“Awake” laparoscopy for the evaluation of equivocal penetrating abdominal wounds

Introduction 

While the majority of surgeons would agree that exploratory laparotomy is essential to the management of abdominal gunshot wounds, opinion varies concerning the evaluation of stab wounds to the abdomen. Appreciation of both the relatively lower incidence of visceral injury and the potential morbidity of negative laparotomy has resulted in a shift toward selective management of abdominal stab wounds.2 Such selective management algorithms vary from institution to institution and rely on diagnostic tests, serial abdominal examination or some combination of the two. A similar practice has evolved around the evaluation of tangential abdominal gunshot wounds. In 1993, we prospectively evaluated the efficacy and safety of diagnostic laparoscopy in the trauma setting and found that our incidence of negative laparotomy was reduced to 17% with the incorporation of laparoscopy into our evaluation of penetrating injury.3 It has since become our practice to evaluate anterior abdominal stab and tangential gunshot wounds with equivocal peritoneal penetration by diagnostic laparoscopy.

In that initial report, we speculated that in selected patients, diagnostic laparoscopy could be performed under local anaesthesia in the resuscitation area (“awake” laparoscopy, AL), reducing costs and allowing expedited discharge for those patients with no evidence of peritoneal penetration. In this study, we report our preliminary experience with awake laparoscopy performed in our emergency department (ED).

Patients and methods 

During the study period, all patients presenting to the Presley Regional Trauma Centre with anterior abdominal stab wounds (anterior to the mid-axillary line) and anterior tangential gunshot wounds were evaluated by the clinical pathway demonstrated in Fig. 1. Tangential gunshot wounds were evaluated for peritoneal penetration with laparoscopy, provided the patient was haemodynamically stable and without evidence of peritonitis on physical examination. For stab wounds, local wound exploration was performed by sharply enlarging the wound to allow evaluation of the underlying fascia. If the anterior fascia was found to be intact, the evaluation was complete and the patient was discharged from hospital following wound care and the management of any additional injuries or medical concerns. If the anterior fascia was found to be violated, the peritoneum was then evaluated by laparoscopy. Peritoneal penetration was not purposefully evaluated during local wound exploration; however, if penetration was obvious, laparoscopy was not performed and the patient was taken to the operating room for exploratory laparotomy. Local wound exploration was omitted at the discretion of the surgeon when it was felt that the procedure would be unrewarding, given the body habitus of the patient and/or trajectory of the wound.

View Large Image Download to PowerPoint

Figure 1. Clinical pathway for the evaluation of anterior abdominal stab and tangential gunshot wounds.

Since 2003, patients requiring laparoscopy according to the pathway were evaluated for awake laparoscopy by the surgeon on duty. Along with the presence of haemodynamic stability and absence of peritonitis, suitability for AL necessitated an alert, cooperative patient. This assessment was performed by the attending surgeon.

Following the attainment of informed consent from the patient, periprocedural sedation (intravenous morphine and midazolam) was administered. A video monitor, lightsource and laparoscopic insufflator were stationed at the bedside. The abdomen was prepped with iodine solution and draped with sterile operating room towels and sheets. Through a 5mm periumbilical incision, an Endopath bladeless 5mm trocar (Ethicon, NJ, USA) was introduced into the peritoneal cavity under direct vision with a zero degree 5mm endoscope (Fig. 2). Pneumoperitoneum was then established and a relatively low intra-abdominal pressure was maintained (7–10mmHg) to allow visualisation of the peritoneum while maintaining patient comfort. The peritoneum was then examined with a 0° and/or 30° endoscope for peritoneal violation. In the absence of peritoneal violation, pneumoperitoneum was evacuated, and the skin wound was closed with an absorbable suture. The patient was then allowed to recover from intravenous sedation in the emergency room prior to dismissal. If peritoneal violation was identified, pneumoperitoneum was evacuated and the patient was transferred to the operating room for exploratory laparotomy.

View Large Image Download to PowerPoint

Figure 2. Access to the peritoneal cavity is obtained under direct vision with an Endopath bladeless 5mm trocar (Ethicon, NJ, USA) and a zero degree 5mm endoscope (Stryker, MI, USA) as assembled above.

The hospital and clinic records of all patients who underwent AL were reviewed. As well, the records of all patients who, during the study period, were deemed to be unsuitable for AL and underwent diagnostic laparoscopy (DL) in the operating room were reviewed. From this group, a matched control group of 24 patients was identified. In the control group, all patients had no evidence of peritoneal penetration on diagnostic laparoscopy (negDL) and had no associated injuries requiring further management. Length of stay and hospital charges were compared between the negDL group and those patients who underwent AL and were found to have an intact peritoneum (negAL). Length of stay was defined as the interval between presentation to the emergency room and hospital dismissal as recorded in the patient's medical record. Statistical evaluation was performed with unpaired Student's t-test.

Results 

Over a 30-month period (May 2003–February 2006), 188 patients with anterior abdominal stab wounds or tangential gunshot wounds underwent laparoscopy (both DL and AL) according to protocol for the evaluation of peritoneal penetration. One hundred and twenty-one patients were found to have no peritoneal violation. Of the remaining 67 patients with laparoscopically proven peritoneal penetration, 31 (46.3%) underwent a therapeutic laparotomy, 21 (31.3%) had a non-therapeutic laparotomy and 15 (22.4%) did not undergo exploratory laparotomy (i.e. absence of organ injury was determined by laparoscopy).

Fifteen patients (8%) in the above group underwent AL. Fourteen of the 15 patients were male. Mean age was 33 (19–52) years. Thirteen wounds were the result of stabs and two were the result of tangential gunshots. Four patients underwent local wound exploration with confirmation of anterior fascial penetration. As per protocol, all were haemodynamically stable, alert and cooperative, without evidence of peritonitis.

On inspection of the peritoneal cavity with the laparoscope, four patients were identified to have peritoneal penetration, and were taken to the operating room for exploratory laparotomy. In two of these patients, no intra-abdominal injuries were identified. The remaining two patients had positive injuries requiring diaphragm repair and gastric repair in one patient, and diaphragm repair and hepatorrhapy in the other. During AL in one patient, penetration of the peritoneum was felt to be marginal. This patient was observed for approximately 16h and then discharged home.

The remaining 10 patients were found to have no evidence of peritoneal penetration with AL (ALneg). The ALneg group included the two patients with tangential gunshot wounds. Nine were subsequently dismissed from the emergency room and one was admitted to the psychiatry service for further management. No patient developed any known complications as a result of AL. Five of the 10 patients in the ALneg group returned for outpatient follow-up.

The ALneg group was compared with the DLneg group with respect to length of stay and hospital charges. The patient admitted to the psychiatry service was excluded from analysis of length of stay. Mean length of stay was 7h (range, 3–10h) in the ALneg group versus 18h (range, 9–38h) in the DLneg group (p=0.0003). The patient with the longest length of stay in the AL group was held for a follow-up chest X-ray to rule out delayed presentation of pneumothorax as a result of a thoracic stab wound. Hospital charges were compared and are demonstrated in Table 1. Cost savings averaged US$ 2227 per patient in the ALneg group.

Table 1. Comparison of average hospital charges between ALneg and DLneg groups (data abstracted from patient billing data, charges in USD)

		ALneg	DLneg
	Anaesthesia charge	NA	1243
	Operating room charge	NA	3085
	Recovery room charge	NA	674
	Hospital bed charge	NA	384
	Procedure charge	3159	NA
	Total	3159	5386

Discussion 

It has become clear that the practice of mandatory laparotomy for all penetrating abdominal wounds can be relegated to surgical history. Series from multiple institutions have demonstrated that the majority of patients with abdominal stab wounds who are haemodynamically stable and without peritonitis do not have associated intra-abdominal injuries warranting operative therapy.1, 5, 7, 8, 10, 11, 12, 13 Selective management of these cases has become standard in most institutions, but there is considerable divergence in just how selective management is defined and implemented.

Observation with serial physical examination has become the dominant selective management strategy, first proposed by Shaftan in 1960.9 Demetriades and Rabinowitz1 reported success with this approach in 306 patients, citing a 4% incidence of delayed laparotomy with minimal complications. Others have incorporated diagnostic peritoneal lavage with observation in their management algorithms.4, 12 Although good results can be obtained with observation, there are inherent limitations to this approach. Observation requires that serial abdominal examinations be performed by an experienced surgeon able to detect subtle changes in physical findings. In the era of resident work hour restrictions and physician extenders, providing both this level of care and continuity of care may be challenging. Observation also requires hospital admission, taxing the resources of overburdened institutions where beds are a scarce commodity. Although most reports have reported minimal morbidity with delayed treatment of injury, it is folly to conclude that delayed treatment of hollow viscus injury may be performed with impunity. Delay in treatment will negatively influence the hospital course of some patients. Martin et al.6 demonstrated that a delay in treatment of colon injury greater than 12h resulted in a complication rate of 18% versus 3%.

Given these limitations of observation, it has been our preference to base selective management on peritoneal integrity, with the philosophy that patients with an intact peritoneal lining may be discharged without further observation or management. Beginning in the early 1990s, we incorporated laparoscopy into our evaluation of peritoneal integrity. With the development of the 5mm bladeless trocar, we have been able to apply laparoscopic technology in the emergency room setting as described in this report. The bladeless 5mm trocar is well suited for placement under local anaesthesia and eliminates the need to close the abdominal fascia. Performance of AL in the emergency room offsets some of the costs of this procedure and expedites dismissal for some patients.

Our experience with AL has demonstrated its limits. Certainly, this procedure is appropriate for a select group of patients; only 8% of patients were deemed suitable for this procedure over the period of study. It is reasonable to expect that many patients with such injuries will be intoxicated and/or uncooperative, necessitating general anaesthesia should laparoscopy be performed. Patient tolerance for the procedure is also unpredictable. While some patients tolerated pneumoperitoneum quite well with minimal analgesic and sedation, other patients required increasing doses of narcotic and benzodiazepine to allow for completion of the procedure. No case, however, was aborted secondary to patient pain and/or non-compliance.

Another theoretical limitation to AL is the difficulty in using additional laparoscopic operating portals or positioning of the patient to facilitate visualisation of the peritoneal lining or diaphragm. AL is not suited to the introduction of multiple operating portals or canted bed positioning. In our experience, however, visualisation was adequate in all cases, facilitated by an angled endoscope and manual pressure on the abdominal wall to better demonstrate the peritoneum to the camera. It is important to note, we reserve awake laparoscopy for the evaluation of wounds anterior to the mid-axillary line.

AL offers little option of proceeding with therapeutic techniques. Although therapeutic laparoscopy has been described for selected traumatic injuries, it has not been embraced at most institutions, including our own, primarily because of the concern for missed injury. It is likely that therapeutic laparoscopy for traumatic injuries will become more prevalent as the next generation of surgeons becomes more skilled with advanced laparoscopic techniques and the technology itself continues to evolve. It is unclear at this time as to what place AL will have therapeutic laparoscopy for trauma become routine.

In summary, we have found AL to be a useful adjunct to our clinical pathway for the evaluation of peritoneal violation, allowing for quick diagnosis and disposition. The primary limitation of AL is its applicability to our patient population, given that only a limited number of patients were deemed suitable for the procedure. At present, we acknowledge this limitation, reserving AL for the cooperative patient who we suspect will have an intact peritoneum, with the goal of expedited hospital dismissal. As our experience with this procedure increases, we expect to perform AL on a greater percentage of patients with the goals of achieving larger scale reductions in cost and length of stay.