Local anaesthetic failure
A Socratic walk-through of local anaesthetic failure — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why does a dental injection that reliably numbs a healthy tooth fail on an abscessed one?
The same drug, the same dose, the same needle, the same nerve. On a healthy tooth it works so dependably that nobody thinks about it. On an inflamed one it can fail, and fail exactly when the person most needs it to work.
That pattern is strange, because nothing about the drug changed. So something about the destination must have. The interesting question is not "why did it fail" but "what does this drug actually have to accomplish, step by step, and which of those steps could inflammation interfere with?"
Reasoning it through
REASONING #Take the steps apart. A local anaesthetic works by blocking voltage-gated sodium channels, so the nerve cannot generate the impulse that would signal pain. But here is the detail that governs everything: the binding site is on the inside of the channel. The drug has to get into the axon before it can do anything.
Now consider what the drug is. These agents are weak bases, and they exist in two interconverting forms: an uncharged base and a protonated cation. Only the uncharged form is lipid-soluble enough to cross the nerve membrane. Only the charged form binds the channel effectively from within. So the molecule needs to be uncharged to arrive and charged to act — and which form dominates is set by the pH of wherever it happens to be.
That relationship is quantitative, so let us do the arithmetic rather than gesture at it. The ratio of base to protonated form is ten raised to the power (pH minus pKa). Lidocaine's pKa is about 7.8 — I am recalling that figure rather than deriving it. In healthy tissue at pH 7.4, the exponent is minus 0.4, giving a ratio of roughly 0.4, so the uncharged share is 0.4 divided by 1.4, near 28%. Take inflamed tissue a full pH unit lower, at 6.4: the exponent is minus 1.4, the ratio about 0.04, and the uncharged share falls to roughly 4%. Treat 6.4 as an illustration rather than a measurement — reported values for inflamed tissue vary. But the shape is the point: a single pH unit cuts the membrane-crossing species by something like sevenfold, without changing the amount injected at all.
That is a clean story, and it is the one most often told. Now let us try to break it, because a mechanism that cannot be tested is not worth much.
The test. If tissue pH is the mechanism, then raising the solution's pH — buffering with bicarbonate before injection — should improve matters. It does speed the onset of block measurably, which supports the pH account for drug injected into the affected tissue. But when the same buffering is tried for failed anaesthesia of an acutely inflamed pulp, the results are mixed at best.
The refuting observation. Here is the one that does real damage. The commonest failure in dentistry is of an inferior alveolar nerve block, where the needle is placed near the mandibular nerve trunk — centimetres away from the abscessed tooth, in tissue of entirely normal pH. Tissue acidosis at the tooth cannot explain a failure at a site the inflammation never reached. So pH alone is not the mechanism, and any account that stops there is incomplete.
What else is going on, then? Several things are documented, and the weighting between them is genuinely unsettled. Inflamed tissue is more vascular, so drug is carried away faster. Inflammatory mediators sensitise nociceptors directly, so a partially blocked nerve still delivers enough traffic to be felt. Nerve fibres serving inflamed tissue alter which sodium channels they express, including subtypes that are relatively resistant to block by these agents — which would produce failure at a remote injection site, exactly where the pH story cannot reach. And the central amplification described elsewhere in this collection means the threshold for reporting pain has itself dropped.
So the honest summary is layered: for drug placed into acidic tissue, the pH partition is real and calculable. For failure at a distant block site, the changed excitability of the nerve is the better candidate, and the field has not fully settled the relative contributions. Whether any of this applies to a given person, and what to do about it, is a clinical judgement requiring examination — the mechanism tells you why it can happen, not what to do next.
The analogy
THE ANALOGY #Think of a delivery that has to pass a wall and then be signed for inside. The parcel can only pass the wall unwrapped, and can only be signed for wrapped. The wrapping is done automatically by the conditions outside — and in acidic surroundings almost everything gets wrapped before it reaches the wall, so very little crosses, even though the same number of parcels were sent.
parcels do not change what the building is willing to accept, whereas an inflamed nerve alters the very channels the drug must bind, so the failure is not only a transport problem at the wall but a change in the lock on the other side.
Clarifying the model
THE MODEL #Two refinements connect the steps. The first is that "the drug did not get in" and "the drug got in but did not work" are different failures with different signatures, and the pH argument only addresses the first. Recognising that is what makes the remote-block observation so informative: it forces the second kind of failure onto the table.
The second is about what "failure" even means to the person in the chair. A nerve block is rarely all-or-nothing; conduction can be substantially reduced while a residual signal remains. Whether that residual is felt as pain depends on how sensitised the receiving system is — and inflammation raises that sensitivity at the same time as it impedes the drug. Two mechanisms therefore pull in the same direction, which is part of why the failure is so much commoner in inflamed tissue than a partition calculation alone would predict.
A picture of it
THE PICTURE #How to readRead top to bottom as one molecule's journey, with each participant a place it must pass through. The first note marks the pH bottleneck: the split set in tissue fluid decides how much of the dose is even eligible to cross the membrane on the next line. The last two lines are the second requirement, that the drug re-acquire its charge inside to bind the channel from within — and the closing note marks the failure mode that no pH calculation predicts, where the channel itself has changed.
What became clearer
WHAT CLEARED #The drug has to satisfy two contradictory requirements in sequence — uncharged to cross, charged to bind — and the balance between them is set by an environment that inflammation changes. That explains a real part of the failure and can be calculated. But the failures that occur at injection sites far from any inflammation say that the nerve's own excitability has shifted too, and that the tidy pH story, taught as though it were the whole answer, is only the part that is easy to compute.
Where to go next
ONWARD #- Why differential block affects thin pain fibres before thick motor fibres, and what that ordering reveals about fibre diameter and myelination.
- How supplementary techniques targeting the tooth directly bypass the transport problem, and what limits them.
Key terms
TERMS #| Term | What it means |
|---|---|
| pKa | the pH at which a weak base is half protonated; it fixes the balance between the two forms at any given pH. |
| Voltage-gated sodium channel | the membrane protein whose opening generates the nerve impulse, and whose inner vestibule these drugs occupy. |
| Tissue acidosis | the lowered pH of inflamed tissue, produced by inflammatory metabolism and local ischaemia. |
| Inferior alveolar nerve block | a dental injection at the mandibular nerve trunk, remote from the tooth being treated. |
Every term the collection defines is gathered in the glossary.