SURGEON AT WORK
A Skin-Sparing Approach to the Treatment
of Necrotizing Soft-Tissue Infections: Thinking
Reconstruction at Initial Debridement
Laura K Tom, MD, Thomas J Wright, BS, Dara L Horn,
Tam N Pham, MD, FACS, Kari A Keys, MD
The rapid progression of a necrotizing soft-tissue infec-
tion (NSTI) makes prompt and aggressive surgical inter-
vention crucial to survival. During the past decade,
increased awareness, earlier diagnosis, advances in critical
care, and appropriate surgical interventions have all
contributed to the overall decrease in mortality of
NSTI, from 10% to 12%1-5 to 4.9%.6 Increasing survivor-
ship demands that the focus expand from purely preser-
ving life to preserving quality of life.
Surgical techniques recommend wide debridement of
all associated skin, subcutaneous, and soft tissue. Such de-
fects can leave survivors with large surface-area wounds
comparable with a full-thickness burn. Skin grafting and
extensive rehabilitation are necessary to mitigate disfigure-
ment, limited joint mobility, and chronic pain. When
debridement focuses only on tissue directly involved in
necrosis, viable skin and subcutaneous tissue can remain
in place despite wide debridement of deeper tissue planes.
We introduce this technique as a skin-sparing
debridement.
We designed incision diagrams for each body region to
allow full access to underlying diseased tissue and were
mindful of future reconstruction through the following
basic principles: maintain unaffected skin, preserve skin
perforator blood vessels, and cover joints and vital struc-
tures. In this article, we describe the rationale and imple-
mentation of this technique, as well as a case series of
skin-sparing debridements and subsequent reconstruction
for patients with NSTIs at our institution. This approach
Disclosure Information: Nothing to disclose.
Presented at the 10th Annual Meeting of the Washington/Oregon State
Chapter of the American College of Surgeons Meeting, Suncadia, WA,
June 2015, and at the 60th Annual Plastic Surgery Research Council, Seat-
tle, WA, May 2015.
Received October 5, 2015; Revised December 18, 2015; Accepted January
6, 2016.
From the Department of Surgery, Harborview Medical Center, University
of Washington, Seattle, WA.
Correspondence address: Kari A Keys, MD, Division of Plastic Surgery,
Department of Surgery, Harborview Medical Center, University of Wash-
ington, Box 359796, 325 Ninth Ave, Seattle, WA 98104. email: karik3@
uw.edu
Published by Elsevier Inc. on behalf of the American College of Surgeons.
e47
MD, Eileen M Bulger, MD, FACS,
has decreased skin graft size and allowed some wounds to
be closed by delayed primary closure (DPC) alone.
HISTORIC RATIONALE
Improved patient survival often drives the adaptation of
surgical techniques. The evolution of the surgical manage-
ment of breast cancer illustrates this concept well. Mortal-
ity was the primary consideration in 1894, when Dr
William Halsted first described the radical mastectomy.7
Despite producing substantial morbidity and disfigure-
ment, this method remained the standard of care for
more than 50 years, until Drs Patay and Dyson described
the modified radical mastectomy.8 This was followed by
the introduction of the skin-sparing mastectomy by Drs
Toth and Lappert in 1984.9 As focus shifted toward qual-
ity of life, skin-sparing and nipple-sparing mastectomies
have demonstrated improved aesthetic and quality of
life outcomes without compromising oncologic
outcomes.7,9,10
Although survival remains a primary goal for patients
with NSTI, we propose that the evolution of its surgical
treatment also follow a skin-sparing strategy. Traditional
surgical technique describes wide debridement of all
necrotic tissue for source control, as this is most closely
associated with favorable outcomes.3-5 With extensive
skin and soft-tissue losses, often the only reconstructive
option is skin graftingdsimilar to burn reconstruction.
Skin grafting is a reliable reconstructive approach, howev-
er, skin grafts have well-described drawbacks, including
poor contour and color mismatch, uneven meshed
appearance, decreased elasticity, pruritus, hypersensitivity
to pain and sunlight, and the propensity for joint contrac-
tures along flexor surfaces and erosive changes over
extensor prominences.
To address these shortcomings, we sought a safe
method for NSTI debridements that would maintain
maximal native tissue to improve reconstructive options
for these patients. The aim was to minimize the surface-
area healing by secondary intention or requiring skin
grafting.
http://dx.doi.org/10.1016/j.jamcollsurg.2016.01.008
ISSN 1072-7515/16
e48 Tom et al Skin-Sparing for Necrotizing Infections J Am Coll Surg

SKIN-SPARING INCISION DESIGNS
Background
The incision designs use the concept of vascular arcades,
perforator blood supply, and delay phenomenon. Similar
to the vascular arcades of the intestines from the mesen-
teric vessels, the skin and subcutaneous tissue have
vascular arcades from main source vessels throughout
the body.11 The skin and subcutaneous tissue vascular ar-
cades are from smaller blood vessels called perforators.
Vascular arcades join adjacent blood supplies through
arterial choke vessels and avalvular veins, which control
the pressure gradient across capillary beds and oxygen de-
livery.11,12 When an arcade is ligateddas with an incision
or undermining a flapdthe adjoining choke vessels dilate
to accommodate additional blood flow based on tissue ox-
ygen demand. In response, these choke vessels dilate and
hypertrophy during the next 48 to 72 hours.13-15 This per-
manent vessel enlargement is termed the delay phenome-
non.13-15 It allows for viability of an increased tissue area
beyond what would normally be perfused by a single
perforator arcade.
The concept of perforators and the delay phenomenon
is of critical importance to the success of skin-sparing
debridement for NSTIs. The spared skin’s underlying
vascular supply might be completely compromised sec-
ondary to myofascial necrosis or subsequent debridement
with large undermining. In this setting, the adjacent intact
perforators and vascular arcades provide the blood supply
to overlying spared skin through increase blood flow via
the delay phenomenon.
Incisions designs
As a combined effort between the Department of Surgery
and Division of Plastic Surgery, we designed incision di-
agrams for each body region for initial rapid and complete
debridement and were mindful of future reconstruction
(Figs. 1 to 4). These approaches allow for full and unre-
stricted access to underlying diseased tissue by following
a set of standard basic principles: maintaining uninvolved
skin, preserving skin perforator blood vessels, and
covering joints and vital structures. Key points are
outlined in Table 1.

Figure 1. Skin-sparing incision diagrams: torso. (A) Anterior. (1) The lateral chest incision runs
along the deltopectoral groove. If needed, an inframammary incision can be used to access tissue
deep to the breast (red line). (2) The neck is approached via a Z or lazy S, similar to that used for a
neck dissection. If the infection does not have laterality or there is concern for concurrent
mediastinitis, a “hangman” approach can be used (blue line). (3) A supraclavicular or infracla-
vicular incision can be useful for a counter incision for a lower chest infection. The shoulder should
be approached through the soft tissue around the deltoid, with attempts to preserve soft tissue
over the acromion. (4) An abdominoplasty incision allows for extensive exposure to abdominal well.
This can be extended inferiorly into the groin for additional debridement, especially in infections
within a large pannus or Fournier gangrene. The incisions should be mindful of the anterior iliac
spine as well as the femoral vessels. (B) Posterior. (5) A lazy-S incision is used to access the un-
derlying latissimus dorsi muscle and provides extensive exposure to the lower back. (6) A curvi-
linear incision allows for access to the upper back, and preserved coverage over the scapula.
(From: Waschke J, Böckers TM, Paulsen F. Anatomie Das Lehrbuch. 1st ed. Munich, Germany:
Elsevier GmbH, Urban & Fischer; 2015, reprinted with permission.)
Vol. 222, No. 5, May 2016
Figure 2. Skin-sparing incision diagrams: upper extremity. (A)
Dorsal. (1) and (9) Fasciotomy techniques should be used in the
hand and forearm. (2) The direct incision over the medial condyle
of the elbow is avoided to preserve soft tissue over the ulnar
nerve. Similarly, direct incision over the olecranon can preserve
soft-tissue coverage. (3) The shoulder is approached with pres-
ervation of tissue over the acromion. (4) A counter incision can be
used to identify additional disease distal to the primary incision
(red line). (B) Volar. (5) Z incisions are used within the axilla and
joint flexion surfaces to prevent joint contracture. (6) Arm in-
cisions should run either radial (blue line) or ulnar to the bicipital
groove to protect underlying brachial artery, median, and ulnar
nerve. (7) Direct incisions over the medial epicondyle are avoided
to protect the ulnar nerve. The flexion surface should be
approached with a z-shaped or a transverse incision (blue line).
(8) A bridge of tissue can be preserved overlying the carpal tunnel
to maintain coverage of the median nerve. (From: Paulsen F,
Waschke J. Sobotta Atlas of Human Anatomy. Vol 2. 15th ed.
Munich, Germany: Elsevier GmbH, Urban & Fischer; 2013,
reprinted with permission).
In a reverse thought process, placement of incisions
began with considering the reconstructive issues that
we have encountered for patients, including flexion
contractures, anatomic distortion, and difficulty
covering joints, boney prominences, large arteries, and
nerves. On extremities, z-shaped or transverse incisions
are used along flexion creases to avoid longitudinal in-
cisions across joints, avoiding future joint flexion con-
tractures. On the trunk, incisions are designed along
anatomic units in the inframammary fold or the
pectoral-deltoid groove to limit malpositioning or en-
bloc loss of the breast and nipple. The lower abdominal
fold is a consistently safe area to undermine and repo-
sition, as demonstrated safely in abdominoplasty. Care
is taken to keep vital structures (eg femoral vessels) and
boney prominences (eg knee, elbow) covered by placing
incisions along their periphery. The design is such that
the entire tissue can be elevated over the vital structure
for evaluation and deep debridement if necessary. This
effort preserves native tissue for reconstruction of these
difficult to cover areas.
Tom et al Skin-Sparing for Necrotizing Infections e49
Debridement
The debridement includes only necrotic tissue. Similar to
previous authors’ descriptions, the necrosis often extends
beyond that which is evident on the external skin exam,
and tends to run in an epifascial or myofascial plane.16
Studies of common monomicrobial infectious organisms,
such as Streptococcus pyogenes and Clostridium perfringens,
have demonstrated the cellular mechanisms that these or-
ganisms use to rapidly progress through tissue planes. The
exotoxins and surface proteins of group A Streptococcus
suppress host responses and induce local vessel throm-
bosis; enzymatic activity allows rapid travel in the supra-
fascial plane.17 Clostridium, on the other hand, is known
for its extensive myofascial necrosis, which is mediated
by a-toxin and other exotoxins effects on vascular endo-
thelium, compromising blood flow and increasing
compartment pressures on the muscles themselves.18
In certain presentations, whether due to specific micro-
biology or delayed presentation, the skin is already
necrotic with thrombosed vessels. We stress that this
dead skin must be removed at initial debridement. How-
ever, we suggest that perfused skin beyond the margins of
the frankly necrotic tissue should be preserved because it
does not appear, in our experience, that this skin or sub-
cutaneous tissue is a source of ongoing septic physiology.
This critical delineation limits excision of viable overlying
skin and subcutaneous tissue, hence, a skin-sparing
debridement.
IMPLEMENTATION AND TECHNIQUE
At Harborview Medical Center, when there is suspicion
for an NSTI, in addition to swift medical interventions
(ie broad-spectrum antibiotics, glucose control, pain man-
agement) and supportive measures (airway, blood pres-
sure support), a surgical consult is immediately called.
Based on history, physical exam, and laboratory studies,
the decision is made by the surgical team whether to pro-
ceed to the operating room. The subjective and objective
criteria for the diagnosis of NSTI has been described pre-
viously.11,19-21 Wong and colleagues20 describes a prog-
nostic score that providers can use to help guide the
concern for NSTI: the Laboratory Risk Indicator for
Necrotizing Fasciitis score. If there is true concern for
an NSTI or the diagnosis cannot be ruled out, proceeding
to the operating room is not delayed.
Once in the operating room, physical exam dictates the
setup for the case. Preparing and draping is done in a wide
fashion to easily access any potential underlying exten-
sion. Extremities are circumferentially prepared and
draped, including the most proximal shoulder or hip
e50 Tom et al Skin-Sparing for Necrotizing Infections J Am Coll Surg

Figure 3. Skin-sparing incision diagrams: lower extremity. Avoiding direct incisions over the
boney prominences, including the trochanter, ischium, patella, fibular head, anterior tibia, and
malleoli, guided the design of the incisions, emphasized with green asterisks. (A) Anterior. (1)
The thigh can be approached from any angle. These incisions allow for optimal extension from
the groin toward the knee. (2) Soft tissue can be elevated completely off the patella using a
generous soft-tissue flap (blue line) based either medially or laterally. (B) Posterior. (3) Trap-door
flaps can be designed by making perpendicular extensions along the longitudinal incision to
allow for elevation both medially and laterally and much larger exposure to underlying tissue (red
line). (4) The knee flexure surface is approached with either a curvilinear approach or with a
z-shaped incision (blue line). This protects the popliteal fossa and minimizes the risk of future
joint contracture. (C) Foot. (5) The distal incisions run anterior to the malleoli, both medial and
lateral. (D) Lateral and medial. (6) Incision over the fibular head and common peroneal nerve is
avoided. (7) Distal incisions are running anteriorly to both the malleoli. (From: Waschke J,
Böckers TM, Paulsen F. Anatomie Das Lehrbuch. 1st ed. Munich, Germany: Elsevier GmbH,
Urban & Fischer; 2015, modified with permission.)

region. Areas on the trunk might require positioning in
lateral decubitus or lithotomy, depending on the area of
concern.
The skin-sparing incision diagrams were designed as
guidelines for the initial markings for debridement, leav-
ing ultimate clinical decisions up to each surgeon. The di-
agrams are readily available in the operating room
through the clinical care algorithm website already in
wide use at our institution.
Step 1: Incision
Incisions are placed over the area of greatest clinical
concern indicated by induration, ecchymosis, blistering,
pain, or fluctuance. The placement of the incision should
be mindful of the need for likely extension to allow for
more exposure as needed. Appropriate intraoperative cul-
tures are sent. Findings indicating an NSTI described by
Anaya and Dellinger21 and others11 include gray necrotic
tissue, lack of bleeding, thrombosed vessels, “dishwater”
pus, noncontractile muscle, and a positive “finger test.”
Step 2: Skin elevation and undermining
Next, full-thickness skin flaps, including subcutaneous
fat, are elevated superficial to the necrotic myofascial
plane. In practice, the finger testdcharacterized by the
lack of resistance to finger dissection seen in normal
healthy adherent tissuedcan be used to help guide
debridement, essentially undermining the skin flaps as
Vol. 222, No. 5, May 2016 Tom et al Skin-Sparing for Necrotizing Infections e51

Figure 4. Skin-sparing incision diagrams: perineum (A) female. (1) Infections of the perineum
necessitate exposure and extension around the labia (red line). This can be extended posteriorly
along the gluteal cleft. (2) Groin, medial thigh, and anal infections are approached a few cen-
timeters away from the thin surrounding skin. The incisions can be extended distally along the
medial thigh (blue line). (B) Male. (3) Scrotal infections are approached through a midline inci-
sion that can be extended laterally at the scroto-penile junction (red line). This can be joined with
a groin incision to explore distally. (4) Similar to the female demonstration, the groin incision can
be extended posteriorly next to the anus or along the gluteal fold (blue line).

far as needed to achieve complete source control. Large
rake retractors assist with gentle elevation of the skin flaps.
Step 3: Debridement
Anything that is necrotic is removed. Overlying frankly
necrotic skin and fat are excised. Myofascial excision
can be performed en bloc. In our experience, this en-
bloc debridement can be done more swiftly and
completely than approaching the region in a piecemeal
fashion.
Perforators
The preserved skin and subcutaneous tissue maintains
vascular supply through uninterrupted local perforators
that arise just outside the dissected plane and vascular ar-
cades within the flaps. If a viable perforating vessel is
identified during the debridement, we attempt to preserve
it, if possible, to increase flap viability, but if salvaging the
perforator inhibits complete debridement the vessel is
sacrificed.
Amputation
The extreme opposite from a skin-sparing approach is
amputation of limbs. This is especially applicable with pa-
tients with extensive extremity NSTIs, particularly if it has
crossed over to infect the trunk as well. In our experience,
if a patient is in septic shock such that their hemody-
namics cannot support a skin-sparing or limb-preserving
operation, then a life-preserving amputation is likely
indicated. Amputation is also considered if the amount
of muscle debrided would effectively defunctionalize the
limb.

Step 4: Additional exposure: adjunct techniques

Table 1. Key Points for Using a Skin-Sparing Approach for
the Treatment of Necrotizing Soft-Tissue Infection
Incision design
Over the most obviously infected, discolored, or necrotic area
Place longitudinally on extremities
Avoid incisions perpendicular to flexion creases
Avoid incisions directly over boney prominences, large vessels,
and nerves
Exposure and troubleshooting
Elevate full thickness skin and subcutaneous flaps
Lengthen linear longitudinal incisions
Use trap-door or counter incisions
Debridement
Excise all necrotic tissue, leave viable tissue
Debride overlying skin if it is necrotic
Plan for repeat examinations under anesthesia
The most common hesitancy in using this skin-sparing
approach is that there could be inadequate source con-
trol. The incision designs purposely provide the surgeon
complete access to the entire subcutaneous and myofas-
cial layers. Elevation continues until healthy, contractile,
bleeding tissue is encountered within the myofascial
plane. If there is inadequate visualization or ability to
physically debride the tissue with the primary incision
over the nidus, then the following adjunct techniques
are critical to the safety and success of the skin-
sparing approach: extension, counter incisions, and
trap doors.
Extension
First, the mindful placement and direction of the primary
incision allows for easy extension. The diagrams illustrate
e52 Tom et al Skin-Sparing for Necrotizing Infections J Am Coll Surg

suggested nuances around joints and vital structures.
Making a longer incision does create a larger wound,
but this provides better access to the underlying tissue
and preserves viable skin.
Counter incision
A counter incision is a second incision placed near, and
often parallel to, the primary incision (Fig. 5A-2 for an
example). It is usually >15 cm from the primary incision
and should be positioned beyond what is comfortably
visualized and debrided through the primary incision. It
can confirm the presence or absence of continuation of
the disease process distant to the primary incision. Ideally,
it is placed along a potential skin-sparing design, such that
if necrosis is encountered, the incision can be extended
and the skin undermined to expose and excise necrotic tis-
sue (return to Step 1). This can create a wide bridge of
skin, which is termed a bipedicled flap, as the undermined
skin has 2 intact lengths of tissue or pedicles. These can
survive on the principles described.
Trap door
A trap-door or hinge flap is created by making 1 or 2
perpendicular incisions along the primary incision to
allow for inspection distal to the primary incision
(Fig. 6A-1). This can be performed on one or both sides
of a primary incision. It is particularly useful when the
primary incision is oriented longitudinally along a convex
surface, such as the thigh or flank, where necrosis can
spread medially and laterally. In essence, it allows the sur-
geon to open the skin like a book and expose the
underlying subcutaneous and myofascial planes and,
again, preserve viable skin.
RECONSTRUCTION
When patients have recovered from acute sepsis and the
wounds are deemed clean, our attention turns to wound
preparation and closure. In addition to skin grafts and
healing by secondary intention, we use the following tech-
niques to aid in reconstruction after skin-sparing debride-
ments: delayed primary closure, external tissue expanders,
and negative pressure wound therapy (NPWT). When
used together, these techniques can minimize the need
for skin grafts.
Using the skin-sparing technique, the spared skin can
be closed primarily in a delayed fashion termed delayed
primary closure (Fig. 7A, B). If possible, the skin edges
are simply reapproximated and closed in 2 or 3 layers.
Sharp excision can be necessary to freshen wound edges
if there is significant granulation tissue present. Drains
are left under skin flaps to decrease seroma formation.
Due to the natural elasticity of the spared skin, contrac-
tion occurs that creates a larger-appearing wound.
Contraction and subsequent edema can result in wound
edges that are not easily reapproximated and closed.
When this occurs, external tissue expanders can be placed
to recruit the spared skin to cover the wound through me-
chanical and biologic creep. Mechanical creep is defined
as the elongation of skin with a constant load over
time.22-24 Biologic creep is the biochemical response to a
stretch stimulus.22-24 Vessel loops can be fashioned in a
criss-cross or Roman sandal design (Fig. 7C, D) to

Figure 5. Skin-sparing case 1: thigh. (A) The debridement used (1) a soft-tissue skin bridge
(bipedicle flap) and (2) a transverse counter incision at the posterior knee. (B) Reconstruction
was accomplished in a staged fashion with (3) external tissue expanders placed. (C) Final
delayed primary closure of the entire defect (90 cm) without the need for a skin graft.
Vol. 222, No. 5, May 2016 Tom et al Skin-Sparing for Necrotizing Infections e53

Figure 6. Skin-sparing case 5: perirectal abscess. (A) Final debridement of the left buttock with
posterior thigh (1) trap-door flaps. (B) Reconstruction completed with delayed primary closure
(48 cm) and a small area (36 cm2) was left to close by secondary intention.

Figure 7. Delayed primary closure and external tissue-expander placement, “Roman sandal.” (A) Debridement result with a skin-sparing
approach for a necrotizing soft-tissue infection after intramuscular heroin injection. (B) Delayed primary closure was obtained for the
entire incision. Drains were placed distally in the dependent position. The knee flexure surface was approached with a z-shaped incision. (C)
External expanders were placed at the time of closure using a Roman sandal configuration. (D) Mechanical creep was used and the external
expanders were tightened, allowing delayed primary closure of the incision line.
e54 Tom et al Skin-Sparing for Necrotizing Infections
externally expand the spared skin. Alternatively, contin-
uous external tissue expansion systems are available as
off-the-shelf products. External tissue expanders are typi-
cally tightened every 3 to 4 days for 1 to 2 weeks to recruit
the necessary length for DPC. These expanders are often
used in conjunction with NPWT. It is important to place
the external tissue expander under the NPWT sponge.
Negative pressure wound therapy is a common modality
to prepare debrided wounds for closure. The precise mech-
anisms, both mechanical and biologic, are still being deter-
mined for open wounds. It appears to decrease wound
exudate and local edema within tissues. The mechanical
force used on the wound can assist in mechanical creep of
the wound edges. There is evidence that the negative pres-
sure can also promote a biologic milieu within the wound
that promotes healing. Additionally, the application and
changing of negative pressure wound dressings every 3 to
4 days can result in fewer dressing changes, which is partic-
ularly appreciated by patients with very large wounds.
CASES
Traditional approach: Abdominal wall
A 22-year-old man with history of IV drug use presented
with an abdominal wall fluid collection consistent with a
hematoma after an injection (Fig. 8). This developed into
Figure 8. Traditional debridement: necrotizing soft-tissue infection of abdomen. (A) The
debridement result, which was done en bloc and removed skin, subcutaneous tissue, fascia,
and muscle. (B) Reconstruction was completed with a split-thickness skin graft over the
875 cm2 defect.
J Am Coll Surg
an NSTI of the abdomen and flank. In addition to broad-
spectrum antibiotics, he was brought emergently to the
operating room for debridement. Skin and subcutaneous
tissue was excised from the right lower chest to the groin,
and from the midline to the right flank. The external obli-
que and inferior pectoralis major were necrotic and
debrided. He returned for a second debridement the
next day, with excision of additional tissue from the
pectoralis major and groin. His cultures grew methi-
cillin-sensitive Staphylococcus aureus. On post-injury day
21, after NPWT, he returned for a split-thickness skin
graft of the 25
35 cm wound.
Skin-sparing case 1: Thigh
A 58-year-old man with type 2 diabetes mellitus sustained
minimal trauma to his right thigh and an NSTI developed
subsequently (Fig. 5). His initial debridement used a skin-
sparing approach with a long, lazy-S incision and counter
incisions. The overlying subcutaneous tissue was viable
and the underlying fascia and muscle were debrided. He
underwent 3 debridements during the first 72 hours of
hospitalization. His antibiotic therapy was tailored based
on methicillin-sensitive Staphylococcus aureus growth in
tissue cultures. On post-injury day 21, he returned to
the operating room for DPC 50 cm in length, external tis-
sue expansion placement with vessel loops, and NPWT.
Vol. 222, No. 5, May 2016 Tom et al Skin-Sparing for Necrotizing Infections e55

The vessel loop ladder was tightened at the time of
NPWT dressing changes to allow recruitment of the sur-
rounding tissue. He returned to the operating room on
post-injury day 35 for DPC of the remaining 40 cm.
Skin-sparing case 2: Chest and axilla
A 41-year-old man presented with severe chest pain and
physical exam findings concerning for an NSTI. His
initial operation was a skin-sparing debridement of his
left chest wall with a midline sternotomy, pericardial
window for exploration of the retrosternal space, and
extension along the left clavicle to evaluate the
supraclavicular space. The left axilla was investigated
through an incision over the deltopectoral groove down
the lateral left chest wall anterior to the latissimus. This
allowed for evaluation of the entire left pectoralis and sur-
rounding soft tissue. He underwent 7 debridements over
10 days. His cultures ultimately grew group A

Figure 9. Skin-sparing case 2: chest. (A) Debridement required a midline sternotomy with
extension along the left clavicle and left axillary counter incision. (B) The midline wound was
reconstructed with bilateral pectoralis muscle advancement flaps and the left clavicle extension
was closed in a delayed primary fashion. (C) The inferior sternal closure failed and was ultimately
reconstructed with a vertical rectus abdominis musculocutaneous flap. (D) Six-month follow-up
photo.
e56 Tom et al Skin-Sparing for Necrotizing Infections J Am Coll Surg

Streptococcus. On hospital day 11, he underwent closure
of his sternum with sternal wires and plating system,
bilateral pectoralis muscle advancement flaps for hard-
ware coverage, and DPC of the overlying midline chest
skin. The left axillary incision was treated with external
tissue expanders and NPWT. On hospital day 32, he un-
derwent DPC of the remaining left axilla wound, which
measured 13 cm. His pectoralis advancement flaps
dehisced due to ongoing delirium and noncompliance
with sternal precautions. He ultimately underwent a
pedicled vertical rectus flap for final soft-tissue coverage
(Figs. 9 and 10).
Skin-sparing case 3: Fournier gangrene
A 41-year-old woman with diabetes mellitus type 2 pre-
sented with left labial cellulitis, progressing into an
NSTI of the perineum and abdominal wall. A skin-
sparing debridement of the left labia to the left groin
identified necrotic skin, subcutaneous tissue, and muscle
fascia with minimal myonecrosis of the external oblique.
The wound was explored 4 times over 5 days, requiring
a counter incision on her right thigh. Cultures demon-
strated a polymicrobial infection. On hospital day 17,
she returned to the operating room for DPC. The major-
ity of the 46
20 cm wound was closed in a delayed pri-
mary fashion with a NPWT dressing placed over an 8 cm2
area. This remaining wound subsequently closed via sec-
ondary intention (Fig. 11).
Skin-sparing case 4: Upper extremity
A 43-year-old woman with diabetes mellitus type 2 and
chronic forearm wounds secondary to IV drug use had
an NSTI of her left arm develop. The initial debridement
operation was performed through an 8-cm incision over
the dorsal aspect of the left upper arm. At subsequent
debridement operations, the incision was extended in a

Figure 10. Skin-sparing case 2: axilla. (A) In addition to the midline sternotomy, debridement
included a left axillary counter incision. (B) Placement of (1) external tissue expanders at the time of
midline closure. (C) The external tissue expanders were tightened during the next 14 days. (D) The
superior aspect of the wound was allowed to heal via the tightened external tissue expanders
reapproximated the edges (15 cm). The inferior aspect was closed in a delayed primary fashion
(15 cm). (E) Six-month follow-up photos with some limited range of motion of the shoulder abduction.
Vol. 222, No. 5, May 2016 Tom et al Skin-Sparing for Necrotizing Infections e57

Figure 11. Skin-sparing case 3: Fournier gangrene. (A) Anterior and (C) lateral views of final
debridement of the left labial and abdominal wall wound. (B) Anterior and (D) lateral views with a
delayed primary closure (46 cm) of the wound with 2 areas remaining open to heal by secondary
intention and allow for egress of fluid.

Z-like fashion over the antecubital fossa with the trans-
verse limb in the joint. A counter incision was made on
the more distal forearm. A soft-tissue bridge, bipedicled
flap, allowed for access to the posterior diseased fascia
and muscle. The patient required 6 subsequent debride-
ment operations removing skin, subcutaneous tissue, fas-
cia, and muscle. The majority of necrotic skin was
debrided superior to the antecubital fossa. Her wound
culture grew group A Streptococcus and her antibiotic ther-
apy was tailored. On day 21, the patient returned to the
operating room for reconstruction, with 60 cm of the
wound closed by DPC after undergoing external tissue
expansion, leaving a 150 cm2 open wound, which was
prepared with allograft skin. She ultimately return to
the operating room for split-thickness skin grafting for
definitive closure (Fig. 12).
necrotizing soft-tissue infection developed, likely from a
fistula in ano. A linear incision was created from the
gluteal clef to the posterior mid-thigh to incorporate the
areas of most concern. Medial and lateral extensions
were designed to allow for additional evaluation of the
thigh in a trap-door fashion. A rectal exam revealed an ab-
scess cavity 2 cm proximal to the anal verge. At the next
operation, a sigmoid loop colostomy was created. He un-
derwent 4 debridement operations. His culture grew
Escherichia coli. One month after the final debridement
his buttock and thigh wounds were closed with DPC.
An NPWT dressing was placed at the previous ischial
pressure sore site, where there was full-thickness skin
loss. This area healed by secondary intention by 3 months
post discharge (Fig. 6).

Skin-sparing case 5: Perirectal abscess CASE SERIES

An 80-year-old man with new-onset malaise became bed- The methods used for this study were reviewed and approved
bound and a right ischial pressure sore and subsequent by the IRB at University of Washington/Harborview
e58 Tom et al Skin-Sparing for Necrotizing Infections
Figure 12. Skin-sparing case 4: upper extremity. (A) Volar view of
the final left arm debridement included (1) counter incision on the
forearm; (2) a soft-tissue bridge (bipedicle flap); and (3) a trans-
verse incision across the flexor elbow. (B) Volar view of the final
reconstruction after multiple stages including external tissue
expanders, split-thickness skin grafting (100 cm2), and delayed
primary closure (60 cm).
Medical Center. Permission to use patient photographs was
obtained through signed consent.
A retrospective review of the medical records of 11
patients treated with a skin-sparing approach for debride-
ment for NSTI was performed. The 5 cases presented
were chosen to represent the fundamental principles of
skin-sparing debridement and reconstruction; the remain-
ing 6 of the case series are the first 6 consecutive patients
after initiating the skin-sparing approach at our institution
in November 2014. The inclusion criteria for the case series
were diagnosis of NSTI treated at Harborview Medical
Center and skin-sparing debridement approach attempted.
Patient demographic data included age and sex. Infec-
tion data collected included nidus of the infection and
microbiology result from cultures obtained. Laboratory
values were collected to calculate a Laboratory Risk Indi-
cator for Necrotizing Fasciitis score. The total score has a
range of 0 to 13. A score of <5 indicates a low risk for
NSTI; a score of 5 to 8 indicates a moderate risk for
NSTI, and a score of >8, indicates a high risk for NSTI.20
The total number of surgical interventions is divided
into debridement procedures and reconstructive proce-
dures. Each debridement operative report was read to
determine whether skin was debrided. The final recon-
structive outcomes were measured with total skin graft
area (cm2) and total length of DPC (cm) and total area
(cm2) allowed to heal by secondary intention. The percent
of the wound closed primarily was determined by review
of the operative reports and clinical photographs by 2 in-
dependent reviewers.
J Am Coll Surg
Results are listed in Table 2. Mean age of the patients
included in this series is 52 years. Four patients suffered
from perineal infections (Fournier disease). Five patients
had disease starting in the extremities. One patient’s infec-
tion started in the chest and one patient’s infections
started with perirectal abscess of the buttock.
The mean Laboratory Risk Indicator for Necrotizing
Fasciitis score at admission was 7.55, indicating a moderate
preoperative risk of an NSTI. The mean number of debride-
ment and reconstructive operations were 5.55 and 1.73,
respectively. Most of the patients (8 of 11 patients) did
require skin excision at some point during their treatment.
The ability to perform DPC was documented in all but
one of the patients. The one patient that was not able to un-
dergo DPC required reconstruction entirely with skin graft.
Of wounds able to be at least partially closed with DPC,
a mean of 87% wound closure was achieved. There were
no deaths or major amputations in this case series.
DISCUSSION
This technique paper offers guidelines for use of a skin-
sparing approach when treating patients with NSTIs. Suc-
cessful pilot of these guidelines at our institution has
shown skin-sparing debridement can provide adequate
source control and maximize future reconstruction op-
tions, as demonstrated in the case series. We recognize
that the safety of this approach is at least partially depen-
dent on the ability to return to the operating room rapidly
if clinical deterioration occurs. This can be readily accom-
plished at busy trauma centers, but might not be possible
at other institutions. In our region, we serve as a referral
center for patients with NSTIs, but frequently a patient’s
initial debridement is performed at the referring center
before transfer. A skin-sparing approach should be safe
in this setting, if the second-look procedure is planned
and necessary transfer is not delayed.
The success of this approach is dependent on the pres-
ervation of the native tissues. We acknowledge that in
subsequent debridement procedures, removing additional
skin along debridement incisions might be necessary if the
skin becomes ischemic or frankly necrotic. This can occur
in critically ill patients with severe peripheral vasoconstric-
tion during early-phase sepsis. We have not found that
this newly ischemic skin contributes substantially to sepsis
physiology if there is continued deterioration; more
frequently, the source of this deterioration is additional
extension of myofascial necrosis, which should be
addressed through extension of skin-sparing incisions or
counter incisions.
Finally, we emphasize early rehabilitation. Patients with
large resulting wound burdens are treated similarly to our
 Vol. 222, No. 5, May 2016
Table 2. Case Series with Demographics, Laboratory Findings, Surgical Data, and Reconstructive Outcomes after Initiating Skin-Sparing Technique for
Necrotizing Soft-Tissue Infection Debridement
Demographics Admit laboratory findings Debridement Reconstruction
Healing by
secondary Closed
Patient Age, LRINEC Sodium, WBC, OR Skin OR STSG, DPC, intention, primarily,
no. y Sex Location score mEq/L
109/L Organism debridement* excised reconstructiony cm2
cm cm2 %
T z, x 22 M Abdomen 5 136 23.06 MSSA 2 Yes 1 875 0 0 0
1x 80 M Buttock 3 138 17.26 Escherichia coli 4 Yes 1 0 48 36 95
2x 41 F Perineum 10 133 39.29 Polymicrobial 4 Yes 1 0 42 8 90
3x 58 M Leg 10 127 25.76 MSSA 3 No 2 0 90 0 100
4 42 M Perineum 10 130 33.39 Polymicrobial 4 Yes 2 2,600 49 0 50
5 31 M Perineum 9 137 17.74 E coli 16 Yes 1 789 99 0 90

Tom et al
6 59 F Leg 4 133 13.96 E coli 4 Yes 2 789 0 0 0
7 50 M Arm 4 134 9.25 Streptococcus milleri 2 No 2 0 36 0 100
8 77 F Perineum 12 124 39.58 Trueperella pyogenes 3 Yes 1 0 30 0 100
9 51 M Leg 10 130 22.82 Polymicrobial 8 Yes 2 1,650 60 0 75
10x 43 F Arm 6 132 23.8 Group A Streptococcus 6 Yes 2 100 60 0 80

Skin-Sparing for Necrotizing Infections

11x 41 M Chest 5 128 2.07 Group A Streptococcus 7 No 3 0 28 0 100jj
*Number of separate operations for source control.
y
Number of separate operations for reconstruction.
z
Traditional debridement patient presented, remaining numbered cases are all skin-sparing debridement cases.
x
Patient case presented in surgical technique series.
jj
Reconstruction involved a vertical rectus abdominis flap.
DPC, delayed primary closure; F, female; LRINEC, Laboratory Risk Indicator for Necrotizing Fasciitis; M, male; MSSA, methicillin-sensitive Staphylococcus aureus; OR, operating room; STSG, split-
thickness skin graft.

e59
e60 Tom et al Skin-Sparing for Necrotizing Infections
burn patients, receiving early interventions with our phys-
ical and occupational therapists soon after admission from
the operating room. Interventions include extension
splinting, range of motion exercises, and edema control.
Integrating activities of daily living and generalized
strengthening is key to enhancing quality of life. We
use a psychologist to assist patients in coping with disease,
trauma, and disfigurement. Often, the inciting event is
due to poorly controlled or previously unknown diabetes
mellitus, and our glycemic control team assists patients in
their transition from acute care to home. Similarly, if the
inciting event was an IV or intramuscular drug injection,
our facility provides addiction counseling and pain medi-
cation expertise. We believe that integrating this multidis-
ciplinary approach is essential to recovery in this complex
patient population. Qualitative and quantitative measures
of quality of life and function need to be evaluated further
to understand the true impact.
CONCLUSIONS
Skin-sparing exposure permits access to all underlying fas-
cia and musculature and preserves potential future recon-
structive avenues. The reconstructive outcomes of
skin-sparing debridement cases surpass that of traditional
wide debridement and extensive skin grafting. Our future
goals are to demonstrate safety and efficacy of this
approach, and to delineate inclusion and exclusion criteria
in specific body locations and microbiology. We will also
be investigating the impact that this has on hospital use,
cost efficiency, and patient quality of life outcomes.
Author Contributions
Study conception and design: Tom, Bulger, Pham, Keys
Acquisition of data: Tom, Wright, Horn, Pham, Keys
Drafting of manuscript: Tom, Wright, Horn, Bulger,
Pham, Keys
Critical revision: Tom, Wright, Horn, Bulger, Pham,
Keys
REFERENCES
1. Mills MK, Faraklas I, Davis C, et al. Outcomes from treat-
ment of necrotizing soft-tissue infections: results from the Na-
tional Surgical Quality Improvement Program database. Am J
Surg 2010;200:790e796; discussion 796e797.
2. Endorf FW, Klein MB, Mack CD, et al. Necrotizing soft-
tissue infections: differences in patients treated at burn centers
and non-burn centers. J Burn Care Res 2008;29:933e938.
3. Bilton BD, Zibari GB, McMillan RW, et al. Aggressive surgi-
cal management of necrotizing fasciitis serves to decrease mor-
tality: a retrospective study. Am Surg 1998;64:397e400;
discussion 400e391.
J Am Coll Surg
4. Lille ST, Sato TT, Engrav LH, et al. Necrotizing soft tissue in-
fections: obstacles in diagnosis. J Am Coll Surg 1996;182:7e11.
5. McHenry CR, Piotrowski JJ, Petrinic D, Malangoni MA. De-
terminants of mortality for necrotizing soft-tissue infections.
Ann Surg 1995;221:558e563; discussion 563e555.
6. Psoinos CM, Flahive JM, Shaw JJ, et al. Contemporary trends
in necrotizing soft-tissue infections in the United States. Sur-
gery 2013;153:819e827.
7. Laronga C, Smith P. Nipple-sparing mastectomy: an oncologic
and cosmetic perspective. Surg Oncol Clin N Am 2014;23:
549e566.
8. Patey DH, Dyson WH. The prognosis of carcinoma of the
breast in relation to the type of operation performed. Br J Can-
cer 1948;2:7e13.
9. Toth BA, Lappert P. Modified skin incisions for mastectomy:
the need for plastic surgical input in preoperative planning.
Plast Reconstr Surg 1991;87:1048e1053.
10. Didier F, Radice D, Gandini S, et al. Does nipple preservation
in mastectomy improve satisfaction with cosmetic results, psy-
chological adjustment, body image and sexuality? Breast Can-
cer Res Treat 2009;118:623e633.
11. Taylor GI, Palmer JH. The vascular territories (angiosomes) of
the body: experimental study and clinical applications. Br J
Plast Surg 1987;40:113e141.
12. Taylor GI, Minabe T. The angiosomes of the mammals and
other vertebrates. Plast Reconstr Surg 1992;89:181e215.
13. Callegari PR, Taylor GI, Caddy CM, Minabe T. An anatomic
review of the delay phenomenon: I. Experimental studies. Plast
Reconstr Surg 1992;89:397e407; discussion 417e418.
14. Dhar SC, Taylor GI. The delay phenomenon: the story un-
folds. Plast Reconstr Surg 1999;104:2079e2091.
15. Morris SF, Taylor GI. The time sequence of the delay phe-
nomenon: when is a surgical delay effective? An experimental
study. Plast Reconstr Surg 1995;95:526e533.
16. Stevens DL, Bisno AL, Chambers HF, et al. Practice guidelines
for the diagnosis and management of skin and soft tissue infec-
tions: 2014 update by the Infectious Diseases Society of Amer-
ica. Clin Infect Dis 2014;59:e10ee52.
17. Shiroff AM, Herlitz GN, Gracias VH. Necrotizing soft tissue
infections. J Intensive Care Med 2014;29:138e144.
18. Stevens DL. The pathogenesis of clostridial myonecrosis. Int J
Med Microbiol 2000;290:497e502.
19. Anaya DA, Dellinger EP. Necrotizing soft-tissue infection:
diagnosis and management. Clin Infect Dis 2007;44:
705e710.
20. Wong CH, Khin LW, Heng KS, et al. The LRINEC (Labora-
tory Risk Indicator for Necrotizing Fasciitis) score: a tool for
distinguishing necrotizing fasciitis from other soft tissue infec-
tions. Crit Care Med 2004;32:1535e1541.
21. Mullangi PK, Khardori NM. Necrotizing soft-tissue infec-
tions. Med Clin North Am 2012;96:1193e1202.
22. Wilhelmi BJ, Blackwell SJ, Mancoll JS, Phillips LG. Creep vs.
stretch: a review of the viscoelastic properties of skin. Ann Plast
Surg 1998;41:215e219.
23. Johnson TM, Lowe L, Brown MD, et al. Histology and phys-
iology of tissue expansion. J Dermatol Surg Oncol 1993;19:
1074e1078.
24. De Filippo RE, Atala A. Stretch and growth: the molecular
and physiologic influences of tissue expansion. Plast Reconstr
Surg 2002;109:2450e2462.