When Local Anaesthetic Does Not Seem To Work

Yes, immediately, and out loud. If you can feel sharpness, pressure that turns into pain, or a hot line of sensation while a tooth is being worked on, the correct response is to stop the surgeon and say so. That report is clinical information nobody else in the room has, and it changes what happens next. Enduring it silently is not toughness; it is a preventable injury to your own recovery.

Patients hesitate for reasons that are entirely human and entirely wrong here. They do not want to seem difficult. They assume the surgeon knows. They read the concentration on the faces around them as a signal not to interrupt. Some have been told, years earlier by someone else, that the injection has been given so the feeling must be imaginary. None of that survives contact with how anaesthesia actually behaves in inflamed tissue, which is unreliably and sometimes not at all.

Under local anaesthetic alone you can talk. Under nitrous oxide you can talk. Under moderate sedation you may be able to talk but you will not reliably remember doing it, which is exactly why a hand signal is agreed before anything starts. Under deep sedation or general anesthesia you are not reporting anything, and depth of anaesthesia is instead judged by a person whose only job is monitoring you.

There is one more reason to speak that has nothing to do with the next five minutes. Pain experienced during a procedure that was expected to be numb is a durable memory. It is one of the more common origins of adult dental avoidance, and people who acquire it postpone care for years and arrive later with worse problems. Interrupting is cheap. What silence costs shows up a decade later.

Infected tissue is acidic, and acid strands the drug outside the nerve. Local anaesthetics are weak bases supplied as an acidic salt solution so they stay stable in the cartridge. Once injected, normal tissue at roughly pH 7.4 converts a workable fraction of the drug into an uncharged form that can cross the fatty nerve membrane. Inflamed tissue can sit well below that, and the fraction collapses.

The arithmetic is unforgiving. Lidocaine has a pKa near 7.7, meaning that at pH 7.4 something on the order of a third of the molecules are in the uncharged, membrane-crossing form. Drop the local pH toward 6 and that share falls to a few percent. Same cartridge, same volume, same technique — a fraction of the drug arriving where it has to act. Articaine, with a pKa nearer 7.8, faces the same physics, though its lipid solubility and diffusion through bone help it in other ways.

Acidity is not the only obstacle. Inflammation raises blood flow through the area, so anaesthetic is carried away faster than it would be from quiet tissue. Sustained inflammation also alters the nerve itself: sodium channel subtypes that are relatively resistant to local anaesthetics are upregulated, and the nerve becomes more easily excited overall, so a partial block that would be adequate in a calm tooth is not adequate in an angry one.

The practical consequence is well known to anyone who does this work. Numbing the soft tissues around a hot lower molar is usually straightforward. Numbing the pulp of that same tooth to a standard that lets it be cut is materially harder, and inferior alveolar nerve blocks in symptomatic irreversible pulpitis fail at rates repeatedly reported in the range of a third to over half of first attempts. That is not a failure of the injection. It is the tissue.

Yes, and it is common enough to be a routine consideration rather than a curiosity. The inferior alveolar nerve is described in textbooks as a single canal running a predictable course through the mandible, but cone-beam imaging of real jaws shows bifid canals, high or low nerve positions, and variable entry points frequently enough that no surgeon treats the textbook picture as a promise.

Bifid mandibular canals — a canal that splits into two branches, each carrying nerve fibres — appear on cone-beam CT in a meaningful minority of scans, with reported prevalence varying widely by population and by how strictly the definition is applied. A block placed accurately for one canal can leave the other branch conducting normally. The lingula, the bony landmark a conventional block is aimed at, also sits higher or lower relative to the occlusal plane in different people.

Then there is accessory innervation, which is the reason a lower third molar can hurt when every soft tissue sign says the block worked. The mylohyoid nerve branches off the inferior alveolar nerve above the point where a standard block deposits solution, and in a proportion of people it supplies sensory fibres to the lower molars from below. The long buccal nerve supplies the cheek-side gum and is not covered by an inferior alveolar block at all. Fibres crossing the midline can supply the front teeth from the opposite side.

• Mylohyoid nerve — branches proximal to a standard block; supplies some lower molars from the lingual and inferior aspect
• Long buccal nerve — supplies buccal soft tissue at the molars and needs its own injection
• Bifid or trifid mandibular canal — a second branch conducting past a correctly placed block
• Retromolar foramen — an accessory opening behind the last molar, present in a small share of jaws
• Cross-midline fibres — relevant at the front of the mouth rather than at third molars, but a real source of unexpected sensation

This is one of several reasons a panoramic film is often not the end of the imaging conversation. A panoramic image flattens a curved jaw onto one plane and cannot reliably show where the canal sits in the third dimension, or that there are two of them. Cone-beam imaging shows the canal in cross-section, and where it changes the plan it changes it before the operation rather than during it.

A frightened nervous system reports more. That is not a metaphor. Anticipatory anxiety lowers the threshold at which a signal is registered as painful, and it changes how non-painful input is interpreted — the pressure of an elevator against bone, the vibration of a handpiece, the sound conducted through the jaw. All of that reaches consciousness whether or not the nerve carrying pain fibres is blocked, because pressure and vibration travel on fibres a local anaesthetic does not silence.

So two things can be true at once: the block is working, and you are having a bad time. A patient who says "I can feel it" is usually describing one of three distinct things, and the surgeon's first job is to work out which, because the response to each is different.

1. Sharp, hot, well-localised sensation at the tooth — the block is incomplete, and more anaesthetic or a different technique is needed
2. Heavy pressure or pushing without sharpness — expected, will not be removed by more local anaesthetic, and is managed by explanation or by adding sedation
3. Diffuse distress, racing heart, an urge to sit up — anxiety dominating, which nitrous oxide or IV sedation manages and more injections will not

Nobody expects you to make that distinction yourself. Describe it in your own words — sharp, hot, dull, pushing, everywhere, right there — and the distinction is the surgeon's to draw. Vague reporting still helps. Saying nothing helps nobody.

It is also worth planning for anxiety honestly at the consultation rather than discovering it in the chair. A person who knows they cannot tolerate awareness of the procedure is better served by an anaesthesia plan built around that from the start than by a local-only appointment that has to be abandoned partway through.

Work stops, and the cause gets diagnosed before another needle appears. The surgeon tests the field to find out what is numb and what is not, decides whether the problem is technique, anatomy, tissue chemistry or awareness, and then chooses from a short list of options. Escalating blindly — more of the same solution in the same place — is the response least likely to help.

Waiting longer is genuinely one of the options, and it is underrated. Onset for an inferior alveolar block is often quoted at three to five minutes, but pulpal anaesthesia in a molar can take ten to fifteen minutes to become profound, and in inflamed tissue longer still. A block declared failed at four minutes was sometimes only a block that had not finished arriving.

• Wait longer and retest, particularly where soft tissue signs suggest the solution is in roughly the right place
• Repeat the block with adjusted landmarks, or use an alternative approach such as a Gow-Gates or Vazirani-Akinosi technique that deposits higher and catches branches a conventional block can miss
• Add a buccal infiltration, commonly with articaine, to cover accessory supply
• Supplementary intraligamentary injection into the periodontal ligament space, or an intraosseous injection delivering solution directly into cancellous bone beside the tooth
• Change the anaesthesia plan altogether — add nitrous oxide, move to IV sedation, or move to general anesthesia
• Stop, treat the infection, and reschedule

Supplementary techniques come with their own trade-offs and they are discussed rather than sprung on you. Intraosseous injection with an adrenaline-containing solution can produce a brief, sometimes marked increase in heart rate that is harmless in most people but matters in some cardiac histories. Intraligamentary injection is effective but short-acting and can leave the tooth tender for a few days. Neither is a free upgrade; both are reasonable when indicated.

There is also a dose ceiling, and it is not negotiable. Local anaesthetics have maximum safe doses calculated on body weight, and lidocaine with adrenaline is conventionally limited to about 7 mg per kilogram up to a fixed absolute ceiling. Systemic toxicity from exceeding it is a serious event involving the central nervous system and the heart. A surgeon who says "I cannot give you more today" is not giving up on you; they have reached a number that exists for your protection.

Sometimes the honest decision is to stop and come back. An acutely infected third molar with a spreading swelling, limited mouth opening and distorted anatomy is a poor operating field and a poor anaesthetic field at the same time. Draining the infection, treating it, and letting the tissue quieten down converts a difficult, half-numb operation into a straightforward one a week or two later.

This is a real cost and it should be stated as one. Rescheduling means another day off work, another fasting period if sedation is planned, another arrangement for someone to drive you, and more time carrying a tooth that hurts. Nobody enjoys being sent home. It is still frequently the better clinical decision, and a practice that never reschedules is not one with unusually good anaesthetic technique.

The alternative worth naming plainly: moving the case to general anesthesia. Under general anesthesia the acidity problem does not disappear, but it stops being decisive, because the anaesthetic managing your awareness is not the anaesthetic being defeated by the tissue pH. That is a different plan with its own preparation, its own fasting rules and its own staffing requirements, and it is decided in advance rather than improvised.

Bring history. If anaesthetic has failed on you before, that is one of the more useful sentences you can say at a consultation, and it should be volunteered rather than waited for. Say which tooth, roughly how long ago, what it felt like, and what the dentist did. Repeat difficulty in the same region points at anatomy; difficulty only when a tooth was already hurting points at inflammation; difficulty everywhere and always is worth investigating on its own terms.

• "Local has not worked well on me before" — say this at the consultation, not on the day
• "What is the signal if I feel something?" — ask it, and confirm it before the first injection
• "How will you test that I am numb before you start?"
• "If the block does not take, what is the next step and does it change what I have to prepare for?"
• "If this tooth is infected on the day, will you still operate?"
• "Which of what I will feel is pressure that will not go away with more anaesthetic?"

Two practical notes that come up constantly. Alcohol the night before does not make you harder to numb by any mechanism worth planning around, but it interacts with sedation and with the fasting rules, so it matters for other reasons. And red hair has a genuine, repeatedly studied association with altered anaesthetic requirement through MC1R variants — the effect is real but modest, and it is a reason to mention it, not a reason to expect the appointment to fail.