Keywords
necrotizing enterocolitis - abdominal compartment syndrome - vacuum-assisted closure
Introduction
Necrotizing enterocolitis (NEC) is the most common gastrointestinal emergency in the
preterm infant and is characterized by variable damage to the intestinal tract, ranging
from mucosal injury to full-thickness necrosis and perforation.[1] The majority of NEC can be managed medically using gastric decompression, bowel
rest, and intravenous antibiotics. Despite medical management, some patients develop
full-thickness necrosis of the bowel wall and perforation, requiring operative intervention.
The standard surgical treatment for perforated NEC is laparotomy and resection of
gangrenous or perforated bowel with the creation of an enterostomy and closure of
the abdomen.[2] Other options include resection with primary anastomosis, proximal diverting jejunostomy,
and the “clip and drop” technique.[3] When using the “clip and drop” technique, the bowel is resected without the creation
of a stoma or anastomosis.[4] Bowel with uncertain viability is retained during the first operation with a final
decision made at a later time.
For the past decade, vacuum-assisted closure (VAC) has become an accepted modality
for children and adults with complex wounds, abdominal wall defects, and damage control
situations for critically ill patients. The wound VAC is an efficient and inexpensive
way to temporarily close the abdomen. It provides sterile coverage of the abdominal
contents, while allowing for reduction in intra-abdominal pressures, observation of
fluid losses, decontamination of intra-abdominal spillage, and easy reentry into the
abdominal cavity. Fenton et al concluded that VAC is a safe and effective temporary
abdominal closure method in infants.[5] The purpose of this study was to compare infants with NEC treated with the “clip
and drop” technique and VAC placement with those infants treated with traditional
bowel resection, ostomy creation, and immediate abdominal closure.
Methods
After the institutional review board approval a retrospective chart review of all
patients with NEC from January 1, 2007 to December 31, 2012 at St Christopher's Hospital
for Children was performed. Data collection included patient demographics, time to
development and diagnosis of NEC, need for mechanical ventilation, time to initiation
of feeds, time to goal feeds, length of hospital stay, and any complications. Analysis
was conducted using a Student's t-test.
The use of the VAC system (VAC, Kinetic Concepts, Inc. [KCI], San Antonio, Texas,
United States) was at the discretion of the operating surgeon. A fenestrated plastic
sheet was placed over the peritoneal contents to protect the bowel. A polyurethane
foam dressing was then placed over the plastic sheeting and adherent tape was placed
over the sponge and connected to the VAC tubing to create suction. The VAC pressure
was set to 25 mm Hg. The VAC was changed every 2 to 4 days depending on the clinical
situation. The LapVac was used for varied lengths of time, depending on the clinical
situation. The VAC system was removed and the abdomen was either closed with an ostomy,
or closed after a bowel anastomosis based on the patient's clinical condition and
the surgeon's judgment.
Results
A retrospective chart review identified 26 patients with NEC requiring operative intervention.
Overall, 17 patients underwent traditional exploratory laparotomy, with intestinal
resection, creation of an ostomy, and immediate abdominal closure (group 1) and 9
patients underwent exploratory laparotomy and bowel resection leaving the bowel clipped,
but in discontinuity, followed by VAC-assisted closure (group 2). The average gestational
age of the traditional group was 29.3 ± 5.3 weeks (range, 24–40 weeks) with a recorded
birth weight of 1,389 ± 906 g (range, 606–3,720 g). The gestational age of the LapVac
group was 27.3 ± 4.6 weeks (range, 23–38 weeks) with a recorded birth weight of 1,118 ± 900 g
(range, 401–3,390 g). There was no significant difference in age or weight at initial
operation between the two groups (see [Table 1]).
Table 1
Demographics of study groups
|
Demographics
|
Group 1 (N = 17)
|
Group 2 (N = 9)
|
p-Value
|
|
Male:female ratio
|
12:5
|
5:4
|
|
|
Gestational age (wks)
|
29.3 ± 5.3 (24–40)
|
27.3 ± 4.6 (23–38)
|
0.35
|
|
Birth weight (g)
|
1,389 ± 906 (606–3,720)
|
1,118 ± 900 (401–3,390)
|
0.47
|
|
Age at index operation (d)
|
30.7 ± 21.4 (2–67)
|
25 ± 19.3 (1–51)
|
0.51
|
|
Weight at index operation (g)
|
1,783.8 ± 778.2 (900–3,700)
|
1,717.8 ± 1,093.8 (920–4,500)
|
0.86
|
Preoperative and postoperative vital signs were assessed to determine how ill patients
were before surgery and how well patients tolerated the procedure. Preoperative heart
rate, mean arterial pressure, Pao
2, and urinary output were statistically similar in both the traditional laparotomy
and LapVac groups. Postoperative vital signs, including heart rate, mean arterial
pressure, Pao
2, and urinary output were also the same between both the groups (see [Table 2]). There was no difference between the two cohorts with respect to the volume of
blood products transfused intraoperatively. Length of bowel removed in the initial
index case and total length of bowel removed for all surgeries were also statistically
similar for both the groups.
Table 2
Clinical parameters of study groups
|
Clinical parameters
|
Group 1 (N = 17)
|
Group 2 (N = 9)
|
p-Value
|
|
Preoperative heart rate (bpm)
|
165.9 ± 18.8 (133–214)
|
167.4 ± 13.3 (150–194)
|
0.83
|
|
Postoperative heart rate (bpm)
|
177.6 ± 24.2 (117–222)
|
172 ± 20.8 (144–213)
|
0.56
|
|
Preoperative mean arterial pressure (mm Hg)
|
44.5 ± 10.9 (31–68)
|
49.6 ± 11.6 (35–70)
|
0.28
|
|
Postoperative mean arterial pressure (mm Hg)
|
44.1 ± 12.0 (25–62)
|
42.6 ± 12.4 (26–61)
|
0.77
|
|
Preoperative mean airway pressure (cm H2O)
|
11.7 ± 2.8 (8–19)
|
17.6 ± 7.6 (11–30)
|
0.02
|
|
Postoperative mean airway pressure (cm H2O)
|
11.1 ± 1.4 (7–13)
|
10.7 ± 3.3 (5–17)
|
0.68
|
|
Blood products received (mL)
|
53.5 ± 49.9 (0–190)
|
41.7 ± 35.5 (0–100)
|
0.54
|
|
Length of bowel resected at index operation (cm)
|
21.6 ± 17.2 (3.5–55), n = 16
|
15.5 ± 16.5 (0–46)
|
0.40
|
|
Total length of bowel resected (cm)
|
24.2 ± 16.7 (3.5–55), n = 16
|
22.0 ± 16.7 (7.1–50)
|
0.75
|
|
Number of VAC changes
|
N/A
|
1.2 ± 1.3 (0–3)
|
N/A
|
|
Days until abdominal closure
|
N/A
|
13.1 ± 19.1 (2–63)
|
N/A
|
|
Days until bowel continuity established
|
99.8 ± 125.9 (25–180), n = 16
|
66.3 ± 39.6 (2–122)
|
0.45
|
Abbreviations: bpm, beats per minute; N/A, not applicable; VAC, vacuum-assisted closure.
There was a statistically significant difference in the two groups' mean airway pressures
both preoperatively and postoperatively. Patients who received LapVac had a higher
preoperative mean airway pressure (17.6 ± 7.6 cm H2O) compared with those in the traditional laparotomy group (11.7 ± 2.8 cm H2O; p = 0.0179). Around 24 hours after the index surgery, the LapVac patients' mean airway
pressure (10.7 ± 3.3 cm H2O) returned to a value similar to those of the traditional group (11.1 ± 1.4 cm H2O; p = 0.67). On average, the LapVac patients took approximately 13.1 days until final
abdominal closure.
Postoperative outcomes were similar for the traditional laparotomy group and the LapVac
group. Both groups had similar total days on the ventilator, similar durations on
total peripheral nutrition (TPN) or similar times until initiation of enteral feeds
(see [Table 3]). Ostomy complications were similar in both the groups. Overall 11 of 17 patients
in the traditional laparotomy group had stoma complications, and 6 of 9 patients in
the LapVac group had stoma complications. However, two of nine patients (22%) in the
LapVac group were placed back into continuity without the need for an ostomy, while
no patients in the traditional group avoided an ostomy.
Table 3
Postoperative outcomes
|
Outcome
|
Group 1 (N = 17)
|
Group 2 (N = 9)
|
p-Value
|
|
Total days on ventilator
|
44.1 ± 37.6 (4–111)
|
45.8 ± 25.6 (10–85)
|
0.90
|
|
Duration on TPN after closure (d)
|
97.7 ± 95.1 (14–360)
|
75.4 ± 69.5 (11–229)
|
0.54
|
|
Time until initiation of enteral feeds (d)
|
38.1 ± 30.7 (8–141)
|
24.9 ± 22.2 (9–74)
|
0.27
|
|
Length of hospital stay (d)
|
173.1 ± 110.3 (25–386)
|
179.6 ± 55.3 (84–2,450)
|
0.87
|
|
Ostomy complications
|
11
|
6
|
0.92
|
Abbreviation: TPN, total peripheral nutrition.
Upon review of the nine LapVac patients, we identified a subset group of patients
that had significantly better outcomes than the rest of the study patients. These
five patients had significantly higher mean airway pressures preoperatively (23.8 ± 5.8
cm H2O) versus the traditional exploratory laparotomy patients (11.7 ± 2.8 cm H2O; p < 0.0001). We considered this group (group 3) to have abdominal compartment syndrome
(ACS) at their index operation (see [Table 4]). Use of the wound VAC allowed normalization in their mean airway pressures over
the 24 hours following their operation, returning their mean airway pressures to values
similar to group 1 (see [Table 5]). Comparing group 3 patients to the traditional laparotomy group, there was no statistical
difference in blood transfusion volume, length of bowel removed in the index operation,
or the total length of bowel removed. Nor was there a difference in total days on
ventilator, or duration on TPN. However, group 3 patients had significantly shorter
times until initiation of enteral feeds and significantly shorter lengths of stay
as compared with the traditional laparotomy group (see [Table 6]). One of the five patients in group 3 did not require an ostomy before abdominal
closure.
Table 4
Demographics of traditional group versus LapVac patients with concomitant abdominal
compartment syndrome
|
Demographics
|
Group 1 (N = 17)
|
Group 3 (N = 5)
|
p-Value
|
|
Male:female ratio
|
12:5
|
3:2
|
|
|
Gestational age (wks)
|
29.3 ± 5.3 (24–40)
|
28.6 ± 5.8 (23–38)
|
0.80
|
|
Birth weight (g)
|
1,389 ± 906 (606–3,720)
|
1,444 ± 1,122 (530–3,390)
|
0.91
|
|
Age at index operation (d)
|
30.7 ± 21.4 (2–67)
|
30.8 ± 16.0 (10–47)
|
0.99
|
|
Weight at index operation (g)
|
1,783.8 ± 778.2 (900–3,700)
|
1,960 ± 1,462 (920–4,500)
|
0.72
|
Abbreviation: LapVac, treated with laparotomy and vacuum-assisted closure.
Table 5
Clinical parameters of traditional group versus LapVac patients with concomitant abdominal
compartment syndrome
|
Clinical parameters
|
Group 1 (N = 17)
|
Group 3 (N = 5)
|
p-Value
|
|
Preoperative heart rate (bpm)
|
165.9 ± 18.8 (133–214)
|
175.0 ± 14.0 (156–197)
|
0.33
|
|
Postoperative heart rate (bpm)
|
177.6 ± 24.2 (117–222)
|
180.0 ± 24.0 (158–213)
|
0.85
|
|
Preoperative mean arterial pressure (mm Hg)
|
44.5 ± 10.9 (31–68)
|
46.2 ± 9.1 (38–60)
|
0.76
|
|
Postoperative mean arterial pressure (mm Hg)
|
44.1 ± 12.0 (25–62)
|
48.2 ± 10.6 (39–61)
|
0.50
|
|
Preoperative mean airway pressure (cm H2O)
|
11.7 ± 2.8 (8–19)
|
23.8 ± 5.8 (16–30)
|
< 0.0001
|
|
Postoperative mean airway pressure (cm H2O)
|
11.1 ± 1.4 (7–13)
|
10.8 ± 4.6 (5–17)
|
0.81
|
|
Blood products received (mL)
|
53.5 ± 49.9 (0–190)
|
34.0 ± 38.5 (0–100)
|
0.43
|
|
Length of bowel resected at index operation (cm)
|
21.6 ± 17.2 (3.5–55), n = 16
|
5.9 ± 6.3 (0–15)
|
0.06
|
|
Total length of bowel resected (cm)
|
24.2 ± 16.7 (3.5–55), n = 16
|
13.9 ± 8.2 (7.1–24.3)
|
0.20
|
|
Number of VAC changes
|
N/A
|
1.0 ± 1.4 (0–3)
|
N/A
|
|
Days until abdominal closure
|
N/A
|
17.6 ± 25.8 (2–63)
|
N/A
|
|
Days until bowel continuity established
|
99.8 ± 125.9 (2–180), n = 16
|
70.2 ± 41.2 (12–122)
|
0.62
|
Abbreviations: bpm, beats per minute; LapVac, treated with laparotomy and VAC; N/A,
not applicable; VAC, vacuum-assisted closure.
Table 6
Clinical outcomes of traditional group versus LapVac patients with concomitant abdominal
compartment syndrome
|
Outcomes
|
Group 1 (N = 17)
|
Group 3 (N = 5)
|
p-Value
|
|
Total days on ventilator
|
44.1 ± 37.6 (4–111)
|
37.8 ± 22.0 (10–64)
|
0.73
|
|
Duration on TPN after closure (d)
|
97.7 ± 95.1 (14–360)
|
33.0 ± 31.0 (11–87)
|
0.16
|
|
Time until initiation of enteral feeds (d)
|
38.1 ± 30.7 (8–141)
|
9.8 ± 1.1 (9–11)
|
0.05
|
|
Length of hospital stay (d)
|
173.1 ± 110.3 (25–386)
|
148 ± 53 (84–201)
|
0.04
|
|
Ostomy complications
|
11
|
3
|
0.11
|
Abbreviation: TPN, total peripheral nutrition.
Discussion
While often treated medically, NEC can become severe and require surgical intervention
if intestinal perforation occurs or patients fail to respond to medical management.
The mainstay of treatment for surgical NEC is exploratory laparotomy with bowel resection,
creation of an ostomy, and closure of the abdomen. However, our institution has been
using abdominal wound VAC therapy as an alternate therapeutic option for infants requiring
surgery for NEC. This is the first report to compare the use of VAC therapy in surgical
NEC patients with those treated with traditional surgical principles for NEC.
We began using VAC therapy with the goal of allowing abdominal decontamination, while
the infant was recovering from the massive systemic inflammatory response seen in
infants with severe NEC. Second, we hoped to be able to reanastomose the infants'
intestine without the need for an ostomy. Avoidance of an ostomy eliminates complications
associated with an ostomy, such as dehydration from high ostomy output, skin breakdown,
and ostomy prolapse or stenosis, and also obviates the need for a second operation
to reanastomose the intestine. Our study shows the high rate of stoma complications
in this patient population with 17 of the 26 (65%) patients having either high ostomy
output or ostomy stenosis. However, use of the wound VAC was successful in preventing
the need for an ostomy in two of the nine LapVac patients (22%), while none of the
traditional laparotomy patients avoided an ostomy.
When using the VAC following intestinal resection and clipping and dropping the intestine
back into the abdomen, we noticed that the infants tolerated the postoperative period
well. Our data show that in comparison to the standard laparotomy group, the LapVac
group had statistically similar preoperative and postoperative vital signs, including
heart rate, mean arterial pressure, Pao
2, and urinary output. This indicates that neither group was more severely ill before
surgery and that both the groups tolerated the surgical interventions similarly. Neither
surgical intervention was superior in reduction of blood transfusions or preserving
bowel length. Both the groups also had similar times on the ventilator, similar times
until the initiation of feeds, and similar lengths of stay in the hospital. This data
leads us to conclude that VAC therapy is an effective and safe option in the surgical
management of NEC, and in some cases, may allow for reestablishment of bowel continuity
without the need for an ostomy.
We identified a subset of five patients in the LapVac group that improved more rapidly
than the rest of the study patients. These patients were judged to have ACS, due to
their significantly elevated mean airway pressures compared with the rest of the patients
in the study. Traditionally, ACS is determined by an elevation in bladder pressures.[6] However, in the neonatal intensive care unit, where bladder pressures are rarely
measured, elevated airway pressures are an established parameter for the diagnosis
of ACS.[7] These five patients had ACS based on elevated mean airway pressures of an average
23.6 cm H2O, which were statistically increased as compared with the rest of the patients in
the study. Use of the wound VAC in this group reduced their postoperative mean airway
pressures to a mean of 10.7 cm of H2O, which was comparable to the rest of the patients.
It is known that ACS has deleterious effects on multiple organ systems, resulting
in potential devastating patient outcomes. In this subset of patients the wound VAC
eliminated the ACS. Subsequently, these five patients had a significantly shorter
time to initiation of feeds and shorter lengths of stay when compared with patients
treated with traditional laparotomy. We believe that by treating the ACS, the wound
VAC improved this group's outcome. Admittedly, this is a small number of patients
and more research is needed to confirm that the treatment of the ACS is the reason
for the improvement in outcomes in patients with surgical NEC. However, the use of
wound VAC therapy has been shown to be an effective treatment for ACS in the critically
ill patients from many other intra-abdominal disease processes in both adults[8]
[9]
[10]
[11] and children,[12]
[13]
[14]
[15]
[16]
[17] so it seems plausible to believe that it could work in infants with NEC.
In summary, our data demonstrate that use of the wound VAC is a safe approach in the
management of premature infants with NEC requiring intervention with outcomes comparable
to standard surgical management. In some cases, use of the wound VAC may allow the
establishment of bowel continuity and abdominal closure without the need for an ostomy.
VAC therapy may also hasten the recovery of NEC patients with concomitant ACS by eliminating
the compartment syndrome. Larger studies are required to confirm this theory.
