RECONSTRUCTIVE

Single-Stage Muscle Flap Reconstruction of the

Postpneumonectomy Empyema Space: The
Emory Experience
Hisham Seify, M.D.
Background: Postsurgical chronic empyema continues to present a complicated
Kamal Mansour, M.D. treatment scenario for thoracic and reconstructive surgeons. Muscle flaps are an
Joseph Miller, M.D. important option in the management of complex thoracic wounds. This study
Trent Douglas, M.D. was designed to report the Emory experience with muscle flaps for the man-
Renee Burke, M.D. agement of complex postsurgical empyema. The authors also present their
Albert Losken, M.D. treatment algorithm for managing empyema thoracis.
John Culbertson, M.D. Methods: The authors retrospectively reviewed the charts of 55 patients requir-
Glyn Jones, M.D. ing different treatment modalities, including muscle flap transposition. Patients
Foad Nahai, M.D. were divided into four groups according to the initial thoracic procedure: group
T. Roderick Hester, M.D. A, no surgical resection; group B, postpneumonectomy; group C, postlobec-
Atlanta, Ga. tomy; and group D, prophylactic postpneumonectomy or postlobectomy. The
study included 42 men (76.4 percent) and 13 women with a mean age of 62 years
(range, 39 to 77 years).
Results: Fifty-one muscle flap procedures were performed in 42 patients (ser-
ratus anterior flaps, 16 patients and 23 flaps; latissimus dorsi flaps, 16 patients
and 18 flaps; pectoralis major muscle flaps, intercostal muscle flaps, and rectus
abdominis flaps, three patients each: omental flap, one patient). The mean
number of ribs resected before flap intervention, usually during the open
window thoracostomy, was three. The average time from initial thoracic oper-
ation to flap intervention was 4 months. Average time from flap intervention to
discharge was 12.5 days. Average hospital stay was 26.6 days. The 51 muscle flaps
represented an average of 1.2 procedures per patient.
Conclusion: Because of the excellent blood supply of extrathoracic muscle flaps
and their ability to reach any place in the pleural cavity, they represent an ideal
tissue with which to fill the contaminated pleural space. (Plast. Reconstr. Surg.
120: 1886, 2007.)

E
mpyema continues to be an uncommon but
potentially lethal complication of resectional
pulmonary procedures in which a pyogenic
infection of the pleural space develops. Post-
pneumonectomy empyema occurs in 1 to 11
percent of patients. These patients are difficult
to manage and can carry mortality rates as high
as 50 percent.1
More than 2000 years ago, Hippocrates rec-
ognized that complete evacuation of the pleural
From the Joseph Whitehead Department of Surgery, Divisions
of Plastic Surgery and Thoracic Surgery, Emory University.
Received for publication May 20, 2005; accepted September
12, 2005.
Presented at the 71st Annual Meeting of the American Society
of Plastic Surgeons, in San Antonio, Texas, October of 2002.
Copyright ©2007 by the American Society of Plastic Surgeons
DOI: 10.1097/01.prs.0000256051.99115.fb
cavity was necessary to effectively treat empyema
thoracis. He later described incision and insertion
of metal tubes into the pleural space to drain an
empyema.2
This principle of dead space obliteration, in
combination with the advent of antibiotic ther-
apy in the 1940s, provides the basis of treatment
of empyema today. Traditional therapy now be-
gins with thoracentesis and culture-directed an-
tibiotic therapy. This is followed by tube thora-
costomy and concludes with either an open
drainage procedure such as a rib resection and
creation of an open window thoracostomy or by
thoracoplasty.3,4
In 1898, J. B. Murphy outlined the surgical
management of chronic empyema by thoracoplasty.5
At the Mayo Clinic in 1915, Robinson described
transposition of skeletal muscles into the chest to

1886 www.PRSJournal.com
Volume 120, Number 7 • Postpneumonectomy Empyema Space
treat empyema6 and, in 1920, Kanavel described
the treatment of the noncollapsing wound by de-
bridement and obliteration of dead space with
vascularized muscle flaps.7 Arnold and Pairolero
further popularized this therapy in the 1980s.8
Miller et al. reported single-stage complete muscle
flap closure of postpneumonectomy empyema
space.9 Others have reported use of the omentum,
myocutaneous flaps, and free flaps.10 –13
Muscle flaps have also been an important op-
tion in the management of bronchopleural fistula
as first described in 1911.14 Often, these complex
wounds are infected, have been previously irradi-
ated, and may have an associated bony defect
and/or open communication with the lung air-
way. Extrathoracic muscle flaps using serratus an-
terior, latissimus dorsi, and pectoralis major mus-
cles have also been described.15
The goal of this study is to report the Emory
experience with the use of muscle flaps for the
management of complex postsurgical empyema.
In addition, we present our treatment algorithm
for the management of empyema thoracis.
PATIENTS AND METHODS
In this study, the authors retrospectively re-
viewed the charts of 55 patients requiring different
treatment modalities, including muscle flap trans-
position, from 1991 to 2002. Data cards were filled
out retrospectively for each patient indicating age,
sex, diagnosis, procedures performed, period of
hospitalization, complications, and outcome. Pa-
tients were divided into four groups according to
the initial thoracic procedure performed: group
A, no surgical resection; group B, postpneumo-
nectomy; group C, postlobectomy; and group D,
prophylactic postpneumonectomy or postlobec-
tomy (Table 1).
The diagnosis of empyema required one of the
following criteria: (1) grossly purulent pleural
fluid documented by thoracentesis or at thoracot-
omy; (2) positive pleural fluid culture or Gram’s
stain; or (3) pleural fluid pH ⬍7.0 and lactate
dehydrogenase greater than 1000 U/liter. Empy-
ema was defined as multiloculated if two or more
Table 1. Cause of Surgical Empyema Divided
According to the Initial Thoracic Procedure
Group A Group B Group C Group D
(n ⴝ 17) (n ⴝ 16) (n ⴝ 13) (n ⴝ 9)
Neoplastic 13 7 8
Inflammatory 15 3 6
Other 2 1
Total 17 16 13 9 Total
pleural fluid collections were seen on chest com-
puted tomographic scan. The results were com-
pared between the different patient groups with
identification of patient morbidity and mortality.
RESULTS
The study included 42 men (76.4 percent) and
13 women; the mean age of the patients was 62
years (range, 39 to 77 years). The initial thoracic
surgery was performed for lung or pleural cancer
in 28 patients, following complicated pneumonia
and other inflammatory conditions in 24 patients,
and following spontaneous esophageal rupture in
two patients. Bronchopleural fistula was present in
20 patients, five in group A (no surgical resection),
seven in group B (postpneumonectomy), and
eight in group C (postlobectomy). The mean
number of ribs resected before flap intervention,
usually during the open window thoracostomy,
was three. Procedures performed included chest
tube thoracostomy/decortication in two patients,
Eloesser flap in 27 patients, and muscle flap trans-
fer in 42 patients (Table 2). A total of 51 muscle
flaps were performed in 42 patients as follows:
serratus anterior flaps in 16 patients (23 flaps),
latissimus dorsi flaps in 16 patients (18 flaps), pec-
toralis flaps in three patients, intercostal flaps in
three patients, rectus abdominis flaps in three pa-
tients, and an omental flap in one patient (Table
3). The average time from initial thoracic surgery
to flap intervention was 4 months. The average
Table 2. Different Types of Procedures Performed
Divided According to the Initial Thoracic Procedure
Group A Group B Group C Group D
(n ⴝ 17) (n ⴝ 16) (n ⴝ 13) (n ⴝ 9)
Chest tube
thoracotomy/
decortication 2
Eloesser flap 13 6 8
Muscle flaps 3 21 18 9
Total 18 27 26 9
Table 3. Different Muscle Flaps Performed Divided
According to the Initial Thoracic Procedure
Group A Group B Group C Group D
(n ⴝ 17) (n ⴝ 16) (n ⴝ 13) (n ⴝ 9)
Intercostal muscle 1 2
Serratus muscle 1 8 7 7
Latissimus muscle 1 9 6 2
Pectoralis muscle 1 2
Omentum 1
Rectus abdominal
muscle 2 1
3 21 18 9
1887
 Plastic and Reconstructive Surgery • December 2007
Table 4. Outcome Measurements Divided According to the Initial Thoracic Procedure
Group A Group B
(n ⴝ 17) (n ⴝ 16)
Average stay 28 days 34 days
Complications 2 (recurrent empyema)
Mortality 1
time from flap intervention to discharge was 12.5
days. The average hospital stay was 26.6 days. In
total, 51 flap procedures were performed, for an
average of 1.2 procedures per patient.
Group A
The nonresectional group included 17 patients
who developed empyema from various causes: 15
patients following pneumonia and two patients fol-
lowing spontaneous esophageal rupture. Three pa-
tients in this group underwent flap coverage (serra-
tus, latissimus, and omentum).
Group B
The postpneumonectomy group included 16
patients: 13 following pneumonectomy for neo-
plasms and three following pneumonectomy for
complicated inflammatory conditions. Eleven pa-
tients in this group underwent muscle flap cover-
age (total, 21 flaps: one intercostal, eight serratus,
nine latissimus, one pectoralis, and two rectus ab-
dominis muscle flaps).
Group C
The postlobectomy group included 13 pa-
tients: seven following lobectomy for neoplasms
and six following lobectomy for complicated in-
flammatory conditions. Ten patients in this group
underwent muscle flap coverage (total, 18 flaps:
two intercostal, seven serratus, six latissimus, two
pectoralis, and one rectus abdominis).
Group D
The prophylactic postpneumonectomy and
postlobectomy group included nine patients, and
all underwent muscle flap coverage.
Complications
Two patients developed persistent empyema
(8.7 percent). One patient required treatment by
means of computed tomography– guided drain-
age and the other patient was treated with open
drainage. One patient developed partial necrosis
of a rectus abdominis muscle flap requiring de-
bridement and local flap closure. One patient
1888
Group C Group D
(n ⴝ 13) (n ⴝ 9)
14 days 6 days
1 (partial flap necrosis)
died 10 days after surgery as a result of sepsis
(Table 4).
DISCUSSION
The optimum treatment strategy for the man-
agement of empyema thoracis remains elusive. As
stated by Thurer, “The goals of appropriate ther-
apy for empyema are to limit morbidity and mor-
tality, shorten hospital stay, and return pulmonary
function to baseline.”16 This is often easier in prin-
ciple than in practice, and surgical input is often
delayed until the failure of “medical manage-
ment” with the advent of image-directed catheter
placement and concomitant use of fibrinolytics.
In a recent report by Thourani et al., patients
undergoing a decortication procedure had the
shortest hospital stay when compared with pa-
tients having image-directed catheters or tube tho-
racostomy. In this series, 45 percent (nine of 20)
of image-directed catheters failed, which resulted
in this group having the longest hospital stay and
the highest hospital charges.17 These data have
persuaded Emory University thoracic surgeons to
opt for early surgical intervention in the manage-
ment of thoracic empyema.
Treatment strategies for empyema thoracis
use a variety of methods. The method is selected
on the basis of the stage of the empyema, the
general condition of the patient, and response to
the initial therapy. Ideally, the patients who had
old, thick pleura and did not respond to tube
thoracostomy would require early surgical thera-
pies such as decortication. However, these oper-
ations often carry an increased risk of perioperative
morbidity to the debilitated patient, as described
by Kaplan and Light.18,19 Although open drainage,
advocated by Eloesser and modified by Clagett
and Geraci20 with instillation of antibiotics at the
time of closure, is an alternative choice, the time
required to sterilize the empyema cavity is long
and additional surgery is necessary to close the
fenestration. Virkkula et al. reported that the in-
terval between the construction of the fenestra-
tion and its closure was on average 6 months
(range, 1.5 to 28 months) for postpneumonec-
tomy chronic empyema.21
Volume 120, Number 7 • Postpneumonectomy Empyema Space
Advances in the transposition of vascularized
tissue offered another alternative for the manage-
ment of chronic wounds of the chest. Arnold and
Pairolero have described their extensive experi-
ence with the intrathoracic transposition of ex-
trathoracic skeletal muscles. Multiple flaps, in-
cluding the omentum, have been described in the
treatment of these patients.22 Michaels et al. re-
cently reviewed the closure of 16 cases of post-
pneumonectomy empyema cavities with flaps. The
average number of flaps per patient was 2.1, and
11 patients required combined thoracoplasty. Two
patients in this series had free flaps transferred
microsurgically, with the remaining tissues trans-
ferred as pedicled flaps.23 Perkins et al. described
the successful management of five patients with
intrathoracic sepsis by the transfer of free flaps.
One transverse rectus abdominis musculocutane-
ous and four latissimus dorsi flaps were used in the
care of these patients.24
Chronic thoracic wounds are prevented from
shrinking by secondary intention by the rigidity of
the surrounding chest wall and are especially com-
plicated when associated with radiation fibrosis
and recurrent infection. Nevertheless, conven-
Fig. 1. Different pedicled flaps used in the obliteration of the
empyema space. Reprinted with permission from Miller, J. I.,
Mansour, K. A., Nahai, F., et al. Single stage complete muscle flap
closure of the postpneumonectomy space: A new method and
possible solution to a disturbing complication. Ann. Thorac. Surg.
38: 227, 1984.
tional thoracic surgical techniques such as the
Clagett procedure are usually successful in resolv-
ing these problems. However, further measures
must occasionally be instituted, such as flap
transposition.13 Ideally, it would be advantageous
to close these patients with tissue transposition
alone and avoid thoracoplasty. In cases of total
pneumonectomy, a significant volume is required
to fill the empyema cavity. Multiple flaps are usu-
ally required to fill the defect (Figs. 1 and 2).
According to our study, the serratus muscle flap
and the latissimus muscle flap were the flaps
most frequently used for empyema cavity oblit-
eration. The choice of flap was a decision made
by the reconstructive surgeon based on the an-
atomical defect and the availability of various
flaps.
Previous thoracic incisions could compromise
flap choice, especially the latissimus dorsi muscle
flap. The omentum is used only when a significant
volume is required together with other flaps or as
a salvage procedure. In this series, the only flap
morbidity was a partial necrosis in a rectus muscle
flap that required local flap closure. There was no
use of free tissue transfer in this series, and it
Fig. 2. Total obliteration of the empyema space with multiple
flaps (PM, pectoralis major; LD, latissimus dorsi; REC, rectus ab-
dominis; SA, serratus anterior). Reprinted with permission from
Miller, J. I., Mansour, K. A., Nahai, F., et al. Single stage complete
muscle flap closure of the postpneumonectomy space: A new
method and possible solution to a disturbing complication. Ann.
Thorac. Surg. 38: 227, 1984.
1889
 Plastic and Reconstructive Surgery • December 2007
would be indicated in cases where no regional
muscle flap option is available.
In this series, muscle flap closure was used in
conjunction with other modalities. It is to be noted
that the number of complicated empyema cases
has declined over the years as a result of early
surgical management and culture-directed antibi-
otic management. The Emory algorithm, as out-
lined by Miller et al., consists of prompt pleural
drainage by closed tube thoracostomy once the
diagnosis of postpneumonectomy empyema, with
or without bronchopleural fistula, has been estab-
lished.
Chest tube drainage is continued until the
mediastinum becomes stabilized, generally after
approximately 2 weeks. Thereafter, open drainage
or another therapy for the empyema space can be
undertaken safely without the mediastinum shifting.
Once the patient is medically stable and has
entered into the chronic phase at 3 to 4 weeks, a
modified Clagett’s procedure is performed if no
bronchopleural fistula is present. This is per-
formed by placement of a second small chest tube
Fig. 3. Algorithm for the management of postpneumonectomy empyema.
1890
inserted into the second intercostal space with a
continuous inflow-outflow irrigation system estab-
lished through the pleural cavity. The irrigant is
based on antibiotic sensitivities to the pleural
drainage. If this method is successful and the re-
turn is culture-negative on three consecutive days
after 2 weeks of irrigation, the chest tubes can be
removed and the pleural fluid is allowed to reac-
cumulate to fill the remaining space. If the mod-
ified Clagett’s technique fails, a complete muscle
flap closure of the pneumonectomy space can be
performed.
If a patient with postpneumonectomy empy-
ema has a bronchopleural fistula, it is likewise
treated during the acute phase with closed chest
tube thoracostomy, with conversion to open drain-
age at the appropriate time when mediastinal sta-
bilization has occurred. If the fistula closes, one
can attempt the modified Clagett’s procedure. If
the fistula persists, the space is managed by sur-
gical closure of the fistula and muscle flap
transposition9 (Fig. 3).
Volume 120, Number 7 • Postpneumonectomy Empyema Space
CONCLUSIONS
The management of postpneumonectomy
empyema remains a challenge in the fields of tho-
racic and reconstructive surgery. It is an uncom-
mon but disturbing complication. Several meth-
ods have been described historically, ranging from
rib resection to complete thoracoplasty and tissue
transfer. Because of the excellent blood supply of
extrathoracic muscle flaps and their ability to
reach any place in the pleural cavity, they repre-
sent an ideal tissue with which to fill a contami-
nated space.
Hisham Seify, M.D.
Joseph Whitehead Department of Surgery
Divisions of Plastic Surgery and Thoracic Surgery
Emory University
92 Via Candelaria
Coto de Caza 92679, Calif.
habdelk@emory.edu hseifymd@gmail.com
DISCLOSURE
None of the authors has a financial interest in any
of the products, devices, or drugs mentioned in this
article.
REFERENCES
1. Ashbaugh, D. G. Empyema thoracis: Factors influencing
morbidity and mortality. Chest 99: 1162, 1991.
2. Hippocrates. Writings. In R. M. Hutchins (Ed.), Great Books
of the Western World. Chicago: Encyclopedia Britannica, 1952.
29: 142.
3. Mandal, A. K., and Thadepalli, H. Treatment of spontaneous
bacterial empyema thoracis. J. Thorac. Cardiovasc. Surg. 94:
414, 1987.
4. Weber, J., Grabner, D., Al-Zand, K., and Beyer, D. Empyema
after pneumonectomy: Empyema window or thoracoplasty.
Thorac. Cardiovasc. Surg. 38: 355, 1990.
5. Milloy, F. The contributions of John B. Murphy to thoracic
surgery. Surg. Gynecol. Obstet. 171: 421, 1990.
6. Robinson, S. The treatment of chronic non-tuberculous em-
pyema. Surg. Gynecol. Obstet. 22: 557, 1916.
7. Kanavel, A. B. Plastic procedures for obliteration of cavities
with non-collapsible walls. Chicago Surgical Society Meeting,
April of 1920.
8. Arnold, P. G., and Pairolero, P. C. Chest wall reconstruction:
Experience with 100 consecutive patients. Ann. Surg. 199:
725, 1984.
9. Miller, J. I., Mansour, K. A., Nahai, F., et al. Single stage
complete muscle flap closure of the postpneumonectomy
space: A new method and possible solution to a disturbing
complication. Ann. Thorac. Surg. 38: 227, 1984.
10. Jurkiewicz, M. J., and Arnold, P. G. The omentum: An ac-
count of its use in the reconstruction of the chest wall. Ann.
Surg. 185: 548, 1977.
11. Iverson, L. I. G., Young, J. N., Ecker, R. R., et al. Closure of
bronchopleural fistulas by an omental pedicle flap. Am.
J. Surg. 152: 40, 1986.
12. Hallock, G. G. Intrathoracic application of the transverse
rectus abdominis musculocutaneous flap. Ann. Plast. Surg.
29: 357, 1992.
13. Mathes, S. J., Alpert, B. S., and Chang, N. Use of muscle flap
in chronic osteomyelitis: Experimental and clinical correla-
tion. Plast. Reconstr. Surg. 69: 815, 1982.
14. Abrashanoff, H. Plastische Methode der Schliessung von
Fistelgangen, welche von inneren Organen kommen. Zen-
tralbl. Chir. 38: 186, 1911.
15. Chen, H., Tang, Y., Noordhoff, M. S., and Chang, C. Micro-
vascular free muscle flaps for chronic empyema with bron-
chopleural fistula when the major local muscles have been
divided: One stage operations with primary wound closure.
Ann. Plast. Surg. 24: 510, 1990.
16. Thurer, R. J. Decortication in thoracic empyema: Indications
and surgical technique. Chest Surg. Clin. North Am. 6: 461,
1996.
17. Thourani, V. H., Brady, K. M., Mansour, K. A., et al. Evalu-
ation of treatment modalities for thoracic empyema: A cost-
effectiveness analysis. Ann. Thorac. Surg. 66: 1121, 1998.
18. Kaplan, D. K. Treatment of empyema thoracis. Thorax 49:
845, 1994.
19. Light, R. W. Pleural Diseases, 3rd Ed. Baltimore, Md.: Williams
& Wilkins, 1995. Pp. 129 –153.
20. Clagett, O., and Geraci, J. E. A procedure for the manage-
ment of postpneumonectomy empyema. J. Thorac. Cardio-
vasc. Surg. 45: 141, 1963.
21. Virkkula, L., Eerola, S., and Varstela, E. Surgical approach to
the chronic empyema: Space sterilization. In J. Deslauriers
and L. K. Lacquet (Eds.), Thoracic Surgery: Surgical Manage-
ment of Pleural Diseases; International Trends in General Thoracic
Surgery, Vol. 6. St. Louis, Mo.: Mosby, 1990. Pp. 263–268.
22. Arnold, P. G., and Pairolero, P. C. Intrathoracic muscle flaps:
A 10-year experience in the management of life-threatening
infections. Plast. Reconstr. Surg. 84: 92, 1989.
23. Michaels, B. M., Orgill, D. P., Decamp, M. M., et al. Flap
closure of postpneumonectomy empyema. Plast. Reconstr.
Surg. 99: 437, 1997.
24. Perkins, D. J., Lee, K. K., Pennington, D. G., et al. Free flaps
in the management of intrathoracic sepsis. Br. J. Plast. Surg.
48: 546, 1995.
1891