Vacuum-Assisted Closure Device: A Useful Tool inthe Management of Severe Intrathoracic InfectionsAlend Saadi, MD,* Jean Yannis Perentes, MD, PhD,* Michel Gonzalez, MD,Adrien Caliera Tempia, MD, Yabo Wang, MD, Nicolas Demartines, MD,Hans-Beat Ris, MD, and Thorsten Krueger, MD
Divisions of Thoracic and Vascular Surgery and Visceral Surgery, Centre Hospitalier Universitaire Vaudois, Lausanne,Switzerland
ias
eiwit
Background. This study is an evaluation of the vacuum-ssisted closure (VAC) therapy for the treatment ofevere intrathoracic infections complicating lung resec-ion, esophageal surgery, viscera perforation, or necrotiz-ng pleuropulmonary infections.
Methods. We reviewed the medical records of all pa-ients treated by intrathoracic VAC therapy betweenanuary 2005 and December 2008. All patients underwenturgical debridement-decortication and control of thenderlying cause of infection such as treatment of bron-hus stump insufficiency, resection of necrotic lung, orlosure of esophageal or intestinal leaks. Surgery wasollowed by intrathoracic VAC therapy until the infec-ion was controlled. The VAC dressings were changednder general anesthesia and the chest wall was tempo-arily closed after each dressing change. All patientseceived systemic antibiotic therapy.
Results. Twenty-seven patients (15 male, median age
64 years) underwent intrathoracic VAC dressings for the
management of postresectional empyema (n � 8) withand without bronchopleural fistula, necrotizing infec-tions (n � 7), and intrathoracic gastrointestinal leaks (n �12). The median length of VAC therapy was 22 days(range 5 to 66) and the median number of VAC changesper patient was 6 (range 2 to 16). In-hospital mortalitywas 19% (n � 5) and was not related to VAC therapy orntrathoracic infection. Control of intrathoracic infectionnd closure of the chest cavity was achieved in allurviving patients.
Conclusions. Vacuum-assisted closure therapy is anfficient and safe adjunct to treat severe intrathoracicnfections and may be a good alternative to the openindow thoracostomy in selected patients. Long time
ntervals in between VAC changes and short course ofherapy result in good patient acceptance.
Severe intrathoracic infection involving the pleuralspace or the mediastinum is a rare but challenging
entity that can complicate lung or esophageal surgery,necrotizing lung infections or viscera perforation. Atypical example is the postpneumonectomy empyema,often related to a bronchopleural fistula, which ismanaged by debridement of the bronchial stump andits closure and reinforcement by a myoplasty. This stepis followed, in early empyema, by continuous irrigationof the pleural cavity until sterilization [1] or, in lateempyema, by an open window thoracostomy. Thelatter was first described by Clagett and Geraci [2] andwidely adopted because of its success rates in control-ling infection. The technique was modified by others[1, 3] and was extended for the management of infectedresidual spaces after lobectomy when conservativetreatment fails [4, 5]. Although the open window tho-
Accepted for publication Jan 10, 2011.
*Both authors have contributed equally to this study.
Presented at the Poster Session of the Forty-sixth Annual Meeting of TheSociety of Thoracic Surgeons, Fort Lauderdale, FL, Jan 25–27, 2010.
Address correspondence to Dr Krueger, Department of Thoracic and
racostomy has demonstrated efficiency and safety forthe management of intrapleural infection, it lasts longperiods of time, involves daily changes of the intracav-itary wound dressings, and does not always allow chestwall closure [5]. Therefore, methods to accelerate thehealing process have been investigated. It was demon-strated that in case of postpneumonectomy empyemathe application of negative pressure to the thoraciccavity filled with povidone-iodine soaked towels de-creases the time necessary for intrathoracic infectioncontrol [6, 7].
Topical negative pressure wound therapy (NPWT) [8]such as the Vacuum Assisted Closure (VAC) therapydevice (KCI Inc, San Antonio, TX) has gained wideacceptance for the management of complex wound infec-tions and has become an established treatment modalityfor a wide range of indications [9–14]. Negative pressurewound therapy promotes healing through the enhance-ment of granulation tissue formation, the removal ofexudates, the reduction of edema, the increased tissueperfusion, and the wound volume reduction [15].
The use of the VAC therapy for the management ofintrathoracic infections has not been widely studied thusfar. Only few studies involving few patients are available
1583Ann Thorac Surg SAADI ET AL2011;91:1582–90 INTRATHORACIC NEGATIVE PRESSURE WOUND THERAPY
GEN
ERA
LT
HO
RA
CIC
Here we report a consecutive case series of severeintrathoracic infections that were managed by NPWTusing the commercially available VAC system as adjunctto surgical management. We applied this therapy forsevere intrathoracic infections including postresectionalempyema, necrotizing lung infections, and empyema andmediastinitis due to intrathoracic esophageal or intesti-nal leaks.
Material and Methods
Study DesignWe reviewed the medical records of 27 consecutivepatients treated by intrathoracic VAC therapy betweenJanuary 2005 and December 2008 for severe intratho-racic infection. The study was reviewed and approvedby the Institutional Ethical Committee and individualpatient consent was waived. Patients were classifiedinto 3 groups: empyema after lung resection with andwithout bronchopleural fistula, necrotizing pleuropul-monary infections, and empyema and mediastinitisrelated to gastrointestinal leakage. The following pa-rameters were determined: patient age and sex, comor-bidity, primary pathology, surgical management of theprimary pathology, surgical treatment of the cause ofinfection, VAC therapy duration, time between VACdressing changes, microbiological analysis of woundsmears, blood cultures, overall morbidity and mortal-ity, VAC-related morbidity and mortality when appli-cable, time to definitive thoracic wall closure, andmean hospital stay.
Surgical ManagementIn all cases, surgical management to control the intratho-racic infection was undertaken first. This consisted ofposterolateral thoracotomy followed by debridement,pleural decortication, and lavage of the thoracic cavity.The specific underlying cause of infection was identifiedand corrected. In case of bronchopleural fistula the bron-chial stump was debrided, and closed and covered by apediculated muscle transposed inside the thoracic cavity(ie, latissimus dorsi or serratus anterior muscles). In caseof residual pleural space infection after lobectomy, thepostresectional space was reduced with a pediculatedmuscle transposed inside the thoracic cavity (ie, latissi-mus dorsi or serratus anterior muscles). Necrotizingpneumonia with destroyed lung was treated by lobec-tomy. Finally, gastrointestinal leaks were repaired by apediculated muscle onlay approach (diaphragm or trans-posed pediculated intercostal muscle or latissimus dorsi)as previously described [20] (Fig 1A;B). In one case, the
sophageal tear could not be repaired and was managedy esophageal exclusion. Finally, in a patient with trau-atic transdiaphragmatic herniation, subsequent perfo-
ation of the jejunum and fecal empyema segmentalejunal resection was performed with primary suture of
he diaphragm.
Intrathoracic VAC TherapyAfter the surgical management of the intrathoracic infec-tion, the VAC dressing was applied. A loose layer ofsterile gauzes was placed on the mediastinum. Theresidual chest cavity was then gently filled out withpolyurethane foam (Granufoam, KCI Inc, San Antonio,TX) fitted to the geometry of the chest cavity. Care wastaken to avoid empty spaces or compression (Fig 1C). Inpatients where a diaphragmatic flap had been used, thediaphragm was substituted by an omentoplasty and theVAC foam was shaped to cover it. One to two suctiontubes were positioned and sutured within the VAC foam(Fig 1D). The chest wall was closed suturing muscle andskin in a single layer using large Vicryl 2 stitches (Ethi-con, Edinburgh, UK) with the VAC suction tubes exitingthe thoracic cavity anteriorly. No chest tubes were in-serted. The skin was covered with an adhesive drape toavoid air leaks. The intrathoracic VAC dressing wasconnected to the vacuum pump and negative pressure,ranging from �50 to �75 mm Hg, was applied.
The subsequent VAC dressings were performed undergeneral anesthesia. The thoracotomy was reopened, theintrathoracic dressing was soaked with saline solution,and the VAC foam and the gauzes were carefully re-moved. A wound smear was taken for microbiologicalanalysis and the cavity was thoroughly rinsed with warmsaline solution. Surgical debridement was performedwhen necessary. A new dressing was applied as de-scribed above. Once the thoracic infection was controlledand granulation tissue was abundant throughout thechest cavity, the residual space, if present, was filled upwith “Clagett” antibiotic solution (gentamicin 80 mg/L,neomycin 500 mg/L, and polymyxin B 100 mg/L), muscleand skin were debrided, and the wound was closed inlayers.
Medical ManagementAll patients with symptoms of severe thoracic infectionwere empirically treated with wide spectrum antibiotics.Based on results of the microbiological analysis, theantibiotherapy spectrum was restrained when possible.Mycotic infections were treated in situations of positivemicrobiological results but not empirically. The antibio-therapy was maintained throughout VAC therapy andstopped two weeks after thoracic wall closure.
Results
Patient CharacteristicsTwenty-seven patients were managed by the adjunct ofan intrathoracic VAC dressing between January 2005 andDecember 2008 for severe intrathoracic infection. Themedian age of patients was 64 years (range, 37 to 77years) and the male to female ratio was 15:12. Patientshad multiple comorbidities and their anesthesia riskscore (American Society of Anesthesiology; ASA) rangedfrom ASA II to ASA IV (Table 1). Based on the cause ofintrathoracic infection, we classified patients into three
groups: postresectional empyema (n � 8), severe necro-
reeipneatae
hwDo
s.
1584 SAADI ET AL Ann Thorac SurgINTRATHORACIC NEGATIVE PRESSURE WOUND THERAPY 2011;91:1582–90
GEN
ERA
LT
HO
RA
CIC
tizing pleuropulmonary infections (n � 7), and intratho-acic gastrointestinal leaks with mediastinitis and empy-ma (n � 12). Of the 8 patients with postresectionalmpyema, 5 had a bronchopleural fistula while 3 hadnfection of a residual space. Of the 7 patients withleuropulmonary infection, 5 had necrotizing pneumo-ia with associated empyema, 1 patient had reluctantmpyema, and 1 patient had empyema originating fromn intraabdominal abscess. Of the 12 patients with gas-rointestinal leaks, 1 had mediastinitis and empyemafter a spontaneous perforation of the esophagus, 10 had
Fig 1. (A) A patient presented after 4 antireflux surgeries with a esopthrough a left posterolateral thoracotomy. The image illustrated the loit. The forceps is placed inside the esophagus. (B) A pediculated diaphtured to the esophagus. An omentoplasty was performed to isolateplaced to fill the thoracic cavity. The mediastinum was first coveremediastinum. Suction tubes were placed within the VAC foam. (D)and the cavity was draped with the transparent film to avoid air leak
sophagopleural leaks after surgery or endoscopy, and 1
ad empyema due to traumatic diaphragmatic herniationith subsequent perforation of the jejunum (Table 2).etails of surgical management of the underlying causef infection are given in Table 2.
VAC Therapy Duration and Dressing Change IntervalAfter the initial surgical management of the intrathoracicinfection, successive VAC dressings were applied. Over-all, VAC therapy lasted 22 (5 to 66) days (median, range)until granulation tissue covered the entire thoracic cavity.The median number of VAC changes was 6 (2 to 14), and
leural fistula. The pleural cavity and mediastinum were debridedsophageal lesion that has been debrided with vital tissue aroundatic flap was used as to cover the esophageal perforation and su-leen from the thoracic cavity. (C) The VAC foam was carefully
h sterile gauzes to avoid direct contact of the VAC foam with thethoracic cavity was closed in one plan by single sutures of Vicryl 2
hagopwer eragm
the spd witThe
the mean interval between VAC changes was 3.9 � 0.3
tVtncwcTwpiacirbo2cTiw
1585Ann Thorac Surg SAADI ET AL2011;91:1582–90 INTRATHORACIC NEGATIVE PRESSURE WOUND THERAPY
GEN
ERA
LT
HO
RA
CIC
days (mean � standard error of the mean). The medianhospital stay of patients treated by VAC therapy was 44.5(range, 20 to 114) days. The median intensive care unitstay and time on ventilator was 8 (range, 0 to 69) and 3(range, 0 to 50) days, respectively. We did not observesignificant differences in VAC therapy duration, VACdressing change interval, or mean hospital stay in thedifferent subgroup of patients (Table 2).
Morbidity and Mortality During and AfterVAC TherapyFive of 27 patients died during (n � 2) and after (n � 3)he course of VAC therapy. All deaths were not related toAC therapy. Of the deaths that occurred during VAC
herapy, one was attributed to a septic shock that origi-ated from an intravenous catheter infection with aontrolled intrathoracic infection while the other deathas related to respiratory failure after massive bron-
hoaspiration in the context of lung resection surgery.hree deaths occurred after thoracic wall closure andere related to bronchoaspiration in a patient with rightneumonectomy, massive central pulmonary embolism
n one patient, and a hemorrhagic shock related tonticoagulation therapy. There was one VAC-relatedomplication which was a nonfatal postoperative bleed-ng originating from the mammary vein after VAC foameplacement that required reoperation for control ofleeding. In the surviving patients, morbidity consistedf pneumonia (n � 8, 30%) and atrial fibrillation (n � 7,6%), pulmonary embolism (n � 4, 15%), renal insuffi-iency (n � 2), and transitory ischemic attack (n � 1).wenty-two of 27 patients (82%) with severe intrathoracic
nfection survived and all of these had a successful chest
Table 1. Patient Characteristics, American Society ofAnesthesiology (ASA) Risk Score, and Comorbidities
Total 27Age, years (median/range) 64 (37–77)Female/male 12/15ASA score
Microbiological Analysis of the Thoracic Cavity andBlood CulturesMicrobiological analyses of intrathoracic wound smearwere performed in all patients during the VAC therapy.Bacterial contamination was found in 23 patients; posi-tive fungal cultures were found in 13 patients. Details ofthe microbiological analysis are given in Table 3. At timeof definitive closure of the chest cavity microbiologicalanalyses were available in 21 of 27 cases. Completesterilization was achieved in 7 patients, and persistentbacterial contamination was observed in 14 patients. Nodifference in thoracic cavity sterilization was observed inthe three different patient subgroups (Table 2). Bloodculture analyses were performed in 24 of 27 patients(89%), and were positive in 5 patients (21%).
Comment
Severe intrathoracic infection is a devastating and life-threatening condition. In this paper, we report the suc-cessful use of VAC therapy as adjunct to the surgicalmanagement of severe intrathoracic infections.
Vacuum-assisted closure therapy is based on negativepressure application to wounds to promote their healing.It was shown to increase blood flow in the treated area,reduce edema, remove excessive fluid from the wound,and impair bacterial proliferation [15]. Negative pressurewound therapy is also known to amplify cell division ratethrough the activation of intracellular signaling pathways[21, 22]. The known clinical benefits experienced in otherfields of surgical wound care are more rapid granulationtissue formation, faster wound healing, reduction ofwound volume, and decreased drainage times, translat-ing in a better control of systemic infection [15].
In thoracic surgery, the management of severe in-trathoracic infection requires thorough control of theinfected pleural space. This can be safely achieved, forcertain indications, by an open window thoracostomythat allows control of infection by repeated debridementand progressive obliteration of the pleural space [3, 5].Here we show that the intrathoracic VAC therapy inpatients with severe intrathoracic infection leads to effec-tive pleural space obliteration and infection control. Thisobservation is in accordance with the results of others[17–19]. Because the lung parenchyma is the most com-pliant structure inside the chest cavity, it is probable thatthe negative pressure used in VAC therapy enhancedreexpansion of the residual lung and promoted pleuralspace obliteration. In situations of postpneumonectomyempyema, as described in our study, partial obliterationof the postresectional space was obtained by transposingan extrathoracic muscle flap into the chest cavity duringinitial surgery. The VAC therapy in these patients ledthen to an additional shrinkage of the postpneumonec-tomy space by inducing a progressive mediastinal shiftand diaphragmatic elevation. However, in cases wherepleural space obliteration was not complete at the end of
VAC therapy, the residual cavity was filled up with
Table 2. Patient Initial Pathology, Initial Surgery When Applicable, Cause of Intrathoracic Infection, Surgical Management, VAC Therapy Length, Number of VACchanges, Length of Intensive Care Unit Stay and Invasive Ventilation (ICU/VT), In-Hospital Stay (IHS), and Closure and Microbiology (M) Status at Closure
Initial Pathology Initial Surgery Cause Infection ManagementVAC
(Days)No.
VACs ICU/VT (Days)IHS
(Days) Closure M
PostresectionalEmyema1 Mesothelioma Pneumonectomy BPF D, SA 44 10 4/0, 3/0, 18/10, 7/5 72 Y P2 NP Lobectomy RIF D 15 2 0 20 Y S3 NSCLC Bilobectomy BPF D, CP, SA 41 14 2/0 45 Y P4 NSCLC Lobectomy RIF D, CP, SA 5 3 8/2 31 Y S5 NSCLC Lobectomy BPF D, CP, SA 10 2 9/5 73 N NA6 NSCLC Bilobectomy BPF SA 12 2 0 24 Y P7 TBC Lobectomy RIF, NP D, CP, SA 23 7 2/0 29 Y NA8 Thymoma Pneumonectomy BPF SA 30 8 6/3, 4/2 45 Y PMedian 19 5 3/1 38Range 5–44 2–14 0–32/0–15 20–73
Severe NecrotizingPleuropulmonaryInfections1 NP — NP Lobectomy NA NA NA NA Y NA2 NP — NP Pn � LD flap 22 4 69/50 69 Y P3 NP — NP Lobectomy 6 2 0 26 Y P4 NP — NP Pn � LD flap 24 6 34/30 42 Y P5 NP — NP CP � SA flap 27 7 32/16 44 Y S6 Pneumothorax — Empyema D 22 4 1/0 24 Y S7 Renal cancer Nephrectomy Empyema D 28 6 2/0, 5/2, 1/0 65 N NAMedian 23 5 20/9 43Range 6–28 2–7 0–69/0–50 24–69
3 Esophageal cancer Esophagectomy AL DP 16 6 24/15 90 Y S4 Esophageal cancer Esophagectomy AL Esophageal exclusion 36 7 28/20 72 Y P5 Esophageal
diverticulumDiverticulum
resectionEPF DP 21 5 0 42 Y P
6 Esophagealstenosis
Endoscopic dilation EPF PM flap 6 2 34/8 75 Y NA
7 Esophagealstenosis
Endoscopic dilation EPF DP 15 3 0 36 Y S
8 Gastro-esophagealreflux
Anti-reflux surgery EPF DP 45 9 7/4 61 Y P
Continued
1586SA
AD
IE
TA
LA
nn
Th
oracSu
rgIN
TR
AT
HO
RA
CIC
NE
GA
TIV
EP
RE
SSUR
EW
OU
ND
TH
ER
AP
Y2011;91:1582–90
GENERALTHORACIC
ble
2.C
ontin
ued
Init
ial
Pat
hol
ogy
Init
ial
Surg
ery
Cau
seIn
fect
ion
Man
agem
ent
VA
C(D
ays)
No.
VA
Cs
ICU
/VT
(Day
s)IH
S(D
ays)
Clo
sure
M
Obe
sity
Byp
ass
AL
DP
227
11/7
114
YN
A0
Obe
sity
Byp
ass
AL
DP
6616
7/2
83Y
P1
Obe
sity
Byp
ass
AL
Eso
ph
agea
lex
clu
sion
3510
19/4
61Y
P2
Th
yroi
dca
nce
rP
arti
ales
oph
agea
lre
sect
ion
EP
FP
rim
ary
rep
air
276
028
YP
edia
n21
610
/461
ange
6–66
2–16
0–34
/0–2
028
–114
tal ed
ian
226
8/3
44an
ge5–
662–
140–
69/0
–50
20–1
14
�an
asto
mot
icle
ak;
BP
F�
bron
chop
leu
ralfi
stu
la;
CP
�co
mp
leti
onp
neu
mon
ecto
my;
D�
deb
rid
emen
t/d
ecor
tica
tion
;D
P�
dia
ph
ragm
pat
ch;
EP
F�
esop
hag
ople
ura
lfist
ula
;L
Dla
tiss
imu
sd
orsi
;N
�n
o;N
A�
not
avai
labl
e;N
P�
necr
otic
pneu
mon
ia;
NSC
LC�
non-
smal
lce
lllu
ngca
ncer
;P
�po
sitiv
e;PM
�pe
ctor
alis
maj
orm
uscl
efla
p;Pn
�um
onec
tom
y;R
IF�
resi
dual
spac
ein
fect
ion;
S�
ster
ile;
SA�
serr
atus
ante
rior
;T
BC
�tu
berc
ulo
sis;
VA
C�
vacu
um
-ass
iste
dcl
osu
re;
Y�
yes.
1587Ann Thorac Surg SAADI ET AL2011;91:1582–90 INTRATHORACIC NEGATIVE PRESSURE WOUND THERAPY
GEN
ERA
LT
HO
RA
CIC
Clagett solution that allowed successful closure of thesecases.
In addition to progressive space obliteration, the effec-tive drainage of the pleural cavity is a key element forinfection control. Our data show that drainage of thechest cavity using the VAC device was efficient in allcases. The time interval between VAC dressing changewas of 3.9 � 0.3 days (mean � standard error of themean). Over the course of the present study, we tendedto increase the intervals between VAC change and thedressings remained functional for more than 5 days.Situations of early VAC foam change because of occlu-sion of the suction tube were exceptional.
It is known that NPWT stimulates the formation ofgranulation tissue, which is another important factor foreffective wound healing [15]. This was confirmed whenapplying VAC therapy to the chest cavity. Granulationtissue formation was observed in all patients, in generalfirst in the lateral aspect of the chest cavity and on thetransposed muscle, then on the mediastinal structuresand the costo-diaphragmatic recess. As soon as the chestcavity was completely covered with healthy granulationtissue, VAC therapy was terminated (median time of 22days, range 5 to 66). It was suggested that the intratho-racic VAC therapy should be terminated once the tho-racic cavity is sterilized [17]. In our study, 14 of 21 patientshad positive microbiological analysis of their intratho-racic wound, which did not affect chest wall closuresuccess. Therefore, positive tissue cultures should notguide VAC therapy termination. White blood cell countsand C-reactive protein blood levels decreased duringVAC therapy (data not shown) but did not entirely
Table 3. Blood Cultures and Microbiology of the PleuralSpace During the Course of Vacuum-Assisted ClosureTherapy
Patients with positive blood cultures 5/24 (21%)Staphylococcus species 2Candida species 2Lactobacillus spp 1
Patients with positive bacteriology(pleural space)
23/27 (85%)
Gram-positive germsStaphylococcus species 19Enterococcus species 9Streptococcus species 8Others 9
Patients with positive fungal cultures 13/27 (48%)Candida species 11Aspergillus fumigatus 2Trichosporon capitatum 1
normalize before chest closure in the majority of patients.Ta
9 1 1 1 M RT
o M R
AL
� pne
2bplpaWfltoigigfictsbisr
bprV
rcsctspopmm
dtoarcmdvcnpowT
1588 SAADI ET AL Ann Thorac SurgINTRATHORACIC NEGATIVE PRESSURE WOUND THERAPY 2011;91:1582–90
GEN
ERA
LT
HO
RA
CIC
This had no influence on outcome after closure. In ourexperience the decision for chest wall closure shouldprincipally be driven by the observation of coverage ofthe chest cavity by granulation tissue and the improve-ment of the patients’ clinical condition, and to a lesserdegree by the normalization of inflammatory parameters.
Our study shows that the intrathoracic VAC therapyhas to be considered as an adjunct to surgery. We firsttreated, in all patients, the underlying cause of infectionby the appropriate surgical procedure. Patients present-ing a postlobectomy empyema had decortication of theresidual lung and intrathoracic muscle transposition inorder to reduce the postresectional space. Patients withpostpneumonectomy empyema with and without BPFhad debridement of the chest cavity followed by intratho-racic muscle transposition for reinforcement or closure ofthe bronchial stump [3, 23]. Patients with empyema andnecrotizing lung infections had resection of the destroyedlung and decortication of the remaining lung, and thosewith mediastinitis and empyema due to esophageal orintestinal perforation had decortication and closure ofthese leaks as recently reported [20]. After surgery theintrathoracic VAC therapy was applied and all soft tissuelayers of the chest wall were readapted. A number ofrethoracotomies were performed under general anesthe-sia to change the VAC dressing, and, if necessary, to doadditional debridement. When the infection was con-trolled, the VAC dressing was removed and the thora-cotomy was closed definitely.
A different concept was recently described by Palmenand colleagues [19]. The intrathoracic VAC was com-bined with an open window thoracostomy, with bedsideVAC dressing changes, and reconstructive surgery usingmuscle flaps for closure of the window once infection wascontrolled. The authors included patients with postresec-tional empyema and cases of empyema after treatment ofrecurrent pneumothorax. Patients with an esophagopleu-ral fistula and postpneumonectomy empyema with bron-chopleural fistulas were excluded from the VAC treat-ment because of the difficulty to obtain negative pressureinside the chest in presence of a fistula [19].
Here we show that primary correction of the underly-ing cause of intrathoracic infection (eg, closure of abronchopleural fistula or an esophageal leak by muscleflap transposition) is important and permits to treat thesepatients successfully with intrathoracic VAC therapy.
In other series reported in the literature [19], intratho-racic VAC therapy did not involve closure of the chestwall. Our patients benefited, during VAC therapy, of anintact chest wall, which resulted in high patient satisfac-tion. However, our method did not allow bedside dress-ing changes, which is an alternative and valid approachavoiding repeated anesthesia. In our study the VACchanges were performed under general anesthesia one totwo times a week, which was convenient for both thepatient and the surgeon as the procedure was painless,short, and efficient. Another advantage was that the finalclosure of the thoracotomy did not require any recon-structive procedure.
The VAC dressings were placed over many types of d
tissues with very few side effects. It has, however, beenrecommended for the safe application of a VAC foam onexposed organs and blood vessels, to apply a separatinglayer of nonadherent material or to use myoplasties [8,4]. In the chest cavity one may consider the heart, theig mediastinal vessels, the esophagus, the lung andulmonary vessels, or a bronchial or vascular stump after
ung resection at risk. Complications of topical negativeressure wound therapy on these structures are bleedingnd fistula formation, as described in the literature [25].
e routinely covered the structures at risk by muscleaps during the primary surgical procedure to control
he source of infection (Table 2). In addition, a loose layerf gauze was used to separate the mediastinum from the
ntrathoracic VAC dressing. This protective layer ofauze was replaced with each change of the VAC dress-
ng until the mediastinum started to be covered byranulation tissue. With this technique no bleeding orstula complications were observed. Others have usedommercially available foam with smaller-sized pores forhis purpose [17], while other authors do not use thiseparating layer [19]. It needs to be mentioned thatefore each replacement of the intracavitary VAC dress-
ng the gauze and the foam were soaked abundantly withaline solution to avoid tears on vital structures whenemoving the dressing.
Interestingly, we observed that the VAC dressing cane safely applied directly on the visceral pleura or lungarenchyma, irrespective of the presence of alveolopleu-al air leaks. Air leaks were drained efficiently by theAC device.In a number of patients, lower esophageal leaks were
epaired using left pedicled diaphragmatic flaps as re-ently described by our group [20]. In patients withevere intrathoracic infection and mediastinitis this pro-edure was combined with intrathoracic VAC therapy. Inhese cases the VAC dressing was separated from thepleen by the omentum, which was sutured to the dia-hragmatic defect. After repeated VAC dressings, thementoplasty became fairly solid, thus creating a neodia-hragm which stiffened over time, and no attempt wasade to replace the diaphragmatic defect by a syntheticesh.It was a concern that intrathoracic application of a VAC
evice may exert compression of the mediastinal struc-ures, or cause an ipsilateral shift of the mediastinumnce the chest cavity is closed and negative pressure ispplied, especially after pneumonectomy. Hemodynamicepercussions have been reported with VAC dressings inontact with the heart, used in sternal wounds for osteo-yelitis [26, 27]. For this reason, the intrathoracic VAC
ressing was applied gently without exerting pressure onital structures and the negative pressure applied to thehest cavity was limited to �50 to �75 mm Hg. We didot observe significant hemodynamic instability in ouratients, which could be attributed to the VAC. On thether hand, hypotension was induced in some patientshen negative pressure was applied to the chest cavity.his was corrected by reducing the volume of the VAC
ressing in these cases.
tt
1589Ann Thorac Surg SAADI ET AL2011;91:1582–90 INTRATHORACIC NEGATIVE PRESSURE WOUND THERAPY
GEN
ERA
LT
HO
RA
CIC
Only a few reports deal with the intrathoracic applica-tion of NPWT. The approach has been described forempyema or deep wound infection [7, 16, 17, 19]. In-trathoracic VAC therapy has been used as a uniquetreatment modality or combined to surgery, most oftencombined to an open window thoracostomy.
Here, we describe the use of intrathoracic VAC therapyas an adjunct to surgery. We demonstrate that it can besuccessfully used for three indications in patients pre-senting severe intrathoracic infection once the underly-ing cause of infection is treated: infected postresectionalspaces after pneumonectomy or lobectomy with or with-out associated bronchopleural fistula, infected residualpleural spaces due to necrotizing pleuropulmonary in-fections, and infected pleural spaces and mediastinitisdue to intrathoracic gastrointestinal leaks. Vacuum-assisted closure therapy is an effective and safe system tomanage patients with severe intrathoracic infections.Based on our limited experience we suggest that thistechnique should be considered in selected patients withsevere intrathoracic infections as an alternative to openwindow thoracostomy. Principle advantages would bethe preserved integrity of the chest wall and a shorttreatment course.
References
1. Gharagozloo F, Trachiotis G, Wolfe A, DuBree KJ, Cox JL.Pleural space irrigation and modified Clagett procedure forthe treatment of early postpneumonectomy empyema.J Thorac Cardiovasc Surg 1998;116:943–8.
2. Clagett OT, Geraci JE. A procedure for the management ofpostpneumonectomy empyema. J Thorac Cardiovasc Surg1963;45:141–5.
3. Pairolero PC, Arnold PG, Trastek VF, Meland NB, Kay PP.Postpneumonectomy empyema. The role of intrathoracicmuscle transposition. J Thorac Cardiovasc Surg 1990;99:958 – 68.
4. Molnar TF. Current surgical treatment of thoracic empyemain adults. Eur J Cardiothorac Surg 2007;32:422–30.
5. Zaheer S, Allen MS, Cassivi SD, et al. Postpneumonectomyempyema: results after the Clagett procedure. Ann ThoracSurg 2006;82:279–87.
6. Schneiter D, Cassina P, Korom S, et al. Accelerated treat-ment for early and late postpneumonectomy empyema. AnnThorac Surg 2001;72:1668–72.
7. Schneiter D, Grodzki T, Lardinois D, et al. Acceleratedtreatment of postpneumonectomy empyema: a binationallong-term study. J Thorac Cardiovasc Surg 2008;136:179–85.
8. Argenta LC, Morykwas MJ. Vacuum-assisted closure: a newmethod for wound control and treatment: clinical experi-ence. Ann Plast Surg 1997;38:563–77.
9. Banwell PE, Musgrave M. Topical negative pressure therapy:mechanisms and indications. Int Wound J 2004;1:95–106.
10. Hardwicke J, Paterson P. A role for vacuum-assisted closure
in lower limb trauma: a proposed algorithm. Int J LowExtrem Wounds 2006;5:101–4.
ions by implementing negative pressure wound therapy
11. Sjogren J, Mokhtari A, Gustafsson R, Malmsjo M, Nilsson J,Ingemansson R. Vacuum-assisted closure therapy for deepsternal wound infections: the impact of learning curve onsurvival and predictors for late mortality. Int Wound J2008;5:216–23.
12. Perez D, Wildi S, Demartines N, Bramkamp M, Koehler C,Clavien PA. Prospective evaluation of vacuum-assisted clo-sure in abdominal compartment syndrome and severe ab-dominal sepsis. J Am Coll Surg 2007;205:586–92.
13. Schintler M, Marschitz I, Trop M. The use of topical negativepressure in a paediatric patient with extensive burns. Burns2005;31:1050–3.
14. Bihariesingh VJ, Stolarczyk EM, Karim RB, van Kooten EO.Plastic solutions for orthopaedic problems. Arch OrthopTrauma Surg 2004;124:73–6.
15. Hunter JE, Teot L, Horch R, Banwell PE. Evidence-basedmedicine: vacuum-assisted closure in wound care manage-ment. Int Wound J 2007;4:256–69.
16. Gharagozloo F, Margolis M, Facktor M, Tempesta B, NajamF. Postpneumonectomy and postlobectomy empyema. Tho-rac Surg Clin 2006;16:215–22.
17. Groetzner J, Holzer M, Stockhausen D, et al. Intrathoracicapplication of vacuum wound therapy following thoracicsurgery. Thorac Cardiovasc Surg 2009;57:417–20.
18. Matzi V, Lindenmann J, Porubsky C, Neuboeck N, Maier A,Smolle-Juettner FM. Intrathoracic insertion of the VAC de-vice in a case of pleural empyema 20 years after pneumo-nectomy. Ann Thorac Surg 2007;84:1762–4.
19. Palmen M, van Breugel HN, Geskes GG, et al. Open windowthoracostomy treatment of empyema is accelerated by vac-uum-assisted closure. Ann Thorac Surg 2009;88:1131–6.
20. Kotzampassakis N, Christodoulou M, Krueger T, et al.Esophageal leaks repaired by a muscle onlay approach inthe presence of mediastinal sepsis. Ann Thorac Surg 2009;88:966–72.
21. Morykwas MJ, Argenta LC, Shelton-Brown EI, McGuirt W.Vacuum-assisted closure: a new method for wound controland treatment: animal studies and basic foundation. AnnPlast Surg 1997;38:553–62.
22. Morykwas MJ, Faler BJ, Pearce DJ, Argenta LC. Effects ofvarying levels of subatmospheric pressure on the rate ofgranulation tissue formation in experimental wounds inswine. Ann Plast Surg 2001;47:547–51.
23. Meyer AJ, Krueger T, Lepori D, et al. Closure of largeintrathoracic airway defects using extrathoracic muscleflaps. Ann Thorac Surg 2004;77:397–405.
24. Clare MP, Fitzgibbons TC, McMullen ST, Stice RC, HayesDF, Henkel L. Experience with the vacuum assisted closurenegative pressure technique in the treatment of non-healingdiabetic and dysvascular wounds. Foot Ankle Int 2002;23:896–901.
25. Sartipy U, Lockowandt U, Gäbel J, Jidéus L, Dellgren G.Cardiac rupture during vacuum-assisted closure therapy.Ann Thorac Surg 2006;82:1110–1.
26. Conquest AM, Garofalo JH, Maziarz DM, et al. Hemody-namic effects of the vacuum-assisted closure device on openmediastinal wounds. J Surg Res 2003;115:209–13.
27. Fuchs U, Zittermann A, Stuettgen B, Groening A, Minami K,Koerfer R. Clinical outcome of patients with deep sternalwound infection managed by vacuum-assisted closure com-
pared to conventional therapy with open packing: a retro-spective analysis. Ann Thorac Surg 2005;79:526–31.
INVITED COMMENTARY
The report by Saadi and associates in this issue of TheAnnals of Thoracic Surgery [1] is an important contributiono the surgical correction of complex intrathoracic infec-
(NPWT). Open-window thoracostomy is a well-established procedure for chronic empyema and otherrefractory intrathoracic infections. In combination with
thoracoplasty and intrathoracic muscle transposition,
