Should the Management Approach to the Anterior Abdominal Stab Wound be Different in Patients with Self-inicted Abdominal Injury?

associated with mortality were investigated using logistic regression. Possible interactions were incorporated into the model. Results are presented as odds ratio and 95% condence intervals (95%CI). Statistical analysis was performed using SAS statistical software Version 9.2 (SAS Institute Inc., Cary, NC). P-values of less than 0.05 were considered statistically signicant.


bing in comparison t
stab wounds from assault, and therefore a more conservative management might be considered.


Methods

All patients with isolated abdominal stab wound (SW) admitted to 19 Trauma Centers in Israel between the years 1997 and 2018 were included in the study.Patients with self-in icted abdominal SW (Group I) were compa

ed in traum
units.Suicide is a common cause of death worldwide [1].However, suicide by self-stabbing is relatively rare and accounts for only 3% of suic

e attempts
1].Several studies showed lower injury severity scores (ISS) in selfin icted injuries compared to victims of violence [2,3].However, mortality rates were not lower.These studies also showed that patients with self-in icted AASW underwent surgery more often and had a higher rate of non-therapeutic laparotomy (NTL).Theoretically, the lower ISS and the higher rate of NTL could be explained in part by the internal defensive instincts of self-preservation that the patient performing suicide might have at the time of in icting self-injury.

Since the data on self-in icted AASW is limited, this study sought to verify previous ndings reported by others by using the data in the Israeli Trauma Registry.It was hypothesized that among the Israeli patients population, self-in icted AASW would result in less serious injuries and probably less mortality.


Material And Methods


Population and sample

All patients older than 18 years of age with AASW who were admitted to 19 Israeli Trauma Centers between the years 1997 and 2018 were included in the study.The patients were divided in t

uded patients with sel
-in icted injury and group II -victims of assault.Patients with concomitant injuries classi ed as AIS > 2 to non-abdominal organs were excluded from the study in order to rule out other causes that might affect the outcome.


Data sources

The data was retrospectively collected from the records of the Israeli National Trauma Registry


Variables

The data collected from the registry included age, sex, injury severity core (Injury Severity Score (ISS) and A

reviated I
jury Scale (AIS)), systolic blood pressure on admission and mortality.It also included information concerning operations.Patient with systolic blood pressure < 90 mmHg were considered hemodynamically unstable.


Statistical analysis

Differences in proportions between Group I and Group II were analyzed with either Chi-square test of i

ependence or Fisher e
act test.Age and hospitalization Length of Stay (LOS) differences were analyzed with Mann-Whitney test.Similar tests were used for categorical and continuous variables when patients who died in Group I were compared to patients who died in Group II.The combined effect of variables that were signi cantly associated with mortality were investigated using logistic regression.

Possible interactions were incorporated into the model.Results are presented as odds ratio and 95% con dence intervals (95%CI).S atistical analysis was performed using SAS statistical software Version 9.2 (SAS Institute Inc., Cary, NC).P-values of less than 0.05 were considered statistically signi cant.

Numbers, percentages and interquartile ranges (IQR) were approximated to the nearest decimal.Odds Ratio (OR), P value and 95% con d nce interval (95%CI) values to the nearest thousandth.

This study was performed without any grants or other type of funding.The authors have not con ict of interest to declare.


Results Overall, 3,324 patients with AASW were included in the study.Of these, 314 (9.4%) patients were included in the self-in ict

injury
Group I).Comparison of sex and age between the two groups is detailed in Table 1.Almost all the patients in Group II were males.Patients in Group II were also younger when compared to patients in Group I.  2 describes the clinical presentation, ISS, incidence of intra-abdominal organs injury, and mortality across studied groups.There were no differences in frequency of the abdominal wall penetration and the frequency of internal organs injuries.When speci c organs were evaluated, the rate of kidney injuries was higher in Group II patients.Though no differences in injury severity and frequency of internal injury were noted between the groups, more patients in Group I underwent abdominal surgery.Mortality was higher in Group I patients compared to Group II (2.9% vs. 0.7%, respectively; p = 0.002).

In order to evaluate whether self-in icted injuries are an independent risk factor for mortality, other possible risk factors for mortality were evaluated (Table 3).The analysis shows that patients who died were older.Age, ISS, and hypotension on presentation were all signi cantly associated with mortality and thus could be considered as covariates.The combined effect of these variables was investigated using logistic regression (Table 4).When the four signi cant covariates were analyzed together, all four remained signi cant, incl

ing self-in
icted injury.Possible interaction effects between the four covariates were all non-signi cant.


Discussion

Anterior Abdominal stab wounds (AASW) is a common cause for admission to surgical E ergency Rooms.Only 28% of patients suffering from AASW will end up having internal abdominal injury [4,5].

Still, due to limitations in arriving to a correct diagnosis, as many as 40-45.8% of said patients will undergo laparotomy.Clear indications for explorative laparotomy in an AASW patient include hemodynamic instability, peritonitis, evisceration, blood in the nasogastric tube, and/or on rectal examination [1,6].Since only about a third of the patients with AASW require surgical exploration, further evaluation is warranted.Management approaches may include local woun exploration, computed tomography, diagnostic laparoscopy, or close clinical observation [7][8][9][10].

In this study, patients with self-in icted AASW had a similar frequency of intra-abdominal injury, as well as similar frequency of high ISS compared to stab wound (SW) patients from an assault.Nevertheless, patients in Group I underwent surgery more frequently and they also had a higher mortality.Relatively similar results were presented by Matsomoto et al [2] who compared self-in icted injuries to those presented due to violence using the Japanese Trauma Data Bank.In their study, 76.4% of abdominal stab wounds were self-in icted.The authors reported lower ISS with the same mortality rate in self-in icted injuries compared to those caused by violence.This result suggests that the association of ISS and mortality may be different in patients with self-in icted injuries compared to assault.Other reports indicate that patients with self-in icted SW were more commonly operated and that these patients had higher rate of NTL (2,11).Higher rate of NTL may explain the discrepancy in the number of operations, and it might explain the discrepancy in the mortality.The data registered in the INTR did not allow to evaluate the rate of TL in this study.The reason for higher NTL observed in other studies is also unclear (2,3,11).

In an effort to explain this outcome, we can propose several possible explanations that should be examined in future studies.First, according to the literature, 60-98% of trauma patients due to suicide attempts suffer from mental problems (10).Those disorders include mostly depression and bipolar affective disorders in the older patients and mental distress due to personality crisis in younger patients (12).In the current study, as is the case in other mentioned studies, the rate of surgical interventions among patients with self-in icted abdominal SW was higher.Higher r te of surgeries, might be due to communication problems with patient with mental disorder.

Second, patients with self-in icted abdominal SW have a lower pain threshold, which might have an effect on the way they present pain (13).Since one of the indications for laparotomy include signs of peritoneal irritation, presentation of hypersensitivity might affect the decision to operate those patients.Incidence of hemodynamically instability, another indication for lapar

omy, was the
same in both groups and accounts for only 5.7-6.8% of the cases (Table 2).


Limitations

This study has several limitations.Though based on data recorded in a very large trauma registry the data that may interpret the results of this study is lacking.We are unable to evaluate why more patients in Group I were operated and why more patients in this group died.We do know, however, that self-in icted injuries to the abdomen results in similar ISS and a similar injury pro le as much as individual abdominal organs are concerned.Future studies dealing with self-in icted injuries to the abdomen should concentrate on th

reasons lea
ing to increased proportion of surgeries and increased mortality rates.


Conclusion

Patients with self-in icted stab injury to the abdomen have a higher frequency of abdominal surgery and higher mortality compared to patients similarly injured by an assault.The hypothesis that basic instincts for self-preservation would lead to better outcome was refuted.Consequently, management approaches of such patients that are based solely on repeated clinical examination may not be appropriate for these patients.

Abbreviations stab wound, AASW -Anterior Abdominal stab wounds, NTL -non-therapeutic laparotomy, ISS injury severity score, AIS -Abbreviated Injury Scale, INTR -Israeli National Trauma Registry, LOS-len

h of stay, CI
-con dence intervals, IQR -interquartile ranges, OR -Odds Ratio,


Declarations

Ethics approval: This study was approved by he Ethical Committee of the Sheba Medical Center (approval number SMC-18-513