Isomerism,
same formula, different molecule.

Two molecules can share an identical molecular formula and still be, pharmacologically, almost unrelated drugs. This happens because the biological targets a drug interacts with — receptors, enzymes, ion channels, transporters, plasma proteins — are themselves three-dimensional and, in the overwhelming majority of cases, chiral. Binding between a drug and its target is a close-fitting, spatially exacting interaction: connectivity (which atoms are bonded to which), spatial arrangement (how those bonds are oriented in space) and even conformation (how a flexible molecule folds at the moment of binding) can all change how well, or how differently, a drug fits.

Isomerism is the general name for this phenomenon: molecules that share an atomic formula but differ in structure. Depending on exactly how they differ, that structural difference can change a drug’s potency, efficacy, receptor selectivity, metabolism, duration of action, adverse-effect profile, or toxicity — sometimes dramatically, sometimes not at all. The rest of this lesson works through exactly which kinds of structural difference matter, and why.

There are two broad classes of isomerism. Structural (constitutional) isomerism exists when the order of atomic bonds differs — the atoms are connected to each other differently. Stereoisomerism exists when the bond order is identical but the three-dimensional arrangement in space differs.

Stereoisomerism itself has two forms. Geometric isomerism arises from restricted rotation around a double bond or within a ring, fixing substituents on the same side (cis-) or opposite sides (trans-) of that restriction. Optical stereoisomerism arises from one or more chiral centres — a carbon atom or quaternary nitrogen with four different attached groups — and splits further by how many such centres a molecule has: a single chiral centre gives a pair of enantiomers; more than one gives a family of diastereoisomers, which are not all mirror images of each other.

Interactive classification tree

Expand each branch for its defining feature and a verified anaesthetic example.

Table 1 — Master classification of isomerism

Structural isomers have identical chemical formulae, but the order of atomic bonds differs. Because the difference is in connectivity itself rather than a subtler spatial rearrangement, the comparative pharmacological effect can range from essentially identical to completely unrelated, and only inspection of the actual structures tells you which.

Isoflurane and enflurane are both volatile anaesthetic agents — structural isomers whose clinical behaviour is broadly comparable. At the other extreme, dihydrocodeine and dobutamine share an identical molecular formula purely by coincidence: one is an opioid analgesic, the other a synthetic inotrope, and their pharmacology has essentially nothing in common. Structural isomerism on its own predicts nothing about pharmacological similarity; it only tells you the atoms are the same.

Structural isomerism is itself subdivided by what kind of connectivity difference is involved: chain isomerism (the carbon skeleton itself branches differently), positional isomerism (a substituent or functional group attaches at a different point on an otherwise identical skeleton), and functional-group isomerism (the same atoms form a chemically different functional group entirely). These are general organic-chemistry distinctions; no anaesthetic drug pair in this lesson’s evidence base is used to illustrate them individually; the isoflurane/enflurane and dihydrocodeine/dobutamine pairs above are the worked anaesthetic-pharmacology examples.

Table 2 — Structural isomers vs stereoisomers

Tautomerism refers to a dynamic interchange between two structural forms of the same molecule, typically triggered by a change in the physical environment such as pH. Unlike the isoflurane/enflurane pair above, the two tautomeric forms are not separately isolable drugs — the same molecule shifts between them.

Midazolam is the clearest anaesthetic example. In an acidic solution (around pH 4, the form it is manufactured and stored in), midazolam is ionised, water- soluble, and has an open-ring structure. At physiological pH 7.4, it changes structure to form a seven-membered, unionised ring, which is lipid-soluble — favouring passage across the blood–brain barrier and increasing speed of onset in the central nervous system. This is why midazolam can be formulated as a stable, water-soluble injection and still act quickly once in the body: the drug itself interconverts to the form each environment favours.

PubChem stores one structural record for midazolam's closed-ring form; the open-ring, protonated form at acidic pH is not a separately registered compound, since it is the same molecule shifting reversibly between the two — the prose above, not a second image, is the accurate way to show that interconversion.

The other common form of tautomerism in anaesthetic practice is the keto-enol transformation seen in barbiturates, and in morphine. The drug most worth understanding it through is thiopentone, still used for induction of anaesthesia.

Thiopentone is a thiobarbiturate — a sulfur substitution at the C2 position of the barbituric-acid ring in place of oxygen. Like other barbiturates, its water solubility depends on which tautomer predominates: the keto form is poorly water-soluble, the enol form is water-soluble, and alkaline conditions favour the enol form. Thiopentone is manufactured and stored as the sodium salt in an alkaline solution around pH 10.5 (produced with sodium carbonate), specifically to hold the enol form in place; the vial is filled with nitrogen rather than air so that atmospheric CO₂ cannot dissolve in, acidify the solution, and pull the equilibrium back toward the keto form. That alkaline enol-form solution is what makes thiopentone storable as a stable preparation at all.

The same tautomeric shift explains one of thiopentone’s best-known complications. Given intravenously, the drug is diluted into blood at physiological pH 7.4 — a large enough drop from the vial’s pH 10.5 that the equilibrium swings back toward the poorly water-soluble keto form, which is expected and harmless once diluted in a vein. Given intra-arterially by accident, the same shift still happens, but arterial vessels downstream are far smaller: thiopentone crystallises out of solution as it reverts to the keto form, and those crystals lodge in small vessels, causing ischaemia and severe pain. This is a tautomerism problem before it is anything else — the same keto–enol switch that lets thiopentone be manufactured as a stable, storable powder is what makes intra-arterial injection dangerous.

Two genuinely separate PubChem records: the free acid and the sodium salt that is actually manufactured and stored. The keto-predominant-vs-enol-favoured pharmacology described above belongs to the drug's behaviour in solution at different pH, not to a claim about which exact tautomer either 2-D depiction draws.

Stereoisomers have both the same chemical constituents and the same bond structure as one another, but a different three-dimensional configuration. This is a subtler kind of difference than structural isomerism: draw either stereoisomer as a flat, two-dimensional structural formula and they can look identical, because the connectivity really is identical. The difference only becomes visible once the molecule is considered in three dimensions — which is exactly the dimension a chiral biological target is sensitive to.

There are two forms of stereoisomerism: geometric, arising from restricted rotation, and optical, arising from one or more chiral centres. This section covers geometric isomerism; chirality and optical isomerism are covered from the next section onward.

Geometric isomerism exists when a molecule has dissimilar groups attached to two atoms (often carbon) linked either by a double bond or within a ring structure. Free rotation of the groups around that link is restricted, so the groups are locked either on the same side of the plane of the double bond or ring — the cis- conformation — or on opposite sides — the trans- conformation.

The bis-benzylisoquinolinium neuromuscular blockers illustrate this directly. Mivacurium has two identical heterocyclic groups linked through an ester-containing carbon chain; each heterocyclic ring can independently sit cis- or trans-, so the molecule needs two prefixes — one per ring. Mivacurium’s marketed mixture contains three such geometric isomers: trans-trans- (58%), cis-trans- (36%) and cis-cis- (6%).

Atracurium has the same two-ring geometric structure, giving rise to its own cis-cis / cis-trans / trans-trans groups — layered on top of its separate optical stereoisomerism, covered in Section 9. The marketed mixture separates approximately 50-55% cis-cis, 35-38% cis-trans and 6-7% trans-trans. Cisatracurium is named for its geometry specifically: it is the single isomer that is cis- at both rings, hence “cis-cis-,” shortened in the trade name to “cis-.”

Table 2b — Geometric isomer ratios: atracurium vs mivacurium

The two drugs' ratios are not interchangeable shorthand for each other — atracurium's mixture is predominantly cis-cis, while mivacurium's is predominantly trans-trans. Do not transpose one drug's percentages onto the other.

The anaesthetic-drug explorer further on lets you browse the full mivacurium/atracurium/cisatracurium comparison alongside the other neuromuscular blockers.

Chirality is a unique subset of stereochemistry: a molecule is chiral if it has a centre — or centres — of three-dimensional asymmetry. This configuration is almost always a consequence of the tetrahedral bonding geometry of carbon. A chiral centre is a carbon atom or a quaternary nitrogen surrounded by four different chemical groups, arranged so that in three dimensions the bonds point to the vertices of a tetrahedron.

Because the four groups are all different, two distinct spatial arrangements are possible — mirror images of each other — and they cannot be superimposed, in the same way a left hand cannot be superimposed on a right hand no matter how you turn it. These two mirror- image, non-superimposable forms are called enantiomers. In every other chemical respect, a pair of enantiomers is identical: same melting point, same solubility, same bond energies. What differs is how each one interacts with anything else that is itself chiral.

Nearly every organic building block in biology — proteins, sugars, lipids, and the receptors, enzymes, transporters and plasma proteins built from them — is chiral. Because these interactions are three-dimensionally exacting, enantiomers can differ in absorption, distribution, clearance, potency and toxicity, and can even antagonise one another’s effects outright. That last point — enantiomers acting as each other’s antagonist rather than simply differing in strength — becomes directly relevant later in this lesson, when methadone, dobutamine and sotalol are covered as drugs whose two enantiomers do pharmacologically different jobs rather than the same job at different strengths.

That mirror-image relationship is exactly what sits at the chiral centre of real anaesthetic drugs — full structural formulas for ketamine and etomidate appear alongside their pharmacology in the intravenous anaesthetics section (Section 13), and the bupivacaine family’s appears in the local anaesthetics section (Section 14), each next to the drug it belongs to rather than separated from it.

The absolute spatial arrangement of the four groups around a chiral centre is what unambiguously distinguishes one enantiomer from the other, and it is assigned using the Cahn-Ingold-Prelog (CIP) priority rules: identify the atom of lowest atomic number directly attached to the chiral centre, and imagine it lying behind the plane of the page. The remaining three substituents now face the viewer; rank them by the atomic number of the atom directly attached at the centre. If their priority descends in a clockwise direction, the configuration is R (rectus); if anticlockwise, S (sinister).

Before CIP nomenclature, the two enantiomers of a chiral drug were distinguished purely by the direction in which they rotated plane-polarised light in solution: clockwise rotation is dextrorotatory, marked (+) or historically d-; anticlockwise is levorotatory, marked (−) or historically l-. This is an entirely empirical, measured property — you observe it with a polarimeter, you do not derive it from the structure. A 50:50 mixture of both enantiomers (a racemic mixture, covered in Section 10) does not rotate light at all, because the two equal-and-opposite rotations cancel.

There is no link between the R/S classification and the dextro-/laevo- (optical-rotation) classification: an S-configured molecule can rotate light clockwise or anticlockwise, and the same is true for R. You genuinely cannot predict one from the other without measuring or calculating it directly.

A third, older system — D/L — labels a molecule’s configuration relative to a reference structure (historically glyceraldehyde), and is still used for sugars and amino acids. Like R/S, D/L is a configurational label fixed by structure, not a measurement of rotation — but it is a separate labelling system from R/S, built on a different reference convention, and it is not interchangeable with (+)/(−) either, despite the superficial resemblance between “D (dextro)” and “(+) dextrorotatory.”

Table 4 — R/S vs (+)/(−) vs D/L

A drug’s marketed or generic name is not a reliable shortcut through any of this. Prefixes such as “dex-,” “levo-” or “es-” often gesture at one of these systems, but they are naming conventions chosen by whoever registered the drug, not a guaranteed indicator of R/S configuration — the nomenclature comparator above works through two concrete examples of this trap.

When a molecule contains more than one chiral centre, multiple stereoisomers become possible, and they are not all mirror images of one another — some pairs are, some are not. The stereoisomers within such a family that are not mirror images of each other are called diastereoisomers. Unlike a pair of enantiomers, diastereomers can differ in ordinary physical properties (melting point, solubility) as well as in pharmacology, because they are not constrained to be identical in every non-chiral respect.

With n chiral centres, the theoretical maximum number of stereoisomers is 2ⁿ. If the molecule has internal symmetry, however, some of those theoretical configurations turn out to be duplicates of one another, so the actual number of distinct three-dimensional structures is fewer. Atracurium is the clearest anaesthetic example: it has four chiral centres (two carbon, two quaternary nitrogen), giving 2⁴ = 16 theoretically possible isomers — but because the molecule is symmetric, only 10 of those are actually distinct structures. Atracurium is therefore described as “a racemic mixture of 10 stereoisomers.”

Table 3 — Enantiomers vs diastereomers

If a drug has a single chiral centre and both enantiomers are present in equal proportions, that preparation is a racemic mixture. Isoflurane, warfarin, bupivacaine, ketamine and adrenaline are all marketed racemic. A racemic mixture has no net optical rotation — the two equal-and-opposite rotations cancel — but that says nothing about its pharmacology: an estimated one-third of drugs in clinical use are racemic, and more than one-third of all synthetic drugs are chiral overall, though most are used clinically as racemates.

Where the pharmaceutical industry has identified the more active, or less toxic, isomer and produced it alone, that is an enantiopure preparation: R-etomidate, S-bupivacaine (levobupivacaine), S-ropivacaine, S-ketamine, and S-medetomidine (dexmedetomidine) are all examples. In nature, chiral molecules are usually produced as a single isomer by stereospecific enzyme synthesis (D-glucose, for instance); when natural agents are purified for medicinal use, the purification process can itself introduce racemisation, so both isomers end up in the pharmaceutical preparation even where nature made only one.

It is tempting to assume a racemic mixture must simply be “half as strong” as the pure active isomer, or that switching to the pure isomer is automatically an improvement. Neither assumption is safe. The two enantiomers of a racemic mixture can differ substantially in both pharmacodynamic and pharmacokinetic contribution, and — as the drug-class comparisons later in this lesson show — the “inactive” or minor enantiomer is sometimes not simply inert: it can be genuinely toxic on its own, or it can act at an entirely different target, or it can complement rather than dilute the other enantiomer’s effect. The therapeutically inactive isomer in a racemic mixture is, at best, best thought of as an impurity from an efficacy standpoint — but “impurity” does not always mean “harmless.”

Table 5 — Racemate vs enantiopure preparation

Chirality is not a peculiarity of drug molecules — it is close to universal in biology. Proteins, sugars and lipids are built from chiral building blocks, and the receptors, enzymes, ion channels, membrane transporters and plasma-protein binding sites a drug encounters are themselves chiral structures. Binding between a drug and any of these is a three-dimensionally exacting interaction, which is why a molecule’s spatial arrangement — not just its formula — can determine absorption, distribution, clearance, potency, selectivity and toxicity.

This distinction matters clinically in two separable ways, covered in the next two sections: differences that arise because a target binds the two enantiomers differently (stereoselective pharmacodynamics), and differences that arise because the body handles the two enantiomers differently — absorbs, distributes, metabolises or excretes them at different rates (stereoselective pharmacokinetics).

Stereoselective pharmacodynamics concerns the target side of the interaction: receptor affinity, intrinsic activity, potency, efficacy and selectivity, ion- channel binding kinetics (how fast a drug associates with and dissociates from a channel), and off-target or organ-toxicity effects. Where a drug acts through a specific receptor or channel, stereoselectivity is common and can be pronounced, because the binding site itself is chiral. Where a drug’s effect instead depends mainly on a non-specific physicochemical property — bulk lipid solubility, for instance — the two enantiomers can behave much more similarly, since that kind of property does not depend on a precise spatial fit.

Stereoselective pharmacokinetics concerns what the body does to the drug: plasma-protein binding, membrane transport, tissue distribution, the specific metabolic enzyme or pathway involved, whether an active or toxic metabolite is formed, clearance and elimination half-life, and susceptibility to drug interactions. Because metabolic enzymes are themselves chiral proteins, it is entirely possible for the two enantiomers in one racemic dose to follow measurably different concentration-time courses in the same patient — effectively behaving as two co-administered drugs with different kinetics, delivered in a single formulation.

Table 6 — Stereoselective pharmacodynamics vs pharmacokinetics

Ketamine has a single chiral carbon and is used both as a racemate and, increasingly, as the pure S(+)-enantiomer (esketamine). Studied separately, S(+)- ketamine is roughly twice as potent an analgesic as the racemate and around four times as potent as R(−)-ketamine — two independently reported ratios that agree with each other. S(+)-ketamine also produces more rapid metabolism and recovery, less salivation, and a lower incidence of emergence reactions than R(−)-ketamine, and the phencyclidine (NMDA-receptor) binding site itself is stereoselective, with S(+) showing the greatest affinity. R(−)-ketamine is not without its own advantage, though: it has greater anticholinergic, bronchodilator activity than S(+), so the racemate can still be the more sensible choice in a bronchospastic patient — a genuine advantage/disadvantage pair, not simply “S is better.”

The isolated R(−)-enantiomer likewise has its own INN, arketamine, mirroring esketamine's naming for S(+) — the figure below labels both.

Etomidate is unusual among intravenous and inhaled anaesthetics in being marketed as a single isomer. Its anaesthetic effect resides predominantly in the R(+) isomer, which is roughly five times as potent as the S(−) isomer — this stereoselectivity is itself part of the evidence that GABA_A receptors are etomidate’s site of action.

The pipecoloxylidide local anaesthetics — mepivacaine, bupivacaine, ropivacaine, levobupivacaine — are chiral at a single carbon substituent on the piperidine nitrogen. Mepivacaine and bupivacaine are marketed as racemic mixtures; ropivacaine and levobupivacaine were deliberately developed as pure S-enantiomers from the outset, rather than purified after the fact.

The clinical motivation is cardiotoxicity, not analgesic potency. The R-enantiomer of bupivacaine is more cardiotoxic than the S-enantiomer, reflecting a difference in affinity for, and dissociation from, myocardial sodium channels. Levobupivacaine (pure S-bupivacaine) is associated with fewer ventricular dysrhythmias than racemic bupivacaine in animal studies, and with a higher rate of successful resuscitation where toxicity did occur. Ropivacaine — a pure S-enantiomer of a bupivacaine homologue — is less lipid- soluble and less cardiotoxic than bupivacaine, though it remains more cardiotoxic than lidocaine. None of this implies levobupivacaine or ropivacaine are without cardiotoxic risk at all: they carry a wider safety margin than racemic bupivacaine, not immunity.

The S-enantiomer's advantage is not unconditional, however: the S-enantiomers of mepivacaine and bupivacaine are also cleared more slowly by the liver than their R counterparts, so on a prolonged infusion the wider cardiotoxic safety margin is partly offset by somewhat greater systemic accumulation over time.

Table 6b — The pipecoloxylidide local anaesthetics compared

Atracurium is a racemic mixture of 10 stereoisomers, arising from four chiral centres with the theoretical 16 reduced by molecular symmetry (Section 9). Those 10 stereoisomers also separate into three geometric groups by ring configuration — approximately 50-55% cis-cis, 35-38% cis-trans and 6-7% trans-trans (Section 6). Cisatracurium is the single 1R-cis–1′R-cis isomer isolated from that mixture: about 15% of the marketed atracurium mixture by weight, but more than half of its neuromuscular-blocking potency — roughly four to five times as potent as atracurium itself. Unlike atracurium, cisatracurium does not cause clinically significant histamine release at clinical doses, which is itself evidence that the histamine-release phenomenon is stereospecific rather than a general property of the drug class.

Mivacurium’s three geometric isomers (Section 6) also differ pharmacokinetically, not just in proportion: as detailed in the drug explorer below, the minor cis-cis isomer is cleared far more slowly than the other two despite contributing only a small fraction of the blocking effect — geometric isomers of the same drug behaving almost like separate drugs kinetically, even though they share the same mechanism of action.

Methadone is a racemic mixture of an opioid μ-receptor agonist (L-methadone) and a potent NMDA-receptor antagonist (D-methadone) — the two enantiomers act at entirely different pharmacological targets, not the same receptor at different strength.

Dobutamine is a 50:50 racemic mixture of two stereoisomers with directly opposing α1-adrenergic actions: the (−) enantiomer is a potent α1-agonist (and weak β-agonist), while the (+) enantiomer is a competitive antagonist at the same α1 receptor (alongside its own potent β-agonism). The drug’s overall haemodynamic effect reflects the balance struck between these two opposing actions delivered together in the same infusion.

Sotalol sits between the two patterns above: both enantiomers share similar class III (potassium-channel, action-potential-prolonging) antiarrhythmic activity, but only the levorotatory isomer additionally acts as a β-adrenergic antagonist (class II activity). The dextrorotatory isomer, on its own, has been associated with increased mortality in patients with ventricular dysfunction after myocardial infarction — so here the “extra” enantiomer in the racemate is not simply weaker, it carries a distinct safety signal of its own.

Table 7 — Anaesthetic-drug comparison by drug class

Across the drug classes covered in this lesson, isomerism’s clinical significance falls into a handful of recurring patterns rather than a single rule. Sometimes the enantiomers share a target and differ mainly in potency (ketamine, etomidate). Sometimes they share a target and differ mainly in a safety margin rather than potency (the bupivacaine family). Sometimes they act at genuinely different targets, or oppose one another at the same target (methadone, dobutamine, sotalol). And sometimes — as with mivacurium’s three geometric isomers — the isomers share essentially the same mechanism of action but differ sharply in how fast the body clears them. Occasionally, as with sevoflurane, propofol and dopamine, isomerism is simply not in play at all: no chiral centre, no geometric restriction, nothing to discuss.

Recognising which pattern applies to a given drug is more useful, and more exam-safe, than trying to memorise each drug’s numbers independently: it tells you what kind of answer a question about that drug’s isomerism is actually looking for.

Table 8 — Common misconceptions and corrected statements

Isomerism splits into structural isomerism (different bond order) and stereoisomerism (same bond order, different 3-D arrangement); stereoisomerism splits into geometric isomerism (restricted rotation) and optical isomerism (chiral centres), and optical isomerism splits again into enantiomers (one centre) and diastereomers (multiple centres). Three separate, non-interchangeable nomenclature systems describe a chiral centre — R/S (configuration), (+)/(−) (measured rotation), and D/L (older relative configuration) — and confusing them is the most consistently examined error in this topic. Layered on top of that classification, the anaesthetic drug examples in this lesson show isomerism affecting potency, safety margin, target selectivity and clearance rate, sometimes in combination within a single drug.