To discuss the difficulties in dealing with infected or exposed
ventral hernia mesh, and to illustrate one solution using an
autogenous abdominal wall reconstruction technique.
Steven R. Szczerba, MD,* and Gregory A. Dumanian, MD†
From the Departments of Surgery and the Divisions of Plastic Surgery, *Emory University, Atlanta, Georgia, and †Northwestern
University, Chicago, Illinois
Objective
To discuss the difficulties in dealing with infected or exposed
ventral hernia mesh, and to illustrate one solution using an
autogenous abdominal wall reconstruction technique.
Summary Background Data
The definitive treatment for any infected prosthetic material in
the body is removal and substitution. When ventral hernia
mesh becomes exposed or infected, its removal requires a
solution to prevent a subsequent hernia or evisceration.
Methods
Eleven patients with ventral hernia mesh that was exposed,
nonincorporated, with chronic drainage, or associated with a
spontaneous enterocutaneous fistula were referred by their
initial surgeons after failed local wound care for definitive management.
The patients were treated with radical en bloc excision
of mesh and scarred fascia followed by immediate abdominal
wall reconstruction using bilateral sliding rectus
abdominis myofascial advancement flaps.
Results
Four of the 11 patients treated for infected mesh additionally
required a bowel resection. Transverse defect size ranged
from 8 to 18 cm (average 13 cm). Average procedure duration
was 3 hours without bowel repair and 5 hours with bowel
repair. Postoperative length of stay was 5 to 7 days without
bowel repair and 7 to 9 days with bowel repair. Complications
included hernia recurrence in one case and stitch abscesses
in two cases. Follow-up ranges from 6 to 54 months (average
24 months).
Conclusions
Removal of infected mesh and autogenous flap reconstruction
is a safe, reliable, and one-step surgical solution to the
problem of infected abdominal wall mesh.
Prosthetic mesh is widely used in the repair of midline
ventral hernias. Mesh ventral hernia repairs have lower
hernia recurrence rates than do primary repairs.
1 This lower2,3
The treatment of infected mesh is a difficult surgical
challenge. Removal of the infected mesh is the clearest
manner of dealing with the problem. Avoidance of postoperative
evisceration and maintenance of a competent abdominal
wall are secondary and important goals of treatment
of patients with infected mesh.
Rectus abdominis myofascial flap closure of the large
midline defect after mesh excision is one potential surgical
solution. This procedure, also known as the “separation of
parts” hernia repair, has been reported as having low hernia
recurrence rates.
rates, this procedure does not seem to be frequently or
widely used in ventral hernia repair. We reviewed our
consecutive series of 11 patients who presented for management
of infected mesh after a previous midline ventral
hernia repair to illustrate one possible solution to this difficult
surgical complication.4–10 Despite low reported hernia recurrence
METHODS
Surgical Technique
Patients receive a mechanical and antibiotic bowel preparation
as an outpatient the day before surgery. A long
midline skin incision is made, generously encompassing
scar and open wounds. In selected cases of infraumbilical
hernias and infected mesh, a panniculectomy incision is
Correspondence: Dr. Gregory A. Dumanian, Division of Plastic Surgery,
675 N. St. Clair, Suite 19-250, Chicago, IL 60611.
E-mail: gdumania@nmh.org
Accepted for publication August 5, 2002.
ANNALS OF SURGERY
Vol. 237, No. 3, 437–441
© 2003 Lippincott Williams & Wilkins, Inc.
437
used.
and direct dissection of adherent bowel and mesh is
avoided. The abdominal cavity is entered above or below
any areas of exposed mesh. The medial borders of the rectus
muscle are identi
medial edge is the guide for cautery dissection through
subcutaneous tissue, mesh, and abdominal wall scar. This
surgical maneuver along both rectus muscles serves to rapidly
open the abdomen. All viscera are dissected off the
posterior abdominal wall, while the infected mesh, in
tissue, and scar located between the rectus muscles will be
removed en bloc. In cases with
entering and leaving the in
isolate, and appropriate bowel work is performed at this
time. Any mesh remnants are dissected free and removed to
minimize the amount of residual prosthetic material. Mesh
removal is easier in cases when it had been used as a11 The incision purposefully straddles the inflammation,fied, and a dissecting finger along thisflamedfistula, single bowel loopsflamed tissue are often easy to
“
been used as a widepatch” closure to the edges of the rectus than when it had“overlay.”
The resultant defect of the abdominal wall is reconstructed
with bilateral rectus abdominis myofascial advancement
fl
of previously described
emphasis on the preservation of skin bloodaps. The technique employed is a modification“separation of parts” repairs in itsflow (Fig. 1).12
The key to this procedure is to release the external oblique
muscle and aponeurosis from its connection to the anterior
rectus fascia from above the rib cage to the iliac crest at a
level just lateral to the semilunar line. The approach to the
semilunar line is either through tunnels created via the
midline incision, or else through two laterally placed transverse
skin incisions. Skin is bluntly elevated off the semilunar
lines bilaterally. Using Deaver retractors for exposure,
the external oblique can be released as in a fasciotomy
incision. The external oblique is bluntly separated off the
internal oblique to allow the muscles to slide relative to each
other.
The medial rectus muscle and fascia are now debrided of
scar. Debridement is complete when the posterior sheath,
the rectus muscle edge, and the anterior sheath are clearly
visible. No other muscle or fascial release is performed. The
fascia is sutured together with interrupted braided nylon
fi
causes the skin to bunch in the midline. When the
procedure is performed correctly, a large amount of medial
skin on both wound edges can and should be excised. The
skin is closed with dermal absorbable sutures and staples
over three drains. One drain is placed along each semilunar
line, while the third drain is in the midline. Drains are left in
an average of 10 days.
When an overhanging pannus was present and when
patients had infraumbilical hernias (four patients), the procedure
was performed through a panniculectomy incision.
The principle of blunt dissection along the semilunar line
with preservation of blood
and skin can likewise be adhered to through this incision.
Mobilization of the patient occurs on the
are used for patient comfort only at their request. When
bowel function resumes and adequate pain control is
achieved, the patient is discharged to home with drains in
place without antibiotics (Figs. 2 and 3).gure-of-eight sutures. The medialization of the rectus musclesflow between the rectus musclefirst day. Binders
Patient Selection
The patients were referred to a single surgeon (G.A.D.)
for management of the abdominal wall. All of the patients
had some combination of exposed mesh, enterocutaneous
fi
All patients had failed conservative measures by their initialstula through or around the mesh, and/or chronic drainage.
Figure 1.“Separation of parts” procedure with perforator preservation. Reprinted with permission.
438Szczerba and Dumanian Ann. Surg. ● March 2003
surgeon (e.g., antibiotics, local wound care, and local removal
of mesh) before referral.
A retrospective chart review included the surgical history,
procedure duration, defect size, length of stay, and
postoperative complications, including hernia recurrence
and infection. The patients have been followed from 6 to 54
months postoperatively.
RESULTS
Table 1 lists patient demographic data, premorbid conditions,
initial surgery, and mesh complication. Three patients
presented with mesh extrusion, three presented with enterocutaneous
fi
chronic abscess/infected seroma. Mesh types encountered
included both Marlex and Prolene. The interval between
previous surgery and the de
month to 2 years (average 8 months). The defect diameter,
as estimated from computed tomography scans, ranged
from 8 to 18 cm (average 13 cm). Patient outcomes are
presented in Table 2.
The follow-up time is currently 6 to 54 months (average
24 months). The one hernia recurrence occurred approximately
1 year after
a computed tomography scan showing a small midline
hernia sac. The patient was explored and the hernia was
closed primarily without the use of mesh; the patient had no
further complications at 2 years of follow-up. This same
patient was treated for
time of her
There were two cases of postoperative stitch abscesses.
Small amounts of midline postoperative incision drainage
occurred months after the hernia repair. Computed tomography
scans did not show a hernia recurrence. Both patients
eventually required a surgical exploration and removal of
the involved interrupted nylon stitches. The wounds were
left open on dressings, and the incisions went on to heal in
both patients.stula through mesh, and five presented withfinitive operation ranged from 1flap hernia repair. Workup consisted ofClostridium difficile colitis at thefirst surgery.
DISCUSSION
The management of chronically exposed or infected mesh
after prosthetic repair of incisional hernias has received
little focused attention. The de
problem in itself. The patients all had chronic open or
draining wounds in association with prosthetic mesh, and all
had failed local wound management.finition of the entity is a
Figure 2.
and exposed mesh. A diverting transverse colostomy had been done at
an outside hospital for wound control due to a midline colonic
through the mesh.Preoperative view of patient with midline abdominal woundfistula
Figure 3.
takedown, andThree-month postoperative view after mesh removal, colostomy“separation of parts” hernia repair.
Vol. 237● No. 3 Treatment of Infected Mesh 439
When faced with a mesh extrusion, one must know how
and why the mesh was placed. If the mesh was used as an
overlay, then local management with mesh excision should
not increase the chance for bowel injury, entry into the
peritoneum, or a delayed hernia. However, when mesh was
used to replace a full-thickness abdominal wall defect, subsequent
attempts at local mesh removal expose the patient to
all of the above possibilities.
We believe that this study demonstrates the feasibility of
a one-step solution to the problem of exposed or infected
mesh. The major difference between our technique and
previously reported methods of autogenous reconstruction
is our focus on the preservation of skin blood
of skin blood
with more reliable wound healing and lower infection
rates.
other series of autogenous reconstructions after mesh removal.
Though not the focus of the article, Girotto et al.flow. Maintenanceflow in other settings is associated13–15 Our techniques differ in other small aspects from9
presented seven patients with infected mesh that was removed
and reconstructed with sliding rectus
patients had postoperative infections. In comparison to
Girotto et al., we avoid unilateral releases of the external
oblique muscle to ensure muscle balance between the right
and left sides of the abdominal wall. Mathes et al. wrote
brie
of removal of infected mesh in a larger series of over 100
abdominal wall reconstructions.
rectus abdominis muscles to the thinner and less
well-vascularized tensor fascia lata. In addition, the tensor
fascia lata is dif
and the pedicle turned over at the groin has been known to
compress the femoral vein and cause deep venous thrombosis.
A more traditional surgical treatment for infected mesh is
mesh removal, bowel work as needed, and placement of a
new prosthetic mesh for abdominal wall reconstruction.
Placement of permanent mesh in heavily contaminatedflaps. Twofly on the utility of the tensor fascia lata flap in cases10 We prefer the wellvascularizedficult to mobilize for supraumbilical defects,
fi
to 90%.
tissue. The two stitch abscesses in this series point to the
dif
keep foreign material to an absolute minimum.
The increased prevalence of laparoscopic ventral hernia
repairs may or may not change the incidence of infected
mesh and the ability to deal with the surgical problem.
Laparoscopic hernia repairs have lower rates of infection than
do open mesh repairs.
impressively low mesh-related complications.
the laparoscope may not be the ideal instrument for the treatment
of infected mesh. Laparoscopic hernia repairs are
avoided in purulent cases and require good skin cover over
the hernia. Simultaneous bowel surgery for an associatedelds has been found to have infection rates as high as 50%16 Much depends on the quality of the overlying softficulty in working in contaminated fields, and the need to17 Series of laparoscopic repairs report18 However,
fi
these reasons, the laparoscope will not tend to be overly
useful in solving the problem of infected mesh.stula would be difficult with laparoscopic techniques. For
References
1. Luijendijk RW, Hop WCJ, van den Tol MP, et al. A comparison of
suture repair with mesh repair for incisional hernia.N Engl J Med.
2000;343:392–397.
Table 1. PATIENT DATA
Sex, Age Premorbid Conditions Prior Surgeries Mesh Complication
M, 69 Prostate cancer Prostatectomy, hernia Chronic infected seroma
F, 55 NIDDM, obesity, COPD Open cholecystectomy, hernia, diverting transverse
colostomy
Mesh extrusion
M, 72 Peripheral vascular disease,
NIDDM
Aortobifemoral bypass, incarcerated hernia Enterocutaneous
F, 52 HTN, HTN, CRI, Takayasu
diverticulitis
Open cholecystectomy, aortic graft, TAH, Hartman
procedure, multiple hernia repairs
Mesh extrusion
F, 45 HTN, obesity Open cholecystectomy, C-sections
hernia repairs
Chronic infected seroma
F, 45 SLE, obesity TAH, hernia repair Chronic abscess
F, 76 Obesity Exploratory laparotomy, hernia repair Chronic abscess
M, 55 HTN, polycythemia, PE/DVT,
GI bleed
Exploratory laparotomy for gastric ulcer, hernia repair Enterocutaneous
M, 31 None Trauma laparotomy, bowel resection, hernia repair Enterocutaneous
F, 70 NIDDM, colon cancer, DVT Exploratory laparotomy for obstructing colon ca.,
multiple hernia repairs
Mesh extrusion
M, 30 Diverticulitis, NIDDM, obesity Colectomy for diverticular disease, multiple hernia repairs Chronic abscess, intermittent drainagefistula through mesh’s,’s3, TAH, multiplefistula through meshfistula through mesh
Table 2. PATIENT OUTCOMES
Procedure
Abd. Reconstruction,
No Bowel Repair
Abd. Reconstruction
With Bowel Repair
# of cases 7 4
Procedure duration 3 hours 5 hours
Length of stay 5
Complications Hernia recurrence, 1
Stitch abscess, 2–7 days 7–9 daysC. difficile colitis, 1
440Szczerba and Dumanian Ann. Surg. ● March 2003
2. Leber GE, Garb JL, Alexander AI, et al. Long-term complications
associated with prosthetic repair of incisional hernias.Arch Surg.
1998;133:378
3. White TH, Santos MC, Thompson JS. Factors affecting wound complications
associated with prosthetic repair of incisional hernias.
Surg.
4. Ramirez OM, Ruas E, Dellon AL.
for closure of abdominal wall defects: an anatomic and clinical study.–382.Am1998;133:378–382.“Components separation” method
Plast Reconstr Surg.
5. Saulis AS, Dumanian GA. Periumbilical rectus abdominis perforator
preservation signi1990;86:519–525.ficantly reduces superficial wound complications in
“
2275
6. Fabian TC, Croce MA, Pritchard FE, et al. Planned ventral hernia.
Staged management for acute abdominal wall defects.separation of parts” hernia repairs. Plast Reconstr Surg. 2002;109:–2280.Ann Surg.
1994;219:643
7. Dibello JN, Moore JH. Sliding myofascial
muscles for the closure of recurrent ventral hernias.
Surg.
8. Lucas CE, Ledgerwood AM. Autologous closure of giant abdominal
wall defects.
9. Girotto JA, Redett R, Muehberger T, et al. Closure of chronic abdominal
wall defects: a long-term evaluation of the components separation
method.
10. Mathes SJ, Steinwald PM, Foster RD, et al. Complex abdominal wall
reconstruction: a comparison of–649.flap of the rectus abdominusPlast Reconstr1996;98:464–469.Am Surg. 1998;64:607–610.Ann Plast Surg. 1999;42:385–395.flap and mesh closure. Ann Surg.
2000;232:586
11. Kohorn EI. Panniculectomy as an integral part of pelvic operation is an
underutilized technique in patients with morbid obesity.
Surg.
12. Sukkar SM, Dumanian GA, Szczerba SM, et al. Challenging abdominal
wall defects.
13. Gibbons GW, Wheelock FC, Hoar CS, et al. Predicting success of
forefoot amputations in diabetics by noninvasive testing.–594.J Am Coll1995;180:279–285.Am J Surg. 2001;181:115–121.Arch Surg.
1979;114:1034
14. Feng LF, Price D, Hohn D, et al. Blood
mobilization in infection: A comparison between ischemic and wellperfused
skin.
15. Chang N, Mathes SJ. Comparison of the effect of bacterial inoculation
on musculocutaneous and random pattern–1037.flow changes and leukocyteSurg Forum. 1983;34:603–604.flaps. Plast Reconstr Surg.
1982;70:1
16. Dayton MT, Buchele BA, Shirazi SS, et al. Use of an absorbable mesh
to repair contaminated abdominal wall defects.
954
17. Ramshaw BJ, Esartia P, Schwab J, et al. Comparison of laparoscopic
and open ventral herniorrhaphy.
18. Heniford BT, Park A, Ramshaw BJ, et al. Laparoscopic ventral and
incisional hernia repair in 407 patients.
645–10.Arch Surg. 1986;121:–690.Am Surg. 1999;65:827–832.J Am Coll Surg. 2000;190:–650.
Vol. 237● No. 3 Treatment of Infected Mesh 441
hernia recurrence rate comes at the price of mesh-related
complications such as infection, extrusion, and enterocutaneous
fistula formation.