Does ketamine maintenance therapy have enough long-term follow-up evidence?

What does the evidence show about long-term benefits of ketamine maintenance?

The strongest evidence for long-term benefit comes from naturalistic follow-up studies. In one study of 108 patients with unipolar and bipolar depression who received six ketamine infusions over 12 days, 65.7% completed a 9-month follow-up, and among those, the response and remission rates were 80.3% and 78.9%, respectively [6]. This means that most patients who stayed in the study maintained significant improvement in depressive symptoms and global functioning (scores over 70 on the Global Assessment of Functioning scale) [6]. However, 46.4% of initial responders sustained their response throughout the 9 months, while 85.7% of relapses occurred within the first two weeks after treatment, suggesting that early relapse is a key risk [6].

For PTSD, a randomized controlled trial of six ketamine infusions over two weeks found that 67% of participants were treatment responders at week 2, compared to 20% in the placebo group, and the median time to loss of response among ketamine responders was 27.5 days [3]. This shows that while ketamine can produce rapid and robust symptom reduction, the effect may not persist without ongoing maintenance. A pilot study combining a single ketamine infusion with exposure therapy found that ketamine reduced amygdala reactivity to trauma memories for at least 30 days, suggesting a potential mechanism for sustained benefit when paired with psychotherapy [5].

In a real-world clinic study of 38 patients with treatment-resistant depression receiving maintenance ketamine or esketamine, the median duration of the longest maintenance cycle was 61 weeks for intravenous ketamine and 48 weeks for intranasal esketamine, with treatment intervals averaging 18.9 days and 10.8 days, respectively [1]. This indicates that maintenance therapy can be sustained for over a year in some patients, though the frequency of dosing varies by route of administration.

What are the caveats and gaps in the long-term evidence?

The most important caveat is that most long-term studies are open-label (not blinded) and lack a control group, which means the observed benefits could be partly due to placebo effects or natural recovery. For example, the 9-month follow-up study [6] and the real-world clinic study [1] are both observational, so they cannot prove that ketamine caused the sustained improvement. The only long-term follow-up of a randomized trial in this set—a 5-year follow-up of a blood pressure intervention—found that initial benefits reversed into harm over time, a pattern that could theoretically apply to any treatment without adequate long-term data [2].

Another gap is high dropout rates and loss to follow-up. In the NIMH long-term follow-up study of 203 participants (average 9 years after discharge), only 20% of the original 1,000 participants completed assessments, and those who had received ketamine were more likely to seek further ketamine treatment afterward, but no signs of abuse were reported [4]. This low follow-up rate limits the reliability of conclusions about long-term safety and efficacy. Additionally, the study found that participants who used ketamine post-discharge reported more depressive symptoms, suggesting that ongoing treatment may be needed for those with more severe illness [4].

Finally, the evidence is almost entirely from depression and PTSD populations, with very little long-term data on other conditions like chronic pain or bipolar disorder. The studies here also vary widely in dosing protocols (e.g., 0.5 mg/kg intravenous vs. intranasal esketamine), follow-up durations (30 days to 9 months), and outcome measures, making it hard to compare results directly. As one commentary notes, long-term follow-up is often underfunded and methodologically challenging, leading to a systematic evidence gap that can let patients down [2].

About These Sources

This answer is built on 6 peer-reviewed studies — published from 2021 to 2025, 2 from 2024 or later, 6 in Q1 journals, collectively cited 314 times — selected as the most relevant from 10 studies that passed quality screening, drawn from 47 papers retrieved from a database of over 500 million.

Sources used in this answer

Related research questions