Title: Management of Sternal Wounds with Bilateral Pectoralis Major Myocutaneous
Advancement Flaps in 114 Consecutively Treated Patients: Refinements in Technique
and Outcomes Analysis
Authors: Jeffrey A. Ascherman, MD,
Sejal M. Patel, BA
The surgical management of sternal wound infections remains an important and
controversial issue. The reduction of local perfusion that occurs when one or both
internal mammary arteries are harvested, or in the presence of diabetes, contributes
to these infections. The spread of infection to grafts, prosthetic valves, or suture
lines in the mediastinum can be life-threatening, and may rapidly lead to sepsis.
The use of pectoralis major muscle flaps to treat these infections has gained
acceptance. However, consensus has not been reached regarding the technique and type
of pectoralis flap to use. The timing for debridement and surgery, and the use of
rectus or omental flaps for inferior wound coverage or filling of mediastinal dead
space, are additional issues that continue to generate discussion.
In 1994, our plastic surgery division reviewed 74 patients treated with immediate
sternal debridement and bilateral pectoralis major myocutaneous advancement flaps
utilizing the anterior rectus sheath fascia for inferior wound coverage. To decrease
morbidity following operative treatment of these patients, we made a number of
refinements in our treatment protocol over the past several years, particularly with
regard to the extent of debridement, method of flap apposition, and management of
drains. The purpose of this study was to obtain specific outcomes data by reviewing
a large series of patients treated by a single surgeon after implementing revisions
to our treatment protocol.
Methods: In this series of 114 consecutive sternal wounds treated by the senior
author, patients were treated almost exclusively with debridement and immediate
closure with bilateral pectoralis major myocutaneous advancement flaps, regardless
of the degree of infection. Inferiorly, the anterior rectus fascia was raised with
the flaps. Culture-positive deep wound infections (100 of 114, 88 percent) were the
most frequent indication for intervention (Table 1). Seventy-eight (62 percent) of
the patients were infected with Staphylococcus. Fifteen patients (13 percent) were
immunosuppressed heart transplant recipients, and 38 (33 percent) were diabetic.
Most of our patients had Pairolero Type II or III infections, as the majority
presented more than one week after their original cardiac surgery (Table 2). All
data were obtained through record and chart review. Minimum follow-up time was one
year.
Procedure: After reopening the sternotomy wound, the sternum of each patient is
debrided back to bleeding bone with a rongeur following removal of all sternal
wires. Cultures of mediastinal fluid and sternal bone are sent. The pectoralis major
myocutaneous flaps are elevated off the chest wall using the electrocautery and
blunt dissection. This dissection begins medially over the third and fourth ribs,
dissecting laterally in the relatively avascular plane deep to the pectoralis major
muscles. Flap elevation is quick with relatively little blood loss. Anterior
intercostal perforators are isolated and cauterized. Superiorly, the dissection
proceeds to just below the level of the clavicles. Laterally, the flaps are elevated
until they can be advanced to the midline with no significant tension, which usually
entails ending the dissection between the midclavicular and anterior axillary lines.
The humeral insertion, thoracoacromial vascular axis, and innervation of the
pectoralis major muscle are preserved. The pectoralis minor muscle origins are also
left intact. Inferiorly, the dissection plane passes from the deep surface of the
pectoralis major muscle to the deep surface of the anterior rectus fascia, and
continues caudally to the level of the xiphoid process. The superior portion of the
anterior rectus sheath is thus elevated with the flap, while the rectus abdominis
muscle is not disturbed. Next, the wound is thoroughly cleansed with a pulse
irrigator using three liters of an antibiotic-containing solution. A single closed
suction drain is placed in the mediastinum and two additional drains are inserted
laterally, with one under each flap. All 3 drains exit the chest wall via separate
stab incisions. The flaps are then apposed to each other in the midline. Although
the underlying sternum is not rewired, there is minimal, if any, mediastinal dead
space. The deep layer of the closure, which includes the pectoralis major muscles
and their overlying fascia, as well as a portion of the anterior rectus sheath
fascia inferiorly, is closed with interrupted #2 Vicryl or Polysorb sutures. The
skin is then closed with 4-0 nylon after placing several 3-0 Vicryl or Polysorb deep
dermal sutures. Retention sutures and chest wall binders are not used, even in very
large patients. Culture results determine the choice and length of postoperative
antibiotic therapy.
Results: There were no intraoperative deaths. The 30-day perioperative mortality
rate was 7.9%, with only one death directly related to sternal infection. Nineteen
patients (16.7%) experienced postoperative morbidity (Table III), including partial
wound dehiscences (5 percent), skin edge necrosis (5 percent), and seromas (3.5
percent). There were no hematomas.
Conclusions: Compared to our previous series of 74 patients (Table IV), morbidity
has been reduced from 39% to 17%. This decrease is attributed to technical
refinements made in the procedure and changes in the postoperative management of
patients.
The median length of time until the final drain was removed after the pectoralis
advancement flap procedure increased from 11 days to 22 days with a decrease in
seroma formation from 24% to 3.5%. In nearly all patients a drain was not removed
until its output was less than 50 cc over 24 hours. Rather than remove the drains
prior to discharge, patients with high drain outputs were sent home with their
drains and seen in the office at weekly intervals until all drains were out. There
were no drain-related infections in any of these outpatients.
The number of partial dehiscences decreased from 9.5 to 5.3 percent, and no complete
dehiscences occurred in this series. This reduction may be related to the larger
sutures (#2 Vicryl or Polysorb) used in this series to appose the flaps, and the
fact that each bite extended 2 to 3 centimeters back from the flap edges.
In addition, meticulous checks for hemostasis following flap elevation and again
after pulse irrigation were felt to have reduced our hematoma rate from 5.4 to 0
percent.
We advocate single-stage management of complicated sternal wounds with immediate
debridement and bilateral pectoralis major myocutaneous advancement flaps that
include the anterior rectus sheath fascia inferiorly. The procedure is rapid and
effective. Refinements in technique have reduced our morbidity, including seroma and
hematoma formation, and partial dehiscences. Unlike procedures using pectoralis
major turnover flaps, intact internal mammary arteries, which are often used for
coronary artery bypass grafts, are not necessary. Detaching the humeral insertion of
the pectoralis muscle for rotation into the mediastinum is not done, as we have not
found it necessary to actually fill the mediastinal dead space. Rectus abdominis
muscle flaps are not necessary for inferior wound coverage as the inclusion of the
anterior rectus fascia with the flaps provides strength and avoids wound healing
problems in the xiphoid region. Furthermore, by utilizing this procedure a symmetric
chest wall contour is preserved, the anterior axillary fold is not disturbed, and
pectoralis major muscle function is maintained.
Table I: Indications for Surgery
Preoperative Condition
Number
%
Wound infection, cultures
positive
True dehiscence, cultures
negative
Inability to close
sternum after initial
cardiac surgery
Congenital sternal
absence
Total
100
(88)
9
(8)
4
(3)
1
(1)
114
Table II: Presentation Characteristics of 100 Infected Sternotomy Wounds
Time from Cardiac
Surgery to Flap
Closure (days)
Less than or equal to
7
8-28
Greater than or equal
to 29
Totals:
Number of Number
Patients
with
Purulent
drainage
4
1
67
37
29
16
100
54
Table III: Morbidity
Diagnosis
Partial dehiscence
Skin edge necrosis
Seroma
Hematoma
Complete dehiscence
Incision line skin atrophy
Subpectoral collection
Rule out hematoma
Recurrent infection
Treated Treated in Observation
at
Operating
Only
Bedside
Room
4
2
4
--1
---11(9.6%)
1
3
----1
-1
6(5.3%)
1
1
-------2(1.8%)
Total,
October
1995 to
January
2001
Total, June 1985
to December
1991 (previously
reported series
of 74 patients)
6 (5.3%)
6 (5.3%)
4 (3.5%)
0
0
1
1
0
1
19(16.7%)
7 (9.5%)
3 (4.1%)
18 (24.3%)
4 (5.4%)
1
0
1
1
0
29(39%)
Table IV: Comparison of current series of 114 patients to series of 74 patients
previously published1
30 Day Perioperative Mortality
(%)
Morbidity (%)
Median Length Of
Hospitalization (Days)
Mean # of Days Until Drain
Removal
Average Estimated Blood Loss
(ml)
Average Operative Time (min)
6/85-12/91
9
10/95-01/01
8
39
19
17
19
11
22
425
315
120
110
1. Hugo, N. E., Sultan, M. R., Ascherman, J. A., et al.
Single-stage management of
74 consecutive sternal wound complications with pectoralis major myocutaneous
advancement flaps. Plast. Reconstr. Surg. 93: 1433, 1994.