Is it possible to prevent relapse in panic disorder?

KEYWORDS:

• Recurrence
• agoraphobia
• cognitive behavioral therapy
• selective serotonin reuptake inhibitor
• serotonin and noradrenaline reuptake inhibitor
• benzodiazepine
• tricyclic antidepressant

1. Introduction

Up to 33.7% of the population is affected by an anxiety disorder during their lifetime according to large population-based surveys. Panic disorder (PD) has a lifetime prevalence of 1.6% to 5.2%, while the lifetime prevalence of agoraphobia is 0.8% to 2.6% in these studies [1]. PD is associated with functional, occupational, and quality of life impairments. Its economic impact on society is also significant. PD has been associated with suicide attempts, greater use of health care services and decreased work productivity [2]. Many treatments are available for PD and agoraphobia, including pharmacological treatments, cognitive-behavioral therapy (CBT)-based treatments, neurostimulation and physical exercise [2,3]. Despite all treatment options, only 64.5% of patients achieve remission in the first year of treatment, and 21.4% of PD patients have recurrences of panic attacks before completing one year of remission [4]. Typically, PD has a waxing and waning course. In the study from Andersch & Hetta [5], patients with PD were followed for 15 years after a clinical trial with imipramine and alprazolam. At the end of the follow-up phase, they found that 51% still presented with panic symptoms, 18% met the diagnostic criteria for PD and 20% had agoraphobia. Relapse after the end of a well-conducted treatment seems to be a common phenomenon regarding both pharmacological and psychological treatments.

2. Relapse after pharmacological treatment

Treatment-as-usual studies lasting from 3 to 8 years showed cumulative relapse rates ranging from 50% to 89% [6]. In a 6-year follow-up study, Freire et al. [7] found similar relapse rates. After a slow programmed discontinuation of a 3-year pharmacological treatment, the cumulative relapse rates were 50%, 63.6%, 68.2%, 83.5%, and 89.4% at the first, second, third, fifth and sixth years, respectively. Participants were allowed to do physical exercise or psychotherapies during the 6-year follow-up period.

Long-term open-label clinical trials lasting from 6 months to 3 years, with ongoing treatment with citalopram, fluoxetine, fluvoxamine, paroxetine, moclobemide, clomipramine, and imipramine, showed continued improvements and maintenance of benefits over time. Most randomized, placebo-controlled, discontinuation studies also show superiority of antidepressants compared to placebo. One-year treatments with sertraline, imipramine, and venlafaxine XR were associated with less relapse, compared to short-term treatment [2].

It was also demonstrated that long-term treatments with alprazolam [8] and clonazepam [9] are safe and effective in the treatment of PD, with a lower risk of relapse compared to short-term treatments. In the study from Freire et al. [7], clonazepam outperformed paroxetine regarding the relapse rates. Relapse rates for paroxetine were 61.8% after 1-year follow-up and 93.3% after 6-year follow-up, while the rates for clonazepam were 34.1% and 83.3% after 1 year and 6 years of follow-up, respectively. Tolerance to the therapeutic dose may occur in some patients after long-term treatment with benzodiazepines, which is more common in patients with substance use disorder, personality disorders or other predisposing factors [3]. Treatment with benzodiazepines is considered effective and safe for short-term use, nevertheless maintenance treatment requires a careful weighing of risks and benefits [3].

The systematic review on relapse in PD from Caldirola et al. [6] included only studies in which patients who had responded to antidepressants were randomized in a double-blind manner to either continue the same antidepressant or switch to placebo. The duration of treatment before randomization ranged from 8 to 52 weeks, and the medications used were fluoxetine, imipramine, sertraline, and venlafaxine. The double-blind discontinuation-phase duration ranged from 8 weeks to 24 months. The rates of relapse after discontinuation of pharmacological treatment ranged from 28.5% to 50%. Four in seven studies demonstrated higher relapse rates in the placebo group, suggesting that 1-year maintenance pharmacotherapy, with constant doses, had a significant impact on relapse rates and time to relapse in these patients. A finding of concern was the duration of maintenance treatment after successful response did not significantly influence the rates of relapse after discontinuation of pharmacological treatment [6]. In other words, the risk of relapse after stopping the medication seems to be the same for those who received shorter or longer treatments. This is in line with the finding from Freire et al. [7] that, despite the long treatment (3 years) and a very slow medication taper off, relapse rates after treatment discontinuation were high.

The recent World Federation of Societies of Biological Psychiatry (WFSBP) guidelines [3] recommend an additional 6 to 12 months of treatment after remission. For some individuals, the treatment should be extended for many years. WFSBP also recommend to continue patients in the same dose used in the acute phase, not decreasing the dose in the maintenance phase [3].

3. Relapse after cognitive behavioural therapy

CBT is recommended as first-line treatment for PD [3], and the relapse rates in 6 months are low, ranging from 0% to 7.1% [10]. However, the relapse rates in longer follow-up studies (2 or more years) show higher relapse rates, ranging from 19% to 30% [11].

It is still controversial if CBT has more enduring effects compared to pharmacological treatments. There are studies showing lower, higher, and equivalent relapse rates for CBT vs. medications [3]. A recent meta-analysis with mixed anxiety disorders [12] indicated that both CBT and pharmacological treatments have long-lasting effects, with no statistical differences between them.

Clinical trials [13–15] demonstrated that CBT can also be a useful tool to decrease benzodiazepine discontinuation symptoms in patients with anxiety disorders. Patients who received CBT had a better chance of following the taper off schedule, discontinuing the benzodiazepine and staying off these medications for up to 12 months, compared to those that did not receive CBT.

Combination of CBT and pharmacotherapy is effective and it is first-line recommendation in the WFSBP guidelines [3], but data about relapse after combined treatment is lacking.

4. Relapse after other treatments

Physical exercise was recommended as second-line adjunctive treatment for PD by the WFSBP guidelines [3], but data on relapse is scant. Transcranial Magnetic Stimulation (TMS) is a promising treatment for PD, with one randomized controlled trial and one open-label trial showing positive results [16], but data on relapse after treatment is also scarce.

5. Expert opinion

Research on relapse in PD is limited, but in clinical practice it is a well-known phenomenon. Studies on duration of pharmacological treatments indicate that selective serotonin reuptake inhibitors (SSRI), serotonin and noradrenaline reuptake inhibitors (SNRI), tricyclic antidepressants (TCA) and benzodiazepines have a protective effect regarding relapse, but studies with reversible and irreversible monoamine oxidase inhibitors, mirtazapine, vortioxetine and other drugs are lacking. A concerning finding is that, as soon as the medications are discontinued, their protective effect slowly wears off and the relapse rates are high after one year. Another limitation of the research done in this field is that, except for one study [7], there are no studies comparing two or more different drugs regarding the risk of relapse.

The risk of relapse after completion of a CBT course seems to be similar to the risk of relapse after pharmacological treatment. The long-term effects of internet interventions based on CBT, other psychotherapies and the combination of CBT and medications deserve to be investigated in future studies.

Newer treatments involving neurostimulation and physical exercise are promising, but it is too early to say if they will have enduring effects or not. Given the extensive list of possible health benefits and the low risk of adverse advents, exercise should be recommended to patients with PD.

In summary, future research should focus on: (1) comparing different drugs regarding risk of relapse in long-term treatment, including MAOI, mirtazapine, and vortioxetine; (2) comparing pharmacological treatment alone to combined treatment (medication plus CBT) regarding relapse prevention; (3) ascertaining long-term effects of adjunctive exercise and (4) ascertaining the long-term effects of TMS and other modalities of neurostimulation.

In conclusion, the best treatments to prevent relapse in PD are roughly the same ones recommended by the guidelines for acute treatment: SSRI, SNRI, TCA, benzodiazepines, and CBT. Pharmacological treatment should be extended for at least 6–12 months after remission is achieved. Maintenance pharmacological treatment should be considered for selected patients and would further decrease the risk of relapse, but it would come with costs such as metabolic effects or persistent of side effects. Patients with chronic PD symptoms and those with high risk of recurrence are more likely to benefit from maintenance pharmacological treatment. If treatments are to be resumed, treatment history must be taken into account. A patient educated about the risks of relapse and aware of factors that could increase them is fundamental to avoid deterioration in clinical presentation. The patient should be the first to recognize early signs of relapse and seek specialized treatment.

Declaration of interest

RC Freire participated and was supported on a sponsored clinical trial by Biohaven Pharmaceuticals and has research grant support from Queen’s University, Canada. AE Nardi meanwhile was supported by grants from the Fundacao de Amparo a Pesquisa do Estado do Rio de Janeiro (FAPERJ), the Coordenacao de Aperfeicoamento de Pessoal de Nivel Superior (CAPES), the National Council for Scientific and Technological Development (CNPq). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript was not funded.