Non-fatal suicide attempt by intentional stab wound: Clinical management, psychiatric assessment, and multidisciplinary considerations.

BACKGROUND: Suicide by means of self-inflicted stab wounds is relatively uncommon and little is known about this population and their management.
MATERIALS AND METHODS: Retrospective review of adult trauma patients admitted to our Level-1 trauma center between January 2005 and October 2009 for management of non-fatal, self-inflicted stab wounds.
RESULTS: Fifty-eight patients were evaluated with self-inflicted stab wounds. Four patients died due to their injuries (mortality, 7%). Of the non-fatal stab wounds, 78% were male ranging in age from 19-82 (mean: 45 years). The most common injury sites were the abdomen (46%), neck (33%), and chest (20%). In terms of operative interventions, 56% of abdominal operations were therapeutic, whereas 100% of neck and chest operations were therapeutic. When assessing for suicidal ideation, 44 patients (81%) admitted to suicidal intentions whereas 10 patients (19%) described accidental circumstances. Following psychiatric evaluation, 8 of the 10 patients with "accidental injuries" were found to be suicidal. Overall, 54 patients (98%) met criteria for a formal psychiatric diagnosis with 48 patients (89%) necessitating inpatient or outpatient psychiatric assistance at discharge.
CONCLUSIONS: Compared to previous reports of stab wounds among trauma patients, patients with self- inflicted stab wounds may have a higher incidence of operative interventions and significant injuries depending on the stab location. When circumstances surrounding a self-inflicted stabbing are suspicious, additional interviews by psychiatric care providers may uncover a suicidal basis to the event. Given the increased incidence of psychiatric illness in this population, it is imperative to approach the suicidal patient in a multidisciplinary fashion.

INTRODUCTION

In 2009, over 374,000 individuals received emergency department based care in the United States for non-fatal self-inflicted injuries.[1] The most common reported methods include self-immolation, chemical or medication ingestion, attempted hanging, jumping from a height, and penetrating trauma. Such mechanisms have been associated with substance intoxication and/or pre-existing psychiatric conditions.[2] Suicide by stabbing is uncommon, constituting only 1.6-3.0% of suicide attempts.[3] Given its relatively low incidence and the fact that mortality associated with self-inflicted stab wounds is surprisingly low,[4] only a few investigations have studied this at risk population. The purpose of this investigation was to evaluate the clinical management, psychosocial profile, and psychiatric diagnoses of those admitted following a non-fatal, self-inflicted stabbing.

MATERIALS AND METHODS

A retrospective chart review was conducted on patients with non-fatal, self-inflicted stab wounds who were admitted to Rhode Island Hospital (a level one trauma center) from January 2005 to October 2009. Inclusion criteria included patient age greater than 18 and admission to the hospital for treatment of traumatic injuries following a self-induced stabbing. Exclusion criteria included adults who have a recurrent history of self-injurious behavior (self-mutilation) that present with intentional superficial self-slashing injuries, and confirmed unintentional (accidental) stab wound injury at the time of primary emergency department assessment. All patients were admitted to the trauma service for acute management and subsequently referred to the psychiatric consult liaison service for suicide assessment. All patients were evaluated and managed by both an attending surgeon and psychiatrist.

A data summary sheet was created that included information on demographics (gender, age, marital status, education, employment, and insurance levels), psychiatric history, substance abuse history, contextual factors related to the suicide act, stab wound site and their management, implement used for self-stabbing, and patient disposition upon hospital discharge. Simple descriptive statistics were used to analyze the data. The Rhode Island Hospital Internal Review Board approved this study.

RESULTS

Fifty-four patients with a presumed suicide attempt by self-inflicted stabbing were admitted during the study period. Ages ranged from 19 to 82, with an average of 45 years of age. Patient demographics can be found in Table 1. In regards to substance abuse, alcohol blood levels ranged from 61-308 MG/ DL with a mean of 169 MG/DL. The most commonly used illicit substances included: benzodiazepines (9), opiates (8), cocaine (5), cannabis (5), and barbiturates (2). Nine (17%) patients used more than one illicit substance together and five (9%) of patients used elicit substances in combination with alcohol. Eight patients overdosed on medications such as insulin, methylphenidate, selective serotonin reuptake inhibitors, aspirin, acetaminophen, ibuprofen, trazodone, or benzodiazepines. Fifteen patients (28%) did not have a blood alcohol level performed and seventeen (32%) patients did not have a urine toxicology screen performed.

Table 1

Patient Demographics

Upon trauma team evaluation, forty patients (74%) presented with a single self-inflicted stab wound whereas 14 (26%) presented with multiple injuries. The most common site of peripheral injury was the wrist (8 patients), all of which necessitated operative intervention. The overall sites of injury and their management are summarized in Table 2. Of the 18 stab wounds to the neck, all were located in zone two. Seven underwent ligation of internal, external, or both jugular veins. Two patients underwent tracheal laceration repair with one individual requiring tracheotomy due to thyroid cartilage fracture. Three patients with penetrating neck injury required a chest tube for hemo-pneumothorax. Another patient was diagnosed with a retropharyngeal hematoma that did not require operative intervention. One patient underwent laceration repair of the sternocleidomastoid muscle. Six additional patients were observed and did not require surgical neck exploration.

Table 2

Self-inflicted stab wound sites of injury, operative interventions, and rates of therapeutic intervention

Of 11 patients with stab wounds to the chest, 5 required operative interventions and all of these were therapeutic. Injuries included myocardial, lung, and internal mammary artery lacerations. One patient required operative removal of a knife from the chest and a subsequent chest tube for hemo-pneumothorax. The remaining 6 patients did not require operative exploration and were managed either with serial chest X-rays or tube thoracostomies.

Of the 25 patients with stab wounds to the abdomen, nine underwent successful non-operative management through our admission protocol for serial exams. Of the 16 that required operative intervention, three patients were noted to have a lacerated liver that was managed non-operatively. One patient sustained a gallbladder laceration, developed bile peritonitis, and underwent a cholecystectomy. Another individual underwent ligation of the right inferior epigastric artery. Three patients had various small and large bowel injuries resulting in resection, primary repair, or ostomy. One of these patients also required a splenectomy. Another patient had an exploratory laparotomy, pericardial window, phrenic artery branch repair, and right ventricular laceration repair. Seven patients were found to have no acute organ injury following laparotomy.

Within 48 hours from admission, all 54 patients underwent a complete psychiatric evaluation. Contextual factors associated with their self-inflicted stab wounds as endorsed by patients are reported in Table 3. Following their interview, fifty-three patients (98%) met criteria for a formal Axis 1 and/or Axis 2 psychiatric diagnosis. Table 4 describes the specific psychiatric diagnoses of our patient population. During the initial interview, forty-four patients (81%) readily described suicidal intentions during their psychiatric evaluation whereas ten patients (19%) reported that their stab wounds were “accidental”. Of these ten patients who reported that their stab wounds were accidental, 8 of the 10 patients’ injuries were subsequently found to be intentional following serial psychiatric interviews while in the post-operative care unit. In this subgroup, patient injuries included the neck (1), chest (2), and abdomen (5). In patient with the neck injury, superficial bilateral neck lacerations were encountered that the patient stated was related to a chain saw accident. The patient denied that his lacerations were self-inflicted, however, the patient had a history of prior suicidality and the wounds were not reflective of a chainsaw related injury. Of the chest traumas, one individual reported that a band saw blade dislodged into his chest, however, the patient could not explain how he sustained a wrist laceration at the same time thus indicating the high likelihood for suicidality. Another patient reported that he accidentally stabbed himself in his chest. Additional information obtained from the patient's family confirmed the accident was in fact a suicide attempt. Of the abdominal traumas, five patients reported accidental abdominal injuries including: Hitting the wrong button on a switchblade knife (1), tripping while putting a knife away (2), or falling on a knife (2). Their stories were subsequently found not to be credible based on the injury pattern and/or conflicting stories later obtained from the families and/ or the patient. Such mechanisms were only fully understood and acknowledged following serial interviews conducted by the consulting psychiatric service.

Table 3

Contextual factors leading to self-inflicted injury

Table 4

Formal psychiatric diagnosis

In the remaining two patients in which non-intentional circumstances were questioned, the first individual was intoxicated and presented with penetrating trauma to the neck. Police later confirmed that an arrest was made in connection with the patient's injuries. The other patient was non-intoxicated and presented with a stab to the abdomen. The patient reported fixing the screw on his glasses at which time the knife slipped, puncturing his abdomen. Despite the suspicious wound and circumstances, information obtained later on supported that the injury may have been accidental.

At discharge, 33 (61%) patients were referred for inpatient psychiatric admission. Eight (15%) were discharged with outpatient follow-up. Five patients were transferred voluntarily to a diversion residential program. Two were referred for partial hospitalization. Three were deemed safe to be discharged home without psychiatric treatment. One patient with a Cluster B Personality Disorder was able to leave against medical advice following a protracted surgical hospital stay. One patient went to jail. Another patient eventually died not of the self-inflicted stab wound, but as a result of a Selective Serotonin Reuptake Inhibitor overdose.

DISCUSSION

Nationally, suicide represents the 11th leading cause of death in the United States. In general, men complete suicide more frequently than women.[1] Adolescents and older adults (>65) present the greatest suicide risk although suicide rates in the elderly have fallen.[2] Caucasians or people of lighter skin color have a higher rate of suicide than those with darker skin color.[5] Qin, Agerbo, and Montensen report that single marital status, unemployment, low income, retirement, disability, sickness-related absence from work, family history of suicide, and/or psychiatric history are also significant risk factors for suicide.[6] In regards to those who abuse elicit substances and alcohol, some studies showed an increased association with suicidal behavior. Sher et al. found that acute alcohol use increased the lethality of suicide attempts in those individuals with mood disorders.[7] Subsequent research demonstrated the association between alcohol and suicidality in the absence or presence of mood disorders.[89] Given that our population shared many of these characteristics, the recognition of risk factors and the implementation of targeted interventions prior to an acute suicidal event continue to challenge our local community.

Compared to previous reports of stab wounds among trauma patients, patients with self-inflicted stab wounds may have a higher incidence of operative interventions and significant injuries depending on the location of the stab wound. Biffl et al. described a 23% incidence of significant injury associated with stab wounds to the neck.[10] Demetriades et al. further characterized the incidence of injuries by structure, with vascular injury being the most common (14% of the presenting stab wounds).[11] In our population, the incidence of significant injury to the neck requiring operative intervention was higher at 66% with all of the operations being therapeutic. In regards to chest injuries, Mattox and Allen reported only 12-15% of patients require operative interventions for penetrating trauma to the chest.[12] Among our patients, 45% required a therapeutic operative intervention. The patient's desire to induce significant self-harm may explain the higher incidence of significant injuries compared to the general population when the site of stab is in the neck or chest.

In terms of those with abdominal stab wounds, it has been well established that those who undergo laparotomy for an anterior stab wound may not receive a therapeutic benefit.[13] In one of the most recent reports, Biffl et al. reports in a multi-institutional study of anterior abdominal stab wounds using a variety of diagnostic modalities to direct operative management, 48% of the patients received a laparotomy with only 74% being therapeutic.[14] In our series, a higher number of patients went to laparotomy (64%), while a lower number received therapeutic benefit (56%). As mentioned previously, our method of deciding operative intervention for stable anterior abdominal stab wounds is that of serial abdominal exams. Given that this population maintains several risk factors including psychiatric illness and poly-substance abuse, potentially unreliable subjective and objective findings elicited at the bedside may have contributed to the higher incidence of non-therapeutic operations. Also, in contrast to neck and chest-based injuries that typically manifest “hard” signs on physical exam that direct operative management, the nature of the abdominal viscera makes significant injury less likely in stabbing. This forces the practitioner to depend on a variety of indirect indicators that may falsely lead to an operation.

Ninety-eight percent of our patients met criteria for a formal psychiatric diagnosis. This high rate of psychiatric disorders underscores the importance of a multidisciplinary approach to patients presenting with non-fatal self-inflicted injuries. This is particularly imperative among patients who deny self-inflicted stabbing despite a presenting history or clinical exam suspicious for intentional self-harm. With 80% of these patients eventually found to have had suicidal ideation, ongoing follow-up with trauma team members, social workers, substance abuse counselors, and psychiatric consultants would contribute to identifying these at risk patients. As a result of identifying suicidal intent, patients were more likely to be transferred to an inpatient psychiatric facility at the time of acute hospitalization discharge. While it remains an ongoing challenge to prospectively prevent suicide, it is left to the trauma care team to truly question suspicious circumstances and provide effective treatment so that any future attempts could be averted.

In regards to study limitations, this study is reflective of the population of the northeastern United States and therefore may be different than other parts of the country. Even though each self-inflicted stabbing patient was interviewed on at least two separate occasions to determine the consistency and veracity of information provided, the information obtained was still subject to substance intoxication, omission, and interpretation at the time of the interview(s). Due to the retrospective nature of the study and the fact that different personnel performed the initial interviews, variations in documentation limited our ability to fully define the specific stressors leading up to each suicidal event. Also, as a result of inconsistent blood alcohol level and urine toxicology screen ordering, the full impact of substance usage and its contribution to the acute event cannot be fully evaluated. Finally, given the lack of long-term follow-up and focus on acute hospitalization, it is unknown how surviving the non-fatal suicide attempt influenced future at-risk behavior.

CONCLUSION

Self-inflicted stabbing is rarely a fatal event during the acute hospitalization; however, the need for operation and its therapeutic benefit may be greater than the general population depending on the location of injury. When managing these patients, a collaborative approach between surgery and psychiatry is imperative to effectively manage the life threatening issues acutely and ultimately treat the psychological factors that led to the suicidal event in the first place.