Injury, Int. J.

Care Injured 41 (2010) 488–491

Contents lists available at ScienceDirect

Injury
journal homepage: www.elsevier.com/locate/injury

An audit of failed non-operative management of abdominal stab wounds

D.L. Clarke *, N.L. Allorto, S.R. Thomson
Pietermaritzburg Metropolitan Complex, Department of General Surgery, University of Kwa-Zulu Natal, Nelson R Mandela School of Medicine, , South Africa

A R T I C L E I N F O A B S T R A C T

Article history: Selective non-operative management based on clinical assessment has been shown to be a generally safe
Accepted 13 October 2009 approach in the management of penetrating stab wounds of the torso. However there will be a subset of
patients who fail selective non-operative management. This audit focuses on the failures.
Keywords: Methods: The metropolitan surgical service in Pietermaritzburg covers 3 hospitals. At the weekly
Failed non-operative management metropolitan morbidity and mortality meeting all trauma patients are reviewed. All cases of failed
Stab abdomen selective non-operative management of penetrating abdominal stab wounds are discussed. Failed non-
operative management is usually defined as any patient who ultimately requires surgical exploration.
We do not subscribe to this as we feel as long as the need for surgical intervention is recognised within a
short period of time (<12 h) there is little additional morbidity. Recognition of the need for surgical
intervention after 12 h would be regarded by us as failed non-operative management as we feel the risk
of delay associated morbidity begins to increase significantly after this time.
Results: A total of 340 patients with a penetrating anterior abdominal stab wound were managed over
the 2 year period under review. A total of 192 (56%) of these patients were subjected to mandatory
laparotomy. Of these mandatory laparotomies 98% were positive. The remaining 148 (44%) patients were
observed. Of the 148 observed patients a total of 30 (20%) subsequently underwent surgery. A total of 13
patients were only taken to surgery after 12 h of observation. In this group of 13 patients the average
delay between admission and recognition of injury was 40 h. There were six gastric injuries, one pyloric
and pancreatic injury, two gallbladder injuries, one liver, one colon and two small bowel injuries. There
were no deaths. 9 patients recovered with no additional morbidity. In the remainder, morbidity included,
relaparotomy (1), open abdomen (1), renal failure (1) and prolonged stay in ICU (3).
Conclusion: Clinical assessment accurately predicts the need for mandatory laparotomy following a stab
wound to the torso. In patients who do not meet the indications for mandatory laparotomy and who are
subjected to non-operative management 20% will come to surgery. A subgroup may only be recognised
as requiring surgery after more than 12 h. These patients are at risk of delay associated morbidity. There
are particular anatomical sites and structures which are prone to error.
ß 2009 Elsevier Ltd. All rights reserved.

Introduction clinical assessment alone.2,12,8,13,21,28 Patients with hypotension,

Mandatory exploration for abdominal stab wounds results in a
negative exploration rate of 20–30%. Although the associated
mortality rate is negligible, there is a significant morbidity rate
and financial cost associated with negative operative explora-
tion.22,23,16 Selective non-operative management of abdominal stab
wounds was proposed by Shaftan in 1960 and has generally been
accepted as a safe and appropriate strategy.25,27,7 This approach
states that the decision to operate or not to operate on a patient with
a penetrating knife wound to the abdomen can be made reliably by
* Corresponding author at: Department of General Surgery, Pietermaritzburg
Metropolitan Service, 201 Town Bush Road, KZN 3200, Pietermaritzburg, South
Africa. Tel.: +27 338973000; fax: +27 338450325.
E-mail address: damianclar@gmail.com (D.L. Clarke).
peritonitis, radiological evidence of hollow visceral injury or hollow
visceral evisceration require surgery. In the absence of these findings
the patient should be provided with adequate analgesia, observed
and repeatedly examined by an experienced clinician. Increasing
abdominal tenderness away from the wound, tachycardia, abdom-
inal distension or vomiting should prompt exploratory laparotomy.
Non-operative management based on clinical assessment of
abdominal stab wounds has been propagated and successfully
implemented in South Africa over the last 30 years.7 In regions
where penetrating trauma is less frequently encountered a number
of modalities have been incorporated into assessment and manage-
ment algorithms. This audit looks at the incidence of failed non-
operative management in a busy South African metropolitan
surgical service with a large trauma burden. The causes of these
failures are investigated and strategies for the appropriate manage-
ment of these patients proposed.

0020–1383/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.injury.2009.10.022
 D.L. Clarke et al. / Injury, Int. J. Care Injured 41 (2010) 488–491 489

Methods to mandatory laparotomy. Of these mandatory laparotomies 98%

The surgical service of the Pietermaritzburg metropolitan
complex covers three hospitals each funded and staffed for
different levels of care. These levels are district, regional and
tertiary. All patients with a stab wound to the anterior abdomen
and who are hypotensive, peritonitic, have eviscerated bowel or
radiological evidence of visceral perforation on presentation are
subjected to operative exploration. Eviscerated omentum is not
regarded as indication for laparotomy at our centre. The
eviscerated omentum is reduced and the patient managed further
according to his merits. In the absence of an indication for
exploration, all patients with an abdominal stab wound are
admitted to the acute wards. The patients are kept nil per mouth
and analgesia and intravenous maintenance fluids are adminis-
tered. Antibiotics are not administered routinely. The managing
surgical trainee reviews the patient at 4 hourly intervals. If clinical
signs of diffuse peritonitis develop then exploration is indicated. If
after 12 h observation there has been no deterioration in clinical
parameters the patient is commenced on graded fluids. All patients
are reviewed by a consultant surgeon the following morning.
Patients will be observed for 24–48h prior to discharge. Clinical
audit is part of the surgical service’s ongoing quality assurance
program and is based around a weekly metropolitan trauma
morbidity and mortality meeting. At this meeting each trauma case
from each of the hospitals is discussed and recorded. Failure of
selective conservatism is recorded as morbidity. We define failed
selective conservatism as an injury requiring surgery which
remains unrecognised after 12 h of in-hospital observation.
Standard demographic data on all these failures were recorded.
The nature of the missed injury was recorded as was the clinical
outcome of the patient. We did not include isolated diaphragmatic
injury as a failure of selective conservatism for the purposes of this
study as it is generally accepted that isolated diaphragmatic breach
will not manifest with clinical signs at all. All patients with an
asymptomatic left thoraco-abdominal stab wounds are offered a
laparoscopy as an elective procedure prior to discharge to exclude
diaphragmatic injury. Logistical pressures prevent this being
offered on an emergency basis. The anatomical area under review
is defined as the area between the nipple line superiorly, the pubic
symphysis inferiorly and the mid-axillary line laterally. We
received approval from the Greys Hospital Ethics Committee to
publish the results of this audit.
Results
A total of 340 patients with a penetrating anterior abdominal
stab wound were managed over the two year period (January
2007–2009). A total of 192 (56%) of these patients were subjected
were positive. The remaining 148 (44%) were subjected to non-
operative management. Of the 148 observed patients a total of 30
subsequently underwent surgery. The breakdown of visceral
injuries discovered at laparotomy as a percentage in our series
is liver 14%, stomach 14%, colon 14%, diaphragm 14%, small bowel
38%, gallbladder 2%, duodenum 2%, pancreas 1%, vascular and renal
1%. The decision to abandon non-operative management was made
after 12 h in 13 cases (Table 1). The average delay between
admission and recognition of injury in this group was 40 h (range
12–120 h). There were five stab wounds in the right upper
quadrant, four in the left upper quadrant and three epigastric stab
wounds and one in the left iliac fossa. There were six gastric
injuries, one combined pyloric and pancreatic injury, two
gallbladder injuries, one liver injury, one colonic injury and two
small bowel injuries. There were confounding variables in the
management of 4 patients. A patient with a through and through
injury of the pylorus and pancreas was observed in a regional
hospital outside of the metropolis for four days prior to discharge
and subsequently presenting to our service. Laparoscopy in this
patient revealed bile staining in the right upper quadrant and a
laparotomy was performed. A patient with a small bowel injury
and a patient with a colonic injury were observed in a district
hospital for 48 h prior to referral. Both patients required multiple
relaparotomies and prolonged ICU courses before ultimately
surviving. A patient with multiple stab wounds to his precordium
and right upper quadrant was thought to have a cardiac injury
initially. His work up included a cardiac echo and the insertion of a
central line. He was observed in high care for 24 h before an
unexplained drop in his haemoglobin prompted a sonar of his
abdomen and a laparotomy. He had a large liver laceration and a
large collection of blood and bile in his abdomen. He developed
renal failure and required dialysis. He survived with no renal
impairment. Significant morbidity was confined to the 4 patients
with confounding variables and included relaparotomy (3), open
abdomen (3), renal failure (1), and a prolonged course in ICU (3).
There were no deaths. Nine patients recovered with no additional
morbidity.
Discussion
Mandatory laparotomy for a penetrating abdominal stab
wound results in a high rate of negative exploration and has
generally been abandoned as an approach. Whilst there is
consensus that patients with active haemorrhage, shock, perito-
nitis, radiological evidence of visceral perforation or evisceration of
hollow viscera require operative exploration the assessment of
stable patients with a penetrating abdominal stab wound is more
controversial. The burden of injury in South Africa is large and the

Table 1
Patients who failed selective conservatism.

Structure Site of wound Delay (h) Grade of staff Hospital stay (days) Outcome

Stomach Epigastrum 72 Resident 6 Survived

Stomach Left upper quadrant 48 Resident 5 Survived
Stomach Left upper quadrant 24 Resident 4 Survived
Stomach and diaphragm Left lower chest 24 Resident 4 Survived
Stomach and pancreasa Right upper quadrant 120 Referring doctor 25 Survived relaparotomy Open abdomen
Stomach and diaphragm Left upper quadrant 72 Resident 4 Survived
Stomach Epigastrum 24 Resident 5 Survived
Gallbladder Right upper quadrant 24 Consultant 4 Survived
Gallbladder Right upper quadrant 24 Consultant 4 Survived
Livera Right upper quadrant 24 Consultant 29 Survived required dialysis
and precordial stab
Ascending colona Right lower chest 24 Referring doctor 36 Survived multiple relaps Open abdomen
Small bowela Epigastrum 24 Referring doctor 39 Survived multiple relaps Open abdomen
Small bowel Left iliac fossa 24 Resident 4 Survived
a
Patients with confounding factors.
490 D.L. Clarke et al. / Injury, Int. J. Care Injured 41 (2010) 488–491
resources to deal with it finite. Consequently selective non-
operative management based on repeated clinical assessment and
detection of clinical deterioration developed as a surgical
philosophy and became established.7,8,12 Since then in South
Africa selective non-operative management has remained clini-
cally driven. A recent report from a local major trauma centre
confirms that it is both a sensitive and specific approach.21 Our
own results also support the findings of these other reports.
Clinical decision making is accurate at identifying patients who
require urgent laparotomy. In our conservatively managed group
20% ultimately required surgery. This is higher than the figure from
Cape Town, however our situation is different as patients were
managed across the metropolis in one of three different hospitals.
In the group with a delay of over 12 h, 4 patients had confounding
variables. Three of the cohorts were managed in institutions
outside the metropolis for more than 48 h. In one patient with
multiple stab wounds a suspected cardiac injury diverted the
clinical focus away from the abdominal injury. Significant
morbidity was confined to these 4 patients. Delay did not translate
into significant morbidity in the remaining 9 patients. A number of
modalities and strategies have been employed in the assessment of
patients with penetrating torso trauma. These include local wound
exploration (LWE), diagnostic peritoneal lavage (DPL), abdominal
CT scan (CT), abdominal ultrasound (AUS) and diagnostic laparo-
scopy (DL).6
DPL has been used to assess penetrating abdominal stab
wounds.10,18–20 However the indications for exploration based on
the analysis of the aspirate are not universal. Muckart and
McDonald demonstrated that the criteria for a positive DPL in blunt
abdominal trauma were not applicable to penetrating trauma.18
They reported better results using protein electrophoresis and
urine dipstick assessment of DPL fluid.19,20 However they
remained unconvinced of the usefulness of DPL in this setting.
Introducing the peritoneal catheter carries a small but significant
morbidity rate and once fluid has been instilled into the peritoneal
cavity it is difficult to re-assess the abdomen as both clinical signs
and radiological findings are altered by the presence of the infused
fluid. Consequently DPL has not been adopted in our centre for the
assessment of penetrating abdominal trauma. Some authors have
used DPL as a screening tool to avoid unnecessary admission to
hospital.11 A negative white cell count and a red cell count of less
than 1000 RBCs/mm3 in a DPL aspirate has been shown to be a
reliable screening tool which allows safe discharge from the
emergency room. LWE aims to identify patients in which
peritoneal breach has occurred. If the LWE confirms peritoneal
breach then laparotomy or laparoscopy is advocated.9 If peritoneal
breach is excluded then the patient may be discharged. We have
not adopted LWE exploration in our algorithms as we do not
believe that peritoneal breach equates to the presence of an
intraperitoneal visceral injury. Furthermore LWE may be difficult
in obese or muscular individuals.24 Abdominal ultrasound is too
insensitive to establish a role in the assessment of stable patients
with abdominal stab wounds.26,4 It is useful in the assessment of
unstable patients who need urgent operative intervention.
Traditionally CT scan has been viewed with a degree of skepticism
as it was believed to be inaccurate in the assessment of enteric
injury.18,24 It has been described as ‘‘superfluous, time consuming,
extravagant and unreliable’’.18 However the rapid evolution of CT
scan technology has necessitated ongoing re-evaluation of trauma
management algorithms.24 Modern triple-contrast enhanced
spiral CT scans can reliably demonstrate peritoneal breach,
extraluminal air and fluid, extravasation of radiographic contrast
and visceral injury. It is a single rapid non-invasive investigation
that accurately predicts the need for surgical exploration. The
limitations include the risk of contrast induced nephropathy, and
the inability to assess the diaphragm and especially in our
environment logistical limitations. Minimally invasive DL allows
the abdominal cavity to be directly inspected without the need for
formal laparotomy.14 This requires a formal general anaesthetic has
a low morbidity and is well tolerated. There are concerns about the
reliability of laparoscopy to detect intra-abdominal injury. In some
algorithms it is used to confirm or refute peritoneal breach. If breach
is confirmed then some algorithms advocate formal laparotomy.
Leppaniemi and Haapiainen showed in patients with proven
peritoneal breach that laparoscopy offered little advantage over
exploration.14 In patients in whom it is unclear that peritoneal
breach had occurred laparoscopy detected more minor injuries than
repeat clinical assessment but was associated with longer hospital
stay and higher costs. They felt that they could not recommend
laparoscopy as a routine diagnostic tool in anterolateral abdominal
stab wounds. Ertekin et al. performed DL in 38 patients with anterior
abdominal stab wounds. DL provided useful information in six.9
Although the predominant indication was for left sided thoraco-
abdominal wounds, these authors also included patients with
equivocal abdominal findings and omental or enteric herniation. DL
diagnosed three small bowel injuries, two colonic injuries one gastric
and two diaphragmatic injuries in their series. Similarly Ahmed et al.
in a study that included patients with gunshot wounds to the
abdomen found twenty injuries using DL which included stomach
(6), colonic (2), small bowel (4), and liver (5).1 The multitude of
approaches to the problem of a penetrating anterior abdominal stab
wound reflects a general lack of consensus about the ideal algorithm
by which to manage these patients.3 All modalities have demon-
strated incidences of false negative and false positive results.
It would appear that most trauma centres in South Africa are
adept at clinical examination. Predicting the need for urgent
laparotomy on clinical grounds alone is a safe approach. Non-
operative management is not a policy that can be applied without
any consideration of available resources as it has always rested
upon active serial examination by experienced clinicians. When
undertaken in a district hospital it is problematic. This is
demonstrated by the two cases in our series where referring
doctors observed two patients with a colonic and small bowel
injury. We have however highlighted a sub group of injuries in
which serial examination may be misleading even for experienced
clinicians. Penetrating trauma to the left upper quadrant and left
thoraco-abdominal region may be considered as the ‘‘Achilles
Heel’’ of selective non-operative management as serial examina-
tion may be misleading. In our series the most common visceral
injury to result in failure of selective non-operative management
was perforation of the stomach. There are several reasons why a
gastric perforation may not manifest overt clinical signs. Depend-
ing on the temporal relationship between the patient’s last meal
and the time of injury, the stomach milieu is sterile. This implies
that leakage of gastric contents may not evoke the same degree of
peritonitis associated with fecal or small bowel content induced
peritonitis. Gastric leakage into the naturally dependent cavities of
the lesser sac, splenic bed or subhepatic space may result in a
localised collection, rather than diffuse peritoneal irritation. In the
six incidents in our series of gastric perforations the operative
findings were of localised fluid collections with limited diffuse
peritonitis. Injury to the extra-hepatic biliary tract may not result
in the development of peritonitis. Delayed recognition of biliary
collections is well documented following laparoscopic cholecys-
tectomy, as sterile bile may not elicit marked peritonitis and tends
to accumulate in naturally dependent cavities.5 Similarly, free
peritoneal haemorrhage secondary to a liver injury may not evoke
a dramatic peritoneal reaction.17
Peritoneal irritation does not appear to be an absolutely reliable
sign of injury to the stomach, gallbladder or liver. Other clinical
parameters may be helpful and signs of an inflammatory response
must raise the suspicion of an intra-abdominal injury. An
 D.L. Clarke et al. / Injury, Int. J. Care Injured 41 (2010) 488–491
inflammatory response has been shown to represent the devel-
opment of a complicated haemoperitoneum requiring intervention
in patients with blunt hepatic trauma.17 Subtle changes in clinical
parameters are occasionally overlooked in our setting leading to
error. Introducing a formal serial observation chart may help
reduce the potential for these errors of judgement. Although we
have excluded isolated diaphragmatic breach from our definition
of failed selective conservatism it is an important consideration in
designing trauma algorithms. Isolated diaphragmatic injury will
escape serial clinical examination as it will not manifest with any
clinical signs. Radiology and DPL are unhelpful. The reported
incidence of diaphragmatic injury following penetrating left
thoraco-abdominal trauma ranges from 10 to 30%.15 In light of
this and the concerns about the possibility of a gastric injury which
this audit highlights we advise mandatory laparoscopy for all
penetrating stab wounds to the left thoraco-abdominal region. For
the vast majority of patients with stab wounds to the right upper
quadrant and epigastric area non-operative management based on
repeated clinical assessment is adequate. In a subgroup with
unexplained signs of inflammation or with multiple distracting
injuries judicious use of DL or CT scan may be helpful. Laparoscopy
must look for peritoneal breach, evidence of haemorrhage, enteric
contents or staining of the tissues. Bile or enteric content staining
implies visceral injury and should trigger laparotomy. Solid organ
injury can be managed on its merits.
Conclusion
Selective non-operative management of abdominal stab
wounds is a safe and effective strategy. However there are specific
areas where a clinician may be misled. Injury to structures in the
upper gastro-intestinal tract may not manifest with clinical
peritonitis. Clinical assessment alone in patients with upper
abdominal stab wounds may be inadequate and judicious use must
be made of special investigations. All stab wounds to the left
thoraco-abdominal region should be subjected to laparoscopy as
there is a significant incidence of isolated diaphragmatic injury and
there is the potential for gastric perforation that may not be
clinically apparent. In cases of penetrating trauma to the
epigastrum and right upper quadrant paying attention to signs
of physiological derangements may alert the physician to the
potential for intra-abdominal injury. Pyrexia, tachycardia, dis-
comfort or leucocytosis should prompt laparoscopy or CT scan. At
laparoscopy the surgeon must confirm peritoneal breach and look
for signs of staining of tissues by either blood or enteric content.
Enteric staining should trigger a laparotomy. Selective non-
operative management of abdominal stab wounds requires
experienced clinicians to be undertaken safely and should not
be adopted in a centre that does not have the necessary human
resources. The use of protocols and formal observation charts may
help to further reduce the low incidence of failed selective non-
operative management in our environment.
Conflict of interests
The authors have no conflicting interests to declare.
491
References
1. Ahmed N, Whelan J, Brownlee J, et al. The contribution of laparoscopy in
evaluation of penetrating abdominal wounds. J Am Coll Surg 2005;201(August
(2)):213–6.
2. Alzamel HA, Cohn SM. When is it safe to discharge asymptomatic patients with
abdominal stab wounds? J Trauma 2005;58(March (3)):523–5.
3. Biffl WL, Kaups KL, Cothren CC, et al. Management of patients with anterior
abdominal stab wounds: a Western trauma association multicenter trial. J
Trauma 2009;66(May (5)):1294–301.
4. Boulanger B, Kearney PA, Tsuei B, Ochoa JB. The routine use of sonography in
penetrating torso injury is beneficial. J Trauma 2001;51:320–5.
5. Braghetto I, Bastias J, Csendes A, Debandi A. Intraperitoneal bile collections after
laparoscopic cholecystectomy: causes, clinical presentation, diagnosis, and
treatment. Surg Endosc 2000;14(November (11)):1037–41.
6. Chihombori A, Hoover EL, Phillips T, et al. Role of diagnostic techniques in the
initial evaluation of stab wounds to the anterior abdomen, back, and flank. J Natl
Med Assoc 1991;83(February (2)):137–40.
7. Clarke DL, Thomson SR, Madiba TE, Muckart DJ. Selective conservatism in
trauma management: a South African contribution. World J Surg 2005;29:
962–5.
8. Demetriades D, Rabinowitz B. Indications for operation in abdominal stab
wounds. ‘A prospective study of 651 patients’. Ann Surg 1987;203:129–32.
9. Ertekin C, Yanar H, Taviloglu K, et al. Unnecessary laparotomy by using physical
examination and different diagnostic modalities for penetrating abdominal
stab wounds. Emerg Med J 2005;22(November (11)):790–4.
10. Feliciano DV, Bitondo-Dyer CG. Vagaries of the lavage white blood cell count in
evaluating abdominal stab wounds. Am J Surg 1994;168:680–3.
11. Gonzalez RP, Turk B, Falimirski ME, Holevar MR. Abdominal stab wounds:
diagnostic peritoneal lavage criteria for emergency room discharge. J Trauma
2001;51(November (5)):939–43.
12. Huizinga WK, Baker LW, Mtshali ZW. Selective management of abdominal and
thoracic stab wounds with established peritoneal penetration: the eviscerated
omentum. Am J Surg 1987;153:564–8.
13. Leppaniemi AK, Haapiainen RK. Selective nonoperative management of abdom-
inal stab wounds: prospective, randomized study. World J Surg 1996;20:
1101–5.
14. Leppaniemi A, Haapiainen R. Diagnostic laparoscopy in abdominal stab
wounds: a prospective, randomized study. J Trauma 2003;55:636–45.
15. Leppaniemi A, Haapiainen R. Occult diaphragmatic injuries caused by stab
wounds. J Trauma 2003;55(4):646–50.
16. Leppaniemi A, Salo J, Haapiainen R. Complications of negative laparotomy for
truncal stab wounds. J Trauma 1995;38:54–8.
17. Letoublon C, Chen Y, Arvieux C, et al. Delayed celiotomy or laparoscopy as part
of the nonoperative management of blunt hepatic trauma. World J Surg
2008;32(June (6)):1189–93.
18. Muckart DJ, McDonald MA. Unreliability of standard quantitative criteria in
diagnostic peritoneal lavage performed for suspected penetrating abdominal
stab wounds. Am J Surg 1991;162(September (3)):223–7.
19. Muckart DJ, McDonald MA. Evaluation of diagnostic peritoneal lavage in
suspected penetrating abdominal stab wounds using a dipstick technique. Br
J Surg 1991;78(June (6)):696–8.
20. Muckart DJ, McDonald MA. Protein electrophoresis of diagnostic peritoneal
lavage in penetrating abdominal trauma: a marker of significant injury. Br J
Surg 1989;76(December (12)):1308–10.
21. Navasaria PH, Berli JU, Edu S, Nicol AJ. Non-operative management of abdom-
inal stab wounds—an analysis of 186 patients. S Afr J Surg 2007;45(November
(4)). 128–30, 132.
22. Renz BM, Feliciano DV. Unnecessary laparotomy for trauma: a prospective
study of morbidity. J Trauma 1995;38:350.
23. Ross SE, Dragon GM, O’Malley KF, Rehm CG. Morbidity of negative coeliotomy
in trauma. Injury 1995;26:393–4.
24. Salim A, Sangthong B, Martin M, et al. Use of computed tomography in anterior
abdominal stab wounds: results of a prospective study. Arch Surg 2006;141
(August (8)):745–50.
25. Shaftan GW. Indications for operation in abdominal trauma. Am J Surg 1960;99:
657–64.
26. Soffer D, McKenney MG, Cohn S. A prospective evaluation of ultrasonography
for the diagnosis of penetrating torso injury. J Trauma 2004;56:953–7.
27. Thompson JS, Moore EE, Wan Duzer-Moore S, et al. The evolution of abdominal
stab wound management. J Trauma 1980;20(June (6)):478–84.
28. Tsikitis V, Biffl WL, Majercik S, et al. Selective clinical management of anterior
abdominal stab wounds. Am J Surg 2004;188(December (6)):807–12.