Triazolam.

The hazards of triazolam are the class hazards, sharpened by speed. Rebound is the best documented one. In 18 insomniacs given nightly or intermittent triazolam across 12 nights, total wake time rose 61 percent and 51 percent above baseline on the first night of each withdrawal period, an effect that appeared after only brief and intermittent use ^[4]. Within a treatment night, rapid elimination produces early morning insomnia: wake time fell 46.6 percent in the first six hours of the night but only 27.7 percent in the last two, a nonsignificant difference, and later in treatment wake time in the final two hours rose above baseline ^[3]. Rebound after stopping can be blunted but not abolished by tapering; in 60 insomniacs, abrupt substitution of placebo after nightly dosing lengthened sleep latency by 57 minutes and cut sleep duration by 1.4 hours relative to baseline, while a stepped taper reduced or eliminated those symptoms ^[5]. Amnesia is not a fringe event: delayed recall was consistently impaired in triazolam studies where temazepam showed no significant effect ^[6], and impairment was still measurable at 24 hours ^[7]. Postmarketing pharmacovigilance is where the drug earned its reputation. In the FDA Spontaneous Reporting System through the mid 1980s, reports attributed to triazolam versus temazepam ran 133 versus 2 for confusion, 109 versus 3 for amnesia, 59 versus 2 for bizarre behaviour, 58 versus 4 for agitation and 40 versus 1 for hallucinations; reporting rate comparisons are subject to notoriety and channelling bias, but the size of the gap drove regulatory action worldwide ^[8]. Abuse liability is real and near the top of the class: triazolam maintained the highest rates of intravenous self injection among the benzodiazepines tested in baboons ^[12], and in human sedative abusers drug liking correlated with self administration measures ^[11]. Interactions are the sharpest practical danger. Systemic ketoconazole or itraconazole plus oral triazolam is described in the primary literature as potentially hazardous ^[9]; grapefruit juice, macrolides and protease inhibitors act on the same CYP3A4 pathway. Alcohol adds directly to the impairment: at high doses the two produce comparable psychomotor impairment, and triazolam contributed memory impairment that ethanol did not ^[10]. Combination with opioids or other central nervous system depressants raises respiratory depression and overdose risk. Older adults, people with sleep apnoea, and anyone with a history of sedative or alcohol use disorder are at elevated risk. Discontinuation after regular use should be supervised and gradual, never abrupt.

Triazolam was developed by Upjohn and first marketed in the Netherlands in 1977, reaching the United States as Halcion in 1982. It became one of the most prescribed hypnotics in the world within a few years, then the most controversial. Reports of confusion, amnesia and psychiatric reactions accumulated in spontaneous reporting systems and in the popular press, and the FDA analysis of its own Spontaneous Reporting System found large excesses of behavioural adverse event reports for triazolam compared with temazepam ^[8]. The United Kingdom Committee on Safety of Medicines suspended the product licence in 1991, and several other countries restricted or withdrew it; the United States kept it on the market with lowered recommended doses, tightened labelling and a short duration of use warning. The sleep laboratory literature that shaped the debate came largely from the Kales group, which described early morning insomnia during treatment ^[3] and rebound insomnia after brief and intermittent use ^[4], while the Greenblatt group argued that gradual withdrawal attenuates the rebound ^[5]. Prescribing has fallen steeply since, and triazolam now survives mainly as a niche hypnotic and as a dental and procedural oral sedative.

Notorious. Halcion is the benzodiazepine most associated in public memory with amnesia and strange nocturnal behaviour, the subject of a genuine international regulatory fight in the early 1990s and of lasting argument over whether the signal reflected the drug, its dose, or reporting bias. Among researchers it holds a different reputation: a clean, fast, well characterized pharmacokinetic probe, which is why it became the standard CYP3A4 substrate in drug interaction studies and the reference hypnotic in abuse liability work. In dentistry it is still valued as a reliable oral premedication. Among people who use benzodiazepines recreationally it is regarded as short, strong and blackout prone.