Gunshot to the abdomen: laparotomy or nonoperative care

The following scene is fictional and was written for teaching. A 30-year-old man arrives at the emergency department with a gunshot wound to the right flank, about four centimeters above the iliac crest, with no exit wound. He is alert, with a blood pressure of 128 over 78 mmHg and a heart rate of 92 beats per minute. He reports pain only at the wound site. The abdomen is soft, with mild tenderness to palpation near the wound and no signs of peritoneal irritation. The radiograph shows the projectile projected over the right lumbar region. The question in the room is an old one: operate on every abdominal gunshot wound or select those who can be observed?

This text summarizes the evolution of treatment for abdominal gunshot wounds, from the dogma of mandatory laparotomy to selective nonoperative management, with the role of CT, laparoscopy, serial examination and damage control surgery. The content is educational, aimed at students and health professionals, and does not replace local protocols or the surgical team's assessment.

Key points

• Hemodynamic instability, peritonitis and evisceration still indicate immediate laparotomy.
• In the stable, alert patient without peritonitis, selective nonoperative management failed in about 7 in every 100 patients, with a mortality of 0.4 in every 100, in a meta-analysis.
• Failure is lower in wounds of the right thoracoabdomen, the flank and the back, and higher in the anterior abdominal wall.
• Contrast-enhanced CT has high sensitivity for injury that requires surgery, but it is poorly specific for hollow viscus injury.
• Serial abdominal examination, preferably by the same professional, is a central part of selective management.
• Diagnostic laparoscopy helps rule out peritoneal penetration and diaphragmatic injury in stable patients.

From dogma to selection

For much of the 20th century, every abdominal gunshot wound was operated on. The logic was simple: the projectile carries enough energy to injure viscera with high probability, and an untreated intestinal injury progresses to peritonitis and sepsis. A review on controversies in penetrating abdominal trauma describes this phase and the gradual shift to selective management (Butt et al., 2009).

The same review notes that stab wounds are about three times more frequent than gunshot wounds in the abdomen, but that about 90 in every 100 deaths from penetrating abdominal trauma are caused by projectiles, because of the greater energy transmitted. Over the last 30 years, the pendulum swung toward selective management, first for stab wounds and later also for gunshot wounds (Butt et al., 2009).

The change came from the realization that many laparotomies were negative or nontherapeutic, and that these operations have complications of their own: ileus, wound infection, adhesions, hernias and longer hospital stays. Better CT, the availability of laparoscopy and experience with serial examination made it possible to select who needs surgery and who can be safely observed.

Who goes straight to laparotomy

Some situations still indicate immediate laparotomy, with no room for observation. In a series of patients with penetrating liver injury, 123 of 178 were operated on without CT because of peritonitis, hypotension or a Glasgow score of 8 or less, and in that group there was no nontherapeutic laparotomy (Schnüriger et al., 2011). In a large series of abdominal war wounds, the indications for laparotomy included peritonitis, hemodynamic instability or shock, evisceration and evident bleeding from the stomach, rectum or urinary tract (Cardi et al., 2019).

The most accepted indications for immediate laparotomy are:

• Hemodynamic instability or shock.
• Peritonitis on physical examination.
• Evisceration of bowel loops or omentum.
• Bleeding from the stomach, rectum or urinary tract.
• Reduced level of consciousness that prevents a reliable examination, in many protocols.
• Free air or evident visceral injury on imaging, depending on the context.

Even these criteria have nuances. The review by Butt and colleagues cites a series in which evisceration, air under the diaphragm, blood on paracentesis and shock on arrival were not absolute criteria for exploration, and selective management was offered with a lower threshold for operating (Butt and colleagues, 2009). In practice, almost all services operate on the unstable patient or the patient with peritonitis, and selection begins with the stable patient.

Selective nonoperative management

The best synthesis on the topic is a meta-analysis of 41 studies with 22,847 patients with abdominal gunshot wounds, of whom 6,777, or 29.7 in every 100, were treated with selective nonoperative management. In stable patients without reduced level of consciousness and without peritonitis, the combined risk of failure was 7.0 in every 100, and the associated mortality was 0.4 in every 100 (Al Rawahi and colleagues, 2018). Among those who failed, 68 in every 100 had a therapeutic laparotomy.

The location of the wound changes the risk. In the same meta-analysis, failure was 3.4 in every 100 in the right thoracoabdomen, 7.0 in every 100 in the flank and 3.1 in every 100 in the back, and reached 13.2 in every 100 in the anterior abdominal wall. With mandatory computed tomography, failure was 4.1 in every 100, versus 8.3 in every 100 with selective computed tomography, a difference that did not reach statistical significance. The mean hospital stay was 6 days with successful management and 10 days when there was failure or a complication (Al Rawahi and colleagues, 2018).

Location of the wound	Failure of selective nonoperative management
Back	3.1 in every 100
Right thoracoabdomen	3.4 in every 100
Flank	7.0 in every 100
Anterior abdominal wall	13.2 in every 100
All stable, without peritonitis and conscious	7.0 in every 100

The liver is the classic example. In the series of penetrating liver injuries, 80.6 in every 100 of isolated liver injuries were successfully treated without surgery (Schnüriger and colleagues, 2011). A tract that passes only through the right liver, in a stable patient, is the most favorable scenario for observation.

Serial examination is the heart of selective management. The review by Butt and colleagues recommends repeated clinical examinations, preferably by the same physician, in the 12 to 24 hours after arrival, and notes that observation periods longer than 24 hours are rarely necessary (Butt and colleagues, 2009). The signs that prompt reconsidering surgery during observation include:

• Abdominal pain that increases or spreads away from the wound.
• Muscle guarding, rebound tenderness or rigidity.
• Persistent tachycardia or a drop in blood pressure.
• Fever that appears in the first hours.
• A drop in hemoglobin without another source of bleeding.
• Vomiting blood, blood on digital rectal examination or in the urine.

None of these signs alone proves injury, but their combination, or worsening between one examination and the next, usually leads to laparotomy or laparoscopy. That is why the patient under observation must stay in a place where the surgical team can examine them often and where the operating room is available at any hour. Observation also requires careful recording of each examination, with time, vital signs and findings, so that worsening is noticed even when the shift changes and another professional takes over the patient.

The role of computed tomography

Contrast-enhanced computed tomography is the tool that makes selection possible. It shows the tract, solid organ injuries, free blood and, in some patients, signs of hollow viscus injury. In the series of liver injuries, admission computed tomography had a sensitivity of 95.7 in every 100 and a specificity of 90.6 in every 100 for predicting a positive laparotomy (Schnüriger and colleagues, 2011).

Hollow viscus injury is the weak point. In a South African study of 437 patients with penetrating abdominal trauma evaluated with venous-contrast tomography alone, sensitivity for hollow viscus injury was 95.1 per 100, but specificity was 44.2 per 100, with a positive predictive value of 57.4 and a negative predictive value of 92 per 100 (Wolmarans et al., 2023). The authors concluded that tomography findings of hollow viscus injury are not very specific and should be interpreted together with the trajectory and serial abdominal examination.

A randomized clinical trial compared contrast tomography and diagnostic laparoscopy in stable patients with penetrating trauma of the anterior abdominal wall. Length of stay was similar, but the rate of non-therapeutic surgery was 65.4 per 100 in the laparoscopy group versus 17.4 per 100 in the tomography group. Tomography-guided selective management was successful in 93.8 per 100 patients (Kaur et al., 2023).

What tomography looks for in abdominal gunshot wounds:

• The projectile trajectory and its relation to bowel loops, colon, kidney, vessels and spine.
• Injuries to the liver, spleen and kidney, with or without contrast extravasation.
• Free fluid without a solid organ injury to explain it, a warning sign for hollow viscus.
• Air outside the bowel loops or bowel wall thickening along the trajectory.
• Diaphragm injury in thoracoabdominal wounds.
• Retroperitoneal hematoma and ureteral injury.

Diagnostic laparoscopy

Laparoscopy entered trauma as a triage tool. In a Dallas series of patients with penetrating abdominal trauma, review of negative laparotomies showed that possibly 8 of 10 gunshot wounds and all 14 stab wounds could have been evaluated by laparoscopy. The authors concluded that the technique is especially useful to rule out peritoneal penetration in wounds of the anterior wall and lower chest (Miles et al., 2004).

A South African pilot study compared early diagnostic laparoscopy with nonoperative management in stable patients who were asymptomatic or had few symptoms. About 66 per 100 in the laparoscopy group had peritoneal penetration, but no significant abdominal injury. Length of stay was 2 days in the observed group versus 3.1 days in the laparoscopy group, with no missed injuries, complications or deaths. The authors concluded that nonoperative management is preferable in these patients, but that laparoscopy accurately diagnoses diaphragm injury (Matsevych et al., 2019).

Laparoscopy has costs: it requires general anesthesia, an operating room and can cause iatrogenic injuries, as Kaur et al. recall (Kaur et al., 2023). Its place is in the stable patient with doubt about peritoneal penetration or diaphragm injury, especially in the left thoracoabdomen.

The thoracoabdomen and the diaphragm

The left thoracoabdominal region is defined, in the review by Butt et al., by the nipple line above and the costal margin below, with the sternum and the spine as medial limits. The same review cites a prospective series in which diaphragm injury was present in 59 per 100 gunshot wounds in this region (Butt et al., 2009).

Diaphragm injury is treacherous. It is often small, may not appear on tomography and causes almost no symptoms at first. On the left, where the liver does not protect, the stomach, colon and spleen can herniate into the chest months or years later, with a risk of strangulation. On the right, the liver tamponades part of the injuries, and that is why the right thoracoabdomen tolerates nonoperative management better, as the meta-analysis showed (Al Rawahi et al., 2018).

In a war setting, minimally invasive surgery has also been described. A Ukrainian series of thoracoabdominal gunshot wounds treated at a field hospital used video-assisted thoracoscopy, laparoscopy and magnetic instruments to remove fragments, combined with resuscitation and damage control surgery in severely injured patients (Lurin et al., 2024).

In this series of 36 patients, video-assisted thoracoscopy was used to remove metal fragments in 28 out of every 100, and 20 unstable patients needed thoracotomy and laparotomy. Among them, the liver was sutured in 8 patients and packed with compresses in 12, and there was one death (Lurin et al., 2024). The numbers show that, even in a low-resource setting, the minimally invasive technique coexists with open surgery and does not replace it in the unstable patient. The topic of occult injuries in high-energy trauma also appears in the text on blast trauma.

Damage control surgery

When the patient arrives in deep shock, with hypothermia, acidosis and coagulopathy, long definitive surgery can kill them on the table. Damage control surgery reverses the logic: the first operation is short and limited to controlling bleeding and contamination, with packing, ligation or stapling of bowel loops and temporary abdominal closure. The patient goes to intensive care for rewarming, correction of coagulopathy and resuscitation, and returns to the operating room for reconstruction when physiology allows.

In the series of 953 patients with abdominal war wounds, patients with uncontrolled abdominal bleeding and multiple intra- or extra-abdominal injuries that would require long operations underwent damage control laparotomy. Overall mortality was 12.8 out of every 100, and the variables associated with death included age above 34 years, injury from a projectile or mine, time since injury above 5 hours, severity score above 17 and associated injuries (Cardi et al., 2019). The most frequent injury was small bowel, in 46.3 out of every 100.

In an Ethiopian hospital, among 352 patients with penetrating abdominal trauma who underwent laparotomy, intensive care admission and damage control surgery were associated with mortality (Abdulkadir et al., 2022). This type of association mainly reflects the severity of the patients who need these measures, not a harmful effect of them. Hemorrhage control before arrival is discussed in the text on tourniquet and the MARCH sequence.

Colon, infection and complications

The treatment of colon injury has changed a great deal. The review by Butt et al. recounts that, after World War II, all colon injuries were treated with colostomy, a policy that lasted until the 1970s. From then on, colostomy was gradually replaced by primary repair in selected cases and, in the 1990s, by liberal primary repair in most injuries (Butt et al., 2009). Destructive injuries in severely injured patients may still require resection and, sometimes, a stoma.

Infection is the most common complication. A classic review on infection risk in penetrating abdominal trauma reports, in a trial comparing two antibiotic regimens, infection in 20 out of every 100 patients with cefoxitin and in 23 out of every 100 with clindamycin plus gentamicin (Rush and Nichols, 1986). The review by Butt et al. cites severe fecal contamination, transfusion of more than 3 units of blood in the first 24 hours and prophylaxis with a single antibiotic as independent risk factors for abdominal complications (Butt et al., 2009).

In the Ethiopian study, surgical site infection occurred in 23.9 out of every 100 patients and in-hospital death in 3.4 out of every 100. Age above 45 years, fluid collection, colostomy, temperature above 37.5 degrees and hemoglobin below 10 were associated with infection (Abdulkadir et al., 2022). The small bowel was the most injured organ, in 55.1 out of every 100 patients, followed by the colon and the liver.

A curious finding comes from a study on obesity and gunshot wounds to the anterior abdominal wall. Obese patients had less fascial penetration, fewer visceral injuries and lower severity scores, which the authors called a cushion effect (Patel et al., 2021). The study is small, with 40 patients, and does not change practice, but it is a reminder that the actual trajectory can be quite different from what is imagined.

How long does it take for nonoperative management to fail

The meta-analysis by Al Rawahi et al. pooled 41 studies with 22,847 patients with abdominal gunshot wounds, of whom 6,777 underwent selective nonoperative management (Al Rawahi et al., 2018). In stable, conscious patients without peritonitis, the pooled risk of failure, defined as laparotomy during the hospital stay, was 7.0 percent, and the mortality associated with management was 0.4 percent (Al Rawahi et al., 2018).

The most useful piece of data for the shift may be time. Among patients who failed, the mean interval between admission and delayed laparotomy was 40.6 hours, with a median of 21.2 hours (Al Rawahi et al., 2018). This means that many failures appear on the first day, but a share only becomes apparent on the second or third. Observation cannot end too early just because the patient did well on the first night.

What is found when management fails also matters. In patients operated after failure, the pooled risk of laparotomy being therapeutic was 68.0 percent, and the risk of it being unnecessary was 28.1 percent (Al Rawahi et al., 2018). In other words, most failures correspond to true injuries, and a relevant minority is operated on without finding an injury that needs repair.

The authors note that the cornerstone of the method is serial physical examination by qualified surgeons or experienced residents, in patients without impaired consciousness, and that CT of the abdomen and pelvis has been used as a complement (Al Rawahi et al., 2018). Nonoperative management for gunshot wounds remains more controversial than for stab wounds, because injuries to viscera and vessels are more frequent and a missed injury is costly (Al Rawahi et al., 2018).

How NEXO helps train this reasoning

Training with fictional clinical cases helps you recognize patterns before you meet them on the shift, as long as each clue is read calmly and compared with the hypotheses that are still open. In abdominal trauma, clues may bring the location of the wound, how the physical examination evolves over the hours and an imaging finding that changes the picture. Training yourself to wait for the next clue before concluding is discussed in the text on diagnostic pause.

The texts on differential diagnosis with clues, diagnostic uncertainty and clinical decision-making also help. To understand points and rewards, see the NEXO Economy page. There are also tracks for those who research (Researcher), for those who review cases (Curator), for those who produce content (Creators) and for those who want to follow results (Closer).

Frequently asked questions

Does every abdominal gunshot wound need surgery?

No. Unstable patients, patients with peritonitis or evisceration go to laparotomy. Stable, conscious patients without peritonitis can be selected for nonoperative management, with CT and serial examination, in services prepared to operate at any time.

Why does the anterior abdominal wall have more failures?

Because there the projectile meets bowel loops just below the wall, without the protection of the liver, kidney or thick musculature. Hollow viscus injuries are more likely and harder to see on CT.

Does a normal CT exclude injury?

Not entirely. Sensitivity for injuries that require surgery is high, but small hollow viscus and diaphragmatic injuries can be missed. That is why serial examination remains mandatory even with a normal CT.

How long should observation last?

Most failures appear in the first 12 to 24 hours. The review by Butt et al. notes that observation periods longer than 24 hours are rarely necessary, but the local protocol should be followed.

When is laparoscopy better than observation?

Mainly in left thoracoabdominal wounds, where diaphragm injury is frequent and hard to see on CT, and when there is doubt about peritoneal penetration in a stable patient.

Does a retained projectile need to be removed?

In general, no. Projectiles lodged in muscle, the retroperitoneum or bone are usually left in place if they cause no symptoms. Removal is considered when the projectile is inside a vessel, joint or spinal canal, when it migrates, or when it causes infection or persistent pain.

Are back and flank wounds less severe?

They have a lower failure rate for selective management, because the thick musculature and the position of the kidney and retroperitoneal colon reduce the chance of intraperitoneal injury. Even so, they can injure the kidney, ureter, colon, major vessels and spine, and contrast-enhanced CT is essential.

Take-home messages

• Operate without delay on the unstable patient, or one with peritonitis or evisceration.
• Select for observation only the stable, conscious patient without peritonitis.
• Use contrast-enhanced CT to define the trajectory and the injuries.
• Do not rely on CT alone to rule out hollow viscus injury.
• Repeat the abdominal examination, preferably by the same professional, within the first 24 hours.
• Investigate the diaphragm in left thoracoabdominal wounds.
• Consider damage control surgery in the patient in deep shock.

If you like reasoning from clues, today's case is waiting on the NEXO home page.

References

1. Abdulkadir A, Mohammed B, Sertse E, et al. Treatment outcomes of penetrating abdominal injury requiring laparotomy at Hiwot Fana Specialized University Hospital, Harar, Ethiopia. Front Surg. 2022. PMID 36081591.
2. Al Rawahi AN, Al Hinai FA, Boyd JM, et al. Outcomes of selective nonoperative management of civilian abdominal gunshot wounds: a systematic review and meta-analysis. World J Emerg Surg. 2018. PMID 30505340.
3. Butt MU, Zacharias N, Velmahos GC. Penetrating abdominal injuries: management controversies. Scand J Trauma Resusc Emerg Med. 2009. PMID 19374761.
4. Cardi M, Ibrahim K, Alizai SW, et al. Injury patterns and causes of death in 953 patients with penetrating abdominal war wounds in a civilian independent non-governmental organization hospital in Lashkargah, Afghanistan. World J Emerg Surg. 2019. PMID 31832085.
5. Kaur S, Bagaria D, Kumar A, et al. Contrast-enhanced computed tomography abdomen versus diagnostic laparoscopy-based management in patients with penetrating abdominal trauma: a randomised controlled trial. Eur J Trauma Emerg Surg. 2023. PMID 35980448.
6. Lurin I, Vorovskiy O, Makarov V, et al. Management of thoracoabdominal gunshot injuries by using minimally invasive surgery at role 2 deployed field hospitals in Ukraine. BMC Surg. 2024. PMID 38877409.
7. Matsevych OY, Koto MZ, Balabyeki M, et al. Diagnostic laparoscopy or selective non-operative management for stable patients with penetrating abdominal trauma: What to choose? J Minim Access Surg. 2019. PMID 30178773.
8. Miles EJ, Dunn E, Howard D, et al. The role of laparoscopy in penetrating abdominal trauma. JSLS. 2004. PMID 15554270.
9. Patel BM, Samsonov AP, Patel JR, et al. Obesity and Anterior Abdominal Gunshot Wounds: A Cushion Effect. Cureus. 2021. PMID 34963852.
10. Rush DS, Nichols RL. Risk of infection following penetrating abdominal trauma: a selective review. Yale J Biol Med. 1986. PMID 3532576.
11. Schnüriger B, Talving P, Barbarino R, et al. Current practice and the role of the CT in the management of penetrating liver injuries at a Level I trauma center. J Emerg Trauma Shock. 2011. PMID 21633569.
12. Wolmarans A, Fru PN, Moeng MS. Accuracy of CT Scan for Detecting Hollow Viscus Injury in Penetrating Abdominal Trauma. World J Surg. 2023. PMID 36859569.