Direct Peritoneal Resuscitation in Trauma Patients Results in Similar Rates of Intra-Abdominal Complications

Abstract

Background:

Trauma patients undergoing damage control surgery (DCS) have a propensity for complicated abdominal closures and intra-abdominal complications. Studies show that management of open abdomens with direct peritoneal resuscitation (DPR) reduces intra-abdominal complications and accelerates abdominal closure. This novel study compares intra-abdominal complication rates and the effect of DPR initiation in patients who received DPR and those who did not.

Patients and Methods:

A retrospective chart review was performed on 120 patients who underwent DCS. Fifty patients were identified as DCS with DPR, and matched to 70 controls by gender, race, age, body mass index (BMI), past medical history, mechanism of trauma, and injury severity score.

Results:

The two groups of patients, those without DPR (−DPR) and those with DPR (+DPR), were similar in their characteristics. The +DPR group was more likely to have a mesh closure than the −DPR (14% and 3%; p = 0.022). The +DPR group took longer to have a final closure (3.5 ± 2.6 days vs. 2.5 ± 1.8; p = 0.020). Infection complications and mechanical failure of the closure technique were similar among the two groups. Timing of DPR initiation had no effect on closure type but did statistically increase the number of days to closure (initiation at first operation 2.8 ± 1.8 days vs. initiation at subsequent operations 6.0 ± 3.3 days; p ≤ 0.001).

Conclusions:

The use of DPR did not result in different outcomes in trauma patients. Therefore, traditional resuscitative measures for DCS may not be inferior to DCS with DPR. When choosing to use DPR, initiating it at the first operation could reduce the number of days to closure.

Damage control surgery (DCS) is the mainstay for the critically ill trauma patient. The goal of DCS is to temporize injured patients who have been taken to the operating room and have severe physiologic derangements that comprise the triad of death. The tenets of DCS are hemorrhage control, contamination control, and temporary closure. This allows the patient to be further resuscitated in an intensive care unit before completing the operation 24 to 72 hours later [1].

This technique of DCS has undoubtedly improved outcomes in trauma by allowing time for restoration of the patient's physiology and for visceral edema from massive resuscitations to subside. Additionally, this technique allows the surgeon interval re-assessment of the abdominal viscera. Damage control surgery has demonstrated the same benefits in emergency general surgery patients with intra-abdominal sepsis [2,3]. However, patients undergoing DCS have a propensity for complicated abdominal closures and intra-abdominal complications [4 –6].

Damage control surgery has changed with the institution of direct peritoneal resuscitation (DPR), first reported in animal studies in 2004 and then in human studies in 2010 [7,8]. Damage control surgery traditionally offered temporary closure with a variety of techniques that permit the fascia to remain open. Direct peritoneal resuscitation has modified this temporary closure through the addition of a 19F Blake drain through which a hypotonic 2.5% dextrose solution is infused. The solution exits via suction applied to the system. Studies have shown that management of open abdomens with DPR reduces visceral edema, improves splanchnic blood flow, reduces both time to closure and ventral hernia rates and improves mortality [9 –11]. Additionally, intra-abdominal complications have been reported to be reduced, however, these studies have been single institutional experiences with small sample sizes that limits the potential for definitive conclusions [8,12,13]. Therefore, a need for further evaluation of how infectious complications are effected by the utilization of DPR exists. This purpose of this study is to evaluate the impact of DPR on infectious complications.

Patients and Methods

This study was conducted at a 945-bed tertiary care medical center facility. The hospital has an American College of Surgeons level 1 trauma center designation and serves the largest catchment area of North Carolina. The study period was from January 2013 to May 2018, and included patients who were traumatically injured, admitted by the trauma services, and who underwent DCS for management of their trauma-related injuries.

A retrospective case-control study was performed utilizing data collected by the trauma registry and additional data were collected via retrospective chart review. Two cohorts were identified: DCS without DPR (−DPR) and DCS with DPR (+DPR). Direct peritoneal resuscitation rate of infusion and fluid type was standardized at this institution prior to the study. The following variables were collected: gender, age, race, body mass index (BMI), mechanism of trauma, injury severity score (ISS), number of trips to the operating room (OR), and surgical procedures. Outcome data included closure type, occurrence of hernia at six months, intensive care unit (ICU) length of stay (LOS), hospital LOS, days to closure, and formation of intra-abdominal abscess, fistulae, wound dehiscence, evisceration, and death.

Statistical analysis was conducted using Fisher exact test for categorical data and Wilcoxon rank sum test was used for non-parametric data. A p value of ≤0.05 was used to indicate statistical significance. The Institutional Review Board of The Brody School of Medicine at East Carolina University and the Center for Research and Grants at Vidant Medical Center approved this study.

Results

One hundred twenty patients were identified as having undergone DCS. Fifty patients (50) were identified +DPR and 70 patients were identified −DPR. The two groups were similar by gender, race, age, BMI, past medical history, mechanism of trauma, and ISS (Table 1).

Table 1.

Clinicodemographic Comparison between −DPR and +DPR

Variables	−DPR (n = 70)	+DPR (n = 50)	p
Gender
Male	52 (74.3)	44 (88)	0.07
Female	18 (25.7)	6 (12)	0.07
Mean age (y)	44	39	0.108
Race
Black	35 (50)	26 (52)	0.071
White	33 (47.1)	17 (34)
Asian	0 (0)	1 (2)
Other	2 (2.9)	6 (12)
BMI (kg/m²)
<18	1 (1)	0	0.28
18–25	19 (27)	19 (38)
26 − 30	23 (33)	10 (20)
>30	27 (39)	21 (42)
Past medical history
COPD	3 (4.3)	1 (2)	0.640
Asthma	6 (8.6)	5 (10)	1.0
Lung cancer	1 (1.4)	0	1.0
Pulmonary TB	0	1 (2)	0.412
Diabetes mellitus	11 (15.7)	6 (12)	0.607
CAD	7 (10)	6 (12)	0.772
Myocardial infarct	0	2 (4)	0.172
Heart failure	3 (4.3)	1 (2)	0.640
Renal disease	3 (4.3)	1 (2)	0.640
ESRD	2 (2.9)	1 (2)	1.0
Liver disease	2 (2.9)	2 (4)	1.0
Immunosuppressed	0	2 (4)	0.172
Chemo/radiotherapy	0	1 (2)	0.417
None	45 (64.3)	31 (62)	0.849
Mechanism of trauma
MVC	22 (31.4)	24 (48)	0.661
Fall	3 (4.3)	1 (2)
GSW	35 (50)	19 (38)
Stab	1 (1.4)	1 (2)
Pedestrian	4 (5.7)	3 (6)
Motorcycle	2 (2.9)	1 (2)
Bicycle	2 (2.9)	0
Other	1 (1.4)	1 (2)
Mean ISS (mean ± SD)	24.9 ± 14.9	24.5 ± 11.0	0.880

Presented as n (%) unless otherwise denoted.

−DPR = without direct peritoneal resuscitation; +DPR = with direct peritoneal resuscitation; BMI = body mass index; COPD = chronic obstructive pulmonary disease; RL = restrictive lung disease; TB = tuberculosis; CAD = coronary artery disease; ESRD = end-stage renal disease; ISS = injury severity score; SD = standard deviation.

The +DPR cohort had substantially greater number of return trips to the OR with 42% having three or more return trips compared with only 11% of −DPR patients (p = 0.018). Those patients with DPR had more coded operations classified for hemorrhage control than −DPR (+DPR vs. −DPR; 60% vs. 36%, p = 0.009). The two groups were similar regarding frequency of preformed splenectomy, hepatorrhaphy, nephrectomy, colectomy, small bowel resection, and cystorrhaphy (Table 2).

Table 2.

Operative Data Comparison

	DCL w/o DPR	DCL w/ DPR	p
Subsequent OR trips after index
1	4 (6)	1 (2)	0.315
2	58 (83)	28 (56)	0.001^a
3	5 (7)	11 (22)	0.018^a
4	2 (3)	5 (10)	0.100
5	1 (1)	5 (10)	0.034^a
Types of surgeries
Colectomy	12 (17)	16 (32)	0.079
Multiple	0	0	–
Anastomosis, final	5 (42)	5 (31)	0.698
Colostomy, final	3 (25)	7 (44)	0.434
Discontinuity, initial	7 (58)	12 (75)	0.432
Small bowel resection	32 (46)	24 (48)	0.854
Multiple	14 (44)	7 (29)	0.403
Anastomosis, final	32 (100)	19 (79)	0.011^a
Ileostomy, final	1 (3)	1 (4)	1
Discontinuity, initial	24 (75)	18 (75)	1
Splenectomy	19 (27)	14 (28)	1
Hepatorrhaphy	12 (17)	11 (22)	0.639
Nephrectomy	3 (4)	3 (6)	0.692
Bladder repair	7 (10)	3 (6)	0.519
Hemorrhage control	25 (36)	30 (60)	0.009^a
Pelvic	5 (20)	9 (30)	0.537
Mesenteric	12 (48)	19 (63)	0.286
Aortic	2 (8)	2 (7)	1
Vena cava	6 (24)	3 (10)	0.273
Other	67 (96)	46 (92)	0.449

Data presented as n (%), unless otherwise denoted.

DCL = damage control laparotomy; DPR = direct peritoneal resuscitation; OR = operating room.

Indicates statistical significance.

The +DPR group was more likely to have a mesh closure than the −DPR (14% and 3%; p = 0.022; Table 3). The formation of all hernias in +DPR was higher than −DPR (30% vs. 9.3%; p = 0.016). However, the frequency of hernia formation after primary fascial closure was not statistically different between the two groups (17% vs. 5%; p = 0.600; Table 3).

Table 3.

Abdominal Closure Type

	−DPR	+DPR	p
Death before closure	9 (13)	7 (14)	1
Primary fascial	58 (95)	36 (84)	0.088
Mesh closure	2 (3)	7 (14)	0.022^a
Hernia after primary closure	2 (5)	6 (17)	0.600

Data presented as n (%), unless otherwise denoted.

−DPR = without direct peritoneal resuscitation; +DPR = with direct peritoneal resuscitation.

Indicates statistical significance.

No difference was found between groups in the rate of formation of intra-abdominal abscesses (28% vs. 33%; p = 0.669) or fistulas (12% vs. 11%; p = 1.000). Furthermore, there was no difference in the rate of dehiscence (7% vs. 15%; p = 0.351) or evisceration (2% vs. 2%; p = 1.0) between groups (Table 4). The frequency of intra-abdominal infectious complications and mechanical wound failures were unaffected by the timing of DPR initiation (Table 4).

Table 4.

Intra-Abdominal Infectious Complications and Mechanical Wound Failure Rates Regardless of When DPR was Initiated

	−DPR	+DPR	p
Intra-abdominal abscess	20 (33)	12 (28)	0.669
Fistula	7 (11)	5 (12)	1.000
Dehiscence	9 (15)	3 (7)	0.351
Evisceration	1 (2)	1 (2)	1.0
Death	5 (8)	7 (16)	0.227

Data presented as n (%), unless otherwise denoted.

DPR = direct peritoneal resuscitation.

The +DPR group had a longer interval to final closure (3.5 ± 2.6 days vs. 2.5 ± 1.8; p = 0.020; Table 5). Delayed initiation increased the number of days to closure (initiation at first operation 2.8 ± 1.8 days vs. initiation at subsequent operations 6.0 ± 3.3 days; p = <0.001; Table 5). Intensive care unit LOS, hospital LOS, and mortality were similar between two groups regardless of when DPR was initiated.

Table 5.

Complications

	−DPR	+DPR	p
ICU LOS	11.4 ± 10.7	13.0 ± 10	0.807
LOS	21.8 ± 17.2	22.6 ± 13.1	0.445
Days to closure	2.5 ± 1.8	3.5 ± 2.6	0.020^a
Intra-abdominal abscess	20 (33)	12 (28)	0.669
Fistula	7 (11)	5 (12)	1.000
Dehiscence	9 (15)	3 (7)	0.351
Evisceration	1 (2)	1 (2)	1.0
Death	5 (8)	7 (16)	0.227

Data presented as n (%), unless otherwise denoted.

DPR = direct peritoneal resuscitation; ICU = intensive care unit; LOS = length of stay.

Indicates statistical significance.

Discussion

It has been suggested that the use of direct peritoneal resuscitation may improve outcomes for damage control laparotomy. However, the existing literature does not draw powered conclusions regarding intra-abdominal infectious complications. It is presumed that because of increased splanchnic blood flow and reduction of systemic cytokines, the inflammatory response is augmented reducing visceral edema [14,15]. The reduction of visceral edema helps to increase the rate of abdominal wall closure and thereby reducing the time and risks associated with the open abdomens [5]. Direct peritoneal resuscitation did not increase rates of intra-abdominal infection or infectious complications leading to wound complications such as dehiscence or evisceration (Fig. 1). Similarly in our study, DPR did not reduce infectious risk as previously published. This may be explained by the higher incidence of penetrating trauma in the two groups as penetrating trauma is associated with higher rates of intra-abdominal contamination [16]. The +DPR group had 40% penetrating trauma and the −DPR group was 51% (Table 1). This exceeds the reported 20% in previous studies and is unique to this study.

FIG. 1.

Infectious complications.

Our institution has a relatively high volume of penetrating injury with between 15% to 20% depending on the year. This could also be a significant factor in our results compared with studies with higher non-penetrating trauma. Although there was an increased incidence of penetrating trauma, our analysis showed no difference in the operations performed, except for a higher level of hemorrhage control operations in the +DPR cohort, which can be used as a surrogate for injury type.

In addition, our study examined mechanical wound complications. Fascial dehiscence and abdominal evisceration are known complications after complicated abdominal wall closures. Previous studies have not delineated an association with the adjunctive use of DPR and mechanical wound complications. Our study demonstrates that there is no statistically significant difference in fascial dehiscence or abdominal evisceration when comparing patients who received DPR to those who did not. Additionally, the timing of DPR initiation had no effect on dehiscence or evisceration outcomes either [17].

The use of DPR at our institution did not result in decreased time to closure compared with DCS without DPR. We suspect this to be because the +DPR group had more severe injuries that was not captured by the ISS, as evidenced by a significantly greater number of coded hemorrhage control procedure performed that suggest they had greater hemodynamic and metabolic compromise at presentation.

Total number of OR trips was also a factor to be considered as a potential confounding variable as the +DPR cohort required a greater number of return trips to the OR compared with the −DPR cohort. Use of DPR can reduce edema and facilitate abdominal closure, but it cannot address other indications for continued delayed closure. In this study, a majority of the DPR cohort did not require further bowel resection or delayed re-anastomosis but were left open strictly because of inability to close the abdomen. However, there were a few occurrences of patients requiring further surgical intervention that further delayed abdominal closure.

Of note, with the implementation of blood product resuscitation in 2007 the resulting incidence of over resuscitation with crystalloid has been reduced dramatically [18]. The current study population is unique compared with previously published studies in that all the patients were admitted after implementation of blood product resuscitation protocols had been well established at our institution.

Additionally, the closure rates may be confounded by the inconsistency of when DPR was initiated (Fig. 2). If DPR was initiated at the index operation, similar closure was observed (2.8 ± 1.8 days) (Fig. 3). Interestingly, our time to closure for both groups was much less than reported in other studies. We demonstrate closure rates of 2.5 ± 1.8 days for −DPR compared with 7 ± 3 days reported by Smith et al. [19]. This trend is also appreciated in the +DPR cohort with our institution rate of closure at 3.5 ± 2.6 days compared with 5.9 ± 3.5 days [19]. Again, this may be related to the use of blood product resuscitation in lieu of the historical crystalloid resuscitation.

FIG. 2.

Closure rates base on initiation of direct peritoneal resuscitation. −DPR = without direct peritoneal resuscitation; +DPR = with direct peritoneal resuscitation TTC = time to closure; OR = operating room.

FIG. 3.

Closure rate and timing. −DPR = without direct peritoneal resuscitation; +DPR = with direct peritoneal resuscitation.

Our study is limited because of the relatively small sample size for retrospective analysis and the retrospective design of the study. There may be differences in our cohorts not captured by the retrospective design because the +DPR group underwent more procedures for hemorrhage control, which could be a marker of more severely injured patients. This is also a single institutional study with a unique patient demographic. As much because two-thirds of our trauma patients are transfers from our rural community and have a relatively higher rate of penetrating trauma. Because of our small cohort of patients undergoing DPR we are unable to conclude if timing of initiation affects infectious complications or mechanical wound failure.

Future studies should include a multi-institutional observation of infectious complications and mechanical wound failure of patients undergoing DCS with DPR. Additionally, studies looking at specific mechanism of injury may demonstrate a difference in infectious complications and closure rates further specifying indications for the use of DPR. No published study data has demonstrated an effect of DPR on shock parameters such as lactic acid levels to evaluate global perfusion. Acknowledging that DPR reduces systemic cytokines and thus blunting the inflammatory response, the degree of distributive shock may be reduced as well. Improved hemodynamic and perfusion states may reduce overall infectious complications.

Conclusions

In conclusion, the use of DPR had no impact on abdominal infectious complications. In addition, if DPR is not initiated at the index operation the time to closure is significantly prolonged. Therefore, if DPR is being considered it should be initiated at the first operating room visit.

Footnotes

Author Disclosure Statement

The authors have no conflicts of interest to disclose.

References

Rotondo

, Schwab

, McGonigal

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

Waibel

, Rotondo

. Damage control for intra-abdominal sepsis. Surg Clin North Am, 2012; 92:243–257.

Khan

, Hsee

, Mathur

, et al. Damage-control laparotomy in nontrauma patients: Review of indications and outcomes. J Trauma Acute Care Surg, 2013; 75(3):365–368.

Fabian

, Croce

, Pritchard

, et al. Planned ventral hernia. Staged management for acute abdominal wall defects. Ann Surg, 1994; 219:643–650.

Miller

, Morris

Jr , Diaz

Jr , et al. Complications after 344 damage-control open celiotomies. J Trauma, 2005; 59:1365–1371.

Smith

, Tweddell

, Bessey

. Alternative approaches to abdominal wound closure in severely injured patients with massive visceral edema. J Trauma, 1992; 32:16–20.

Garrison

, Conn

, Harris

, et al. Direct peritoneal resuscitation as adjunct to conventional resuscitation from hemorrhagic shock: A better outcome. Surgery, 2004; 136:900–908.

Smith

, Garrison

, Matheson

, et al. Direct peritoneal resuscitation accelerates primary abdominal wall closure after damage control surgery. J Am Coll Surg, 2010; 210:658–664.

Hurt

, Zakaria el

, Matheson

, et al. Hemorrhage-induced hepatic injury and hypoperfusion can be prevented by direct peritoneal resuscitation. J Gastrointest Surg, 2009; 13:587–594.

10.

Smith

, Ghazi

, Cain

, et al. Direct peritoneal resuscitation improves inflammation, liver blood flow, and pulmonary edema in a rat model of acute brain death. J Am Coll Surg, 2014; 219:79–87.

11.

Zakaria el

, Li

, Garrison

. Mechanisms of direct peritoneal resuscitation-mediated splanchnic hyperperfusion following hemorrhagic shock. Shock, 2007; 27:436–442.

12.

Smith

, Nash

, Procter

, et al. Not all abdomens are the same: A comparison of damage control surgery for intra-abdominal sepsis versus trauma. Am Surg, 2016; 82:427–432.

13.

Weaver

, Smith

. Direct peritoneal resuscitation: A review. Int J Surg, 2016; 33(Pt B):237–241.

14.

Crafts

, Hunsberger

, Jensen

, et al. Direct peritoneal resuscitation improves survival and decreases inflammation after intestinal ischemia and reperfusion injury. J Surg Res, 2015; 199:428–434.

15.

Matheson

, Hurt

, Franklin

, et al. Obesity-induced hepatic hypoperfusion primes for hepatic dysfunction after resuscitated hemorrhagic shock. Surgery, 2009; 146:739–747.

16.

Croce

, Fabian

, Stewart

, et al. Correlation of abdominal trauma index and injury severity score with abdominal septic complications in penetrating and blunt trauma. J Trauma, 1992; 32:380–387.

17.

Zakaria el

, Garrison

, Kawabe

, et al. Direct peritoneal resuscitation from hemorrhagic shock: effect of time delay in therapy initiation. J Trauma, 2005; 58:499–506.

18.

Holcomb

, Jenkins

, Rhee

, et al. Damage control resuscitation: Directly addressing the early coagulopathy of trauma. J Trauma, 2007; 62:307–310.

19.

Smith

, Matheson

, Franklin

, et al. Randomized controlled trial evaluating the efficacy of peritoneal resuscitation in the management of trauma patients undergoing damage control surgery. J Am Coll Surg, 2017; 224:396–404.