Abstract
This observational study compares outcomes of patients with torso stab wounds who present to level I vs level II trauma centers.
Stab wounds result in more than 400 000 injuries treated annually nationwide in the United States.1 Management includes operative and selective nonoperative management (SNOM) depending on hemodynamic stability, stab wound location, presence of peritoneal signs, and surgeon preference.2
The American College of Surgeons verifies trauma centers, and the clinical care rendered at level I and level II centers is expected to be equivalent.3 Previous authors have demonstrated better survival and functional outcomes in patients with specific high mortality injuries at level I vs level II trauma centers.4 However, a 2017 analysis found no difference in mortality among patients with penetrating trauma presenting to level I vs level II centers.5
Owing to the increased number of surgical personnel, clinical volume, and resources at level I trauma centers,3 we hypothesized that patients with torso stab wounds presenting to level I centers would have a shorter time to surgical intervention (laparotomy or thoracotomy) and a lower mortality, compared with level II centers.
Methods
The research protocol was approved by the University of California, Irvine Human Research Protections, Regulatory Compliance institutional review board committee, and the Trauma Quality Improvement Program was queried (January 2010 to January 2016) for patients 18 years or older admitted to the American College of Surgeons–verified level I or II centers after torso stab wounds. No informed consent was required because data were deidentified and available in the registry. Patients with Abbreviated Injury Scale grade more than 1 of the head, neck, or extremities were excluded to select for patients with isolated torso injuries. Analysis was conducted from January 2010 to January 2016. Covariates with P < .20 on univariable analysis were selected for entry in a multivariable logistic regression model to report the adjusted risk of mortality. All P values were 2-sided, with statistical significance level of P < .05.
Results
Of 17 245 patients with torso stab wounds, 12 628 (73.2%) were treated at level I centers and 4618 (26.8%) were treated at level II centers. Both groups had a similar Injury Severity Score (median [interquartile range], 10 [9-14]; P = .11). Level I centers had a higher rate of laparotomy (3220 of 12 628 [25.5%] vs 984 of 4617 [21.4%]; P < .001) with shorter time to laparotomy (median [interquartile range], 60 [40-120] vs 67 [45-120] minutes; P = .01) and a higher rate of thoracotomy (614 of 12 628 [5.9%] vs 173 of 4617 [3.7%]; P = .01) with no difference in time to thoracotomy (median [interquartile range] time to thoracotomy for level I: 60 [26-138] minutes vs level II: 60 [23-142] minutes; P = .86) (Table 1). Patients with stab wounds at level I centers treated with thoracotomy (148 [24.3%] vs 59 [34.1%]; P = .01) and SNOM (388 [3.1%] vs 169 [3.7%]; P = .04) had a lower rate of mortality, compared with level II centers. Adjusting for covariates, patients treated at level I centers had a decreased risk of mortality compared with similar patients at level II centers. Patients undergoing thoracotomy or SNOM at level I centers had a decreased risk of mortality as well (Table 2).
Table 1. Characteristics of Adult Trauma Patients Presenting After Torso Stab Wounds to ACS Level I and II Trauma Centers.
| Characteristic | No. (%) | P Value | |
|---|---|---|---|
| ACS Level I Center (n = 12 628) | ACS Level II Center (n = 4617) | ||
| Age, median (IQR), y | 31.0 (24-43) | 30.0 (23-41) | .01 |
| Male | 11 192 (88.6) | 4127 (89.5) | .12 |
| ISS, median (IQR) | 10.0 (9-14) | 10.0 (9-14) | .11 |
| Lowest SBP within 24 h, median (IQR), mm HG | 81.0 (46-105) | 88.0 (60-112) | .02 |
| Comorbidities | |||
| Congestive heart failure | 47 (0.4) | 11 (0.2) | .18 |
| End-stage renal disease | 14 (0.1) | 4 (0.1) | .66 |
| Smoker | 3717 (29.4) | 1184 (25.6) | <.001 |
| Diabetes | 401 (3.2) | 136 (2.9) | .44 |
| Hypertension | 1123 (8.9) | 361 (7.8) | .03 |
| COPD | 503 (4.0) | 185 (4.0) | .94 |
| AIS (grade >3) | |||
| Thorax | 1463 (11.6) | 477 (10.3) | .02 |
| Abdomen | 673 (5.3) | 252 (5.5) | .74 |
| Injuries | |||
| Lung | 8791 (70.7) | 3265 (69.6) | .16 |
| Heart | 789 (6.2) | 248 (5.4) | .03 |
| Liver | 1578 (12.5) | 545 (11.8) | .22 |
| Spleen | 646 (5.1) | 260 (5.6) | .18 |
| Small bowel | 991 (7.8) | 317 (6.9) | .03 |
| Colorectal | 769 (6.1) | 293 (6.3) | .54 |
| Pancreas | 129 (1.0) | 47 (1.0) | .98 |
| Treatment | |||
| Exploratory (laparotomy) | 3220 (25.5) | 984 (21.3) | <.001 |
| Exploratory (thoracotomy) | 614 (4.9) | 173 (3.7) | .002 |
| SNOM | 8796 (69.7) | 3460 (74.9) | <.001 |
| Failed SNOM for exploratory (laparotomy)a | 102 (3.8) | 27 (2.9) | .19 |
| Failed SNOM for exploratory (thoracotomy)b | 71 (0.8) | 17 (0.5) | .11 |
Abbreviations: ACS, American College of Surgeons; AIS, Abbreviated Injury Scale; COPD, chronic obstructive pulmonary disease; ISS, Injury Severity Score; IQR, interquartile range; SBP, systolic blood pressure; SNOM, elective nonoperative management.
Failure defined as exploratory laparotomy occurring more than 6 hours from admission; patients with AIS grade higher than 2 for the abdomen were excluded.
Failure defined as exploratory thoracotomy occurring more than 6 hours from admission; patients with AIS grade higher than 2 for the thorax were excluded.
Table 2. Median Operative Time, Mortality Rate, and Adjusted Analysis for Risk of Mortality in Adult Trauma Patients Presenting After Stab Wound at Level I vs II Trauma Centersa.
| Risk Factor | Operating Time, Median (IQR), min | Mortality, No. (%) | Risk of Mortality | |||||
|---|---|---|---|---|---|---|---|---|
| Level I Center | Level II Center | P Value | Level I Center | Level II Center | P Value | OR (95% CI) | P Value | |
| All patients | 60 (52-68) | 66 (55-69) | .002 | 429 (3.4) | 182 (3.9) | .09 | 0.67 (0.54-0.84) | <.001 |
| Treated with exploratory (laparotomy) | 60 (54-67) | 67 (58-75) | .002 | 101 (3.1) | 40 (4.1) | .16 | 0.73 (0.48-1.12) | .15 |
| Treated with exploratory (thoracotomy) | 60 (50-68) | 60 (51-68) | .86 | 149 (24.3) | 59 (34.1) | .01 | 0.53 (0.32-0.88) | .01 |
| Treated nonoperatively | NA | NA | NA | 388 (3.1) | 169 (3.7) | .04 | 0.65 (0.52-0.82) | .001 |
| Subanalysis of thoracotomyb | 82 (72-98) | 68 (58-73) | .29 | 75 (15.1) | 32 (23.0) | .03 | 0.54 (0.29-0.97) | .03 |
Abbreviations: IQR, interquartile range; NA, not applicable; OR, odds ratio.
Each model controlled for age 65 years or older; hypotension on admission; Injury Severity Score of 25 or higher; injuries to the lung, heart, spleen, or small intestine; severe Abbreviated Injury Scale thorax and abdomen (grade >3); and history of congestive heart failure, smoking, or hypertension.
Excluded patients who underwent thoracotomy within 20 minutes of arrival (0.9% of all thoracotomies) that likely occurred in the emergency department.
Discussion
Patients with stab wounds requiring thoracotomy had a nearly 50% lower risk of mortality when treated at level I centers vs those admitted to level II centers. Those with SNOM had a more than 30% lower risk of mortality compared with patients at level II centers. While patients underwent laparotomy more often at level I centers compared with level II, there was no significant difference found in risk of mortality. Time to surgery had no clinically significant association with mortality.
Studies have reported a strong positive association between trauma center volume and outcomes in patients with severe wounds and knife-related injuries.6 Given the higher patient volume and rate of thoracotomies performed at level I centers,3 it is possible that greater experience in treatment may have resulted in better outcomes. With regard to patients who underwent SNOM, while statistically significant, the absolute difference was 0.6%. As a result, it is difficult to correlate clinically. Further studies are necessary to assess this finding and determine the reasons behind the difference in risk of mortality.
Limitations to our study include lack of data on the location of stab wounds contributing to variation in management, clinician experience, time from injury to arrival, personnel availability, variations in SNOM, and number of stab wounds.
Conclusions
This study demonstrates that patients with torso stab wounds treated at level I centers have a lower risk of mortality when treated with thoracotomy or SNOM, compared with patients at level II centers. Future prospective research is needed to evaluate these findings and support the institution of optimal prehospital trauma designation for this population.
References
- 1.Smith GA. Knife-related injuries treated in United States emergency departments, 1990-2008. J Emerg Med. 2013;45(3):315-323. doi: 10.1016/j.jemermed.2012.11.092 [DOI] [PubMed] [Google Scholar]
- 2.Ball CG. Current management of penetrating torso trauma: nontherapeutic is not good enough anymore. Can J Surg. 2014;57(2):E36-E43. doi: 10.1503/cjs.026012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Rotondo M, Cribari C, Smith R. Resources for optimal Care of the Injured Patient. Chicago, Illinois: American College of Surgeons; 2014:6. [Google Scholar]
- 4.Demetriades D, Martin M, Salim A, Rhee P, Brown C, Chan L. The effect of trauma center designation and trauma volume on outcome in specific severe injuries. Ann Surg. 2005;242(4):512-517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kaji AH, Bosson N, Gausche-Hill M, et al. Patient outcomes at urban and suburban level I versus level II trauma centers. Ann Emerg Med. 2017;70(2):161-168. doi: 10.1016/j.annemergmed.2017.01.040 [DOI] [PubMed] [Google Scholar]
- 6.Nathens AB, Jurkovich GJ, Maier RV, et al. Relationship between trauma center volume and outcomes. JAMA. 2001;285(9):1164-1171. doi: 10.1001/jama.285.9.1164 [DOI] [PubMed] [Google Scholar]
