Few concepts have proved as persistently tantalising, or as stubbornly unrealised, as Rick Strassman's proposal for an extended-state DMT infusion protocol: a continuous intravenous drip designed to hold a human subject in a sustained DMT experience for hours rather than the compound's customary handful of minutes.

Origins: The Rapid Off-Switch as Both Feature and Limitation

Strassman's early clinical work at the University of New Mexico in the 1990s established what remains the canonical pharmacokinetic portrait of intravenous DMT in humans. A bolus injection of 0.4 mg/kg produced intense subjective effects within seconds; those effects peaked at roughly two minutes and were largely resolved by twenty to thirty. The brevity was, in one sense, a clinical virtue — it made DMT the most controllable of the classic psychedelics for research purposes, with none of the eight-to-twelve-hour commitment demanded by LSD or psilocybin. But it also meant that subjects were, in Strassman's own framing, launched and returned before they could do much more than register astonishment. The experience was too short for sustained introspection, therapeutic dialogue, or careful phenomenological reporting in real time.

The extended-state concept followed logically. DMT is rapidly metabolised by monoamine oxidase, principally MAO-A. If one replaced the single bolus with a carefully titrated continuous infusion — a loading dose followed by a maintenance rate adjusted to keep plasma DMT concentrations within a target window — one could, in principle, sustain the fully psychedelic state for as long as the infusion ran. Strassman discussed the idea publicly and in his 2001 book DMT: The Spirit Molecule, and outlined preliminary pharmacokinetic modelling suggesting it was feasible. The mathematics were not especially exotic: standard one- or two-compartment models, the sort of thing anaesthetists do routinely with propofol.

Why It Was Never Pursued — At Least Not by Strassman

The reasons the protocol did not proceed during Strassman's active research period are multiple and not entirely pharmacological. By his own account, the New Mexico work was terminated in 1995 owing to a confluence of factors: institutional friction, the absence of a therapeutic framework he felt comfortable applying, and his own growing unease about the ontological claims some subjects were making about their DMT encounters — particularly those involving contact with apparently autonomous entities. An extended-state protocol, which would presumably deepen and elaborate precisely those sorts of experiences, sat uneasily with those concerns.

There were also practical hurdles of a regulatory and ethical character. Holding a human being in a full psychedelic state for ninety minutes or several hours raises consent questions that a fifteen-minute experience does not. The capacity for real-time withdrawal of consent becomes philosophically murky when the subject is, by any conventional measure, in a profoundly altered state of consciousness. Monitoring requirements would be substantially more demanding, and the psychiatric risk profile — particularly around prolonged anxiety or psychotic-spectrum reactions — would need careful characterisation in shorter exposures first.

The Idea's Afterlife

The extended-state concept did not die with Strassman's withdrawal from active research. It has resurfaced repeatedly. Andrew Gallimore, a computational neurobiologist, collaborated with Strassman on a 2016 paper presenting a formal pharmacokinetic model for target-controlled DMT infusion, drawing explicit parallels with target-controlled infusion systems used in anaesthesiology. The Gallimore–Strassman model proposed that a target plasma concentration could be selected, a loading bolus calculated, and an exponentially decreasing infusion rate applied to maintain steady state. The paper was a theoretical exercise — no human data from an actual extended infusion were presented — but it lent the concept mathematical respectability.

More recently, the notion appears to have found a partial practical foothold. Several research groups investigating DMT's therapeutic potential — notably those exploring its application in depression — have employed intravenous infusion protocols rather than single bolus doses, though typically at sub-breakthrough doses or for relatively modest durations. These are not quite the full Gallimore–Strassman vision of a sustained, titrated, deep psychedelic immersion, but they represent steps along the same pharmacokinetic reasoning. The DMTx project, an initiative that has explicitly taken up the extended-state banner, has discussed protocols aimed at maintaining the state for longer periods, though published human data from such efforts remain sparse at the time of writing.

What the Field Still Wrestles With

Several fundamental questions remain open, and they are worth enumerating plainly.

First, tolerance. Strassman's own bolus studies noted that four closely spaced doses of DMT at thirty-minute intervals did not produce marked tachyphylaxis — unlike the rapid tolerance seen with LSD. This was a crucial observation, since the entire extended-state concept collapses if receptor desensitisation renders the infusion ineffective within minutes. However, the longest interval Strassman tested was two hours of repeated boluses, which is not the same as a genuinely continuous infusion. Whether 5-HT2A receptor internalisation or downstream signalling attenuation would erode the experience over longer periods remains an empirical question without a confident answer.

Second, phenomenological stability. Brief DMT experiences are characterised by an extraordinarily rapid sequence of perceptual and cognitive shifts. Whether extending the state would produce a stable plateau — a kind of sustained visionary environment one could explore methodically — or merely prolong the kaleidoscopic flux is unknown. Subjective reports from ayahuasca, which provides a pharmacologically distinct but conceptually adjacent model of extended DMT activity (via MAO inhibition rather than continuous infusion), suggest that something like a navigable state does emerge, but the pharmacokinetic profiles are sufficiently different that direct extrapolation is risky.

Third, purpose. The original Gallimore–Strassman framing leant heavily on the idea that extended immersion would permit more careful phenomenological investigation — particularly of the entity-encounter phenomenon. Whether this constitutes a scientifically productive research aim, a therapeutic one, or something else entirely depends rather heavily on one's prior commitments. The more recent clinical interest in DMT infusion is motivated by therapeutic pragmatism — can we deliver a psychedelic-assisted therapy session in an hour rather than a day? — which is a decidedly different question from the one Strassman was asking.

Marginalia

There is something faintly poignant about the extended-state protocol's career: pharmacokinetically straightforward, clinically demanding, ethically complex, and philosophically loaded, all at once. It has spent three decades being almost-but-not-quite attempted. One suspects that when it is finally executed with proper rigour, the results will confound enthusiasts and sceptics alike in ways neither camp has adequately imagined.