Efficacy and Safety of Aripiprazole in Borderline Personality Disorder: A Systematic Review

Aripiprazole is an atypical antipsychotic medication, and its use in treating borderline personality disorder (BPD) is debatable because it is not FDA-approved for treating BPD. This study aimed to investigate the efficacy and safety of aripiprazole in patients with BPD. On July 2, 2021, the protocol (CRD42021256647) was registered in PROSPERO. PubMed, Scopus, Web of Science, Ovid-Medline, Embase, PsycINFO, and Cochrane (CENTRAL) were searched without regard for language or publication date. We also searched trial registries on ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform. Randomized clinical trials with adult patients diagnosed with BPD met the inclusion criteria. The Cochrane risk of bias for randomized trials (RoB-2) method was used to assess the quality of the included studies. We included two previously published randomized clinical trials. There were 76 patients with BPD, with 38, 12, and 26 assigned to the aripiprazole, olanzapine, and placebo groups, respectively. Most patients (88.16%) were females, with ages ranging from 22.1 to 28.14 yr. Aripiprazole has been proven to reduce anxiety, depression, anger, hostility, clinical severity, and obsessive–compulsive behavior, insecurity, melancholy, anxiety, aggressiveness/hostility, phobic anxiety, paranoid thinking, psychoticism, and somatization. The adverse effects were headache, insomnia, restlessness, tremor, and akathisia. The risk of bias was considerable in both trials, which is somewhat problematic considering that prejudice can lead to incorrect outcomes and conclusions. Aripiprazole has demonstrated encouraging outcomes in the treatment of patients with BPD. More randomized controlled studies are needed.


Background
Borderline personality disorder (BPD) is defined by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), as "a dominant pattern of instability of selfimage, relationships between people, and affections, as well as intense impulsivity present in various contexts and usually appearing in early adulthood" [1].This mental condition affects 1% of the general population and 22% of hospital psychiatric patients [2].According to the National Institute of Mental Health, 5% of Peruvians may have BPD, which can be diagnosed as early as the age of 16 yr [3].
Aside from the significant suffering caused by BPD, it is likely that getting a diagnosis and selecting the proper therapy will be challenging due to the variability of symptoms in each patient.Structured psychotherapy is the primary recommended treatment for BPD [4].Its availability, however, is limited due to high economic costs, the need for skilled human resources, and the time commitment [5].In this case, the treating physician has few therapeutic options.Psychotropic medications are commonly used in the clinical practice of BPD despite their usage not being recommended due to the lack of specific evidence [6].Over 70% of patients with BPD also have other comorbidities, such as mood disorders, anxiety disorders, posttraumatic stress disorder, substance use disorders, or eating disorders [7].The National Institute for Health and Care Excellence (NICE) guidelines in the United Kingdom indicate that psychotropic medication be used for psychiatric comorbidities and the shortest feasible term [8].A study on prescription practice in the United Kingdom, on the other hand, revealed that at least four out of five patients with BPD received psychotropic medication and three out of five patients received antipsychotics [6], suggesting the extensive use of antipsychotics for the treatment of BPD.
Aripiprazole is an atypical antipsychotic medication previously studied to treat schizophrenia [9][10][11].It was FDA approved for diseases like schizophrenia or bipolar I disorder, major depressive disorder, autistic disorder, and Tourette syndrome, and its use were expanded to children and adolescents [12].This medication has been demonstrated to have strong short-and long-term tolerance, be less likely to alter the metabolic profile, to lower prolactin levels, to raise the QT interval, and to generate extrapyramidal symptoms, as well as to have few adverse effects, such as headache, nausea, insomnia, or anxiety [13].
According to an Italian survey, 22.8% of psychiatrists believe that off-label usage of aripiprazole is adequate for treating BPD [14].Similarly, in a follow-up cohort of 2,195 hospitalized patients with BPD, a significant increase in the trend of aripiprazole prescription was reported [15].These trends contradict NICE guidelines, which do not advocate aripiprazole for treating BPD; however, it is crucial to note that this recommendation has not yet been updated [8].As a result, the use of aripiprazole for the treatment of BPD is still disputed, and the available data must be reassessed.This study aims to compile evidence on the efficacy and safety of aripiprazole in patients with BPD.

Methods
The "Preferred Reporting Items for Systematic Reviews and Meta-Analysis" [16] statement and the Cochrane Manual of Systematic Reviews were used to conduct this systematic review.On July 2, 2021, the protocol was registered in PROSPERO (CRD42021256647), which analyzed studies that assessed the efficacy and safety of aripiprazole in treating BPD.

Data Source
We searched PubMed, Scopus, Web of Science, Ovid-Medline, Embase, PsycINFO, and the Cochrane Central Register of Controlled Trials (CENTRAL) without regard to language or publication date until June 9, 2021.Additional file 1 contains the complete search technique for each database.Furthermore, trial registries (https:// www.clini caltr ials.gov) and the World Health Organization's International Clinical Trials Registry Platform (http:// apps.who.int/ trial search/) were searched.

Eligibility Criteria
The following studies were included: (a) randomized controlled trials (RCTs) and (b) adult patients (> 18 yr) diagnosed with BPD based on DSM criteria.Case-control studies, cohort studies, and case reports were eliminated, as were nonrandomized clinical trials or clinical trials without a control group.

Data Selection and Extraction
Rayyan QCRI (https:// rayyan.qcri.org/) was used to pool searches and remove duplicates.Four investigators (D.A.P.Z., S.K.C.R., B.O.S., and M.F.F.F.) independently assessed the titles and abstracts to choose possibly relevant studies for inclusion.Any disagreements were handled by consensus with the other investigators (V.B.Z. and G.V.J.).
The data were extracted independently in a standardized format by four investigators (D.A.P.Z., S.K.C.R., B.O.S., and M.F.F.F.) and contained the following information: investigators, year of publication, baseline characteristics of patients, dose of aripiprazole, route of administration, frequency, and time of the intervention, score of scales used for BPD, and number of adverse effects.Any disagreements were discussed and resolved.

Risk of Bias
The Cochrane risk of bias for randomized trials (RoB-2) [17] tool was used to assess the methodological quality of the included studies.Four investigators (D.A.P.Z., S.K.C.R., B.O.S., and M.F.F.F.) independently evaluated and classified the risk of bias as low, unclear, or high based on five domains: randomization process, deviations from the intended interventions, missing outcome data, outcome measurement, and selection of the reported result.Disagreements among the authors were resolved by discussion and consensus.
Using qualitative synthesis, the characteristics of the selected studies were discussed in detail.Because of the small number of studies chosen and the fact that these studies used different comparators, had very other follow-up times, and assessed the outcomes using different scales, a quantitative analysis (meta-analysis) was not undertaken.

Study Selection and Study Characteristics
Figure 1 shows the study selection procedure.The original search produced 507 articles, which were later removed due to duplicates.A total of 311 publications were screened based on title and abstract, with 12 potentially eligible for full-text review.Nine were excluded based on the selection criteria (Additional file 2), leaving only two investigations, one published in two complementary papers.
The general characteristics of the included studies are summarized in Table 1.The study conducted by Nickel et al. [25,26] was a double-blind RCT, whereas Shafti and Kaviani [27] did an open-label RCT.Seventy-six patients diagnosed with BPD using DSM III and IV were included in this study, published between 2006, and 2014.Of these patients, 38, 12, and 26 were assigned to the aripiprazole, olanzapine, and placebo groups, respectively.Most participants (88.16%) were females, with a mean age of 22.1-28.14yr.
Both trials used different aripiprazole doses, with Nickel et al. [25,26] using 15 mg/ day and Shafti and Kaviani [27] using 30 mg/day.Begin with 2.5 mg/day and progressively increase to a maximum of 10 mg/day by 2.5 mg per week.Both trials had different durations; the former had a follow-up time of up to 18 months, whereas the latter had a followup term of up to 12 months.

Risk of Bias
Nickel et al. [25,26] conducted a study using the Cochrane Collaboration technique to assess the risk of bias that showed some reservations about the claimed result's selection.On the other hand, the study of Shafti and Kaviani [27] obtained a high-risk score due to many problems in the randomization procedure, deviations from the intended intervention domains, and a high risk in outcome measurement and domain selection (Fig. 2).

Outcomes
Both studies used multiple measures to assess symptoms, including the SCL-90-R, HAM-D, HAM-A, STAXI, BDHI, and BPRS.The SCL-90-R Global Severity Index and Clinical Global Impressions-Severity Scale were used to determine the severity of clinical symptoms.Table 2 summarizes the key findings.

Psychiatric Symptoms
Nickel et al. [25,26] used the SCL-90-R scale to assess psychological and psychopathological symptoms in 52 patients, 26 of whom received aripiprazole at 15 mg/day, and the r, remaining patients received a placebo.The results were measured at 2, 6, 12, and 18 months, with the aripiprazole group showing a significant reduction in obsessive-compulsive behavior, insecurity in social contacts, depression, anxiety, aggressiveness/hostility, phobic anxiety, paranoid thoughts, and psychoticism.The only symptom that did not show a significant reduction at 2 months was somatization, but it did at 6, 12, and 18 months.At 18 months, none of the symptoms in the placebo group showed a significant change in the mean score.
In the study by Shafti and Kaviani [27], psychiatric symptoms were assessed using the BPRS scale in 24 patients.Twelve of these patients were given aripiprazole at starting dose of 2.5 mg, which was gradually increased to 10 mg, and the rest were given olanzapine.A 2-month examination revealed that both groups experienced a considerable reduction in mental symptoms, with the olanzapine group experiencing a modest increase.This assessment, however, did not consider long-term results.

Anxiety and Depression
Anxiety and depression were assessed using specific Hamilton measures (HAM-A and HAM-D) in the study by Nickel et al. [25,26].The mean score for both scales was significantly lower in the aripiprazole group but not in the placebo group.This difference continued throughout the study's 18-month period.

Anger/Hostility
Nickel et al. [25,26] explicitly used the STAXI to assess rage in their study.Aripiprazole significantly reduced the mean score of the following anger expression scales: state anger, trait anger, anger in, anger out, and anger management.The placebo group, on the other hand, demonstrated no improvement on these scales.Furthermore, Shafti and Kaviani [27] used the BDHI to assess "hostility" and found that the olanzapine group significantly improved the mean score on this scale, whereas the aripiprazole group did not.

Severity
Nickel et al. [25,26] assessed "severity" with the SCL-90-R Global Severity Index.On a 5-point Likert scale, this subscale assesses an individual's basic mental tension.At 2 to 18 months of follow-up, the aripiprazole group demonstrated significant changes in their score.The placebo group did not show a significant reduction in their mean score.Similarly, Shafti and Kaviani [27] used the CGI-S to assess the disease's overall severity, with the aripiprazole group showing no improvement and the olanzapine group showing a significant reduction in the mean score.

Adverse Effects
In all of the studies examined, no major adverse events, or suicidal behavior was documented.
In the study by Nickel et al. [25,26], no weight changes were seen; nevertheless, various side symptoms, such as headache, insomnia, nausea, numbness, restlessness, anxiety, and constipation, were observed.Weight gain, drowsiness, dizziness, and tremor were observed in the olanzapine group, whereas headache, insomnia, restlessness, tremor, and akathisia were observed in the aripiprazole group.Table 3 summarizes the adverse effects identified in both studies.

Discussion
This comprehensive analysis examined two published trials with a total of 76 patients diagnosed with BPD using DSM III and IV criteria, with 38 patients assigned to the aripiprazole group, 12 patients assigned to the olanzapine group, and 26 patients assigned to the placebo group.Women outnumbered men (88.16%), with ages ranging from 22.1 to 28.14 yr.
There is little evidence that aripiprazole is effective in the treatment of BPD.Only one of the included studies in the subgroup that referred to the use of aripiprazole coincided with our findings in a systematic review and meta-analysis of the efficacy of antidepressants, mood stabilizers, and antipsychotics in alleviating symptoms, such as depression, and anger in BPD conducted by Mercer et al. [28].This was the study Nickel et al. [25] conducted, with an 8-week patient follow-up, showing that it had the largest effect size on anger, depression, and aggressiveness.Although the findings are comparable, our analysis incorporated the work of Nickel et al. [26], who used a study conducted in 2006 as a reference for the use of a sample of patients who were examined over a period of more than 18 months.As a result, this study was included, whereas the other one was not.
Similarly, additional systematic reviews on pharmacological treatment for BPD that used the same study by Nickel et al. [25,26] reached the same conclusion about the insufficient evidence for using aripiprazole and produced the same results [29][30][31][32].Taking this into account, one of the strengths of our study was the inclusion of a sample study of a larger sample of participants with a longer follow-up period and another one that compared aripiprazole and olanzapine, where previous evidence suggested that they had similar effects on the aggressiveness and anger of BPD when studied separately [30,31].
The neurobiology of BPD is complicated, comprising multiple neurobiological and genetic variables [33].It has been proposed that BPD involves decreased serotonergic activity and dopaminergic dysfunction, which would explain the prevalence of specific BPD symptoms, such as impulsivity and emotional dysregulation [34,35].Similarly, polymorphisms associated with BPD were found in the genes encoding the 5-hydroxytryptamine receptor type 1A (5HT1A), dopamine receptor type 4 (D4), and dopamine transporter type 1 (DAT1) [35][36][37].Aripiprazole is a partial agonist at dopamine receptors (D2-D4) and serotonin receptors (5HT1A, 5HT2A, 5HT2C, and 5HT7) [38].Therefore, its partial agonist action may improve regulation in both pathways after engaging with their respective receptors, hence alleviating BPD symptoms.The evidence on the influence of polymorphisms on the pharmacodynamics of aripiprazole is still limited, even though the results obtained in this review [38] would indirectly support this theory.
When the results of each trial are reviewed, a significant reduction in the intensity of BPD symptoms can be detected, although the evidence for using aripiprazole has been insufficient.However, both trials have a risk of bias, which limits clinical decision-making for using aripiprazole for BPD.This conclusion is significant because treating personality disorders is still a developing topic.Therefore, studying the efficacy, and safety of antipsychotic medicine, such as aripiprazole, will allow us to make better therapeutic decisions and investigate various approaches, both pharmaceutical, and involving psychotherapy.To properly assess aripiprazole as a treatment alternative, well-designed clinical studies with high sample sizes, standardized instruments to measure outcomes, and long-term follow-up are required.
The limitations of this study were related to the small number of RCTs available, the risk of bias, and the small number of participants.Furthermore, a quantitative synthesis (meta-analysis) was not possible due to variability in many elements of both studies and data limitations.

Fig. 1
Fig. 1 PRISMA 2020 flow diagram of search process

Fig. 2 a
Fig. 2 a Risk of bias distribution, b Risk of bias summary.Methodological quality of the studies included in this systematic review is shown.a Risk of bias graph for the studies included in this systematic review.b Domain 1: Randomization process; 2: Deviations from the intended interventions; 3: Missing outcome data; 4: Measurement of the outcome; 5: Selection of the reported result

and
Climate Change Research Unit (EMERGE-UPCH) and the Institute for the Evaluation of Technologies in Health and Research of EsSalud (IETSI -ESSALUD).Susan K. Campos-Rodriguez My name is Susan Kimberly Campos Rodriguez, I was born in Peru on January 7, 1999.She is currently a 6th year medical student at the National University of Trujillo.In recent years, I have been gaining experience in research by participating in scientific conferences and congresses, as well as in internships at the Systematic Reviews Unit of the Peruvian Epidemiological Research Group (GRUPIED) and the Emerging Diseases and Climate Change Research Unit (EMERGE-UPCH).Likewise, I have held the position of president of the Scientific Society of Medical Students of the National University of Trujillo during the 2021 term.Brando Ortiz-Saavedra My name is Brando Ortiz-Saavedra.I am currently 23 years old and I have just completed my professional studies in Human Medicine at the National University of San Agustin in Arequipa, Peru.During the last years of his degree, he was part of the Scientific Society of Augustinian Medicine Students (SOCIEMA), as an active member.Likewise, I was able to participate in different student congresses organized by SOCIMEP.I completed my medical internship at the Honorio Delgado Regional Hospital, located in the city of Arequipa.María F. Fernández President of the Peruvian Student Medical Scientific Society (SOCIMEP) 2023.Student of Human Medicine at the Private University of Tacna.Full Member of the Medical Student Research Center (CIESMED) 2022-2023, Tacna, Peru.President of the XXXV SOCIMEP National Scientific Congress -Tacna, Peru 2021.Vicente Aleixandre Benites-Zapata Physician specialized in Medical Auditing, Master's degree in Epidemiological Research and PhD studies in Nutrition and Food.Researcher in the Office of the Vice Chancellor of the Universidad San Ignacio de Loyola, professor in the medical school of the Universidad Privada de Ciencias Aplicadas and professor in the master's program in Epidemiology and Biostatistics at the Universidad Científica del Sur.Areas of Interest Social Epidemiology, Human Papilloma Virus and Uterine Cervical Cancer, Evaluation of Health Programs, Cardiovascular Diseases, Diabetes, Chronic Kidney Disease and Clinical Nutrition.

Table 1
General characteristics of the studies included

Table 2
Main results of the studies included

Table 3
Reporting of adverse events