What If Ketamine Doesn't Work? Honest Look at Next Steps

Any responsible conversation about ketamine has to include what happens when it doesn't work. The enthusiasm around ketamine's rapid effects is earned, but "meaningful response for 60-70% of treatment-resistant depression patients" also means 30-40% don't get the improvement they hoped for. Patients deserve to know that before they start, not after.

I want to talk about what non-response actually looks like (it's often not what patients assume), what adjustments can still be tried before concluding ketamine has failed, and what evidence-based options exist when it truly hasn't worked.

The response rate in context

A 60-70% response rate in treatment-resistant depression is strong, not weak. For comparison: a third or fourth SSRI trial after two have failed has response rates in the 10-30% range. TMS lands around 50-60% for TRD. Even ECT, the most effective single intervention for severe depression, is 50-70%. Ketamine's numbers hold up against those benchmarks.

That said, a 30-40% non-response rate isn't trivial, and having a plan for that possibility is part of careful clinical work, not pessimism.

What "not working" actually looks like

Before concluding that ketamine has failed, the question to answer is: what kind of non-response is this? They aren't all the same.

Complete non-response. No meaningful change in symptoms after an adequate trial (four to six sessions at appropriate dose). Depression feels the same in intensity, frequency, and impact on daily life. This is the clearest form of treatment failure and the shortest path to concluding ketamine isn't the right tool for you.

Partial response. The patient notices some improvement (better sleep, slightly more energy, brief periods of lighter mood) but the overall depression burden is still substantial. This is more common than complete non-response, and it isn't failure. It's information that the treatment is doing something and may need optimization before you know its ceiling.

Initial response then relapse. Substantial improvement in the first few sessions that doesn't hold. This pattern suggests the treatment works but the maintenance schedule isn't keeping pace.

Unmet expectations. Sometimes what looks like non-response is actually a patient who expected complete elimination of depression after two sessions. Measurable improvement that would continue to build with additional sessions gets mislabeled as "didn't work." Standardized assessment tools (PHQ-9, GAD-7) help distinguish perception from measurable change, which is part of why we use them.

Adjustments worth trying before you give up

If you're not responding as hoped, there are several things to try before concluding ketamine has failed.

Dose optimization is usually first. Starting doses are conservative by design because ketamine's side effects are dose-dependent. If you've tolerated your current dose without trouble but aren't seeing adequate response, a gradual dose increase is the next step. Some patients need to land higher than the starting dose to access therapeutic effect.

Session frequency is worth revisiting. Twice-weekly during the loading phase is standard, but some patients respond better to a compressed initial course of three sessions per week for a short window. More sessions in less time isn't for everyone, but it sometimes unlocks response that slower cadence didn't.

How you're dosing matters more than patients realize. Sublingual absorption varies based on technique. Are you holding the medication under your tongue for the full recommended time? Swallowing saliva cuts bioavailability roughly in half. Some of the "non-responders" I see are actually underdosed because of how they're taking the medication, not because ketamine isn't working for them.

Environmental factors affect outcomes more than the literature formally tracks. Patients who treat sessions casually (distracted by their phone, in a cluttered room, not blocking the day) do worse on average than patients who prepare the space, put the phone away, and give the session the seriousness it deserves. If sessions have been chaotic, the fix may be procedural rather than pharmacological.

Interfering medications are worth reviewing. Benzodiazepines (Xanax, Klonopin, Ativan) can significantly blunt ketamine's therapeutic effect through their action on the same GABAergic systems ketamine partly bypasses. Daily or frequent benzodiazepine use is one of the most common reasons patients underrespond. Ongoing heavy alcohol use has a similar effect. Neither is a hard contraindication to ketamine treatment, but they can suppress response in ways worth addressing first.

What predicts non-response

Some factors are associated with lower response rates in the literature:

Longer continuous depressive episodes (many years without meaningful remission) tend to respond less robustly. Active substance use disorders interfere with response. Daily benzodiazepine use blunts response. Personality disorder comorbidity complicates the clinical picture. Emerging research suggests high systemic inflammation may predict lower response.

None of these are absolute predictors. Some patients with every risk factor do well; some with none of them don't respond. But they're a useful framework for thinking about what to try next if ketamine doesn't work for you.

Evidence-based alternatives

If an adequate ketamine trial with appropriate adjustments hasn't produced meaningful improvement, several other options deserve serious consideration.

TMS works through an entirely different mechanism: electromagnetic stimulation of the prefrontal cortex rather than pharmacology. Patients who don't respond to ketamine sometimes respond to TMS, because the two are targeting different aspects of depression's biology. Response rates are 50-60% in TRD, FDA-cleared since 2008, covered by most insurance with prior authorization.

ECT remains the most effective single intervention for severe depression, with response rates 50-70% even in patients who've failed multiple prior treatments. Modern ECT is not the procedure its historical reputation suggests; anesthesia, refined techniques, and careful protocols have substantially reduced the memory effects that defined earlier generations of the treatment. For severe or psychotic depression that hasn't responded to anything else, ECT is the option I'd want a loved one to have access to.

Augmentation strategies (adding lithium, thyroid hormone, or an atypical antipsychotic like aripiprazole to an existing antidepressant) can sometimes produce response that single-agent treatment couldn't. Response rates are modest but not zero, especially in patients who haven't tried these before.

Intensive psychotherapy matters more for some patients than medication adjustments. Cognitive behavioral therapy, dialectical behavior therapy for emotion dysregulation, EMDR for trauma-driven depression: these have real evidence when the primary driver is psychological rather than purely neurochemical.

Lifestyle medicine is often dismissed but the evidence is actually strong. Regular vigorous exercise, sleep regularization (see our post on sleep and depression), anti-inflammatory dietary changes, and sustained stress reduction have individual effect sizes comparable to some pharmacological interventions in mild-to-moderate depression. These alone rarely fix severe TRD, but as the substrate on top of which other treatments work, they matter.

Clinical trials are worth considering if you've exhausted standard options. Novel NMDA modulators, psilocybin-assisted therapy (in trial settings), and other investigational treatments are enrolling. Your clinician can help identify trials appropriate to your situation.

The hopelessness that failed treatments create

Patients who've been through multiple treatment failures often arrive at my office having internalized the losses as evidence nothing will work. That belief is part of the depression. It's not a trustworthy read on reality.

The outcome data tell a different story. Most patients with treatment-resistant depression eventually find a treatment or combination that produces meaningful relief. The road is longer and harder than anyone would choose, but "untreatable depression" is rarer than it feels from inside a depressive episode.

Every treatment you try provides information. A medication that didn't work rules out one mechanism. A therapy that partly helped identifies what's amenable to what. Non-response to ketamine doesn't mean non-response to everything; it means one pathway has been tested and you can point the next effort somewhere with better odds.

How we think about this at Tovani Health

Non-response isn't a patient failure or a brand failure. It's clinical information that tells you where to go next. If ketamine doesn't produce the results we hoped for, I'll have a direct conversation with you about what we learned, which adjustments might help, and which alternatives are worth trying.

We don't continue treatment indefinitely without evidence of benefit. If after an adequate trial with appropriate optimization you aren't improving, I'll tell you so and help you find the most promising next option, even when that option isn't us.

Frequently Asked Questions

How do I know if ketamine isn't working for me?

Don't conclude failure too early. Roughly a third of responders don't notice change until session 3 or later. Use objective measures (PHQ-9, sleep quality, ability to engage daily activities) tracked weekly, not just subjective feel. After 4-6 sessions with no measurable improvement on validated scales, that's the meaningful signal. Common patterns mistaken for failure: partial response (improvement that's real but short of remission), plateau after early gain (need adjustment, not abandonment), and response masked by an active life stressor.

What protocol adjustments can be tried before giving up on ketamine?

Several. First-line: dose adjustment, since sublingual ketamine has a wide therapeutic range, and modest dose increases sometimes convert partial responders to full responders. Second: extending the loading phase, since some patients need 8-10 sessions rather than the typical 4-6. Third: switching delivery route, since patients who don't respond to sublingual sometimes respond to IV with its different pharmacokinetic profile. Fourth: adding structured integration therapy or psychotherapy alongside dosing, particularly helpful when partial response plateaus.

What are the evidence-based alternatives when ketamine truly doesn't work?

Several proven options. Spravato (esketamine nasal spray) is same mechanism family but different delivery and pharmacokinetics; some patients respond to one and not the other. TMS (transcranial magnetic stimulation) is non-medication, well-tolerated, response rate around 40-50% in TRD. ECT (electroconvulsive therapy) has the highest response rate (60-80%), particularly for severe or psychotic depression, with cognitive trade-offs. Pharmacological augmentation: adding lithium, an atypical antipsychotic, or thyroid hormone to existing antidepressants. Each has different risk/benefit profiles worth discussing.

Does non-response to ketamine mean my depression is untreatable?

No. It means the NMDA-receptor / glutamate-pathway approach isn't your axis, and the next mechanism worth trying is different. Depression is heterogeneous; different patients respond to different mechanisms. About 60-70% of people who failed two SSRIs respond to ketamine. Of the ~30% who don't, most respond to TMS, ECT, augmentation, or pharmacologic combinations. The number of patients with truly treatment-resistant depression who fail ALL evidence-based options is small. Non-response to ketamine is a directional signal, not a verdict.

Ready to find out if ketamine is worth trying?

If you want to find out whether ketamine is worth trying for your situation, here's the entry point. Our approach is honest, thorough, and focused on finding what actually works for you, whether that turns out to be ketamine or something else.

• Eligibility check: tovanihealth.com/eligibility (5 minutes, FL and NJ residents)
• Phone: 561-468-6981
• What you get back: an honest answer, including the next mechanism worth trying if ketamine isn't your fit.

Benjamin Soffer, DO — Tovani Health

Related reading: treatment-resistant depression, after failed antidepressants, ketamine vs. TMS, ketamine vs. Spravato.