Phenobarbital order-set PK simulator

Severity — a high CIWA, prior delirium tremens or withdrawal seizures, heavy chronic use, or a large benzodiazepine requirement — predicts the total phenobarbital a patient will need, not a single fixed number. More severe withdrawal means a higher cumulative mg/kg and/or more PRN redosing. A weight-based front-load (commonly ~10 mg/kg) covers most patients, and symptom-triggered PRN handles breakthrough. The long half-life means that load self-tapers, which is especially useful in severe or DT-prone patients. Most protocols cap cumulative dosing around 15 mg/kg IBW per 24 h to limit oversedation.

Moderate-to-limited quality — mostly observational cohorts plus a few randomized trials. The key ED RCT (Rosenson 2013) added a single 10 mg/kg IV phenobarbital load to symptom-triggered benzodiazepines and found fewer ICU admissions with no increase in adverse events. Front-loaded or monotherapy phenobarbital in the ED/ICU is associated in cohort studies with less benzodiazepine use, less mechanical ventilation, and shorter stays. For scheduled-vs-symptom-triggered strategy, the strongest RCT evidence is with benzodiazepines (symptom-triggered lowers total dose and duration — Saitz 1994); phenobarbital protocols extrapolate from it. Bottom line: a weight-based load ± PRN is well supported for moderate–severe withdrawal, and fixed oral tapers are reasonable for milder or floor/outpatient care — but no single regimen is proven superior.

Contraindications: barbiturate hypersensitivity; porphyria (acute intermittent / variegate — barbiturates can precipitate an attack); severe respiratory compromise or airway obstruction; significant hepatic failure is a strong caution.
Interactions: the acute danger is additive CNS/respiratory depression with opioids, benzodiazepines, alcohol, and other sedatives. With continued use phenobarbital is a potent hepatic enzyme inducer (CYP3A4, 2C, 1A2, UGT, P-gp), lowering levels of warfarin, hormonal contraceptives (→ contraceptive failure), many anticonvulsants, immunosuppressants (tacrolimus, cyclosporine), antiretrovirals, corticosteroids, and methadone. Valproate inhibits its metabolism and can raise phenobarbital levels. Induction matters most for chronic therapy — a short withdrawal course is lower-risk, but still counsel on contraception and anticoagulation.

Usually you pick one — stacking two GABAergic sedatives raises the risk of oversedation and respiratory depression. Reasonable times to combine: (1) a one-time phenobarbital load added on top of symptom-triggered benzodiazepines in the ED for moderate–severe withdrawal (the Rosenson approach); (2) benzodiazepine-resistant withdrawal or delirium tremens — because benzodiazepines increase the frequency of GABA-A channel opening while phenobarbital increases its duration (and at high dose opens the channel directly and blunts glutamate), phenobarbital keeps working when escalating benzodiazepines plateau; (3) breakthrough symptoms during a phenobarbital taper. When you do combine, use a monitored setting, watch the airway / respiratory rate / RASS, and be especially cautious in the non-tolerant, the elderly, or anyone on opioids. Many phenobarbital-monotherapy protocols deliberately avoid concurrent scheduled benzodiazepines for exactly this reason.

Two mechanistic advantages: it acts at a different GABA-A site (duration vs frequency) and dampens excitatory glutamate/AMPA signalling, addressing the up-regulated glutamate drive of withdrawal that benzodiazepines don't; and its ~4-day half-life gives a smooth, self-tapering level that avoids the peaks, troughs, and repeated re-dosing of benzodiazepines. Trade-offs: a narrower therapeutic margin (genuine respiratory-depression and overdose potential — no ceiling effect like benzodiazepines), enzyme-induction interactions, and it is hard to reverse (flumazenil doesn't touch it).

Steady state takes ~4–5 half-lives. At a ~90–100 h half-life that's roughly 2–3 weeks if you only give a maintenance dose — far too slow for acute withdrawal. So you give a weight-based loading dose to reach a therapeutic level in hours, then let the long half-life hold and self-taper. Scheduled maintenance doses mainly extend the plateau; the taper (halving) simply walks the level down a bit faster than natural decay. Use the sliders above to see it: the load sets the peak, the half-life sets the slope.

IV works within minutes (brain peak ~15–30 min) and is preferred for acute, severe symptoms. Oral is almost completely absorbed (F ≈ 1) but peaks over ~1–4 h, so it's fine for maintenance and tapering. IM is used in some classic protocols (e.g., the MGH load, split 40/30/30). Because oral bioavailability is near-complete, IV→PO conversion is roughly 1:1.

For a short withdrawal course, usually no — dosing is weight-based and titrated to clinical sedation (e.g., RASS) and symptom control, not to a number. Levels help mainly when the picture is unclear (unexpected oversedation, renal/hepatic impairment, suspected toxicity) or with prolonged therapy. The mcg/mL values here are a teaching estimate from a one-compartment model — not a substitute for a measured level.