Dental Literature Watch · Systematic review · Endodontics
For a hot mandibular tooth, primary intraosseous anesthesia failed about half as often as an inferior alveolar nerve block across five trials
TL;DR
- Primary intraosseous anesthesia roughly halved anesthetic failure compared with an inferior alveolar nerve block in mandibular teeth with symptomatic irreversible pulpitis: risk ratio 0.46 (95% CI 0.32 to 0.64, P < 0.001).
- Five randomised trials, 265 patients: 27 failures in 132 intraosseous patients against 65 in 133 nerve-block patients. From those counts, failure fell from about 49% to about 20%.
- No heterogeneity between trials (I² = 0%), so the five studies pointed the same way.
- Articaine subgroup RR 0.30 (0.16 to 0.58); lidocaine subgroup RR 0.42 (0.16 to 1.07, P = 0.07). The difference between the two agents was not statistically significant.
- "Primary" means the intraosseous injection was the first and only technique, not a rescue after a failed block.
- The authors temper the headline: small evidence base, methodological concerns, and trials that differed in agent and delivery system.
The study
| Element | Detail |
|---|---|
| Design | Systematic review and meta-analysis of randomised clinical trials; PRISMA; PROSPERO CRD420251180658; search to July 2026; risk of bias with RoB 2 |
| Population | Patients with symptomatic irreversible pulpitis in a mandibular tooth undergoing endodontic treatment; 265 participants across 5 trials |
| Intervention | Primary intraosseous anesthesia (anesthetic delivered directly into cancellous bone as the first technique), n = 132 |
| Comparator | Conventional inferior alveolar nerve block, n = 133 |
| Outcome | Anesthetic failure, pooled as a risk ratio with a Mantel-Haenszel random-effects model |
| Result | 27 failures with intraosseous versus 65 with the block; RR 0.46 (0.32 to 0.64), P < 0.001; I² = 0% |
| Subgroups | Lidocaine RR 0.42 (0.16 to 1.07, P = 0.07); articaine RR 0.30 (0.16 to 0.58, P = 0.0003); no significant interaction |
💎 Insight — what this means chairside
The hot mandibular molar is where blocks fail, and this is the strongest pooled answer so far
A nerve block alone failed in about one patient in two in these trials.
Going straight to an intraosseous injection cut that to about one in five.
Derived from the abstract's counts, that is roughly 3 to 4 patients treated intraosseously to prevent one failed anesthetic.
A nerve block alone failed in about one patient in two in these trials.
Going straight to an intraosseous injection cut that to about one in five.
Derived from the abstract's counts, that is roughly 3 to 4 patients treated intraosseously to prevent one failed anesthetic.
Primary, not rescue
Most clinicians reach for intraosseous or intraligamentary injection only after a block has already failed and the patient is already distressed.
These trials used it as the opening move, which is a different workflow: one injection, faster onset, no waiting to find out whether the block took.
Most clinicians reach for intraosseous or intraligamentary injection only after a block has already failed and the patient is already distressed.
These trials used it as the opening move, which is a different workflow: one injection, faster onset, no waiting to find out whether the block took.
Articaine carried the clearer signal
The articaine subgroup had the tighter interval and the smaller RR; the lidocaine interval crossed 1.
The review found no significant interaction between agents, so this is a hint, not a head-to-head verdict.
The articaine subgroup had the tighter interval and the smaller RR; the lidocaine interval crossed 1.
The review found no significant interaction between agents, so this is a hint, not a head-to-head verdict.
⚠️ The catch
Why this sits at 3 of 5
It sharpens one decision, how to anesthetise a symptomatic mandibular tooth, for practices that own an intraosseous delivery system.
Five small trials totalling 265 patients, with methodological concerns the authors flag, is not enough to call it a new standard.
It sharpens one decision, how to anesthetise a symptomatic mandibular tooth, for practices that own an intraosseous delivery system.
Five small trials totalling 265 patients, with methodological concerns the authors flag, is not enough to call it a new standard.
The abstract does not define failure
Trials in this field usually score failure as pain on access or instrumentation above a threshold, but the pooled definition is not given.
Nor does the abstract report onset time, duration of pulpal anesthesia, or how often a second injection was needed.
Trials in this field usually score failure as pain on access or instrumentation above a threshold, but the pooled definition is not given.
Nor does the abstract report onset time, duration of pulpal anesthesia, or how often a second injection was needed.
Different drugs, different devices
Agents varied (lidocaine, articaine) and so did the delivery system, which the authors name as a limit on the size of the effect.
Perforator-and-needle systems and motor-driven systems are not interchangeable in technique or learning curve.
Agents varied (lidocaine, articaine) and so did the delivery system, which the authors name as a limit on the size of the effect.
Perforator-and-needle systems and motor-driven systems are not interchangeable in technique or learning curve.
Safety was not pooled
The abstract reports no adverse-event data.
Intraosseous injection of an epinephrine-containing solution is known, outside this review, to produce a transient rise in heart rate, and the perforation site can be sore afterwards.
Patients in whom a vasoconstrictor surge matters were not addressed here.
The abstract reports no adverse-event data.
Intraosseous injection of an epinephrine-containing solution is known, outside this review, to produce a transient rise in heart rate, and the perforation site can be sore afterwards.
Patients in whom a vasoconstrictor surge matters were not addressed here.
Abstract only
No free full text was reachable through Europe PMC at the time of writing.
Per-trial sizes, risk-of-bias judgements and the GRADE rating, if one was made, are not in the abstract.
No free full text was reachable through Europe PMC at the time of writing.
Per-trial sizes, risk-of-bias judgements and the GRADE rating, if one was made, are not in the abstract.
⚡ Bottom line
- Mandibular tooth, symptomatic irreversible pulpitis, intraosseous system in the drawer → primary intraosseous injection roughly halved failure versus a block alone; articaine had the clearer signal.
- No intraosseous system → this review offers nothing on buccal infiltration, intraligamentary injection or a second block as alternatives; it compared only primary intraosseous against the block.
- Patient in whom a transient tachycardia matters → the trials report no safety data; the known intravascular-uptake effect of intraosseous epinephrine is the trade-off to weigh.
- Maxillary teeth, or a mandibular tooth without symptomatic pulpitis → outside the population; the block-failure problem this addresses is specific to the hot mandibular tooth.
🧪 Questions
Tap an answer — instant grade, deciding line, and the trap.
A 41-year-old has spontaneous, lingering pain in a mandibular first molar, cold testing confirms symptomatic irreversible pulpitis, and the practice owns an intraosseous delivery system. According to this meta-analysis, which single opening technique carried the lowest risk of anesthetic failure?
Primary intraosseous injection roughly halved failure against the block (RR 0.46), and the articaine subgroup carried the clearer signal (RR 0.30). The nerve block alone is the comparator, failing in about 49% of patients. Block plus buccal articaine is a common everyday supplement, but no arm in these five trials tested it, so the review offers it no support. Primary intraligamentary injection is a different technique that was not part of this comparison. Trap: block plus buccal articaine, the reflex supplement that feels evidence-based but was not studied here.
The pooled result was 27 failures in 132 intraosseous patients against 65 in 133 nerve-block patients, RR 0.46. Which reading of those numbers is correct?
65 of 133 is about 49% and 27 of 132 is about 20%; the absolute difference of about 28 points gives a number needed to treat of roughly 3.5. Succeeded in 46% confuses the risk ratio with a success rate; intraosseous success was about 80%. The block failed in 54% swaps the relative reduction for the comparator's failure rate, which was 49%. Fell by 46 percentage points reads the ratio as an absolute difference. Trap: the 46 percentage-point drop, the classic relative-to-absolute slip.
Which feature of this review most limits how confidently the headline effect can be applied?
A small evidence base with mixed agents and devices, and a failure definition the abstract does not state, is exactly the limit the authors name. High heterogeneity is the opposite of what was found: I² was 0%. Lidocaine proved ineffective over-reads a subgroup whose interval crossed 1 (0.16 to 1.07) with no significant interaction between agents. Rescue rather than primary misstates the design; every trial used intraosseous injection as the opening technique. Trap: the lidocaine over-read, turning a wide subgroup interval into a negative verdict.