Rapid Source-Control Laparotomy: Is There a Mortality Benefit?

Abstract

Background:

The purpose of this study was to determine the influence rapid source-control laparotomy (RSCL) has on the mortality rate in non-trauma patients with intra-abdominal infection. The hypothesis was that RSCL reduces deaths and hospital lengths of stay (LOS) in patients compared with definitive repair and primary fascial closure (PFC).

Methods:

The International Classification of Diseases-10 codes for sepsis, gastric and duodenal ulcer perforation or hemorrhage, incisional or ventral hernia with obstruction, intestinal volvulus, ileus with obstruction, diverticulitis with perforation or abscess, vascular disorder of intestine, non-traumatic intestinal perforation, peritoneal abscess, and unspecified peritonitis were used to query the 2015 National Surgical Quality Improvement Project (NSQIP) database for all patients treated with either RSCL or PFC. The two groups of patients were compared on the basis of LOS and deaths. Collected data included age, gender, body mass index (BMI), site classification, American Society of Anesthesiologists (ASA) class, operative time, number of risk factors, and pre-operative septic state.

Results:

After adjusting for the aforementioned variables, propensity score-matched cohorts (n = 210 in each cohort) were used to evaluate the influence of incision closure type on LOS and mortality rate. The odds of death (31.4% vs. 21.4%) with RSCL was 1.78 (95% confidence interval 1.08–2.95; p = 0.02) times that of PFC. Closure type was not significantly associated with an increased LOS (median 14 vs. 11 days; p = 0.35).

Conclusions:

This retrospective cohort analysis demonstrated that RSCL is associated with higher odds of death in general surgical patients with intra-abdominal infection. There is a need for further studies to delineate what, if any, physiologic parameters indicate a need for RSCL.

The premise of damage-control laparotomy (DCL) in abdominal trauma was first documented by Ogilvie in 1940 [1]. The initial report described Vaseline-impregnated canvas as a temporary closure mechanism in trauma patients with an abdominal wall too tense to be closed primarily. Ogilvie proceeded, in 1945, to apply the principle of primary abdominal closure with planned re-exploration to general surgical patients with intra-abdominal infection [2]. Thereafter, this management protocol was abandoned until 1983, when Stone et al. [3] demonstrated a mortality benefit for an abbreviated laparotomy in patients with blunt or penetrating abdominal trauma who had the lethal triad of hypothermia, metabolic acidosis, and coagulopathy. The initial, temporizing procedure is followed by aggressive resuscitation in the intensive care unit (ICU), and definitive closure is performed within 48 h pending successful resuscitation [4,5]. The term, DCL, for this procedure was coined in 1992 by Rotondo et al. [6]. The extension of DCL to control the metabolic acidosis, coagulopathy, and body temperature derangement of septic shock in general surgery patients has been termed “rapid source-control laparotomy” (RSCL) [7]. Although some studies demonstrate a role for RSCL in the management of necrotizing pancreatitis [8], intra-abdominal hemorrhage [9], and complicated diverticulitis [10], its general acceptance in the management of other intra-abdominal infections is unclear.

The purpose of this study was to determine the influence RSCL has on the mortality rate and hospital length of stay (LOS) in non-trauma patients with intra-abdominal infection. The hypothesis was that RSCL reduces deaths and morbidity in patients with intra-abdominal infection compared with definitive repair and primary fascial closure (PFC).

Patients and Methods

Patient selection criteria

This was a retrospective cohort analysis based on the 2015 NSQIP annual database. Inclusion criteria were performance of urgent RSCL or PFC for the International Classification of Disease (ICD)-10 codes listed in Table 1. Patients with an incision classification of clean were excluded.

Table 1.

Indications for Intervention in RSCL vs. PFC Cohorts

ICD-10 Code	Diagnosis	RSCL cohort	% of total^a	PFC cohort	% of total^b
A41.90	Sepsis	77	36.67	77	36.67
K25.10	Acute gastric ulcer perforation	3	1.43	3	1.43
K43.00	Incisional hernia with obstruction	5	2.38	5	2.38
K43.60	Ventral hernia with obstruction	3	1.43	3	1.43
K55.10	Chronic vascular disorder of intestine	2	0.95	2	0.95
K55.90	Vascular disorder of intestine, not specified	21	10.00	21	10.00
K56.20	Intestinal volvulus	6	2.86	6	2.86
K56.50	Intestinal adhesions with obstruction	8	3.81	8	3.81
K56.60	Intestinal obstruction not specified	3	1.43	3	1.43
K56.69	Ileus with obstruction	9	4.29	9	4.29
K57.20	Diverticulitis of large intestine with perforation or abscess	14	6.67	14	6.67
K57.80	Diverticulitis of intestine with perforation or abscess	3	1.43	3	1.43
K63.10	Non-traumatic perforation of intestine	45	21.43	45	21.43
K65.10	Peritoneal abscess	6	2.86	6	2.86
K65.90	Peritonitis, not specified	5	2.38	5	2.38

Expressed as percent total of RSCL cohort.

Expressed as percent total of PFC cohort.

ICD = International Classification of Diseases; PFC = primary fascial closure; RSCL = rapid source-control laparotomy.

Collected data were age (binary variable: <65 vs. ≥65 y), gender, body mass index (BMI), incision classification (binary variable: clean/contaminated vs. contaminated or dirty/infected), American Society of Anesthesiologists (ASA) class (binary variable: 0–2 vs. 3⁺), operative time, presence of pre-operative sepsis or post-operative pneumonia, and number of risk factors (binary variable: 0–3 vs. 4⁺). The risk factors examined were the presence of diabetes mellitus, alcohol or tobacco abuse, blood dyscrasias, disseminated cancer, or cardiac, gastrointestinal, pulmonary, hepatobiliary, or renal dysfunction. The primary outcomes were LOS and death before 30 d. Given the influence of these variables on death and LOS, RSCL patients were matched with PFC patients based on propensity scores for ICD-10 code, number of risk factors, and presence of pre-operative sepsis.

Statistics

All statistical analyses were conducted using the Statistica software package, version 13 (Dell Inc., 1984-2015) and R (The R Foundation for Statistical Computing, 2015). Baseline patient demographic characteristics, incision classification, ASA class, and operative time were compared for the RSCL and PFC groups using Mann Whitney U or χ² tests, depending on variable classification. Similar bivariable analyses were conducted to evaluate associations between incision closure type, LOS, and death. The results of these analyses were used to build generalized linear regression models for death (using the logit link) and LOS (using the identity link) to examine the influence of incision closure type on these outcomes while controlling for additional covariables. Mortality model selection was based on the Hosmer-Lemeshow Goodness of Fit statistic, whereas model selection for LOS relied on Akaike's Information Criterion. Statistical significance was defined as α = 0.05.

Results

A total of 420 propensity score-matched patients were identified for analysis, of whom 210 (50%) each underwent RSCL and PFC.

Diagnostic codes

Table 1 lists the ICD-10 codes and indications for intervention in the study cohorts. The most common indications for intervention were sepsis (36.67%) and non-traumatic intestinal perforation (21.43%).

General demographics

There were no significant differences in median age (p = 0.05; 95% confidence interval [CI] 0.46–1.0), male:female ratio (p = 0.43), or BMI (p = 0.36) between the study cohorts (Table 2).

Table 2.

Baseline Patient Characteristics in RSCL vs. PFC Cohorts

	RSCL cohort	PFC cohort	Statistical significance
No. of patients	210 (50.00)^a	210 (50.00)^a	–
Age (y)			P = 0.05^d
<65	109 (51.90)^b	89 (42.38%)^c	OR 0.68
≥65	101 (48.10)^b	121 (57.62%)^c	(0.46–1.00)
M:F ratio (%)	105:105 (50:50)	97:113 (46.2:53.8)	P = 0.43^d
			OR 0.86
			(95% CI 0.59–1.26)
BMI	21.18 (13.08–31.37)^e	24.13 (15.18–31.27)^e	p = 0.36^f
Number of risk factors			p = 1.00^d
0–3	120 (57.14)^b	120 (57.14%)^c	OR 1.00
4⁺	90 (42.86)^b	90 (42.9%)^c	(95% CI 0.68–1.47)

Expressed as percentage of total patient population.

Expressed as number of patients (percent of total RSCL group).

Expressed as number of patients (percent of total PFC group).

Calculated using Pearson χ² test.

Expressed as median (interquartile range).

Calculated using Mann-Whitney U test.

CI = confidence interval; PFC = primary fascial closure; OR = odds ratio; RSCL = rapid source-control laparotomy.

Operative risk

Pre-operative incision classification demonstrated a greater prevalence of contaminated/dirty incisions and 3⁺ ASA class in the RSCL cohort, although the difference for ASA was not significant (89.50% vs. 74.80%; p < 0.0001 and 60.50% vs. 51.90%; p < 0.08, respectively). Consistent with the rapid source-control approach, there was a significant reduction in median operative time (82 vs. 104.5 min; p = 0.0008). Results are presented in Table 3.

Table 3.

Operative Risk in RSCL vs. PFC Cohorts

	RSCL cohort	PFC cohort	Statistical significance
Site classification			p < 0.0001^a
Clean/contaminated	22 (10.48)^c	53 (25.24)^d	OR = 2.88
Contaminated or dirty/infected	188 (89.52)^c	157 (74.76)^d	(CI 1.68–4.95)
ASA class			p = 0.08^a
1–2	83 (39.52)^c	101 (48.10)^d	OR = 1.42
3⁺	127 (60.48)^c	109 (51.90)^d	(CI 0.96–2.09)
Operative time	82 (49–125)^e	104.5 (70–141)^e	p = 0.0008^b
Presence of sepsis	186 (88.57)^c	186 (88.57)^d	p = 1.00
			OR = 1.00
			(CI 0.55–1.82)

Calculated using Pearson χ² test.

Calculated using Mann-Whitney U test.

Expressed as number of patients (percent of total RSCL group).

Expressed as number of patients (percent of total PFC group).

Median (interquartile range) operative time (min).

ASA = American Society of Anesthesiologists; OR = odds ratio; PFC = primary fascial closure; RSCL = rapid source-control laparotomy.

Rate of re-operation

Bivariable comparisons of the rate of re-operation between propensity score-matched RSCL and PFC patients did not reveal a significant difference (12.86 vs. 15.24%; p = 0.57).

Hospital LOS

Bivariable comparisons of LOS (d) between propensity score-matched RSCL and PFC patients did not a reveal significant difference (median 14 vs. 11 d; p = 0.35). For this reason, a generalized linear model adjusting for demographic and operative risk variables was not constructed (Table 4).

Table 4.

Primary Outcomes

	RSCL cohort	PFC cohort	Statistical significance
Rate of reoperation	27 (12.86)^a	32 (15.24)^b	p = 0.57
Length of stay	14 (7–22)^c	11 (6–20)^c	0.35
Mortality			p = 0.02
Alive	144 (68.57)^a	165 (78.57)^b	OR = 1.68
Dead	66 (31.43)^a	45 (21.43)^b	(CI 1.08–2.61)

Expressed as number of patients (percent of total RSCL group).

Expressed as number of patients (percent of total PFC group).

Expressed as median number of days (interquartile range).

CI = confidence interval; OR = odds ratio; PFC = primary fascial closure; RSCL = rapid source-control laparotomy.

Mortality outcome

A logistic regression model adjusting for age, ASA class, number of risk factors, presence of pre-operative sepsis, operative time, and incision classification demonstrated the odds of death among patients undergoing RSCL was 1.78 (95% CI 1.08–2.95; p = 0.02) times that of PFC patients. The independent risk factors for death are presented in Table 5.

Table 5.

Logistic Regression: Independent Risk Factors for Death

Variable	Death (n = 111) N (% of category)	p	Odds ratio	95% CI
ASA score		<0.0001
0–2	20 (10.87)		Ref
≥3	91 (38.56)		3.72	2.13–6.51
Risk factors		0.0008
0–3	42 (17.50)		Ref
≥4	69 (38.33)		2.31	1.41–3.77
Age (y)		0.0009
<65	34 (17.17)		Ref
≥65	77 (34.68)		2.37	1.42–3.93
Incision closure type		0.02
PFC	45 (21.43)		Ref
RSCL	66 (31.43)		1.78	1.08–2.95
Incision classification		0.23
Clean contaminated	12 (16.00)		Ref
Contaminated or dirty/infected	99 (28.70)		1.56	0.75–3.24
Operative time	–	0.27	–	–
Pre-operative sepsis		0.30
No	6 (12.50)		Ref
Yes	105 (28.23)		1.68	0.64–4.43

ASA = American Society of Anesthesiologists; CI = confidence interval; PFC = primary fascial closure; RSCL = rapid source-control laparotomy.

Discussion

The results of this retrospective study demonstrated that despite a reduction in operative time and presumed reduction in further physiologic derangement, after adjustment for confounding variables (age, ASA score, site classification, post-operative pneumonia, number of risk factors, and pre-operative sepsis), RSCP conferred an odds of death approximately 1.7 times that of PFC. The principles of DCL arose from a need to correct the coagulopathy associated with the lethal triad. The metabolic acidosis is a result of reduced tissue perfusion and concomitant anaerobic metabolism. The resultant pH decline increases fibrinogen consumption and decreases pro-coagulant complex interaction. Hypothermia alters the von Willebrand factor and glycoprotein Ib complex interaction, impairing platelet aggregation. Coagulopathy intrinsic to hemorrhagic shock arises from consumption and dilution of clotting factors secondary to crystalloid resuscitation. It is likely that this triad of coagulation dysfunction does not apply to the pro-inflammatory state of intra-abdominal infection [11 -13].

Sepsis, unlike hemorrhagic shock, is a hypercoagulable state characterized by release of pro-inflammatory cytokines, activation of the extrinsic cascade through tissue factor, inhibition of fibrinolysis through the protein C/S cascade, and consumption of anti-thrombin [14]. The result is metabolic acidosis secondary to microthrombus formation, infarction, and tissue hypoxia [15]. The cardiac function in sepsis is classically hyperdynamic, with a relatively preserved cardiac output despite decreases in stroke volume and ejection fraction [16,17]. The massive peripheral vasodilation and myocardial depression, mediated by bacterial endotoxin and cytokine release, respectively, permit adequate end-organ perfusion. It is not until septic shock, with evidence of systolic and diastolic myocardial dysfunction, intervenes that end-organ perfusion is compromised [15]. These hemodynamic changes necessitate massive resuscitation with crystalloid products to maintain circulatory volume and promote dilutional coagulopathy. This differs vastly from the fluid resuscitation protocol in trauma, which promotes permissive hypotension and restricted fluid resuscitation to limit dilutional coagulopathy [18].

The results of this study validate the intra-abdominal infection guidelines set forth by Mazuski et al. [19]. Both reports suggest there are limited observational data to support any role for RSCL in the management of intra-abdominal infection. Several studies have compared the observed mortality rate in RSCL with that predicted by the Simplified Acute Physiologic Score (SAPS II) and Acute Physiology and Chronic Health Evaluation (APACHE) scoring systems. These studies failed to demonstrate a mortality benefit from RSCL, although they are limited by use of mortality models that allow clinical data to be obtained within 24 h of ICU admission [9,20,21]. This is not ideal, as there is a need to identify pre-operative physiologic factors that necessitate an RSCL in general surgery patients. This issue was partially addressed with the application of the Physiological and Operative Severity Score for the Enumeration of Mortality and Morbidity (POSSUM) scoring system. In one study, a case series by Finlay et al. demonstrated a mortality reduction by the POSSUM and P-POSSUM scoring systems (7% vs. 64%; p = 0.002 and 7% vs. 50%; p = 0.038, respectively) [22]. However, a larger retrospective cohort by Stawicki et al. [9] failed to demonstrate a mortality benefit.

To the best of our knowledge, there are three studies in the Western literature that provide a direct comparison between the RSCL and PFC techniques. Two are retrospective cohort studies, by Becher et al. and Person et al. [7, 23], both of which demonstrate an increase in the in-hospital mortality rate in patients undergoing RSCL. Becher et al. [7] performed a further subgroup analysis examining only patients in septic shock who underwent either RSCL or PFC. Even in these critically ill patients, RSCL failed to show a mortality benefit. The authors went on to recommend that the indication for RSCL is not a mortality reduction but rather ease of re-operation, as 50% of patients with PFC require un-planned re-exploration. There is one prospective study regarding RSCL and PFC in general surgery patients. In this study, Christou et al. could not demonstrate a mortality benefit in their 18 patients who underwent RSCP [24]. The results of all these studies are summarized in Table 6.

Table 6.

Literature Review on RSCL versus PFC in Non-Trauma Patients

Author	Study type	Number of patients	Primary outcome	Result
Adkins et al. [20]	Retrospective cohort with matched historical controls	81	In-hospital death in RSCL vs. age/sex/APACHE III-matched historical controls	33% vs. 25%; p > 0.05
Becher et al. [7]	Retrospective cohort	215 (53)^a	1) In hospital death in RSCL vs. PFC cohorts	1) 45% vs. 20%; p = 0.0004
			2) In hospital death in subgroup analysis for septic shock (RSCL vs. PFC) cohorts	2) 60% vs. 53%; p = 0.68
Cipolla et al. [5]	Case control	17	28-day deaths	6%
Christou et al. [24]	Prospective cohort	239 (18)^a	In-hospital death in RSCL vs. PFC cohorts	44% vs. 31%; p = 1.33
Finlay, Edwards & Lambert [22]	Case series	14	In-hospital death vs. POSSUM and P-POSSUM predicted deaths	1) 7% vs. 64%; p = 0.002
Finlay, Edwards & Lambert [22]	Case series	14	In-hospital death vs. POSSUM and P-POSSUM predicted deaths	2) 7% vs. 50%; p = 0.038
Jusoh and Yanzie [29]	Case series	6	In-hospital vs. POSSUM predicted mortality rate	33% vs. 50%; statistical significance calculations not performed
Person et al. [23]	Retrospective cohort	291 (31)^a	In-hospital death in abbreviated laparotomy (AL) vs. definitive laparotomy (DL) cohorts	55% vs. 17%; p < 0.0001
Stawicki et al. [9]	Retrospective cohort	16	Observed 28-day vs. predicted mortality rate from POSSUM and APACHE II	1) 43% vs. 75%; p = 0.07
Stawicki et al. [9]	Retrospective cohort	16		2) 43% vs. 63%; p = 0.24
Stawicki, Cipolla, and Bria [21]	Retrospective cohort	60 (35)	Observed 28-day mortality rate vs. predicted rate from SAPS II	17% vs. 40%; statistical significance calculations not performed
Tsuei et al. [30]	Case control	46	In-hospital death	43%

Reported as total patient number (number of patients in RSCL cohort).

APACHE = Acute Physiology and Chronic Health Evaluation; PFC = primary fascial closure; POSSUM = Physiological and Operative Severity Score for the Enumeration of Mortality and Morbidity; RSCL = rapid source-control laparotomy.

Several factors that could influence death in general surgery patients undergoing RSCL have not been evaluated adequately. There is, to the best of our knowledge, one study comparing the 30-d mortality rate for early, less than seven d, vs. delayed, greater than seven d, fascial closure. This study did not demonstrate a mortality reduction with early fascial closure, although there was a reduction in median LOS [25]. A meta-analysis of the literature, by Chen et al., encompassing trauma, vascular, and general surgery patients, demonstrated a risk reduction in the mortality rate of 0.53 (p < 0.0001) with early fascial closure. This meta-analysis is limited by its heterogeneous definition of early fascial closure, with the majority of studies defining it as closure within three wks of the initial operation, and its unclear applicability to general surgery patients undergoing RSCP [26].

The median operative times for the RSCL and PFC cohorts in this study were 82 and 104.5 minutes, respectively. It is unclear why a 22.5-minute median difference was all that was required for a definitive operation and fascial closure. There is no literature regarding operative duration and death in general surgery patients undergoing RSCP.

The NSQIP database does not contain information regarding the temporary abdominal closure system employed, nor is it clear that this is of any significance. A meta-analysis by van Hensbroek et al. [27] on trauma, vascular, and general surgery patients who underwent RSCL demonstrated no mortality difference between the artificial burr, dynamic retention sutures, and vacuum-assisted closure methods.

The limitations of this study are, as alluded to, a lack of data regarding: the method employed for temporary abdominal closure, the time from the initial procedure to fascial closure, and pre-operative physiologic parameters. The reliance of NSQIP data on institution-reported incision classification is a significant limitation, as the Western literature debates classification accuracy and subsequent unintended bias [28]. There is a need to determine what, if any, physiologic parameters in general surgery patients should indicate the need for a damage-control procedure.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Ogilvie

. The late complications of abdominal war-wounds. Lancet, 236(6105):253–257.

Ogilvie

. Surgical lessons of war applied to civil practice. Br Med J, 1945; 1(4400):619–623.

Stone

, Strom

, Mullins

. Management of the major coagulopathy with onset during laparotomy. Ann Surg, 1983; 197:532–535.

Godat

., Kobayahsi

, Constantini

, Coimbria

. Abdominal damage control surgery and reconstruction: World Society of Emergency Surgery position paper. World J Emerg Surg, 2013; 8:53.

Cipolla

, Staicki

, Hoff

, et al. A proposed algorithm for managing the open abdomen. Am Surg, 2005; 71:202–207.

Rotondo

, Garcia

, Schwab

, et al. “Damage control”: An approach for improved survival in exsanguinating penetrating abdominal injury. J Trauma, 1993; 35:375–382.

Becher

, Peitzman

, Sperry

, et al. Damage control operations in non-trauma patients: Defining criteria for the staged rapid source control laparotomy in emergency general surgery. World J Emerg Surg, 2016; 11:10.

Garcia-Sabrido

, Tallado

, Christou

, et al. Treatment of severe intra-abdominal sepsis and/or necrotic foci by an “open-abdomen” approach: Zipper and zipper-mesh techniques. Arch Surg, 1988; 123:152–156.

Stawicki

, Brooks

, Bilski

, et al. The concept of damage control: Extending the paradigm to emergency general surgery. Injury, 2008; 39:93–101.

10.

Kafka-Ritsch

, Birkfellner

, Perathoner

, et al. Damage control surgery with abdominal vacuum and delayed bowel reconstruction in patients with perforated diverticulitis Hinchey III/IV. Gastrointest Surg, 2012; 16:1915–1922.

11.

Morrison

, Carrick

, Norman

, et al. Hypotensive resuscitation strategy reduces transfusion requirements and severe postoperative coagulopathy in trauma patients with hemorrhagic shock: Preliminary results of a randomized controlled trial. J Trauma, 2011; 70:652–663.

12.

Hoffman

, Monroe

3rd , A cell-based model of hemostasis. Thromb Haemost, 2001. 85:958–965.

13.

Kashuk

, Moore

, Sawyer

, et al. Primary fibrinolysis is integral in the pathogenesis of the acute coagulopathy of trauma. Ann Surg, 2010; 252:434–442.

14.

Petaja

. Inflammation and coagulation: An overview. Thromb Res, 2011; 127(Suppl 2): S34–S37.

15.

King

, Bauzá Mella

, Remick

, et al. Pathophysiologic mechanisms in septic shock. Lab Invest, 2014; 94:4–12.

16.

Parker

, Shelhamer

, Bacharach

, et al. Profound but reversible myocardial depression in patients with septic shock. Ann Intern Med, 1984; 100:483–490.

17.

Bone

. Gram-negative sepsis: Background, clinical features, and intervention. Chest, 1991; 100:802–808.

18.

Chatrath

, Khetarpal

, Ahuja

. Fluid management in patients with trauma: Restrictive versus liberal approach. J Anaesthesiol Clin Pharmacol, 2015; 31:308–316.

19.

Mazuski

, Sawyer

, Nathens

, et al. The Surgical Infection Society guidelines on antimicrobial therapy for intra-abdominal infections: An executive summary. Surg Infect, 2002; 3:161–173.

20.

Adkins

, Robbins

, Villala

, et al. Open abdomen management of intra-abdominal sepsis. Am Surg, 2004; 70:137–140.

21.

Stawicki

, Cipolla

, Bria

. Comparison of open abdomens in non-trauma and trauma patients: A retrospective study. OPUS 12 Scientist, 2007;(1):1–8.

22.

Finlay

, Edwards

, Lambert

. Damage control laparotomy. Br J Surg, 2004; 91:83–85.

23.

Person

, Dorfman

, Bahouth

, et al., Abbreviated emergency laparotomy in the non-trauma setting. World J Emerg Surg, 2009; 4:41.

24.

Christou

, Barie

, Delinger

, et al. Surgical Infection Society intra-abdominal infection study: Prospective evaluation of management techniques and outcome. Arch Surg, 1993; 128:193–199.

25.

Khan

, Hsee

, Mathur

, et al. Damage-control laparotomy in nontrauma patients: Review of indications and outcomes. J Trauma Acute Care Surg, 2013; 75:365–368.

26.

Chen

, Ye

, Song

, et al, Comparison of outcomes between early fascial closure and delayed abdominal closure in patients with open abdomen: A systematic review and meta-analysis. Gastroenterol Res Pract, 2014; 2014:8.

27.

Boele van Hensbroek

, Wind

, Dijkgraaf

, et al. Temporary closure of the open abdomen: A systematic review on delayed primary fascial closure in patients with an open abdomen. World J Surg, 2009:33:199–207.

28.

, Cohen

, Billmorla

, et al. Effect of wound classification on risk adjustment in American College of Surgeons NSQIP. J Am Coll Surg, 2014; 219:371–781.

29.

Jusoh

, Yanzie

, Damage control surgery/laparostomy in nontrauma emergency abdominal surgery: A new concept of care. Saudi Surg J, 2014; 2:75–79.

30.

Tsuei

, Skiner

, Bernard

, et al. The open peritoneal cavity: etiology correlates with the likelihood of fascial closure. Am Surg, 2004; 70:652–656.