Journal Article Annotations
2019, 4th Quarter
Annotations by Scott Simpson, MD, MPH
The investigators randomized 18 patients in an emergency department (ED) presenting with nonspecific depression diagnosis (DSM-IV-TR depression nos, major depressive disorder, or bipolar depression) and active suicidal ideation. A lower ketamine dose was used to enable administration in 5 minutes in the ED setting. On the primary outcome of intensity of suicidal ideation, 88% of ketamine patients experienced reduced suicidal ideation at 90 minutes compared to 33% of controls. This statistically significant effect did not persist beyond 180 minutes, and there was no difference between groups in secondary outcomes related to depressive symptoms. The study ended early when the principal investigator left the institution.
Strength and weaknesses:
Suicidality is a challenging and intensive condition to treatment in the ED; patients often experience direct observation, psychiatric hospitalization, and prolonged lengths of stay. This novel study aims to deliver a treatment for suicidality in an important treatment setting. Indeed, suicidality appears to loose intensity for many patients, and the number needed to treat is impressive (roughly 2). The investigators also selected a low dose that could feasibly be delivered. However, there are several methodological limitations. The authors do not describe the recruitment process, so it is unknown how many patients were approached and excluded; patients on medications affecting the NMDA system (eg, lithium and lamotrigine) were also excluded, which may have made subjects less sick than the larger population of presenting patients. Also, it is unclear that patients or clinicians could be effectively blinded from a subject having received ketamine versus placebo. Finally, the outcomes are relatively subjective, and they do not report other important outcomes such as length of stay, need for or length of admission, or self-harm at follow-up.
Around three-quarters of patients presenting to the ED with suicidal ideation are admitted to the hospital—roughly the inverse of patients with medical complaints. Better treatments for acute suicidal ideation are required, and aggressive treatments like ketamine hold promise for enhancing treatment in this setting. This study deserves notice for its attempts to deliver a practical and novel treatment in a general medical setting that treats many acute psychiatric patients. Methodological limitations temper this study’s impact. Nevertheless, ketamine is becoming more accessible (and available in oral and intranasal forms) and ED physicians are very familiar with the agent; psychiatrists may begin seeing a larger push to provide ketamine in EDs.
Type of study:
(http://ebm.bmj.com/content/early/2016/06/23/ebmed-2016-110401): randomized controlled trial
The investigators examined mortality among patients within one year after presentation to the emergency department (ED). All EDs in California were included. The investigators hypothesized that there would be a gradient of suicide risk based on the diagnostic code from the patient’s index ED visit: indeed, patients presenting with self-harm had a higher suicide incidence (693 per 100,000 person-years) than those with suicidal ideation but no self-harm (385 per 100,000 person-years) or other, non-suicide diagnoses (23.4 per 100,000 person-years). Patients with self-harm or suicidal ideation were also far more likely to die by unintentional injury or homicide.
Strength and weaknesses:
This large cohort describes a population-based sample of all ED visits in California, thereby providing a sense of the scale and morbidity among patients seen in the ED. The study’s major limitation lies in its use of diagnostic codes, which may be inconsistently applied and introduce selection bias that exaggerates correlations. For instance, only the sickest patients may be included in the suicidal ideation and self-harm cohorts, whereas many more patients may screen positive in emergency departments or carry co-morbid diagnoses yet not receive the specific diagnostic codes necessary for inclusion.
There are several significant findings from this study beyond simply the substantial observed mortality. First, patients with self-harm and suicidal ideation in the ED have elevated risks of death not only by suicide but also by accident and homicide in the year after ED discharge. These risks may reflect important confounders such as impulsivity, chaotic psychosocial environments, or substance use. Second, even patients presenting to EDs without suicidal ideation carry an increased risk of death by suicide relative to the general population. Thus, screening and intervention strategies should consider how to mitigate risk among this population of patients who are not necessarily presenting for suicidal ideation.
Type of study:
(http://ebm.bmj.com/content/early/2016/06/23/ebmed-2016-110401): Retrospective cohort study