Drug Therapy Prevents Panic Attacks, but Relapse Very Common

Daniel M. Keller, PhD

March 30, 2015

VIENNA — Panic disorder (PD) responds well to drug therapy, but almost all patients relapse when drugs are withdrawn, a recent long-term study shows.

Benzodiazepines and selective serotonin reuptake inhibitors (SSRIs) each have their advantages and disadvantages in the treatment of PD. Until now, there has been no head-to-head comparison of drugs in the two classes.

Antonio Nardi, MD, of the Institute of Psychiatry at the Federal University of Rio de Janeiro, Brazil, conducted a long-term study of the use of clonazepam (Klonapin, Roche) or paroxetine (multiple brands) or a combination of both.

He presented the results here at the European Psychiatric Association (EPA) 23rd Congress.

The study included men and women aged 18 to 65 years (average age, 34.8 ± 9.3 years) who met DSM-IV criteria for PD-SCID-1 [structured clinical interview for DSM-IV axis 1 disorders]. About two thirds were women, and 37% of patients had previously been treated without lasting success. They completed regular neuropsychological tests and self-reported any adverse effects (AEs).

Clonazepam (n = 63) was started at 1 mg/day and was escalated to 2 mg/day after the second week. Paroxetine (n = 57) was started at 10 mg/day and was escalated to 20 mg/day or 40 mg/day after 2 weeks. Patients received acute treatment with either drug for 8 weeks and continued if they had a good response. For those with a partial response, combination therapy was initiated.

Participants were treated for 3 years and then began a slow tapering of drug during a 2- to 4-month period. They were followed for more than 6 years using a systematic methodology that included an annual in-person interview and a telephone interview.

Good Acute Results

"The clonazepam had a faster response," Dr Nardi reported, "but at the end of the eighth week, both drugs had a similar response" (Table 1).


Table 1. Results of Acute Treatment of Panic Disorder*

  Clonazepam Paroxetine Combination
Panic attacks/month      
Before 5.1 ± 2.5 5.4 ± 3.3 5.5 ± 2.0
After 0.1 ± 0.31 0.2 ± 0.43 0.3 ± 0.8

*Intention to treat; last value carried forward.


AEs were much more prevalent in the paroxetine group than in the clonazepam group (drowsiness, 81% vs 57%, respectively; sexual dysfunction, 70% vs 11%, respectively; both P < .01). Other AEs more common in the paroxetine group were nausea/vomiting, appetite/weight change, dry mouth, excessive sweating, and diarrhea/constipation.

Forty-seven clonazepam patients entered the long-term study along with 21 patients receiving combination therapy and 37 patients receiving paroxetine. After 3 years of treatment, the frequency of panic attacks was greatly reduced for all three groups (see Table 2).

Table 2. Frequency of Panic Attacks After 3 Years

  Clonazepam Paroxetine Combination
Panic attacks/month      
Before 5.1 ± 2.5 5.4 ± 3.3 5.5 ± 2.0
After 0.1 ± 0.31 0.2 ± 0.43 0.3 ± 0.8


For patients who were treated for 3 years and who were essentially free of panic attacks for at least 1 year, drugs were tapered very gradually until they were withdrawn entirely unless AEs appeared or panic attacks returned. Complete, delayed, or partial drug withdrawal was achieved by 88.9% in the clonazepam arm, 60.7% in the paroxetine group, and 75.0% in the combination group.

During the 6-year follow-up period, almost all patients relapsed. The cumulative relapse rates at years 1, 3, and 6 were 41%, 77%, and 94%, respectively.

Dr Nardi concluded that "clonazepam and paroxetine were highly effective and well tolerated during acute and chronic treatment and after relapse." However, "clonazepam was consistently better tolerated than paroxetine during all periods." Clonazepam also showed efficacy more quickly. He also noted that panic disorder is a chronic condition with relapses common if therapy is discontinued.

Session moderators Jean-Marie Sengelen, MD, of the psychiatric hospital in Rouffach, France, and Benjamin Lavigne, who is in his final year of medical training at the University of Limonges, discussed the study findings with Medscape Medical News.

Dr Sengelen said that it would have been interesting to study clonazepam with one or more other drugs, such as pregabalin (Lyrica, PF Prism CV) or the serotonin-norepinephrine reuptake inhibitor venlafaxine (multiple brands), in addition to paroxetine.

Lavigne added that it was an interesting study but that he expected more AEs with the benzodiazepine clonazepam and less than with paroxetine. "But...paroxetine is maybe not the [best] SSRI for the panic disorders or anxiety disorders," he said.

In light of the high relapse rate even after 3 years of treatment, Dr Nardi had suggested stopping drugs earlier.

Lavigne agreed. "I don't see why we would continue too long the treatment," he said. "Stop at the end of the first year, and then do it again in case of relapse."

Dr Sengelen advised not thinking of therapy just in terms of drugs but also to employ some form of psychotherapy, which he predicted would increase the time before relapse. Lavigne added that psychotherapy helps in achieving a better recovery faster and in preventing relapse.

There was no commercial funding for the study. Dr Nardi has received research funding from FAPERJ, CAPES, CNPq FINEP, and INCT-TM. He is on advisory boards of Aché, CNPq, GlaxoSmithKline, Grupo A, and Lundbeck and is on the speakers bureau of Solvay, Cristália, GlaxoSmithKline, Roche, and Aché. Dr Sengelen and Benjamin Lavigne report no relevant finanacial relationships.

European Psychiatric Association (EPA) 23rd Congress. Abstract 0170. Presented March 29, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.