When a Root Canal Fails — and What Can Still Be Done

A Root Canal Is Not a Permanent Guarantee

Root canal treatment has a high long-term success rate, and most treated teeth serve for decades without further trouble. But "high" is not "absolute." A proportion of treated teeth develop problems later — sometimes within a year, sometimes fifteen years on.

When that happens, the conversation a patient usually has is short: the tooth hurts, the root canal failed, it needs to come out. And that conclusion is sometimes correct. But it skips a step, because a failed root canal and an unsalvageable tooth are not the same thing.

The purpose of retreatment is to answer a specific question: is there something inside this tooth that the first treatment did not reach? Surprisingly often, there is.

The Five Reasons Root Canals Fail

Failure is rarely mysterious. In practice it traces back to one of the following.

• A missed canal — The most common cause. Root anatomy is not standardised. Upper first molars very frequently have a second canal in the mesiobuccal root — the MB2 — which is narrow, curved and easy to overlook without magnification. If a canal is never located, it is never cleaned, and bacteria remain inside a tooth that appears treated
• Incomplete cleaning or sealing — A canal that was found but not fully cleaned to its end, or not sealed densely enough, leaves space for bacteria to persist or re-enter. On a radiograph this can sometimes be seen as filling material that stops short of the root tip
• Coronal leakage — This one surprises patients. The root canal itself may have been done well, but if the tooth was left with only a temporary filling for months, or the permanent crown developed a gap at its margin, saliva and bacteria travel back down into the canals from above. The treatment did not fail; the seal above it did
• New decay beneath a crown — A crowned tooth can still decay at the margin where crown meets tooth. The decay is hidden, progresses quietly, and eventually reaches the canal system
• Vertical root fracture — A crack running lengthwise down the root. This is the one category where retreatment does not help, because the problem is structural rather than microbial. It is more common in teeth that were heavily restored, that carry a post, or that have been under long-term clenching forces

The first four are addressable. The fifth is not, and distinguishing it from the others is a large part of what the initial assessment is for.

How You Would Know

A failing root canal does not always announce itself. Symptoms, when present, tend to be:

• Tenderness when biting on that specific tooth — often the earliest sign, and often dismissed because it comes and goes
• A dull ache or pressure in the area rather than sharp pain
• A small bump on the gum near the root — a sinus tract, which is infection finding a route to drain. It may discharge, relieve pressure, and then reappear. Patients frequently think it is an unrelated gum problem
• Swelling, intermittent or persistent
• Nothing at all — a substantial proportion of failures are silent and are found on a radiograph taken for another reason

Diagnosis rests on imaging. A periapical radiograph shows whether there is a radiolucent area at the root tip, how far the existing filling extends, and whether the canal outline suggests anatomy that was not treated. Where the picture is unclear — a suspected missed canal, a possible fracture, an unusual root shape — a CBCT scan gives a three-dimensional view that a flat radiograph cannot.

What Retreatment Actually Involves

Retreatment is methodical rather than dramatic. The sequence:

• Access — The existing crown or filling is removed, or an opening is made through it. If a post is present, it must be carefully removed without stressing the root
• Removing the old filling material — Gutta-percha is softened with solvent and heat and taken out of each canal. This is the slow part, and it must be complete; material left behind shields bacteria from disinfection
• Re-exploring the anatomy — The canal system is re-examined for anything missed. This is where magnification matters. An MB2 canal is often a fraction of a millimetre wide and sits beneath a shelf of dentine; under loupes or a microscope it can be located, while by eye it can be walked past entirely
• Disinfection — Canals are cleaned mechanically and irrigated. Between visits, an antibacterial dressing is usually left inside the tooth to reduce the bacterial load further
• Re-sealing — Once the tooth is symptom-free, the canals are filled and sealed to the root tip
• Restoring the tooth properly — This step is not optional. A retreated tooth needs a well-sealed final restoration, usually a crown. Restoring the seal above is what prevents the same failure repeating

Expect two appointments in most cases, occasionally three, with healing then monitored radiographically over the following months. Bone at the root tip regenerates slowly; a follow-up image at six to twelve months is what confirms success, not the absence of pain at two weeks.

When Retreatment Is Not the Answer

Two situations lead elsewhere.

• The problem sits outside the canal system — If infection persists at the root tip despite well-executed canal treatment, or if the canal cannot be safely renegotiated (a blockage, a separated instrument, an unusually curved root), the alternative is apicoectomy. A small flap is raised, the root tip and surrounding inflamed tissue are removed, and the root end is sealed from the outside. It is a day procedure under local anaesthetic
• The root is vertically fractured — No amount of disinfection addresses a split root. Signs include a narrow deep pocket beside the tooth, a sinus tract close to the gum margin, and a characteristic radiographic pattern. Here extraction is the correct recommendation, and continuing to treat would waste both time and money

Being told a tooth is fractured is disappointing, but it is a real diagnosis rather than a shortcut. The distinction matters: extraction because a root is split is sound clinical reasoning; extraction because retreatment was never considered is not.

Retreatment or Implant — an Honest Comparison

Patients increasingly arrive having already decided that an implant is the modern option and retreatment is the outdated one. The comparison deserves more care than that.

• A natural tooth has a periodontal ligament. That ligament provides fine pressure feedback — the reason you can sense a grain of sand while chewing. Implants fuse directly to bone and have no such feedback
• Bone behaviour differs. A retained root maintains the surrounding bone. Extraction is followed by resorption of the socket, which is why grafting is often needed before or during implant placement
• Treatment burden differs. Retreatment is non-surgical and usually two visits. Implant placement involves extraction, healing, surgery, integration time and a crown — a longer path with more stages
• But retreatment is not always the better choice. If the remaining tooth structure is minimal, if the root is fractured, or if the tooth has already been retreated unsuccessfully, an implant is the more predictable outcome. Persisting with a tooth that cannot be restored serves no one

The reasonable sequence is to assess whether the tooth is restorable first, then decide — rather than deciding first and assessing afterwards.

If You Are Travelling for This

Retreatment is well suited to a planned trip, with some caveats worth knowing in advance.

• Allow for two visits several days apart. A dressing is usually left between appointments; compressing both into twenty-four hours is not ideal for a tooth with active infection
• Factor in the final restoration. A retreated tooth needs a proper crown. With same-day milling this can often be completed in the same trip, but it should be planned rather than assumed
• Send imaging ahead. Existing radiographs allow much of the assessment to happen before you arrive, including a preliminary view on whether the tooth looks restorable at all
• Understand that follow-up is radiographic. Healing is confirmed months later. A dentist at home can take that image, and a copy of your treatment record makes this straightforward

The Short Version

• A treated tooth that hurts again has not automatically reached the end of its life
• Most failures come from a missed canal or a leaking restoration above — both addressable
• A vertical root fracture is the exception, and identifying it early prevents wasted treatment
• Retreatment keeps extraction available as a later option; extraction does not keep anything available
• Ask what specifically failed before agreeing to remove the tooth. If the answer is vague, that itself is informative

Frequently Asked Questions