- Email: [email protected]
The utility of diagnostic laparoscopy in the evaluation of anterior abdominal stab wounds
The American Journal of Surgery (2008) 196, 871– 877
The Southwestern Surgical Congress
The utility of diagnostic laparoscopy in the evaluation of anterior abdominal stab wounds Tammy R. Kopelman, M.D., F.A.C.S.a,*, Patrick J. O’Neill, Ph.D., M.D., F.A.C.S.a, Luis H. Macias, M.D.a, Jordy C. Cox, M.D., F.A.C.S.a, Marc R. Matthews, M.D., F.A.C.S.a, David A. Drachman, Ph.D.b a
Division of Burns, Trauma Surgery and Surgical Critical Care, Department of Surgery, and bDepartment of Research, Maricopa Medical Center, Phoenix, AZ 85008, United States KEYWORDS: Anterior abdominal stab wounds; Diagnostic laparoscopy; Computed tomography; Diagnostic peritoneal lavage; Hospital charges; Laparotomy; Length of stay; Local wound exploration; Nontherapeutic; Serial abdominal examinations; Therapeutic
Abstract BACKGROUND: To assess if diagnostic laparoscopy (DL) is superior to nonoperative modes (serial abdominal examination with/without computed axial tomography [CAT] and diagnostic peritoneal lavage) in determining the need for therapeutic laparotomy (TL) after anterior abdominal stab wound (ASW). METHODS: Retrospective review of ASW patients. Patients were divided into group A (DL/ exploratory laparotomy) to identify peritoneal violation (PV) and group B (initial nonoperative modes). RESULTS: Seventy-three patients met inclusion criteria. In group A (n ⫽ 38), 29 patients (76%) had PV by DL and underwent exploratory laparotomy. Only 10 (35%) underwent TL (sensitivity for PV ⫽ 100%; specificity and positive predictive value of PV in determining need for TL ⫽ 29% and 33%, respectively). In group B (n ⫽ 35), 7 patients (20%) underwent TL, yielding an improved specificity (96%) and positive predictive value (88%). CONCLUSIONS: We find no role for DL in the evaluation of ASW patients solely to determine PV. © 2008 Elsevier Inc. All rights reserved.
There is little argument that patients who present with abdominal stab wounds (ASWs) and hemodynamic compromise, evisceration, or generalized peritonitis require immediate laparotomy. However, only approximately one third of patients presenting with anterior ASWs actually sustain an injury that requires surgical intervention. This
Presented at the Southwestern Surgical Congress (60th Annual Meeting), Acapulco, Mexico, March 30 to April 2, 2008 * Corresponding author. Tel.: ⫹1-602-344-5637; fax: ⫹1-602-3445705. E-mail address: [email protected] Manuscript received May 3, 2008; revised manuscript July 3, 2008
0002-9610/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2008.07.031
finding is related to the fact that ⬎25% of anterior ASWs do not penetrate the peritoneal cavity,1 and only approximately one half of wounds that do violate the peritoneum cause visceral injury requiring surgical repair.2 In contrast, major complications and significant mortality rates caused by missed injuries after penetrating abdominal trauma may be as high as 83% and 17%, respectively.3 Therefore, the decision to operate on the asymptomatic patient after anterior ASW injury remains a challenge. The efficacy of diagnostic laparoscopy in decreasing nontherapeutic rates by excluding peritoneal violation has been studied in patients with abdominal ballistic injuries of uncertain trajectory.4 Using diagnostic laparoscopy, evi-
872
The American Journal of Surgery, Vol 196, No 6, December 2008
dence of peritoneal violation reliably confirmed the need for therapeutic laparotomy in 82% of patients and drastically decreased the incidence of negative laparotomy to 1.2%. This approach produced a positive predictive value of a surgical abdomen in a patient sustaining a ballistic injury to 97.6%.4 Since that time, many centers have attempted to expand the indication of diagnostic laparoscopy to include the evaluation of anterior ASWs with the similar goal of using peritoneal violation to determine the need for further exploration.5– 8 The use of diagnostic laparoscopy, however, still exposes the patient to general anesthesia and the risk of iatrogenic injury during exploration, not to mention a great monetary cost burden. Thus, even minimally invasive diagnostic laparoscopy is not without finite risk. To be successful, the optimal diagnostic evaluation for patients sustaining anterior ASWs must accomplish 2 things: First, negative laparotomy rates should be minimized, and second, sensitivity for surgically significant intra-abdominal injury must be maximized to avoid delays in diagnosis and subsequent morbidity and mortality. The optimal “nonoperative” method to accomplish this goal will likely include a multitude of diagnostic modalities. With this in mind, the objective of this study was to assess if diagnostic laparoscopy to assess for violation of peritoneum would be superior to nonoperative modes (serial abdominal examination with/without computed axial tomography (CAT) and diagnostic peritoneal lavage [DPL]) for determining which patient sustaining anterior ASW injury would require therapeutic laparotomy.
Abdominal Stab Wounds (n=244)
Hypotension ? Peritonitis? Evisceration ?
Yes
To Operating Room for Exploratory Laparotomy (n=65)
No
Local Wound Exploration?
No
CT Scan / Diagnostic Peritoneal Lavage Uncertain Fascial Status (n=38)
Yes
Negative Local Wound Exploration ?
Yes
No Intra-abdominal Injury (n=18)
No
Concern for Diaphragm Injury?
Yes
To Operating Room for Exploratory Laparoscopy or Laparotomy (n=50)
No
Study Group (n=73) Group A: + LWE plus Laparoscopy / Laparotomy (n = 38) Group B: + LWE plus CT / Serial Exam / DPL (n = 35)
Methods
Figure 1 Flow diagram of study patients. ⫹LWE: positive local wound exploration.
This is a retrospective cohort study performed at a 449bed urban teaching level I adult and pediatric trauma center. After obtaining Institutional Review Board approval, the trauma registry was queried for all anterior ASW patients presenting during a 4-year period (October 2003 to 2007) (Fig. 1). The anterior abdomen was defined as the area bordered by the costal margins superiorly, the inguinal ligaments inferiorly, and the anterior axillary lines laterally. Patients were excluded on the following basis: (1) indication for emergent laparotomy (ie, hemodynamic instability, peritonitis, evisceration, etc); (2) lack of evidence indicating anterior fascial violation (either by failure to perform or by identifying a negative local wound exploration); or (3) concern for left-sided diaphragm injury. Patients were then divided into 2 groups based on their diagnostic evaluation and positive local wound exploration. Group A included those patients who underwent diagnostic laparoscopy or exploratory laparotomy to identify peritoneal violation. Group B included patients whose initial nonoperative management was composed of serial physical examination and/or CAT as well as DPL. After we identified the final study groups, a systematic review of electronic medical records was performed. De-
identified demographic data were gathered using a closedresponse data collection form that included patient demographics, physical examination and laboratory findings, non–psychiatric-influenced length of stay (LOS), intra-abdominal injuries, complications, and hospital charges. DPL was performed in the emergency department by way of a percutaneous method. A positive lavage was signified by aspiration of 10 mL gross blood, an effluent erythrocyte count greater than or equal to 20,000 cells/mm3, leukocyte count ⱖ500 cells/mm3, and/or the presence of bacteria on Gram’s stain. CAT was performed by certified technologists with a helical 16-detector scanner (GE LightSpeed 16 Pro, Waukesha, Wisconsin). Intravenous contrast (150 mL Omnipaque) was injected by a power injector at a rate of 3 mL/s with a scanning delay of 60 seconds. Oral contrast was not administered. Radiographic interpretations were performed by in-house, attending radiologists. Diagnostic laparoscopy was performed in the standard fashion with an initial infraumbilical 5- or 12-mm camera port. Negative diagnostic laparoscopy was defined as the absence of peritoneal penetration, whereas positive diagnostic laparoscopy noted the presence of peritoneal penetration.
T.R. Kopelman et al. Table 1
Anterior abdominal stab wounds
873
Patient admission demographics and injury pattern characteristics
Patient demographics Average age in years (range) Male/female Location of stab wound (%) Self-inflicted injury (%) Nontherapeutic laparotomy rate (%) Therapeutic laparotomy rate (%) Survival (%)
Group A (initial surgical management) (n ⫽ 38 patients)
Group B (initial nonoperative management) (n ⫽ 35 patients)
30.8 (16–80) 32/6
30.3 (13–60) 29/6
EPI 4 LLQ 15 RLQ 22 UMB 19 RUQ 33 MULT 7 10 (26)
EPI 10 LLQ 10 RLQ 10 UMB 20 RUQ 30 MULT 0 16 (46)
19/29 (65.5)a
11/8 (12.5)
10/38 (26) 37/38 (97.4)
7/35 (20) 35/35 (100)
EPI ⫽ epigastrum; LLQ ⫽ left lower quadrant; MULT ⫽ multiple; RLQ ⫽ right lower quadrant; RUQ ⫽ right upper quadrant; UMB ⫽ umbilical. a Based on peritoneal violation.
Laparotomy was defined as therapeutic if there was a visceral injury that required surgical repair, nontherapeutic if a visceral injury was identified that did not require surgical repair (ie, nonbleeding solid-organ injury), and negative if no visceral injury was noted. All data were entered into an electronic database (Microsoft Excel, Microsoft Corporation, Redmond, Washington) and were analyzed using SPSS version 15.0.
Results During the 48-month study period, 244 patients presented to our urban level I trauma center with ASWs (Fig. 1). After the appropriate patients were excluded, 73 patients
Table 2
(30%) met our inclusion criteria and underwent further analysis. Within this final study group, 38 patients (52%) underwent diagnostic laparoscopy to exclude peritoneal violation (group A), and 35 (48%) patients underwent either DPL or serial abdominal exams with or without abdominal/ pelvic CAT (group B). Demographic, injury patterns, and survival characteristics of the study populations are listed in Table 1. Among patients in group A, 29 of 38 (76%) were noted to have peritoneal violation and underwent exploratory laparotomy to exclude intra-abdominal injury. At laparotomy, approximately one third (10 of 29 patients) were noted to have injuries requiring therapeutic intervention (Table 2). Diagnostic laparoscopy therefore had a sensitivity of 100% for determining peritoneal violation and a specificity and
Summary of surgical findings Findings at laparotomy (%)
Diagnostic evaluation
Laparotomy (%)
Group A (n ⫽ 38)
29 (76)
Group B (n ⫽ 35)
8 (23)
Sm intest ⫽ small intestine. a Positive peritoneal violation.
a
Negative
Nontherapeutic
Positive
Missed injury
18 (62)
1 (3.5)
0 (0)
0 (0)
1 (12.5)
10 (34.5) Liver: 4 Kidney: 0 Sm intest: 3 Colon: 1 Mesentery: 2 Omentum: 1 Ovary: 1 7 (87.5) Liver: 4 Kidney: 1 Sm intest: 3 Colon: 1 Mesentery: 4 Omentum: 0 Ovary: 0
0 (0)
874
The American Journal of Surgery, Vol 196, No 6, December 2008
Table 3
Postsurgical complications
Complications
Group A (n ⫽ 38)
Group B (n ⫽ 35)
Ileus Pneumonia Wound dehiscence Wound infection Emesis Fever Small intestinal obstruction Acute renal failure Acute respiratory distress syndrome Cardiac ectopy Myocardial infarction Iatrogenic splenic injury Iatrogenic small intestinal injury Total patients with morbidity (%) Mortality (%)
2 1 1 1 1 1 1 1 1 1 1 1 2 9a (24) 1 (2.6)
2 0 1 0 1 0 0 0 0 0 0 0 0 4 (11) 0 (0)
a
Multiple complications may have occurred in a single patient.
positive predicative value for determining the need for therapeutic laparotomy of 29% and 33%, respectively. All patients with negative laparoscopy were discharged home and without complications. Furthermore, no patient in group A was found to have a missed injury during their hospital stay or in follow-up (Table 2). The only patient death in this study occurred because of postsurgical complications after negative laparotomy was performed for peritoneal violation that was noted on laparoscopy. Conversion to exploratory laparotomy was complicated by both intraoperative and postsurgical complications, and a greater morbidity rate was noted in the nontherapeutic subgroup (Table 3). Of the 35 group B patients who underwent initial nonoperative diagnostic management, 19 patients (54%) underwent serial abdominal examinations with or without abdominopelvic CAT scan, and 16 patients (46%) underwent DPL. In this group, 7 patients (20%) subsequently required therapeutic laparotomy (Table 1). One patient underwent a nontherapeutic laparotomy after a stab wound to the right upper quadrant. CAT showed air in the soft tissues through the fascia with extension into the peritoneal cavity but no evidence of intra-abdominal fluid, air, or injury. Despite this, the attending surgeon’s preference was to evaluate this
Table 4
patient surgically because he had mild abdominal discomfort, and a small nonbleeding hematoma was noted in the mesentery of the colon. No intervention was required. In contrast, the 1 false-positive finding in this group was a female patient with isolated free fluid noted on CAT and a clinically benign examination. Subsequent serial physical examinations failed to identify clinical indications for surgical intervention, and she was discharged to home without surgical intervention. As in group A, no delayed identification of injuries was noted on serial physical examination in any patient. Thus, nonoperative diagnostic management in group B showed a sensitivity of 100%, a specificity of 96%, and a positive predictive value of 88% while maintaining a negative predictive value of 100%. The average length of stay for those undergoing nonoperative management or negative diagnostic laparoscopy was equivalent (Table 4). Similarly, the length of stay in all patients undergoing laparotomy, therapeutic or not, was similar in both groups. It should be noted, however, that 19 of 29 patients (66%) in group A had peritoneal violation and underwent nontherapeutic exploratory laparotomy (Table 1), which increased their length of stay by 2.5 days and charges by ⬎80% compared with the nonoperative arm (Table 4).
Comments There is little doubt that patients with peritonitis, evisceration, or hemodynamic instability following anterior ASW require surgical intervention. It is the stabbed patient with minimal to no symptoms that remains a diagnostic enigma to the trauma surgeon. Reports on the incidence of unnecessary laparotomy cite rates of 23% to 65% in patients presenting with ASWs.6,8 –10 Because of this nontherapeutic surgery, about 40% of patients will have complications, including atelectasis, postsurgical hypertension, pleural effusion, pneumothorax, prolonged ileus, pneumonia, surgical wound infection, small-bowel obstruction, urinary infection, and others.11 In addition, these avoidable surgeries lead to increased inpatient length of stay and subsequent greater health care costs. Clearly, complication rates and increased expense must be weighed against the mortality and morbidity of a missed injury when deciding how to evaluate a patient after anterior ASW.
Non–psychiatric-influenced hospital LOS and charges Group A
Length of stay in days (range) Charges
Group B
No PV/DL alone (n ⫽ 4)
⫹ PV nontherapeutic laparotomy (n ⫽ 15)
⫹ PV therapeutic laparotomy (n ⫽ 9)
No laparotomy required (n ⫽ 15)
Therapeutic laparotomy (n ⫽ 4)
1.4 (1–3) $20,310
4.5 (2–15) $30,275
4.2 (2–10) $32,190
1.4 (1–4) $16,620
5.4 (4–8) $30,585
PV ⫽ peritoneal violation; DL ⫽ diagnostic laparoscopy.
T.R. Kopelman et al.
Anterior abdominal stab wounds
As far back as the 1960s, Shaftan12 described “selective conservatism” in the management of patients after ASW using serial physical examination to identify those patients requiring laparotomy, hereby decreasing the incidence of unnecessary surgery. Since that time, many studies have confirmed this finding, citing the use of serial abdominal examination to safely and reliably predict the need for laparotomy.13–15 In this management arm, patient admission is necessary because hemodynamic monitoring, frequent serial abdominal examinations, white blood cell count and hemoglobin follow-up, as well as serial temperature measurements are typically required for at least 24 hours. The onset of hemodynamic instability, increasing abdominal pain or tenderness, increasing temperature, decreasing hemoglobin, or leukocytosis would warrant laparotomy or further investigation. Ertekin et al9 reported on 117 patients with penetrating stab wounds to the thoracoabdominal region (67% anterior abdomen), all of whom underwent nonoperative management. Clinical examination was the primary tool to differentiate those patients requiring surgical intervention, but this was often combined with other modalities, such as DPL, ultrasound, or CAT. DPL was used in patients whose physical examinations were not reliable because of alcohol or drug use or unconsciousness. In their study, nonoperative management was successful in 79% of patients with 11 early (within 8 hours of admission) and 14 delayed (⬎8 hours after admission) laparotomies performed. They had no negative laparotomies in the entire series, and the rate of nontherapeutic laparotomy remained low at 9% in the early group and 14% in the delayed group. No morbidity or mortality was attributed to delays in surgical intervention, which ranged from 1.5 hours to 20 hours. This group concluded that physical examination alone and/or coupled with different diagnostic methods allowed for a decrease in nontherapeutic laparotomies and eliminated negative laparotomy.9 Another option that exists in the nonoperative management of anterior ASW is CAT scanning. Eastern Association for the Surgery of Trauma guidelines emphasize that the use of triple-contrast abdominal/pelvic CAT should be strongly considered as a diagnostic tool in the initial evaluation of penetrating abdominal trauma to predict the need for surgical intervention.16 Despite this recommendation, there still remains a reluctance to rely on CAT because of its varying sensitivity, specificity, and accuracy in the diagnosis of bowel and mesenteric injuries.17–19 These concerns may be unfounded, however, because accuracy appears to be improving with advancing technology and the advent of helical and multislice CAT. Evidence of improved imaging may be found in the study by Shanmuganathan et al,20 which used CAT to assess penetrating abdominal injuries in 200 patients. They found that triple-contrast abdominopelvic CAT had a sensitivity of 97%, a specificity of 98%, and a negative predictive value of 98%. Similar results were noted by Demetriades and Rabinowitz21 in a prospective study of 156 stable patients in
875 who all sustained anterior ASWs and were admitted for serial abdominal examinations with or without CAT. Excluding patients with thoracoabdominal stab wounds, the results of this study demonstrated that CAT using intravenous contrast only demonstrated a high sensitivity (100%) with a moderately high specificity (81%) in identifying those patients with injuries. More importantly, it showed a negative predictive value of 100%. In this study as well, CAT correctly identified all patients with bowel injury, and no false-negative scans were noted.21 Group B, our nonoperative diagnostic arm, was a compilation of all the above-cited modes of evaluation, using a combination of serial examination, DPL, and CAT to determine the need for surgical intervention. The results of our study, namely, a sensitivity of 100%, a specificity of 96% and a positive predictive value of 88% while maintaining a negative predictive value of 100%, only serve to confirm the efficacy of nonoperative diagnostic modalities in the evaluation of patients with anterior ASWs. More recent literature has identified diagnostic laparoscopy as an important tool in the evaluation of ASWs. Compared with mandatory laparotomy, this minimally invasive evaluation allows for lower rates of nontherapeutic laparotomy by identifying the patient at risk for intra-abdominal injury after penetrating injury.8 In the current study, we chose to evaluate our experience with diagnostic laparoscopy to assess patients with anterior ASWs and to determine the need for exploratory celiotomy by identifying peritoneal violation. In our experience, approximately 25% of patients avoided exploratory laparotomy by excluding intra-abdominal penetration by the stab wound, and an additional 25% underwent therapeutic laparotomy after peritoneal violation was noted on laparoscopy. However, this mode of screening for the need for open exploration still led to an unnecessary laparotomy rate of 62% and an overall nontherapeutic laparotomy rate of 66%, which was no different from historic controls. Other studies confirm our results when peritoneal penetration was identified as the impetus to convert to exploratory laparotomy, citing nontherapeutic rates of 50% to 65%.6,8 In our experience, peritoneal violation was noted to have a sensitivity of 100% but a specificity and positive predictive value for determining the need for therapeutic laparotomy of only 29% and 33%, respectively. In certain patients, investigators advocate not only screening for peritoneal violation with laparoscopy but actually using this minimally invasive technique for therapeutic endeavors when peritoneal violation is encountered.5,22 Ivatury et al23 concluded that in carefully selected patients, it can be practical and feasible. However, in their study of 207 patients undergoing celiotomy after laparoscopy, 9.2% were found to have additional injuries requiring surgical intervention. In addition, 15 patients were found to have additional bowel injuries on laparotomy that were not visualized on laparoscopic evaluation, and 4 patients had significant vascular injuries (vena cava, common iliac vein) at laparotomy that were not suspected at laparoscopy.23 Fur-
876
The American Journal of Surgery, Vol 196, No 6, December 2008
thermore, in an outcome analysis performed by Villavicencio and Aucar,24 37 studies with ⬎1,900 trauma patients were reviewed, and the use of laparoscopy as a screening, diagnostic, and therapeutic tool was analyzed. Laparoscopy was regarded as a diagnostic tool only in studies that mandated a laparotomy (“gold standard”) after laparoscopy to confirm the diagnostic accuracy of the laparoscopic finding (rather than as a screening tool to identify the need for laparotomy). However, he noted that in this setting that laparoscopy had a 41% to 77% missed-injury rate/patient and therefore concluded that its value in this context was limited. Another distinct disadvantage of diagnostic laparoscopy is that it is performed in the operating room and requires general anesthesia. A 1999 review of 37 studies, which included more than 1,900 patients, demonstrated a procedure-related complication rate of 1%.24 Intraoperative complications may occur during creation of the pneumoperitoneum or trocar insertion or during the diagnostic examination. Our nonoperative arm is not without its own inherent complications. DPL is also an invasive procedure and carries a small but significant risk of iatrogenic intraabdominal injury in the range of 0.6%,25 but it may be performed safely in the emergency department. CAT with intravenous contrast administration carries the risk of a radiation exposure as well as contrast reaction, extravasation of contrast material at the site of intravenous injection, and contrast medium–induced nephropathy. The development of contrast medium–induced nephropathy is low in people with normal renal function, varying from 0% to 5%,26,27 but it has been reported in several prospective controlled studies to increase to rates of 12%–27% in patients with preexisting renal dysfunction.27,28 In the current study, no complications were noted related to DPL or CAT. There were, however, intraoperative complications related to the examination during diagnostic laparoscopy with the creation of enterotomies during lysis of adhesions in an attempt to visualize peritoneal violation. From a cost perspective, those advocating screening laparoscopy have reported decreased mean hospital LOS in negative laparoscopy patients compared with negative laparotomy patients with significant decrease in mean LOS.6,10,29 In the current study, the charges of negative laparoscopy was comparable with that of a negative nonoperative evaluation and considerably less than that of the negative laparotomy group, but that is exactly the issue at hand. It is our opinion that this additional group must be considered when contemplating cost, namely, the patients with false-positive laparoscopic evaluations who undergo nontherapeutic laparotomy simply on the basis of peritoneal violation. In our study, this group increased their average length of stay by during 2.5 days and charges by ⬎80% compared with their nonoperative counterparts who did not undergo surgical intervention. This accounted for a charge difference of approximately $14,000/patient.
In conclusion, our data indicate that in patients without obvious indication for emergent celiotomy, diagnostic laparoscopy was inferior to initial nonoperative management after anterior ASW in determining the need for therapeutic laparotomy. As discussed previously, to be successful the optimal diagnostic evaluation must minimize negative laparotomy rates while maximizing sensitivity for surgically significant intra-abdominal injury to avoid delays in diagnosis and subsequent morbidity and mortality. Nonoperative initial management had an improved specificity and positive predictive value compared with diagnostic laparoscopy. At the same time, although there were no missed injuries in either group, nonoperative initial management maintained lower morbidity and mortality as well as a lower cost overall. In our experience, we find no role for diagnostic laparoscopy in the evaluation of anterior ASWs solely to determine the presence or absence of peritoneal violation.
References 1. Thal ER. Evaluation of peritoneal lavage and local exploration in lower chest and abdominal stab wounds. J Trauma 1977;17:642– 8. 2. Cayton CG, Nassoura ZE. Abdomen. In: Ivatury RR, Cayton CG, Textbook of Penetrating Trauma. 1st ed. Baltimore, MD: Williams & Wilkins; 1996: 281–99. 3. Sung CK, Kim KH. Missed injuries in abdominal trauma. J Trauma 1996;41:276 – 82. 4. Sosa JL, Arrillaga A, Puente I, et al. Laparoscopy in 121 consecutive patients with abdominal gunshot wounds. J Trauma 1995;39:501– 4. 5. Miles EJ, Dunn E, Howard D, et al. The role of laparoscopy in penetrating abdominal trauma. JSLS 2004;8:304 –9. 6. Ditmars ML, Bongard F. Laparoscopy for triage of penetrating trauma: The decision to explore. J Laparoendosc Surg 1996;6:285–91. 7. Taner AS, Topgul K, Kucukel F, et al. Diagnostic laparoscopy decreases the rate of unnecessary laparotomies and reduces hospital costs in trauma patients. J Laparoendosc Adv Surg Tech A 2001;11:207–11. 8. Hallfeldt KK, Trupka AW, Erhard J, et al. Emergency laparoscopy for abdominal stab wounds. Surg Endosc 1998;12:907–10. 9. Ertekin C, Yanar H, Taviloglu K, et al. Unnecessary laparotomy by using physical examination and different diagnostic modalities for penetrating abdominal stab wounds. Emerg Med J 2005;22:790 – 4. 10. Fabian TC, Croce MA, Stewart RM, et al. A prospective analysis of diagnostic laparoscopy in trauma. Ann Surg 1993;217:557– 64. 11. Renz BM, Feliciano DV. Unnecessary laparotomies for trauma: A prospective study of morbidity. J Trauma 1995;38:350 – 6. 12. Shaftan GW. Indications for operation in abdominal trauma. Am J Surg 1960;99:657– 64. 13. Navsaria PH, Berli JU, Edu S, et al. Non-operative management of abdominal stab wounds—An analysis of 186 patients. S Afr J Surg 2007;45:128 –30, 32. 14. Zubowski R, Nallathambi M, Ivatury R, et al. Selective conservatism in abdominal stab wounds: The efficacy of serial physical examination. J Trauma 1988;28:1665– 8. 15. Demetriades D, Rabinowitz B. Indications for operation in abdominal stab wounds. A prospective study of 651 patients. Ann Surg 1987; 205:129 –32. 16. Como JJ, Bokhari F, Chiu WC, et al. Practice management guidelines for non-operative management of penetrating abdominal trauma. 2007; http:// www.east.org/tpg/archive/html/nonoppene.html. 17. Marx JA, Moore EE, Jorden RC, et al. Limitations of computed tomography in the evaluation of acute abdominal trauma: A prospective comparison with diagnostic peritoneal lavage. J Trauma 1985;25: 933–7.
T.R. Kopelman et al.
Anterior abdominal stab wounds
18. Sherck JP, Oakes DD. Intestinal injuries missed by computed tomography. J Trauma 1990;30:1–5. 19. Butela ST, Federle MP, Chang PJ, et al. Performance of CT in detection of bowel injury. AJR Am J Roentgenol 2001;176:129 –35. 20. Shanmuganathan K, Mirvis SE, Chiu WC, et al. Penetrating torso trauma: Triple-contrast helical CT in peritoneal violation and organ injury—A prospective study in 200 patients. Radiology 2004;231:775– 84. 21. Salim A, Sangthong B, Martin M, et al. Use of computed tomography in anterior abdominal stab wounds: Results of a prospective study. Arch Surg 2006;141:745–50. 22. Cherry RA, Eachempati SR, Hydo LJ, et al. The role of laparoscopy in penetrating abdominal stab wounds. Surg Laparosc Endosc Percutan Tech 2005;15:14 –7. 23. Zantut LF, Ivatury RR, Smith RS, et al. Diagnostic and therapeutic laparoscopy for penetrating abdominal trauma: A multicenter experience. J Trauma 1997;42:825–9. 24. Villavicencio RT, Aucar JA. Analysis of laparoscopy in trauma. J Am Coll Surg 1999;189:11–20. 25. Hodgson NF, Stewart TC, Girotti MJ. Open or closed diagnostic peritoneal lavage for abdominal trauma? A meta-analysis. J Trauma 2000;48:1091–5. 26. Berns AS. Nephrotoxicity of contrast media. Kidney Int 1989;36: 730 – 40. 27. Morcos SK, Thomsen HS, Webb JA. Contrast-media-induced nephrotoxicity: A consensus report. Contrast Media Safety Committee, European Society of Urogenital Radiology (ESUR). Eur Radiol 1999;9:1602–13. 28. Solomon R. Contrast-medium-induced acute renal failure. Kidney Int 1998;53:230 – 42. 29. Marks JM, Youngelman DF, Berk T. Cost analysis of diagnostic laparoscopy vs laparotomy in the evaluation of penetrating abdominal trauma. Surg Endosc 1997;11:272– 6.
Discussion Daniel Margulies, M.D. (Los Angeles, California): I have 2 comments and 2 questions. There are really 3 ways to look at laparoscopy for peritoneal penetration: the first is to look for peritoneal violation; the second is to rule out an injury that needs a repair; and the third is as a therapeutic modality. So, first, what the investigators are really saying here is that if one takes all patients with what appears to be a positive wound exploration and use laparoscopy only to look for peritoneal violation, one will identify those patients in whom the peritoneum was not violated. However, by converting to laparotomy all those patients in whom the peritoneum was violated, one will have a high nontherapeutic laparotomy rate. In other words, laparoscopy used in this way is nothing more than an expanded wound exploration. However, before dismissing laparoscopy entirely, another way to interpret these findings is seeing that the real advantage of laparoscopy is to identify which patients have
877 an injury that needs repair and which ones do not, not whether the peritoneum was violated. The whole basis of nonoperative management is the concept that despite knowing that the knife went into the peritoneum, often surgery is still not necessary. Because of the low kinetic energy transmitted by the knife and the fact that the bowel that can be pushed away rather than pierced and that liver injuries may stop bleeding on their own, often surgery is not needed. If one opens just for peritoneal penetration, there will be many nontherapeutic operations and consequent comorbidities. This study was not a randomized comparison of the full use of laparoscopy versus CAT scan or of serial examination, so we must be wary of making conclusions about that from this work. Therefore, my questions are as follows: 1. If you had included the 38 patients with no local wound exploration who underwent CAT or peritoneal lavage, how would that have changed the results? I do not know if you have that information with you, but that would be interesting to know. 2. Were any cases completed laparoscopically? Or was every patient in whom peritoneal violation was present converted to open? Luis Macias. M.D. (Phoenix, Arizona): Thank you so much for the review of the manuscript, Dr. Margulies. Regarding your 2 questions: 1. The first question was about the 38 patients that we excluded for wound exploration. Unfortunately, I do not have that information, but that would be interesting to look at. We figured that we needed to ensure if there was evidence of it or not to properly evaluate if there was peritoneal violation, ie, it would be solely used as completely an expensive wound exploration, so there needed to be evidence that there was anterior fascial penetration. 2. And the second question regarding therapeutic interventions laparoscopically, unfortunately, no, none of the patients were treated therapeutically with laparoscopy. It was used solely as PV-mandating exploratory laparotomy. There are of course several papers, including 1 of yours, which show that using diagnostic laparoscopy and then using it therapeutically would decrease the unnecessary laparotomy rate. However, not everyone is facile with the laparoscopic use laparoscopically, I mean treatment of traumatic injuries laparoscopically. Some are using it solely for peritoneal violation and then converting it. Our role was simply to dismiss that use.
The Southwestern Surgical Congress
The utility of diagnostic laparoscopy in the evaluation of anterior abdominal stab wounds Tammy R. Kopelman, M.D., F.A.C.S.a,*, Patrick J. O’Neill, Ph.D., M.D., F.A.C.S.a, Luis H. Macias, M.D.a, Jordy C. Cox, M.D., F.A.C.S.a, Marc R. Matthews, M.D., F.A.C.S.a, David A. Drachman, Ph.D.b a
Division of Burns, Trauma Surgery and Surgical Critical Care, Department of Surgery, and bDepartment of Research, Maricopa Medical Center, Phoenix, AZ 85008, United States KEYWORDS: Anterior abdominal stab wounds; Diagnostic laparoscopy; Computed tomography; Diagnostic peritoneal lavage; Hospital charges; Laparotomy; Length of stay; Local wound exploration; Nontherapeutic; Serial abdominal examinations; Therapeutic
Abstract BACKGROUND: To assess if diagnostic laparoscopy (DL) is superior to nonoperative modes (serial abdominal examination with/without computed axial tomography [CAT] and diagnostic peritoneal lavage) in determining the need for therapeutic laparotomy (TL) after anterior abdominal stab wound (ASW). METHODS: Retrospective review of ASW patients. Patients were divided into group A (DL/ exploratory laparotomy) to identify peritoneal violation (PV) and group B (initial nonoperative modes). RESULTS: Seventy-three patients met inclusion criteria. In group A (n ⫽ 38), 29 patients (76%) had PV by DL and underwent exploratory laparotomy. Only 10 (35%) underwent TL (sensitivity for PV ⫽ 100%; specificity and positive predictive value of PV in determining need for TL ⫽ 29% and 33%, respectively). In group B (n ⫽ 35), 7 patients (20%) underwent TL, yielding an improved specificity (96%) and positive predictive value (88%). CONCLUSIONS: We find no role for DL in the evaluation of ASW patients solely to determine PV. © 2008 Elsevier Inc. All rights reserved.
There is little argument that patients who present with abdominal stab wounds (ASWs) and hemodynamic compromise, evisceration, or generalized peritonitis require immediate laparotomy. However, only approximately one third of patients presenting with anterior ASWs actually sustain an injury that requires surgical intervention. This
Presented at the Southwestern Surgical Congress (60th Annual Meeting), Acapulco, Mexico, March 30 to April 2, 2008 * Corresponding author. Tel.: ⫹1-602-344-5637; fax: ⫹1-602-3445705. E-mail address: [email protected] Manuscript received May 3, 2008; revised manuscript July 3, 2008
0002-9610/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2008.07.031
finding is related to the fact that ⬎25% of anterior ASWs do not penetrate the peritoneal cavity,1 and only approximately one half of wounds that do violate the peritoneum cause visceral injury requiring surgical repair.2 In contrast, major complications and significant mortality rates caused by missed injuries after penetrating abdominal trauma may be as high as 83% and 17%, respectively.3 Therefore, the decision to operate on the asymptomatic patient after anterior ASW injury remains a challenge. The efficacy of diagnostic laparoscopy in decreasing nontherapeutic rates by excluding peritoneal violation has been studied in patients with abdominal ballistic injuries of uncertain trajectory.4 Using diagnostic laparoscopy, evi-
872
The American Journal of Surgery, Vol 196, No 6, December 2008
dence of peritoneal violation reliably confirmed the need for therapeutic laparotomy in 82% of patients and drastically decreased the incidence of negative laparotomy to 1.2%. This approach produced a positive predictive value of a surgical abdomen in a patient sustaining a ballistic injury to 97.6%.4 Since that time, many centers have attempted to expand the indication of diagnostic laparoscopy to include the evaluation of anterior ASWs with the similar goal of using peritoneal violation to determine the need for further exploration.5– 8 The use of diagnostic laparoscopy, however, still exposes the patient to general anesthesia and the risk of iatrogenic injury during exploration, not to mention a great monetary cost burden. Thus, even minimally invasive diagnostic laparoscopy is not without finite risk. To be successful, the optimal diagnostic evaluation for patients sustaining anterior ASWs must accomplish 2 things: First, negative laparotomy rates should be minimized, and second, sensitivity for surgically significant intra-abdominal injury must be maximized to avoid delays in diagnosis and subsequent morbidity and mortality. The optimal “nonoperative” method to accomplish this goal will likely include a multitude of diagnostic modalities. With this in mind, the objective of this study was to assess if diagnostic laparoscopy to assess for violation of peritoneum would be superior to nonoperative modes (serial abdominal examination with/without computed axial tomography (CAT) and diagnostic peritoneal lavage [DPL]) for determining which patient sustaining anterior ASW injury would require therapeutic laparotomy.
Abdominal Stab Wounds (n=244)
Hypotension ? Peritonitis? Evisceration ?
Yes
To Operating Room for Exploratory Laparotomy (n=65)
No
Local Wound Exploration?
No
CT Scan / Diagnostic Peritoneal Lavage Uncertain Fascial Status (n=38)
Yes
Negative Local Wound Exploration ?
Yes
No Intra-abdominal Injury (n=18)
No
Concern for Diaphragm Injury?
Yes
To Operating Room for Exploratory Laparoscopy or Laparotomy (n=50)
No
Study Group (n=73) Group A: + LWE plus Laparoscopy / Laparotomy (n = 38) Group B: + LWE plus CT / Serial Exam / DPL (n = 35)
Methods
Figure 1 Flow diagram of study patients. ⫹LWE: positive local wound exploration.
This is a retrospective cohort study performed at a 449bed urban teaching level I adult and pediatric trauma center. After obtaining Institutional Review Board approval, the trauma registry was queried for all anterior ASW patients presenting during a 4-year period (October 2003 to 2007) (Fig. 1). The anterior abdomen was defined as the area bordered by the costal margins superiorly, the inguinal ligaments inferiorly, and the anterior axillary lines laterally. Patients were excluded on the following basis: (1) indication for emergent laparotomy (ie, hemodynamic instability, peritonitis, evisceration, etc); (2) lack of evidence indicating anterior fascial violation (either by failure to perform or by identifying a negative local wound exploration); or (3) concern for left-sided diaphragm injury. Patients were then divided into 2 groups based on their diagnostic evaluation and positive local wound exploration. Group A included those patients who underwent diagnostic laparoscopy or exploratory laparotomy to identify peritoneal violation. Group B included patients whose initial nonoperative management was composed of serial physical examination and/or CAT as well as DPL. After we identified the final study groups, a systematic review of electronic medical records was performed. De-
identified demographic data were gathered using a closedresponse data collection form that included patient demographics, physical examination and laboratory findings, non–psychiatric-influenced length of stay (LOS), intra-abdominal injuries, complications, and hospital charges. DPL was performed in the emergency department by way of a percutaneous method. A positive lavage was signified by aspiration of 10 mL gross blood, an effluent erythrocyte count greater than or equal to 20,000 cells/mm3, leukocyte count ⱖ500 cells/mm3, and/or the presence of bacteria on Gram’s stain. CAT was performed by certified technologists with a helical 16-detector scanner (GE LightSpeed 16 Pro, Waukesha, Wisconsin). Intravenous contrast (150 mL Omnipaque) was injected by a power injector at a rate of 3 mL/s with a scanning delay of 60 seconds. Oral contrast was not administered. Radiographic interpretations were performed by in-house, attending radiologists. Diagnostic laparoscopy was performed in the standard fashion with an initial infraumbilical 5- or 12-mm camera port. Negative diagnostic laparoscopy was defined as the absence of peritoneal penetration, whereas positive diagnostic laparoscopy noted the presence of peritoneal penetration.
T.R. Kopelman et al. Table 1
Anterior abdominal stab wounds
873
Patient admission demographics and injury pattern characteristics
Patient demographics Average age in years (range) Male/female Location of stab wound (%) Self-inflicted injury (%) Nontherapeutic laparotomy rate (%) Therapeutic laparotomy rate (%) Survival (%)
Group A (initial surgical management) (n ⫽ 38 patients)
Group B (initial nonoperative management) (n ⫽ 35 patients)
30.8 (16–80) 32/6
30.3 (13–60) 29/6
EPI 4 LLQ 15 RLQ 22 UMB 19 RUQ 33 MULT 7 10 (26)
EPI 10 LLQ 10 RLQ 10 UMB 20 RUQ 30 MULT 0 16 (46)
19/29 (65.5)a
11/8 (12.5)
10/38 (26) 37/38 (97.4)
7/35 (20) 35/35 (100)
EPI ⫽ epigastrum; LLQ ⫽ left lower quadrant; MULT ⫽ multiple; RLQ ⫽ right lower quadrant; RUQ ⫽ right upper quadrant; UMB ⫽ umbilical. a Based on peritoneal violation.
Laparotomy was defined as therapeutic if there was a visceral injury that required surgical repair, nontherapeutic if a visceral injury was identified that did not require surgical repair (ie, nonbleeding solid-organ injury), and negative if no visceral injury was noted. All data were entered into an electronic database (Microsoft Excel, Microsoft Corporation, Redmond, Washington) and were analyzed using SPSS version 15.0.
Results During the 48-month study period, 244 patients presented to our urban level I trauma center with ASWs (Fig. 1). After the appropriate patients were excluded, 73 patients
Table 2
(30%) met our inclusion criteria and underwent further analysis. Within this final study group, 38 patients (52%) underwent diagnostic laparoscopy to exclude peritoneal violation (group A), and 35 (48%) patients underwent either DPL or serial abdominal exams with or without abdominal/ pelvic CAT (group B). Demographic, injury patterns, and survival characteristics of the study populations are listed in Table 1. Among patients in group A, 29 of 38 (76%) were noted to have peritoneal violation and underwent exploratory laparotomy to exclude intra-abdominal injury. At laparotomy, approximately one third (10 of 29 patients) were noted to have injuries requiring therapeutic intervention (Table 2). Diagnostic laparoscopy therefore had a sensitivity of 100% for determining peritoneal violation and a specificity and
Summary of surgical findings Findings at laparotomy (%)
Diagnostic evaluation
Laparotomy (%)
Group A (n ⫽ 38)
29 (76)
Group B (n ⫽ 35)
8 (23)
Sm intest ⫽ small intestine. a Positive peritoneal violation.
a
Negative
Nontherapeutic
Positive
Missed injury
18 (62)
1 (3.5)
0 (0)
0 (0)
1 (12.5)
10 (34.5) Liver: 4 Kidney: 0 Sm intest: 3 Colon: 1 Mesentery: 2 Omentum: 1 Ovary: 1 7 (87.5) Liver: 4 Kidney: 1 Sm intest: 3 Colon: 1 Mesentery: 4 Omentum: 0 Ovary: 0
0 (0)
874
The American Journal of Surgery, Vol 196, No 6, December 2008
Table 3
Postsurgical complications
Complications
Group A (n ⫽ 38)
Group B (n ⫽ 35)
Ileus Pneumonia Wound dehiscence Wound infection Emesis Fever Small intestinal obstruction Acute renal failure Acute respiratory distress syndrome Cardiac ectopy Myocardial infarction Iatrogenic splenic injury Iatrogenic small intestinal injury Total patients with morbidity (%) Mortality (%)
2 1 1 1 1 1 1 1 1 1 1 1 2 9a (24) 1 (2.6)
2 0 1 0 1 0 0 0 0 0 0 0 0 4 (11) 0 (0)
a
Multiple complications may have occurred in a single patient.
positive predicative value for determining the need for therapeutic laparotomy of 29% and 33%, respectively. All patients with negative laparoscopy were discharged home and without complications. Furthermore, no patient in group A was found to have a missed injury during their hospital stay or in follow-up (Table 2). The only patient death in this study occurred because of postsurgical complications after negative laparotomy was performed for peritoneal violation that was noted on laparoscopy. Conversion to exploratory laparotomy was complicated by both intraoperative and postsurgical complications, and a greater morbidity rate was noted in the nontherapeutic subgroup (Table 3). Of the 35 group B patients who underwent initial nonoperative diagnostic management, 19 patients (54%) underwent serial abdominal examinations with or without abdominopelvic CAT scan, and 16 patients (46%) underwent DPL. In this group, 7 patients (20%) subsequently required therapeutic laparotomy (Table 1). One patient underwent a nontherapeutic laparotomy after a stab wound to the right upper quadrant. CAT showed air in the soft tissues through the fascia with extension into the peritoneal cavity but no evidence of intra-abdominal fluid, air, or injury. Despite this, the attending surgeon’s preference was to evaluate this
Table 4
patient surgically because he had mild abdominal discomfort, and a small nonbleeding hematoma was noted in the mesentery of the colon. No intervention was required. In contrast, the 1 false-positive finding in this group was a female patient with isolated free fluid noted on CAT and a clinically benign examination. Subsequent serial physical examinations failed to identify clinical indications for surgical intervention, and she was discharged to home without surgical intervention. As in group A, no delayed identification of injuries was noted on serial physical examination in any patient. Thus, nonoperative diagnostic management in group B showed a sensitivity of 100%, a specificity of 96%, and a positive predictive value of 88% while maintaining a negative predictive value of 100%. The average length of stay for those undergoing nonoperative management or negative diagnostic laparoscopy was equivalent (Table 4). Similarly, the length of stay in all patients undergoing laparotomy, therapeutic or not, was similar in both groups. It should be noted, however, that 19 of 29 patients (66%) in group A had peritoneal violation and underwent nontherapeutic exploratory laparotomy (Table 1), which increased their length of stay by 2.5 days and charges by ⬎80% compared with the nonoperative arm (Table 4).
Comments There is little doubt that patients with peritonitis, evisceration, or hemodynamic instability following anterior ASW require surgical intervention. It is the stabbed patient with minimal to no symptoms that remains a diagnostic enigma to the trauma surgeon. Reports on the incidence of unnecessary laparotomy cite rates of 23% to 65% in patients presenting with ASWs.6,8 –10 Because of this nontherapeutic surgery, about 40% of patients will have complications, including atelectasis, postsurgical hypertension, pleural effusion, pneumothorax, prolonged ileus, pneumonia, surgical wound infection, small-bowel obstruction, urinary infection, and others.11 In addition, these avoidable surgeries lead to increased inpatient length of stay and subsequent greater health care costs. Clearly, complication rates and increased expense must be weighed against the mortality and morbidity of a missed injury when deciding how to evaluate a patient after anterior ASW.
Non–psychiatric-influenced hospital LOS and charges Group A
Length of stay in days (range) Charges
Group B
No PV/DL alone (n ⫽ 4)
⫹ PV nontherapeutic laparotomy (n ⫽ 15)
⫹ PV therapeutic laparotomy (n ⫽ 9)
No laparotomy required (n ⫽ 15)
Therapeutic laparotomy (n ⫽ 4)
1.4 (1–3) $20,310
4.5 (2–15) $30,275
4.2 (2–10) $32,190
1.4 (1–4) $16,620
5.4 (4–8) $30,585
PV ⫽ peritoneal violation; DL ⫽ diagnostic laparoscopy.
T.R. Kopelman et al.
Anterior abdominal stab wounds
As far back as the 1960s, Shaftan12 described “selective conservatism” in the management of patients after ASW using serial physical examination to identify those patients requiring laparotomy, hereby decreasing the incidence of unnecessary surgery. Since that time, many studies have confirmed this finding, citing the use of serial abdominal examination to safely and reliably predict the need for laparotomy.13–15 In this management arm, patient admission is necessary because hemodynamic monitoring, frequent serial abdominal examinations, white blood cell count and hemoglobin follow-up, as well as serial temperature measurements are typically required for at least 24 hours. The onset of hemodynamic instability, increasing abdominal pain or tenderness, increasing temperature, decreasing hemoglobin, or leukocytosis would warrant laparotomy or further investigation. Ertekin et al9 reported on 117 patients with penetrating stab wounds to the thoracoabdominal region (67% anterior abdomen), all of whom underwent nonoperative management. Clinical examination was the primary tool to differentiate those patients requiring surgical intervention, but this was often combined with other modalities, such as DPL, ultrasound, or CAT. DPL was used in patients whose physical examinations were not reliable because of alcohol or drug use or unconsciousness. In their study, nonoperative management was successful in 79% of patients with 11 early (within 8 hours of admission) and 14 delayed (⬎8 hours after admission) laparotomies performed. They had no negative laparotomies in the entire series, and the rate of nontherapeutic laparotomy remained low at 9% in the early group and 14% in the delayed group. No morbidity or mortality was attributed to delays in surgical intervention, which ranged from 1.5 hours to 20 hours. This group concluded that physical examination alone and/or coupled with different diagnostic methods allowed for a decrease in nontherapeutic laparotomies and eliminated negative laparotomy.9 Another option that exists in the nonoperative management of anterior ASW is CAT scanning. Eastern Association for the Surgery of Trauma guidelines emphasize that the use of triple-contrast abdominal/pelvic CAT should be strongly considered as a diagnostic tool in the initial evaluation of penetrating abdominal trauma to predict the need for surgical intervention.16 Despite this recommendation, there still remains a reluctance to rely on CAT because of its varying sensitivity, specificity, and accuracy in the diagnosis of bowel and mesenteric injuries.17–19 These concerns may be unfounded, however, because accuracy appears to be improving with advancing technology and the advent of helical and multislice CAT. Evidence of improved imaging may be found in the study by Shanmuganathan et al,20 which used CAT to assess penetrating abdominal injuries in 200 patients. They found that triple-contrast abdominopelvic CAT had a sensitivity of 97%, a specificity of 98%, and a negative predictive value of 98%. Similar results were noted by Demetriades and Rabinowitz21 in a prospective study of 156 stable patients in
875 who all sustained anterior ASWs and were admitted for serial abdominal examinations with or without CAT. Excluding patients with thoracoabdominal stab wounds, the results of this study demonstrated that CAT using intravenous contrast only demonstrated a high sensitivity (100%) with a moderately high specificity (81%) in identifying those patients with injuries. More importantly, it showed a negative predictive value of 100%. In this study as well, CAT correctly identified all patients with bowel injury, and no false-negative scans were noted.21 Group B, our nonoperative diagnostic arm, was a compilation of all the above-cited modes of evaluation, using a combination of serial examination, DPL, and CAT to determine the need for surgical intervention. The results of our study, namely, a sensitivity of 100%, a specificity of 96% and a positive predictive value of 88% while maintaining a negative predictive value of 100%, only serve to confirm the efficacy of nonoperative diagnostic modalities in the evaluation of patients with anterior ASWs. More recent literature has identified diagnostic laparoscopy as an important tool in the evaluation of ASWs. Compared with mandatory laparotomy, this minimally invasive evaluation allows for lower rates of nontherapeutic laparotomy by identifying the patient at risk for intra-abdominal injury after penetrating injury.8 In the current study, we chose to evaluate our experience with diagnostic laparoscopy to assess patients with anterior ASWs and to determine the need for exploratory celiotomy by identifying peritoneal violation. In our experience, approximately 25% of patients avoided exploratory laparotomy by excluding intra-abdominal penetration by the stab wound, and an additional 25% underwent therapeutic laparotomy after peritoneal violation was noted on laparoscopy. However, this mode of screening for the need for open exploration still led to an unnecessary laparotomy rate of 62% and an overall nontherapeutic laparotomy rate of 66%, which was no different from historic controls. Other studies confirm our results when peritoneal penetration was identified as the impetus to convert to exploratory laparotomy, citing nontherapeutic rates of 50% to 65%.6,8 In our experience, peritoneal violation was noted to have a sensitivity of 100% but a specificity and positive predictive value for determining the need for therapeutic laparotomy of only 29% and 33%, respectively. In certain patients, investigators advocate not only screening for peritoneal violation with laparoscopy but actually using this minimally invasive technique for therapeutic endeavors when peritoneal violation is encountered.5,22 Ivatury et al23 concluded that in carefully selected patients, it can be practical and feasible. However, in their study of 207 patients undergoing celiotomy after laparoscopy, 9.2% were found to have additional injuries requiring surgical intervention. In addition, 15 patients were found to have additional bowel injuries on laparotomy that were not visualized on laparoscopic evaluation, and 4 patients had significant vascular injuries (vena cava, common iliac vein) at laparotomy that were not suspected at laparoscopy.23 Fur-
876
The American Journal of Surgery, Vol 196, No 6, December 2008
thermore, in an outcome analysis performed by Villavicencio and Aucar,24 37 studies with ⬎1,900 trauma patients were reviewed, and the use of laparoscopy as a screening, diagnostic, and therapeutic tool was analyzed. Laparoscopy was regarded as a diagnostic tool only in studies that mandated a laparotomy (“gold standard”) after laparoscopy to confirm the diagnostic accuracy of the laparoscopic finding (rather than as a screening tool to identify the need for laparotomy). However, he noted that in this setting that laparoscopy had a 41% to 77% missed-injury rate/patient and therefore concluded that its value in this context was limited. Another distinct disadvantage of diagnostic laparoscopy is that it is performed in the operating room and requires general anesthesia. A 1999 review of 37 studies, which included more than 1,900 patients, demonstrated a procedure-related complication rate of 1%.24 Intraoperative complications may occur during creation of the pneumoperitoneum or trocar insertion or during the diagnostic examination. Our nonoperative arm is not without its own inherent complications. DPL is also an invasive procedure and carries a small but significant risk of iatrogenic intraabdominal injury in the range of 0.6%,25 but it may be performed safely in the emergency department. CAT with intravenous contrast administration carries the risk of a radiation exposure as well as contrast reaction, extravasation of contrast material at the site of intravenous injection, and contrast medium–induced nephropathy. The development of contrast medium–induced nephropathy is low in people with normal renal function, varying from 0% to 5%,26,27 but it has been reported in several prospective controlled studies to increase to rates of 12%–27% in patients with preexisting renal dysfunction.27,28 In the current study, no complications were noted related to DPL or CAT. There were, however, intraoperative complications related to the examination during diagnostic laparoscopy with the creation of enterotomies during lysis of adhesions in an attempt to visualize peritoneal violation. From a cost perspective, those advocating screening laparoscopy have reported decreased mean hospital LOS in negative laparoscopy patients compared with negative laparotomy patients with significant decrease in mean LOS.6,10,29 In the current study, the charges of negative laparoscopy was comparable with that of a negative nonoperative evaluation and considerably less than that of the negative laparotomy group, but that is exactly the issue at hand. It is our opinion that this additional group must be considered when contemplating cost, namely, the patients with false-positive laparoscopic evaluations who undergo nontherapeutic laparotomy simply on the basis of peritoneal violation. In our study, this group increased their average length of stay by during 2.5 days and charges by ⬎80% compared with their nonoperative counterparts who did not undergo surgical intervention. This accounted for a charge difference of approximately $14,000/patient.
In conclusion, our data indicate that in patients without obvious indication for emergent celiotomy, diagnostic laparoscopy was inferior to initial nonoperative management after anterior ASW in determining the need for therapeutic laparotomy. As discussed previously, to be successful the optimal diagnostic evaluation must minimize negative laparotomy rates while maximizing sensitivity for surgically significant intra-abdominal injury to avoid delays in diagnosis and subsequent morbidity and mortality. Nonoperative initial management had an improved specificity and positive predictive value compared with diagnostic laparoscopy. At the same time, although there were no missed injuries in either group, nonoperative initial management maintained lower morbidity and mortality as well as a lower cost overall. In our experience, we find no role for diagnostic laparoscopy in the evaluation of anterior ASWs solely to determine the presence or absence of peritoneal violation.
References 1. Thal ER. Evaluation of peritoneal lavage and local exploration in lower chest and abdominal stab wounds. J Trauma 1977;17:642– 8. 2. Cayton CG, Nassoura ZE. Abdomen. In: Ivatury RR, Cayton CG, Textbook of Penetrating Trauma. 1st ed. Baltimore, MD: Williams & Wilkins; 1996: 281–99. 3. Sung CK, Kim KH. Missed injuries in abdominal trauma. J Trauma 1996;41:276 – 82. 4. Sosa JL, Arrillaga A, Puente I, et al. Laparoscopy in 121 consecutive patients with abdominal gunshot wounds. J Trauma 1995;39:501– 4. 5. Miles EJ, Dunn E, Howard D, et al. The role of laparoscopy in penetrating abdominal trauma. JSLS 2004;8:304 –9. 6. Ditmars ML, Bongard F. Laparoscopy for triage of penetrating trauma: The decision to explore. J Laparoendosc Surg 1996;6:285–91. 7. Taner AS, Topgul K, Kucukel F, et al. Diagnostic laparoscopy decreases the rate of unnecessary laparotomies and reduces hospital costs in trauma patients. J Laparoendosc Adv Surg Tech A 2001;11:207–11. 8. Hallfeldt KK, Trupka AW, Erhard J, et al. Emergency laparoscopy for abdominal stab wounds. Surg Endosc 1998;12:907–10. 9. Ertekin C, Yanar H, Taviloglu K, et al. Unnecessary laparotomy by using physical examination and different diagnostic modalities for penetrating abdominal stab wounds. Emerg Med J 2005;22:790 – 4. 10. Fabian TC, Croce MA, Stewart RM, et al. A prospective analysis of diagnostic laparoscopy in trauma. Ann Surg 1993;217:557– 64. 11. Renz BM, Feliciano DV. Unnecessary laparotomies for trauma: A prospective study of morbidity. J Trauma 1995;38:350 – 6. 12. Shaftan GW. Indications for operation in abdominal trauma. Am J Surg 1960;99:657– 64. 13. Navsaria PH, Berli JU, Edu S, et al. Non-operative management of abdominal stab wounds—An analysis of 186 patients. S Afr J Surg 2007;45:128 –30, 32. 14. Zubowski R, Nallathambi M, Ivatury R, et al. Selective conservatism in abdominal stab wounds: The efficacy of serial physical examination. J Trauma 1988;28:1665– 8. 15. Demetriades D, Rabinowitz B. Indications for operation in abdominal stab wounds. A prospective study of 651 patients. Ann Surg 1987; 205:129 –32. 16. Como JJ, Bokhari F, Chiu WC, et al. Practice management guidelines for non-operative management of penetrating abdominal trauma. 2007; http:// www.east.org/tpg/archive/html/nonoppene.html. 17. Marx JA, Moore EE, Jorden RC, et al. Limitations of computed tomography in the evaluation of acute abdominal trauma: A prospective comparison with diagnostic peritoneal lavage. J Trauma 1985;25: 933–7.
T.R. Kopelman et al.
Anterior abdominal stab wounds
18. Sherck JP, Oakes DD. Intestinal injuries missed by computed tomography. J Trauma 1990;30:1–5. 19. Butela ST, Federle MP, Chang PJ, et al. Performance of CT in detection of bowel injury. AJR Am J Roentgenol 2001;176:129 –35. 20. Shanmuganathan K, Mirvis SE, Chiu WC, et al. Penetrating torso trauma: Triple-contrast helical CT in peritoneal violation and organ injury—A prospective study in 200 patients. Radiology 2004;231:775– 84. 21. Salim A, Sangthong B, Martin M, et al. Use of computed tomography in anterior abdominal stab wounds: Results of a prospective study. Arch Surg 2006;141:745–50. 22. Cherry RA, Eachempati SR, Hydo LJ, et al. The role of laparoscopy in penetrating abdominal stab wounds. Surg Laparosc Endosc Percutan Tech 2005;15:14 –7. 23. Zantut LF, Ivatury RR, Smith RS, et al. Diagnostic and therapeutic laparoscopy for penetrating abdominal trauma: A multicenter experience. J Trauma 1997;42:825–9. 24. Villavicencio RT, Aucar JA. Analysis of laparoscopy in trauma. J Am Coll Surg 1999;189:11–20. 25. Hodgson NF, Stewart TC, Girotti MJ. Open or closed diagnostic peritoneal lavage for abdominal trauma? A meta-analysis. J Trauma 2000;48:1091–5. 26. Berns AS. Nephrotoxicity of contrast media. Kidney Int 1989;36: 730 – 40. 27. Morcos SK, Thomsen HS, Webb JA. Contrast-media-induced nephrotoxicity: A consensus report. Contrast Media Safety Committee, European Society of Urogenital Radiology (ESUR). Eur Radiol 1999;9:1602–13. 28. Solomon R. Contrast-medium-induced acute renal failure. Kidney Int 1998;53:230 – 42. 29. Marks JM, Youngelman DF, Berk T. Cost analysis of diagnostic laparoscopy vs laparotomy in the evaluation of penetrating abdominal trauma. Surg Endosc 1997;11:272– 6.
Discussion Daniel Margulies, M.D. (Los Angeles, California): I have 2 comments and 2 questions. There are really 3 ways to look at laparoscopy for peritoneal penetration: the first is to look for peritoneal violation; the second is to rule out an injury that needs a repair; and the third is as a therapeutic modality. So, first, what the investigators are really saying here is that if one takes all patients with what appears to be a positive wound exploration and use laparoscopy only to look for peritoneal violation, one will identify those patients in whom the peritoneum was not violated. However, by converting to laparotomy all those patients in whom the peritoneum was violated, one will have a high nontherapeutic laparotomy rate. In other words, laparoscopy used in this way is nothing more than an expanded wound exploration. However, before dismissing laparoscopy entirely, another way to interpret these findings is seeing that the real advantage of laparoscopy is to identify which patients have
877 an injury that needs repair and which ones do not, not whether the peritoneum was violated. The whole basis of nonoperative management is the concept that despite knowing that the knife went into the peritoneum, often surgery is still not necessary. Because of the low kinetic energy transmitted by the knife and the fact that the bowel that can be pushed away rather than pierced and that liver injuries may stop bleeding on their own, often surgery is not needed. If one opens just for peritoneal penetration, there will be many nontherapeutic operations and consequent comorbidities. This study was not a randomized comparison of the full use of laparoscopy versus CAT scan or of serial examination, so we must be wary of making conclusions about that from this work. Therefore, my questions are as follows: 1. If you had included the 38 patients with no local wound exploration who underwent CAT or peritoneal lavage, how would that have changed the results? I do not know if you have that information with you, but that would be interesting to know. 2. Were any cases completed laparoscopically? Or was every patient in whom peritoneal violation was present converted to open? Luis Macias. M.D. (Phoenix, Arizona): Thank you so much for the review of the manuscript, Dr. Margulies. Regarding your 2 questions: 1. The first question was about the 38 patients that we excluded for wound exploration. Unfortunately, I do not have that information, but that would be interesting to look at. We figured that we needed to ensure if there was evidence of it or not to properly evaluate if there was peritoneal violation, ie, it would be solely used as completely an expensive wound exploration, so there needed to be evidence that there was anterior fascial penetration. 2. And the second question regarding therapeutic interventions laparoscopically, unfortunately, no, none of the patients were treated therapeutically with laparoscopy. It was used solely as PV-mandating exploratory laparotomy. There are of course several papers, including 1 of yours, which show that using diagnostic laparoscopy and then using it therapeutically would decrease the unnecessary laparotomy rate. However, not everyone is facile with the laparoscopic use laparoscopically, I mean treatment of traumatic injuries laparoscopically. Some are using it solely for peritoneal violation and then converting it. Our role was simply to dismiss that use.