Characteristics of Combat-Associated Small Bowel Injuries

Abstract

Introduction

Although there are multiple studies regarding the management and outcomes of colonic injuries incurred in combat, the literature is limited with regard to small bowel injuries. This study seeks to provide the largest reported review of the characteristics of combat-associated small bowel injuries.

Materials and Methods

The Department of Defense Trauma Registry was queried for U.S. Armed Forces members who sustained hollow viscus injuries in the years 2007–2012 during Operations Enduring Freedom, Iraqi Freedom, and New Dawn. Concomitant injuries, procedures, and complications were delineated. Fisher’s exact test was used to analyze the relationship of bowel injury pattern to rates of repeat laparotomy, fecal diversion, and complications.

Results

One hundred seventy-one service members had small bowel injuries. The mean age was 25.8 ± 6.6 yr with a mean injury severity score of 27.9 ± 12.4. The majority of injuries were penetrating (94.2%, n = 161) as a result of explosive devices (61.4%, n = 105). The median blood transfusion requirement in the first 24 h was 6.0 units (interquartile range 1.0–17.3 units). The most frequent concomitant injuries were large bowel (64.3%, n = 110), pelvic fracture (35.7%, n = 61), and perineal (26.3%, n = 45). Fifty patients (29.2%) had a colostomy, and nine patients (5.3%) had an ileostomy; 62.6% (n = 107) of soldiers underwent more than one laparotomy. The mortality rate was 1.8% (n = 3). The most common complications were pneumonia (15.2%, n = 26), deep vein thrombosis (14.6%, n = 25), and wound infection (14.6%, n = 25). The need for repeat laparotomy and fecal diversion was found to be significantly associated with injury pattern (p = 0.00052 and p < 0.0001, respectively).

Conclusion

We found that two-thirds of service members with small bowel injuries also had a large bowel injury. One-third of the patients required diversion and two-thirds had more than one laparotomy. The pattern of bowel injury significantly affected the need for repeat laparotomy and fecal diversion.

Introduction

The nature and management of military trauma are continuously evolving. In the 10-yr period of 2001–2011, 4,596 U.S. battlefield fatalities were enumerated by Eastridge et al.¹ Eighty-seven percent (n = 4,016) of these deaths occurred before arrival at a medical treatment facility (MTF). Seventy-four percent of the injuries were the result of explosive trauma. In this study, an expert review panel analyzed the pre-MTF deaths and reported that 24% (n = 976) of the deaths were potentially survivable, with hemorrhage (91%) and airway compromise (8%) being the predominant causes of death.

When comparing case fatality rates (the ratio of deceased to wounded) for U.S. military personnel for current and past conflicts, the rates have decreased through the years. A study from 2006 found that the case fatality rates decreased from 19.1 (World War II) to 15.8 (Vietnam) to 9.4 (Iraq/Afghanistan).² Interestingly, the rates of those killed in action (death before MTF) have decreased, whereas the rates of those who die of their wounds (death after reaching a MTF) have increased when comparing these three conflicts. One potential explanation is that these trends reflect improved pre-MTF care and evacuation of severely injured patients; in other words, severely injured patients who would have been killed in action before improved care and system measures now die of their wounds at a MTF. This underscores the importance of continued improvements in care for those who do survive until arrival at a MTF.

For survivors of combat injuries, there is an additional imperative to ensure that injury management is held to the highest standards, informed by expertise and quality data. Although traumatic bowel injuries are not isolated to military trauma, the circumstances of combat trauma and the delivery of trauma care are unique. Although pre-MTF care is crucial, best practices in continuing care must not be ignored. Questions abound in regard to multiple dimensions of the care of these injuries, including techniques of repair and anastomosis, decision-making surrounding fecal diversion, and timing of abdominal closure after damage control laparotomy. These issues can have a profound effect on subsequent complications such as anastomotic leak and hernia formation as well as length of stay (LOS) and quality of life.

One challenging area of care is the optimal management of hollow viscus injuries. These challenges are compounded by the transfer of the patient from the battlefield often through multiple echelons of care, making the smooth continuity of care and standardization of best practices particularly tasking. Publications during the recent major military operations (Operation Enduring Freedom [OEF], Operation Iraqi Freedom [OIF], and Operation New Dawn [OND]) have investigated issues relevant to management and outcomes, including fecal diversion, injury pattern, and complications.^3–8 Much attention has been paid to colon and rectal trauma; however, there is a distinct paucity of data regarding small bowel injuries. This study seeks to address that void by providing the largest reported review of the characteristics of combat-associated small bowel injuries.

Methods

The Department of Defense Trauma Registry (DoDTR), formerly the Joint Theater Trauma Registry, was utilized for this study after approval by the United States Army Medical Research and Materiel Command Institutional Review Board. The DoDTR was queried for active service members of the U.S. Armed Forces who sustained hollow viscus combat injuries in the 6-yr period of January 1, 2007, to December 31, 2012, during OEF, OIF, and OND. Patients with other backgrounds, including North Atlantic Treaty Organization (NATO) troops and local nationals, were excluded. Service members were included if they were older than 17 yr and were evacuated from a combat theater to Landstuhl Regional Medical Center in Landstuhl, Germany. Patients who died within 48 h of injury and who died secondary to a severe neurologic injury (i.e., severe traumatic brain injury or paralysis) within 14 d of injury were excluded.

Given the ever-evolving body of literature on combat-associated trauma during the data procurement phase, the scope of the study was subsequently narrowed to look specifically at service members with small bowel trauma. Background information was collected including age and gender. Date and mechanism of injury, military operation, injury severity score, blood transfusions, and hospital LOS were tracked in addition to bowel injury pattern (small bowel [SB], SB + large bowel [LB], SB + LB + rectum [Rect], and SB + Rect). Concomitant intra-abdominal (diaphragm, liver, spleen, pancreas, and kidney) and extra-abdominal (pelvic fracture, pelvic organ, and perineal) injuries were delineated. For procedures, repeat laparotomy and fecal diversion (colostomy or ileostomy) were recorded. A repeat laparotomy was recorded if the procedure code for repeat laparotomy was used (54.12) or if an intra-abdominal operation was coded at multiple levels of care. Designated echelons, or levels, of care utilized in the military were noted for procedures (described in Fig. 1).⁹ Complications were also enumerated.

Figure 1.

Military echelons of care (levels I–V).

Fisher’s exact test, a procedure to compare observed vs expected values when expected frequencies are small, was used to analyze the relationship of bowel injury pattern to rates of repeat laparotomy, fecal diversion, and complications. Statistical significance is considered with a p < 0.05. Post hoc analysis was conducted using chi-square with p-value adjustment by the FDR method. Normally distributed data are reported as mean (standard deviation [SD]), whereas nonparametric data are reported as median (25th and 75th interquartile range [IQR]).

Results

In the studied 6-yr period, DoDTR had a record of 343 service members with at least one hollow viscus injury. One hundred seventy-one of those patients (49.6%) had a small bowel injury. The incidence of hollow viscus injuries over the course of the 6 yr is shown in Figure 2. There was a mean of 57 injuries per year (range 32–80). Between 43.0% and 70.0% of patients with a hollow viscus injury had a small bowel injury, dependent on the year.

Figure 2.

Combat-associated bowel injuries from years 2007–2012.

For service members with small bowel injuries, the mean age was 25.8 ± 6.6 yr with a mean injury severity score of 27.9 ± 12.4 (Table I). The majority of injuries were penetrating (94.2%, n = 161) as a result of explosive devices (61.4%, n = 105). The median packed red blood cell transfusion requirement in the first 24 h was 6.0 units (IQR 1.0–17.3 units) with 28.1% of patients (n = 48) requiring more than 10 units. The median LOS was 13 d (IQR 5–38 d).

Table I.

Demographics and Characteristics of Personnel with Small Bowel Injuries

Variable	n (%) or Mean ± SD or Median (IQR)
Age (yr)	25.7 ± 6.6
Gender
Male	170 (99.4)
Female	1 (0.6)
Military operation
Operation Enduring Freedom	119 (69.6)
Operation Iraqi Freedom	51 (29.8)
Operation New Dawn	1 (0.6)
Injury mechanism
Bullet/GSW/firearm	66 (38.6)
Explosive devices	105 (61.4)
Injury type
Blunt	6 (3.5)
Burn	4 (2.3)
Penetrating	161 (94.2)
Injury severity score	27.9 ± 12.4
LOS (d)	13 (5–38)
Blood transfusion (units)a
FFP	4.5 (0–13.3)
Platelets	0 (0–2.0)
PRBC	6.0 (1.0–17.3)
Whole bloodb	0 (0–0)
Mortality	3 (1.8)

^aWithin first 24 h of admission.

^bMean ± SD for whole blood was 1.1 ± 3.8.

SD, standard deviation; IQR, interquartile range; GSW, gunshot wound; FFP, fresh frozen plasma; PRBC, packed red blood cells.

Injuries

The patterns of bowel injury are delineated in Table II. The majority of our patients with a small bowel injury also had a large bowel injury (64.3%). Only 12 patients had a gastric injury, and 27 patients also had a rectal injury. The most frequent concomitant intra-abdominal injuries apart from a hollow viscus injury were liver (20.5%, n = 35) and spleen (18.1%, n = 31). In regard to extra-abdominal injuries, pelvic fracture was associated in 35.7% of cases (n = 61) and perineal injury in 26.3% (n = 45). Injuries outside the abdomen and pelvis were not tracked.

Table II.

Injury Pattern and Associated Injuries

	n (%)
Bowel injury
Gastric	12 (7.0)
Small bowel	171 (100.0)
Large bowel	110 (64.3)
Rectum	27 (15.8)
Bowel injury pattern
SB	52 (30.4)
SB + LB	91 (53.2)
SB + Rect	8 (4.7)
SB + LB + Rect	20 (11.7)
Associated injuries
Intra-abdominal
Diaphragm	10 (5.9)
Liver	35 (20.5)
Pancreas	13 (7.6)
Spleen	31 (18.1)
Kidney	22 (12.9)
Extra-abdominal
Pelvic fracture	61 (35.7)
Pelvic organ	34 (19.9)
Perineal	45 (26.3)

SB, small bowel; LB, large bowel; Rect, rectum.

Procedures

Fecal diversion was performed in 57 of the patients with small bowel trauma. Fifty patients (29.2%) had a colostomy, and nine patients (5.3%) had an ileostomy; two patients had both a colostomy and an ileostomy (Table III). Only three patients with isolated small bowel trauma underwent diversion; the remaining patients also had a large bowel or rectal injury. Fecal diversion procedures occurred in similar numbers at level III (29 patients) and level IV (26 patients) facilities. Only four patients were diverted at a level V facility. Procedure codes for nine patients (one ileostomy and eight colostomy) described the diversion as temporary.

Table III.

Procedures

	n (%)
Colostomy	50 (29.2)
Ileostomy	9 (5.3)
Repeat laparotomy	107 (62.6)

62.6% (n = 107) of service members underwent more than one laparotomy. For patients with a repeat laparotomy, the mean number of laparotomies was 2.7 ± 0.9.

Complications

The overall complication rate was 44%. The most common complications were pneumonia (15.2%, n = 26), deep vein thrombosis (14.6%, n = 25), wound infection (14.6%, n = 25), and pulmonary embolus (12.9%, n = 22) (Table IV). The mortality rate was 1.8% (n = 3).

Table IV.

Complications

Complication	n (%)
Abscess	7 (4.1)
Anastomotic leak	1 (0.5)
Bacteremia	11 (6.4)
Bowel obstruction	6 (3.5)
Clostridium difficile infection	1 (0.5)
Deep vein thrombosis	25 (14.6)
Dehiscence	7 (4.1)
Hernia	3 (1.8)
Pneumonia	26 (15.2)
Pulmonary embolus	22 (12.9)
Sepsis	10 (5.8)
Wound infection	25 (14.6)

Outcomes by Injury Pattern

The outcomes of (1) repeat laparotomy, (2) fecal diversion, and (3) presence of one or more complications were analyzed with respect to the pattern of injury (SB, SB + LB, SB + Rect, and SB + LB + Rect) (Table V). Repeat laparotomy occurred in 39.6% of those with a SB injury compared with 71.4%, 75.0%, and 78.9% of those with SB + LB, SB + Rect, and SB + LB + Rect injury patterns, respectively. Patients with rectal injuries were most likely to undergo fecal diversion (62.5% of SB + Rect, 68.4% of SB + LB + Rect).

Table V.

Repeat Laparotomy, Fecal Diversion, and Complications with Respect to Injury Pattern

Outcome	n (%)	p-Value
Repeat Laparotomy		0.00052
SB	21 (39.6)
SB + LB	65 (71.4)
SB + Rect	6 (75.0)
SB + LB + Rect	15 (78.9)
Fecal diversion		<0.0001
SB	3 (5.7)
SB + LB	36 (39.7)
SB + Rect	5 (62.5)
SB + LB + Rect	13 (68.4)
Complications		0.089
SB	16 (30.2)
SB + LB	46 (50.5)
SB + Rect	4 (50.0)
SB + LB + Rect	10 (52.6)

SB, small bowel; LB, large bowel; Rect, rectum.

Through Fisher’s exact test, the need for repeat laparotomy and fecal diversion (ileostomy and/or colostomy) was found to be significantly associated with injury pattern (p = 0.00052 and p < 0.0001, respectively) (Table V). Post hoc analysis results are presented in Table VI. There was a significant difference in need for repeat laparotomy between patients with SB only vs SB + LB (p = 0.0013) and SB + LB + Rect (p = 0.0197) injury patterns. In addition, presence of fecal diversion was significantly different between SB only vs SB + LB (p < 0.0001), SB + Rect (p = 0.0009), and SB + LB + Rect (p < 0.0001) as well as SB + LB vs SB + LB + Rect (p = 0.0373) injury patterns.

Table VI.

Post Hoc Analysis of Repeat Laparotomy and Fecal Diversion with Respect to Injury Pattern

Outcome	p-Valuea
Repeat laparotomy
SB vs SB + LB	0.0013
SB vs SB + Rect	0.2458
SB vs SB + LB + Rect	0.0197
SB + LB vs SB + Rect	1.0000
SB + LB vs SB + LB + Rect	0.8755
SB + Rect vs SB + LB + Rect	1.000
Fecal diversion
SB vs SB + LB	<0.0001
SB vs SB + Rect	0.0009
SB vs SB + LB + Rect	<0.0001
SB + LB vs SB + Rect	0.3238
SB + LB vs SB + LB + Rect	0.0373
SB + Rect vs SB + LB + Rect	1.0000

^aAdjusted.

SB, small bowel; LB, large bowel; Rect, rectum.

Discussion

A number of studies have added to the body of literature regarding combat-associated and traumatic bowel injury in an effort to improve management and outcomes. The vast majority of this literature involves colorectal trauma, recently enabled by the development of the Joint Surgical Transcolonic Injury or Ostomy Multi-Theater Assessment (J-STOMA) database.

In contrast to colorectal injuries, the data relevant to combat-associated small bowel injuries are much more sparse. Salim et al conducted a 29-mo prospective observational study of non-military patients with penetrating stomach or small bowel injuries.¹⁰ One hundred and seventy-eight patients were identified during this time period, and the most frequent associated intra-abdominal injury was the colon (39%). Sixty-four percent of the small-bowel-injured patients in our series had a colonic injury; this number increases to 80% if rectal injuries are included as well.

The American Association for the Surgery of Trauma grades small bowel injuries on a scale of I–V, ranging from a simple hematoma to transection with tissue loss.¹¹ Injuries to the small bowel are frequently encountered with penetrating trauma, but blunt trauma can also result in such injuries, particularly to the jejunum and ileum given the relatively shorter mesentery. For small bowel injuries, primary repair is generally favored over diversion especially in light of complications that may ensue from high-output small bowel ostomies.¹²

Examining the colorectal literature provides insight into combat-associated bowel injury management. Glasgow and colleagues reported on colorectal trauma occurring during an 8-yr span (2003–2011) of OIF/OEF.³ The study, which included all coalition military personnel, characterized the colorectal trauma of 977 individuals; just over half of these patients were U.S. military personnel. The predominant mechanism of injury was a gunshot wound (57.6%), which contrasts with our finding of the majority of injuries attributable to explosive devices (61.4%); however, NATO and U.S. troops were more likely to have explosion and blast injuries compared with non-NATO forces. Colon-injured patients were more likely to have concomitant chest injury, whereas extremity and skin/soft tissue injuries were more frequent in patients with rectal injuries. Concomitant small bowel injuries were not detailed. Another study identified 5.1% of patients at one combat hospital had a colorectal injury, and rectal and transverse colon injuries were predictors of mortality.⁴

In an effort to capture differences in patients and management between the military and civilian population with colon injuries, a retrospective review was conducted comparing traumatic colon injuries managed at a civilian hospital (n = 30) to such injuries incurred in combat (n = 59) during the same 2-yr time period (2005–2006).⁵ The predominant mechanism of injury was blunt (70%) in civilians compared with 51% blast injuries in combat. An important endpoint examined in this study that has limited available literature was colostomy reversal. Seventeen percent of the civilian population was diverted with a reversal rate of 40% compared with 69% diversion in the military with a reversal rate of 84%. Colon-related complications (i.e., leak and abscess) were not significantly different between the populations. The study also reports that 80% of the patients in the military group were managed with open abdomen techniques, which is comparable to our rate of 62.6% who underwent repeat laparotomy. Our rate increases to 75.1% if only small-bowel-injured patients with a concomitant colorectal injury are included (excluding patients with isolated small bowel injuries).

Whereas the historical mainstay of bowel injury management – particularly for battlefield trauma – was fecal diversion, this dogma has been challenged by advocates of primary repair in recent decades.^13–15 The increased practice of damage control laparotomy contributes to the evolution of this debate as well. The most comprehensive studies of recent major military operations (OIF/OEF) report an overall diversion rate of 36–37% for colorectal injuries commensurate with previous findings of a diversion rate of 33% (referring to initial surgical management).^3,4,8 Furthermore, one study found that the mortality rate was lower in diverted patients compared with those who were not diverted (3.7% vs 10.8%, p < 0.0001) even in light of a slightly higher injury severity score in diverted patients. A review of studies examining the surgical management of colorectal injuries after combat-associated blast injury found diversion rates of 0–78% (mean 49.5%) with no significant differences in mortality or complications between primary repair and diversion.⁶ Another review of 133 patients with military colonic injuries from the years 2005–2006 found a diversion rate of 44% compared with primary repair (32%) and damage control (23%).⁷ There was not a significant difference in complications between the groups. These military-specific studies provide mounting evidence that it is reasonable to replace the previous dictum of diversion with careful consideration of primary repair as an alternate approach, in keeping with changes in civilian practice as well.

In comparison, our study found that 33.3% of the patients with small bowel injuries were diverted. The vast majority (94.7%) of these patients had a colorectal injury in addition to small bowel trauma. Sixty-seven percent (n = 51) of complications occurred in non-diverted patients compared with 33% of the complications occurring in diverted patients.

Much of the literature comparing traumatic small bowel injuries and colon injuries revolves around infectious complications. Surgical site infections are a commonly observed complication of penetrating abdominal injuries. In a study of civilian penetrating stomach and small bowel injuries, the overall complication rate was 27%, with 20% of the patients experiencing a surgical site infection, including wound infections, fascial dehiscence, and intra-abdominal abscess.¹⁰ Our overall complication rate was 44%, and 27% of the patients had a surgical site infection, as defined by the aforementioned study. Previous civilian research has found damage control procedures as well as associated pancreaticoduodenal injuries to be predictors of anastomotic-related complications in civilian traumatic small bowel injuries.¹⁶ Furthermore, not surprisingly, large bowel anastomoses have higher anastomotic-related complications (occurring in 27% of anastomoses) than small bowel anastomoses (9% of anastomoses).¹⁷

Pneumonia was much more common in our population (15.6%) compared with the civilian study (1.7%).¹⁰ A DVT/PE rate of 23% has previously been reported, which is similar to our rate of 24.0% (patients with both a diagnosed DVT and a PE only counted once).⁵ One reason for differences in complication rates is likely the challenges in the provision of combat casualty care; however, further study is warranted. One review of combat-related infections describes increased time between injury and hospital admission to be associated with higher infection rates.¹⁸ During OIF/OEF, time from injury to MTF admission averaged 45 min. Rapid evacuation, diagnosis, and treatment of infections are essential for improved outcomes in patients with combat-associated injuries. In addition, patients with combat-associated injuries are similarly susceptible to antimicrobial resistant bacteria exposures that accumulate with prolonged hospitalization.

There are limitations to the present study. Diagnosis and procedure codes do not capture condition and events precisely and are subject to the interpretation of the provider and/or coder. This can lead to inaccurate recognition of injuries as well as difficulty in elucidating procedures, such as repeat laparotomies. In addition, further analysis of management decision-making would be useful, such as a thorough review of operative reports and hospital course, with particular attention to timing and types (i.e., stapled vs hand-sewn) of anastomoses. These particular details were not available for the current study.

Conclusion

The characteristics of combat-associated small bowel trauma have not previously been reported. We found that two-thirds of service members with small bowel injuries also had a large bowel injury. One-third of the patients in this study required diversion, and two-thirds required at least one repeat laparotomy. The pattern of bowel injury significantly affected the need for repeat laparotomy and fecal diversion. Further investigation is warranted to elucidate how patients with small bowel injuries are compared with those with other hollow viscus trauma and how different methods of operative management affect outcomes.