is toerform continuous irrigation and drainage; most otheruthors report using irrigation with intermittent drainage,llowing the antibiotic to dwell in the thoracic cavity for aet period of time.15,17-19 Additional studies are needed tovaluate which method of irrigation is superior.

Finally, our study does not provide a firm conclusion aso when treatment failure should be declared in patientsith persistently positive pleural fluid cultures. The irriga-

ion process can be labor intensive and typically requires ann-hospital stay. Only 1 patient (patient 3, Table 1) was ableo complete the irrigation process at home. We set treat-ent failure at 3 months, because we believed that a longer

reatment period would not only be expensive but would

hpOfp3ntsot

arotpmw

ASAA

DC

R

1182 Ng et al Treatment of Postpneumonectomy Empyema J Am Coll Surg

ave a negative impact on the patient’s quality of life, es-ecially for those with advanced stage lung malignancies.ur technique does not preclude subsequent rib resection

or open drainage in patients with treatment failure. All ouratients achieved thoracic space sterilization withinmonths, and no patient had recurrent empyema. But theumber of patients in our series was small, and more pa-ients are needed to confirm the efficacy of our treatmenttrategy for PPE. In addition, comparison with other meth-ds of therapy with regard to financial cost and impact onhe quality of life is needed.

In conclusion, patients with PPE can successfullychieve postpneumonectomy space sterilization by tho-acic debridement followed by continuous thoracic antibi-tic irrigation, avoiding the morbidity of rib resection andhoracic space-reducing procedures. BPF closure, ifresent, is a necessary component of this strategy. Debride-ent can be effectively performed by VATS in patientsithout BPF.

uthor Contributionstudy conception and design: Ng, Ryder, Maziak, Shamjicquisition of data: Ng, Rydernalysis and interpretation of data: Ng, Ryder, Maziak,Shamjirafting of manuscript: Ng, Ryder, Maziak, Shamji

Figure 2. Treatment algorithm for postpneumVATS, video-assisted thoracic surgery.

ritical revision: Ng, Ryder, Maziak, Shamji

EFERENCES

1. Vallieres E. Management of empyema after lung resections(pneumonectomy/lobectomy). Chest Surg Clin North Am2002;12:571–585.

2. Deschamps C, Bernard A, Nichols FC 3rd, et al. Empyema andbronchopleural fistula after pneumonectomy: factors affectingincidence. Ann Thorac Surg 2001;72:243–247.

3. Zaheer S, Allen MS, Cassivi SD, et al. Postpneumonectomyempyema: results after the Clagett procedure. Ann Thorac Surg2006;82:279–286.

4. Eerola S, Virkkula L, Varstela E. Treatment of postpneumonec-tomy empyema and associated bronchopleural fistula. Experi-ence of 100 consecutive postpneumonectomy patients. ScandJ Thorac Cardiovasc Surg 1988;22:235–239.

5. Clagett OT, Geraci JE. A procedure for the management ofpostpneumonectomy empyema. J Thorac Cardiovasc Surg1963;45:141–145.

6. Stafford EG, Clagett OT. Postpneumonectomy empyema. Neo-mycin instillation and definitive closure. J Thorac CardiovascSurg 1972;63:771–775.

7. Miller JI, Mansour KA, Nahai F, et al. Single-stage completemuscle flap closure of the postpneumonectomy empyema space:a new method and possible solution to a disturbing complica-tion. Ann Thorac Surg 1984;38:227–231.

8. Regnard JF, Alifano M, Puyo P, et al. Open window thoracos-tomy followed by intrathoracic flap transposition in the treat-ment of empyema complicating pulmonary resection. J ThoracCardiovasc Surg 2000;120:270–275.

9. Massera F, Robustellini M, Pona CD, et al. Predictors of success-ful closure of open window thoracostomy for postpneumonec-

tomy empyema. BPF, bronchopleural fistula;
onec
tomy empyema. Ann Thorac Surg 2006;82:288–292.

1

1

1

1

1

1

1

1

1

1


0. Jadczuk E. Postpneumonectomy empyema. Eur J CardiothoracSurg 1998;14:123–126.

1. Icard P, Le Rochais JP, Rabut B, et al. Andrews thoraco-plasty as a treatment of post-pneumonectomy empyema:experience in 23 cases. Ann Thorac Surg 1999;68:1159–1163.

2. Shamji FM, Ginsberg RJ, Cooper JD, et al. Open window tho-racostomy in the management of postpneumonectomy empy-ema with or without bronchopleural fistula. J Thorac Cardio-vasc Surg 1983;86:818–822.

3. Gharagozloo F, Trachiotis G, Wolfe A, et al. Pleural space irri-gation and modified Clagett procedure for the treatment of earlypostpneumonectomy empyema. J Thorac Cardiovasc Surg1998;116:943–948.

4. Schneiter D, Cassina P, Korom S, et al. Accelerated treatment forearly and late postpneumonectomy empyema. Ann Thorac Surg
2001;72:1668–1672.
5. Hollaus PH, Lax F, Wurnig PN, et al. Videothoracoscopic de-bridement of the postpneumonectomy space in empyema. EurJ Cardiothorac Surg 1999;16:283–286.

6. Gossot D, Stern JB, Galetta D, et al. Thoracoscopic manage-ment of postpneumonectomy empyema. Ann Thorac Surg2004;78:273–276.

7. Bagan P, Boissier F, Berna P, et al. Postpneumonectomy empy-ema treated with a combination of antibiotic irrigation followedby videothoracoscopic debridement. J Thorac Cardiovasc Surg2006;132:708–710.

8. Karkola P, Kairaluoma MI, Larmi TK. Postpneumonectomyempyema in pulmonary carcinoma patients. Treatment with an-tibiotic irrigation and closed-chest drainage. J Thorac Cardio-vasc Surg 1976;72:319–322.

9. Hakim M, Milstein BB. Empyema thoracis and infected pneu-monectomy space: case for cyclical irrigation. Ann Thorac Surg
1986;41:85–88.

Treatment of Postpneumonectomy Empyema with Debridement Followed by Continuous Antibiotic Irrigation

Documents

Transcript of Treatment of Postpneumonectomy Empyema with Debridement Followed by Continuous Antibiotic Irrigation