Every psychiatrist has seen it: a patient stable on olanzapine for months develops new sedation or toxicity after an antibiotic, or lithium levels jump after an anti-inflammatory. The mechanism behind these surprises is cytochrome P450 metabolism — and understanding what CYP450 inhibition in psychiatric drug interactions actually means is the difference between predicting an interaction and discovering it in an adverse-event report.
This article explains inhibition versus induction in plain terms, maps the four enzymes that matter most in psychiatry with real drug examples, and shows you the direction trick that exam questions quietly test.
Inducer versus inhibitor: the one-direction rule
Cytochrome P450 enzymes are hepatic (and intestinal) oxidases that metabolise most psychotropic drugs. Two things can alter their activity:
- Inhibitor — blocks the enzyme (competitively or irreversibly). Metabolism slows. Substrate levels rise. Onset: hours to days (immediate for reversible, longer for irreversible). Examples: fluvoxamine, fluoxetine, paroxetine, bupropion, grapefruit juice.
- Inducer — upregulates enzyme synthesis. Metabolism speeds up. Substrate levels fall. Onset: days to weeks (needs protein turnover). Examples: carbamazepine, phenytoin, rifampicin, St John's wort, chronic alcohol use.
Memory hook: an inhibitor is a traffic jam (drug accumulates); an inducer clears the road (drug vanishes). Interaction effects persist until the inhibitor washes out — with irreversible inhibitors like fluoxetine, functional enzyme regeneration takes weeks.
The four enzymes you must know
| Enzyme | Key psychiatric substrates | Notable inhibitors | Notable inducers |
|---|---|---|---|
| CYP3A4 | Most atypicals (quetiapine, aripiprazole, midazolam), cyclosporine | Grapefruit juice, ritonavir, ketoconazole | Carbamazepine, rifampicin, St John's wort |
| CYP2D6 | Risperidone, many antidepressants, codeine → morphine | Fluoxetine, paroxetine, bupropion | Little clinically relevant induction |
| CYP1A2 | Olanzapine, clozapine, theophylline | Fluvoxamine, ciprofloxacin, smoking cessation | Heavy smoking, grilled food, carbamazepine |
| CYP2C19 | Some SSRIs, diazepam, PPIs | Fluvoxamine, fluoxetine | Rifampicin |
Two genetic twists worth knowing: CYP2D6 poor metabolisers (~5–10% of Europeans) accumulate risperidone-like drugs and convert codeine poorly; ultra-rapid metabolisers get dangerously high morphine from normal codeine doses. Psychiatric prescribing increasingly treats enzyme phenotype as part of the prescription.
Clinical examples that actually happen
- Fluvoxamine + olanzapine/clozapine — potent CYP1A2 inhibition can double or triple levels; doses of clozapine often need reduction when fluvoxamine is added.
- Smoker stops → clozapine toxicity — polycyclic hydrocarbons in smoke induce CYP1A2; when a patient quits, clozapine levels rise substantially and dose must be cut (often 20–30%) within a week or two.
- Carbamazepine + almost anything — broad induction drops antipsychotic and oral contraceptive levels.
- St John's wort — a genuine inducer that has triggered psychotic relapses by crashing antipsychotic levels.
- Fluoxetine + tamoxifen — CYP2D6 inhibition blocks conversion to endoxifen, the active metabolite — the reason oncologists avoid paroxetine/fluoxetine.
Applying it: a four-step interaction check
- Name the substrate — which enzyme handles the drug you're starting/stopping?
- Name the partner — inhibitor or inducer of that enzyme? Check a reputable interaction checker (BNF, Lexicomp, or the SmPC) rather than memory for anything unusual.
- Predict direction — inhibitor ⇒ levels up (watch toxicity); inducer ⇒ levels down (watch treatment failure).
- Act with timing — measure levels after ~4–5 half-lives (sooner for lithium/digoxin-type monitoring), adjust, and document the plan.
Some pairs don't wait for enzymes: serotonergic combinations (SSRI + MAOI + lithium/tramadol) carry direct receptor-level toxicity — that's serotonin syndrome pharmacology, a separate mechanism from CYP metabolism.
Exam trap
The direction puzzle: "Adding fluvoxamine to clozapine. What happens to clozapine levels?" — answer: rises (CYP1A2 inhibition, toxicity risk). Then the mirror: "Patient quits smoking while on clozapine" — levels rise too (induction is removed). Students invert one of these routinely. Second trap: paroxetine/fluoxetine are mechanism-based CYP2D6 inhibitors — ask what the question is really enzyme-ing at.
Pharmacogenomics and the practical toolkit
Enzyme phenotype is now a prescribing variable. Commercial CYP2D6 and CYP2C19 genotyping is widely available, and the clinical vignettes increasingly reference it: a CYP2D6 poor metaboliser given codeine gets no analgesia (no morphine conversion) while accumulating risperidone to prolactin-toxic levels; an ultra-rapid metaboliser gets dangerous morphine from an ordinary codeine dose. CPIC (Clinical Pharmacogenetics Implementation Consortium) publishes dosing guidance tiers — "use alternative," "reduce dose," "normal dose" — for the antidepressant-enzyme pairs most commonly tested.
Enzyme-drug relationships worth committing to memory beyond the table:
- Irreversible inhibition is why washouts matter — fluoxetine's CYP2D6/2C19 inhibition outlives the drug itself, weeks of functional enzyme being resynthesised.
- Time-dependent interactions — an inducer's effect builds over days to weeks (protein turnover sets the pace) and decays the same way; an inhibitor's hits within days. Stopping a concomitant prescription requires the same directional thinking as starting one: the levels move opposite to intuition.
- Proton pump inhibitors raise pH-dependent absorption drugs, and gastrointestinal motility changes (from anticholinergics or metoclopramide) shift absorption windows — interactions don't all run through the liver.
- Herbals are not inert — St John's wort is a potent CYP3A4 inducer with documented relapse cases behind it; ask about supplements in every interaction review.
For day-to-day practice, one reliable workflow: check the BNF interaction section or a clinical decision-support tool (Lexicomp, Epocrates) for the specific pair, predict the direction before looking, then confirm with trough levels after roughly five half-lives when a sensitive drug (lithium, clozapine, anticonvulsants, warfarin) is involved. Write the prediction and the plan in the notes — it converts a reactive check into teachable prescribing reasoning, which is exactly what exam questions and clinical reviews reward.
One final habit separates fluent prescribers: when you start or stop any interacting drug, write down the predicted direction of change and the review date. "Adding fluvoxamine → clozapine levels expected to rise → check trough at day 10" converts pharmacology from trivia into monitored care — and it is precisely the sentence an examiner, a supervisor, or a reviewer wants to read.
And remember the non-enzymatic interactions: additive anticholinergic burden, QT-prolongation stacking and serotonergic convergence each produce their own toxicity syndromes regardless of CYP status.
Frequently asked questions
What does CYP450 inhibition do to drug levels?
It raises them. An inhibitor slows the enzyme's metabolism of the substrate drug, so the drug accumulates toward toxicity — for example fluvoxamine inhibiting CYP1A2 metabolism of clozapine. Inducers do the opposite: they accelerate metabolism and lower levels toward therapeutic failure.
Which cytochrome enzymes matter most in psychiatry?
CYP3A4 (quetiapine, aripiprazole), CYP2D6 (risperidone, many SSRIs, codeine activation), CYP1A2 (olanzapine, clozapine — strongly affected by smoking) and CYP2C19 (several SSRIs and diazepam). Learning these four covers the majority of clinically important psychotropic interactions.
Why does smoking affect psychiatric medication levels?
Tobacco smoke induces CYP1A2, so smokers metabolise clozapine and olanzapine faster. When a patient stops smoking, induction reverses over days to weeks, levels rise, and clozapine doses commonly need reduction of 20-30% to avoid toxicity.
How long does an enzyme inhibitor's effect last after stopping?
Depends on mechanism: reversible inhibitors act within hours and wash out over days, while irreversible (mechanism-based) inhibitors like fluoxetine require new enzyme synthesis — functionally weeks, matching the 4-6 week period its metabolite norfluoxetine persists.
Key takeaways
- Inhibitor = levels rise; inducer = levels rise or fall accordingly — inducers always lower levels.
- Know CYP3A4, 2D6, 1A2, 2C19 substrates-inhibitors-inducers cold — that set covers most psychiatry interactions.
- CYP1A2 + smoking: cessation raises clozapine/olanzapine levels; initiation lowers them.
- Fluvoxamine is the classic clozapine interaction; fluoxetine/paroxetine hijack CYP2D6 (risperidone, tamoxifen).
- Check direction before acting: predict first, confirm with levels after ~5 half-lives.
Enzyme tables and interaction drills live in Preppsy's MCQ mock exams — because CYP questions reliably appear across MRCPsych Paper A and USMLE Step 1/2 pharmacology.
Cover image: “CYP450 Interactions (cover)” by Preppsy Editorial Team (Original artwork) via Preppsy.
