Serotonergic/norepinephrinergic antidepressants.
The Food and Drug Administration (FDA) has approved several selective serotonin reuptake inhibitors (SSRIs) and serotonin norepinephrine reuptake inhibitors (SNRIs) for PD, GAD, and SAD. Despite these classifications, medications not approved for a condition are commonly used “off-label” in clinical practice. The European Union has similar indications for the use of SSRIs and SNRIs for the treatment of anxiety disorders as the FDA but with broader indications of SSRIs (
12). See
Table 1 for a list of FDA-approved and off-label medications for anxiety.
Selective serotonin reuptake inhibitors and SNRIs are both first-line treatments for PD, GAD, and SAD and have been shown to be efficacious for the treatment of anxiety disorders (
13–
16). A recent meta-analysis reported that most SSRIs and SNRIs are more efficacious than placebo in GAD, with escitalopram and duloxetine potentially having the largest effect sizes (
17). The recommended duration of treatment can vary but may be as short as 3–6 months, or up to 1–2 years or even longer. Although there may be concern about tachyphylaxis, there is limited evidence of adverse outcomes with the chronic use of SSRIs or SNRIs (
18). These medications also tend to be well-tolerated, with usually manageable or short-lived adverse effects such as nausea, headache, dry mouth, diarrhea, or constipation. Sexual dysfunction tends to be a more durable and problematic adverse effect of SSRIs and SNRIs but can be managed with adjunctive treatments. There is the possibility of patients developing antidepressant-induced jitteriness or anxiety, potentially due to initial surge of serotonin, although this anxiety can be mitigated by slower titration or adjunctive use of benzodiazepines (
19).
The tricyclic antidepressants (TCAs), which act as reuptake inhibitors of serotonin and norepinephrine transporters, were one of the first classes of medications used for anxiety disorders (
20). Despite comparable efficacy to SSRIs, they are now less frequently prescribed due to concerns about side effects including weight gain, dry mouth, sedation, urinary hesitancy or retention, arrhythmias, and risk of mortality with overdose (
20). Clomipramine and imipramine (both TCAs) are FDA-approved for PD. Monoamine oxidase inhibitors (MAOIs) are also older antidepressant medications which are now typically used only as a third-line option because of side effects and dietary restrictions. They are not FDA-approved for anxiety disorders but may be considered in patients with SAD who are non-responsive to SSRIs (
21).
Buspirone, a 5-HT
1A partial agonist classified under the azapirones, is FDA-approved for use in anxiety, and is commonly used as an adjunctive treatment with SSRIs or SNRIs primarily for GAD (
22). It is the only azapirone currently approved in the United States. A Cochrane review of buspirone for GAD found that it was superior to placebo but had a smaller effect size in GAD compared to benzodiazepines and antidepressants (
22). Moreover, it was not as well-tolerated (nausea and dizziness) and less effective in those with past benzodiazepine use (
22). A subsequent Cochrane review compared buspirone to placebo for PD and found buspirone to be less efficacious than placebo but the review was limited by the dearth of high-quality studies (
23). Buspirone is generally dosed two to three times a day and has a gradual onset of action of around 10 days to 4 weeks. Adverse effects include nausea, dizziness, and headache, and there are reports of buspirone-induced movement disorders (
24). There is also anecdotal reporting for using buspirone to offset sexual side effects from SSRIs but there are few studies offering empirical support of this practice (
25,
26).
Mixed antidepressants.
Mirtazapine has a broad pharmacological effect, with presynaptic antagonism of the alpha-2 adrenergic receptor, postsynaptic blockade of 5-HT
2 and 5-HT
3 receptors, and antagonism of histamine-1 (H
1) receptors (
27). Mirtazapine is FDA-approved for the treatment of MDD in adults. Its benefits include positive effects on sleep and appetite and its general safety for elderly patients, fewer drug-drug interactions, and less likelihood of sexual side effects compared to SSRIs and SNRIs. Adverse effects include weight gain and other antihistamine effects like sedation and dry mouth. There are very few clinical trials assessing mirtazapine for anxiety disorders. In PD, one small randomized controlled trial (RCT) reported that mirtazapine was comparable in efficacy to escitalopram (
28). In SAD, one RCT of women showed a significant improvement in anxiety symptoms compared to placebo (
29), while a subsequent study failed to show separation from placebo (
30). There are no controlled studies of mirtazapine in GAD to date. Overall, in the absence of further trials, the evidence has suggested that mirtazapine may have efficacy in improving anxiety but primarily as an adjunctive agent.
Bupropion is a dopamine norepinephrine reuptake inhibitor approved for the treatment of MDD, attentiondeficit/hyperactivity disorder (ADHD), and smoking cessation (
31). Although bupropion has been used in patients with anxiety who are being treated with SSRIs as an adjunct to offset sexual side effects, there has been limited investigation of this medication as a monotherapy for anxiety. Although there is a common perception that bupropion can worsen anxiety, this may not be entirely accurate based on previous research of bupropion on anxiety symptoms in MDD when compared to SSRIs (
31–
33). To date, there is only one controlled trial of bupropion in anxiety disorders, a RCT comparing bupropion XL to escitalopram in GAD, which found that the two drugs had comparable anxiolytic efficacy (
34). The evidence regarding efficacy of bupropion for the treatment of PD is conflicting (
35,
36). Further work is needed to determine if bupropion and similar dopamine-enhancing agents are efficacious for the treatment of anxiety disorders.
Nefazodone, a serotoninergic modulating antidepressant thought to inhibit 5-HT reuptake and block postsynaptic 5-HT
2 receptor (
37), is only FDA-approved for the treatment of MDD. There have been several open-label studies suggesting potential benefit in PD and GAD but no controlled studies have been conducted (
37–
40). One RCT of nefazodone in SAD did not report separation from placebo (
41). Overall, its use has been limited by concerns related to very rare but severe cases of liver toxicity.
Gamma aminobutyric acid (GABA).
Benzodiazepines have been a longstanding treatment for anxiety and are still among the most widely prescribed class of psychiatric medications in the world (
42,
43) although there has been increasing stigma surrounding the use of benzodiazepines in clinical practice (
44). Critics of benzodiazepines cite their being prescribed as first-line treatments for anxiety in primary care settings before SSRIs, potential risks of tolerance, dependence, abuse or misuse, and concerns about falls in the elderly (
45). However, there is a lack of strong evidence that SSRIs and other first-line treatments are superior to, or better-tolerated than, benzodiazepines for anxiety disorders, in particular GAD (
46), especially for short-term treatment (
44,
47), and possibly beyond 8 weeks as well (
48). Benzodiazepines, which act as GABA-A agonists, are highly versatile medications that can be prescribed for a wide range of conditions including alcohol withdrawal, agitation or aggression, anesthesia, catatonia, mania, insomnia, muscle spasms, epilepsy or seizures, and REM sleep behavior and movement disorders (
44). Although some reports suggest a potential risk of dementia associated with the chronic use of benzodiazepines, these have been called into question and it appears there is not an increased risk of neurocognitive disorders (
49). Benzodiazepines are no longer considered firstline monotherapy for PD or other anxiety disorders but can be used in the short-term on either a standing or as-needed basis for PD, GAD, and SAD in conjunction with SSRIs and SNRIs (
Table 1 ) (
14–
16). Caution is needed in children, geriatric patients, those with medical comorbidities, and individuals with substance use disorders, especially those using other central nervous system depressants like opioids and/or alcohol. Additionally, chronic use of benzodiazepines to treat anxiety with comorbid depression may result in reduced efficacy of antidepressants (
50).
Anticonvulsants, some of which have GABAergic properties, include medications like pregabalin, gabapentin, tiagabine, lamotrigine, and topiramate. There is scant research on the use of this class of medications for anxiety disorders (
51), with the strongest evidence for the use of pregabalin in GAD, including a meta-analysis of multiple RCTs reporting superiority to placebo and comparable effects to benzodiazepines (
52). Pregabalin is thought to have anti-epileptic effects by its activity on the alpha-2 delta subunit of calcium channels to reduce neurotransmitter release (
51). Pregabalin has FDA approvals for neuropathic pain, post-herpetic neuralgia, fibromyalgia, and as an adjunctive treatment for partial seizures. It was approved for GAD by the European Union in 2006, although since it was not approved by the FDA in the United States in 2009 the FDA application was withdrawn in 2010. Pregabalin was also shown to have potential efficacy in SAD (
53), but only at doses of 450 or 600 mg, based on three randomized, double-blind, placebo-controlled trials (
54–
56). Pregabalin is generally well-tolerated, with the most common adverse effects being sedation, dizziness, and weight gain. Since there is the potential for abuse and dependence with pregabalin, it is listed as a Schedule V medication by the United States Drug Enforcement Administration (DEA), and providers need to be mindful of tapering the medication to prevent withdrawal, and to monitor prescribing in patients with substance use problems, especially opioids, for which there is an increased risk of overdose death (
57–
59).
Gabapentin, much like pregabalin, acts to modulate neurotransmitter release on voltage-dependent calcium channels (
51). It is FDA-approved for the treatment of neuropathic pain, post-herpetic neuralgia, and partial seizures, but has been widely used off-label for various indications including fibromyalgia. There is increasing evidence for its use in alcohol use disorder, primarily for the treatment of withdrawal (
60). Gabapentin has also been prescribed off-label for anxiety despite a lack of research evidence supporting such use (
61). It was found to be efficacious in a small (
N = 69) randomized, double-blind, placebo-controlled study in SAD (
62). Another RCT of patients with PD found a difference between gabapentin and placebo but only in patients with severe panic symptoms (
62). A third study found that gabapentin may help with anxiety related to public speaking (
63). There are also several trials of gabapentin showing efficacy in perioperative anxiety (
61). Gabapentin has similar adverse effects as pregabalin including sedation, dry mouth, constipation, weight gain, and pedal edema. Although gabapentin is not listed as a DEA controlled substance in the US, it may vary on how it is scheduled between states, and there is, like pregabalin, a risk of withdrawal and abuse potential, meaning caution must be used when prescribing this medication to patients taking opioids or those with substance use disorders (
57,
59,
64). There are currently no known ongoing trials of gabapentin for anxiety disorders.
Another anticonvulsant, tiagabine, is FDA-approved for the treatment of partial seizures and has been shown to have potential anxiolytic effects in preclinical studies (
65). Its mechanism of action is unknown, although it is thought to increase GABA activity by inhibiting GABA uptake in presynaptic neurons (
51). While there have been promising open-label studies for PD, GAD, and PTSD (
51), several RCTs do not support the efficacy of tiagabine in GAD (
66) or PD (
67). A small, randomized, doubleblind crossover study with gabapentin and tiagabine in SAD reported that both drugs may be effective in reducing anxiety scores (
68). There are no known active studies of tiagabine in anxiety disorders.
Lamotrigine is an anticonvulsant thought to inhibit voltage-dependent sodium channels causing decreased glutamate release (
51). It is FDA-approved for the maintenance treatment of bipolar I disorder and for several types of seizure disorders. Lamotrigine was shown to have anxiolytic properties in preclinical studies (
69). There are, however, very few studies of lamotrigine in anxiety disorders. One small case series reported improvement in symptoms in PD with agoraphobia (
70). To date, no studies of lamotrigine for anxiety disorders are underway. Other anticonvulsants, like topiramate (thought to inhibit voltage-dependent calcium sodium channels to enhance GABA, block glutamate and inhibit carbonic anhydrase) have very limited data (
51), mostly in open-label trials for SAD (
71), although there is a trial listed on the Clinical Trials database for topiramate augmentation in treatment-refractory SAD (NCT00182455). There is only one known RCT of valproate, also known as valproic acid (which blocks voltage-dependent sodium channels and is thought to increase GABA by inhibiting glutamate-mediated excitation), in which is primarily used to treat bipolar disorder (
51). In terms of GAD (
72), valproic acid showed separation from placebo, and an open-label trial in SAD suggesting potential efficacy (
73). Levetiracetam, approved only for seizure disorders (and with unclear mechanism, but thought to reduce hyperexcitation of brain cells through binding to the synaptic vesicle protein SV2A, but to not interfere with normal electrical activity) (
51), has undergone two RCTs for SAD in which the medication failed to show a statistically significant difference compared to placebo (
74,
75). There are no known RCTs of carbamazepine or oxcarbazepine in anxiety disorders.
Antipsychotics.
There is currently only one antipsychotic, trifluoperazine, a first-generation antipsychotic (FGA), which is FDA-approved for the treatment of anxiety. In spite of this, antipsychotics, most of which are dopamine-2 (D
2) receptor antagonists, have been utilized on an off-label basis for multiple indications other than psychosis including anxiety (
84). Several systematic reviews of antipsychotics in anxiety have reported that the majority of studies were of quetiapine, a second-generation antipsychotic (also with antagonism of the 5-HT
2 and H
1 receptors), in GAD, and showed the potential utility of quetiapine monotherapy in GAD despite poor tolerability (
85,
86). The Canadian guidelines for anxiety and related disorders recommend olanzapine (D
2, 5-HT
2 and H
1 antagonist), aripiprazole (a partial D
2 and 5-HT
1A agonist and 5HT
2A antagonist) and risperidone (D
2, and 5-HT
2 antagonist), as augmentation strategies for GAD and PD (
87). Risperidone and aripiprazole are also recommended as adjunctive drugs in the treatment of SAD (
87). There is reasonable concern about the short- and long-term risks of using antipsychotics in anxiety disorders. First, there are limited studies to date in other anxiety disorders such as SAD and PD (
85,
88). Second, it is unclear whether patients receive appropriate psychoeducation about the risks of tardive dyskinesia, extrapyramidal symptoms, neuroleptic malignant syndrome, weight gain, and metabolic syndrome. Further large-scale research and longitudinal studies of antipsychotics in anxiety are needed before these medications can be recommended.