DentAI – All-on-4: A Clinical Summary

An edentulous patient with insufficient posterior bone who declines regenerative surgery is the primary All-on-4 candidate: four implants anteriorly, two axial and two distal tilted, with a fixed provisional loaded the same day. This summary is built on Soto-Peñaloza’s 2017 systematic review and walks from patient selection through prosthetic and biologic complications.

Treatment concept

In the All-on-4 protocol, four implants are placed in the anterior zone of an edentulous jaw. The two anterior implants are axial; the two posterior implants are tilted distally. Distal tilt moves the implant exit farther back, shortens the cantilever, and lets the prosthesis cover up to 12 teeth. The whole design maximizes residual bone so sinus lift and bone graft can be avoided, lowering morbidity, cost, and treatment time. A fixed provisional is immediately loaded on those four implants.

Patient selection

The protocol is designed for a fully edentulous jaw, or a jaw with only a few hopeless teeth planned for extraction in the same visit. Posterior bone status drives the choice. A patient who lacks enough posterior bone for conventional implant placement and declines sinus lift or bone graft is the main candidate; several studies list that reluctance to regenerative surgery as an explicit indication. On systemic health, most studies enrolled only ASA I–II patients—healthy or with mild, controlled disease.

Practical selection in this protocol is about bone dimensions, not density. Bone quality was assessed with Lekholm & Zarb in some studies, but only during drilling and not as a preoperative gate. In contrast, two studies set minimum interforaminal dimensions: one required at least 6 mm height; another required more than 5 mm width and more than 8 mm height.

To predict surgical difficulty, Lopes stratified residual ridge width: more than 5 mm as low difficulty, irregular 4–5 mm as moderate, and irregular less than 4 mm as high.

One study used the Cawood & Howell classification of residual ridge resorption. Class III is still a broad convex ridge; class IV is a knife-edge ridge with adequate height but inadequate width; class V is a flat ridge with inadequate height and width; class VI is a depressed ridge with basal bone loss. That study found All-on-4 safe and effective in classes IV, V, and VI—exactly the atrophic jaws where conventional implant placement without grafting is not feasible.

Surgery

Anesthesia and sedation

All studies used local infiltration anesthesia; some added oral or intravenous sedation. No sedation-related complication was reported for this protocol, but one pharmacologic note matters. Benzodiazepines such as triazolam are metabolized in the liver via CYP3A, so inhibitors of that pathway raise drug levels and deepen and prolong sedation. The most important inhibitors are macrolide antibiotics (erythromycin, clarithromycin), azole antifungals (ketoconazole, itraconazole), and protease inhibitors. Grapefruit juice, cyclosporine, and calcium-channel blockers (nifedipine, verapamil, diltiazem) have a similar but milder effect. Triazolam is contraindicated in pregnancy, lactation, and in people who drink alcohol.

Incision

In both jaws a full crestal incision runs from first molar to contralateral first molar. In the maxilla some surgeons add a distal vertical releasing incision for access. When surgery is guided and flapless, no incision is made and implants are placed through a computer-designed splint.

Why the surgeon exposes sinus and nerve

The main anatomic limits are posterior: the maxillary sinus above and the mental nerve exit below. Distal tilt is meant to place the implant just anterior to these structures and along their walls, so the surgeon must see them rather than guess.

In the maxilla a small window in the anterior sinus wall exposes the mesial sinus border so the distal implant sits just in front of it and parallel to the wall. In the mandible the mental foramen is exposed so the distal implant stays anterior to it. The prosthetic payoff is direct: the posterior implant exit moves as far back as possible and the cantilever shortens.

When hopeless teeth remain, they are extracted in the same visit before implant placement, sockets are cleaned, and if needed the crest is leveled so sharp edges are removed and the prosthesis seats on a regular bed.

Surgical guides

Seventeen of 24 studies used some form of guide to improve implant position and angulation, most often the Nobel Biocare system and the All-on-Four guide. Guided surgery still has its own learning curve and complications, including guide fracture, implant loss, and low primary stability.

Distal implant angulation

Most studies used 30°, with an overall reported range of 25–45° depending on local anatomy. Collateral evidence suggests angulation differences do not meaningfully change implant survival or marginal bone loss.

Primary stability and the immediate-loading gate

Immediate loading is allowed only when the implant has enough mechanical grip at surgery—primary stability. The practical metric is insertion torque of 30–50 Ncm.

Surgeons reach that torque in two ways. First, the osteotomy is deliberately prepared slightly narrower than the implant diameter (under-preparation) so the implant locks under pressure. Second, countersinking is skipped so crestal cortical bone is preserved and provides maximum grip.

Site preparation follows the same logic: the drill sequence stops earlier or later according to bone density. In soft bone, preparation stops at smaller diameters so the implant seats with more compression; larger drills are reserved for dense bone and, even then, mainly for the cortical layer. The implant neck is placed at bone level and, when possible, engages both cortices (bicortical anchorage) to increase primary stability.

Two clinical caveats matter. Excessive torque can abrade the implant surface and release titanium particles and ions that trigger a foreign-body response. Preclinical evidence also suggests that under-drilling with high torque may reduce crestal bone–implant contact in early healing. Higher primary stability is not always a better outcome.

None of the studies used resonance frequency analysis such as Osstell, which scores stability numerically and independently. In every study the only gate for immediate loading was the number read on the ratchet while seating the implant.

Implant dimensions

The shortest length reported was 7 mm with 95.4% survival at three years; the longest was 18 mm; the average was about 10 mm. Diameters ranged from 3.3 to 5 mm.

Prosthesis

Provisional loading timing varied. Six studies loaded the same surgical day between 2 and 8 hours, three after 24 hours, and six after 48 hours. Provisionals were mostly acrylic, often reinforced with a titanium framework or titanium cylinders, with 10–12 teeth; in several studies they were built without a cantilever.

The definitive prosthesis was delivered at 4–6 months in 14 studies, at 2 months in one, and at 3 months in two. Definitive materials were mainly CAD/CAM with a titanium framework and resin teeth, metal-ceramic in some studies, and zirconia in one.

Abutments are multiunit. Anterior implants usually receive a straight or 17° abutment and distal implants a 30° abutment, with an overall angulation range of 17–35° to compensate for non-parallel implants. Only a few studies reported prosthetic screw torque, in the 10–20 Ncm range.

The occlusal goal is the same across studies: reduce load on the implants in the early months and redirect force to the anterior zone where the axial implants sit. In excursive movements, contacts stay within the intercanine zone with canine guidance or canine-and-premolar guidance—mutually protected occlusion. Centric contacts are distributed across functional units, with the cantilever kept out of contact; in one protocol that lasts at least the first three months. When the opposing arch is a complete removable denture, balanced occlusion replaces mutually protected occlusion because denture stability depends on simultaneous bilateral contacts and anterior guidance would unseat it.

Prosthesis and abutment seating are checked with panoramic and periapical radiographs using a paralleling technique.

Survival and failure pattern

Data covered 11,743 implants, of which 175 failed. Timing matters more than the raw count: 134 failures in year one, 9 by the end of year two, and 31 between years three and ten. The large majority fail in early healing, not late. The sensitive part of the case is reaching primary stability and controlling load in the first months.

Reported survival ranged from a minimum of 97.6% at 36 months to 94.8% at 10 years in 245 patients; a few small series reported 100%. The 99.8% figure in the abstract is not a pooled estimate—it belongs to one study. For counseling, a 95–98% range over five- to ten-year horizons is more accurate.

Mechanical complications

The most common prosthetic complication was acrylic prosthesis fracture, reported in nine studies. Next was detachment of an element of the definitive prosthesis (23.2% of patients in two studies), then screw loosening or fracture in several studies. Five studies reported no prosthetic complications. Most authors argue these events do not change implant or prosthesis survival and stay manageable with maintenance.

Management includes repairing the acrylic on the framework, adjusting occlusion, fabricating a nightguard, retightening screws, and advising the patient to avoid overload.

A major risk factor for prosthetic tooth fracture is the opposing arch. Natural teeth or fixed ceramic restorations raise failure through higher force and enamel abrasion. High failure is also reported when both arches are full-arch implant-supported, attributed to reduced proprioception.

Biological complications

Loss of at least one implant is the most common biologic complication; peri-implantitis, appearing after about two years, is second and was reported in six studies. Mucositis in two studies, infection of at least one implant in two, and one case of paresthesia that resolved within six months were also reported.

These events were recorded without precise definitions or criteria. Without a standard definition of success versus survival, true peri-implant disease prevalence under this protocol cannot be estimated—absence of a report is not absence of disease. A history of periodontitis is linked to mucositis, peri-implantitis, and implant loss, so why the prior teeth were extracted matters in risk assessment.

Patient satisfaction

Reported satisfaction was high. After treatment, patients were pleased with phonetic, esthetic, psychological, and functional outcomes; in one study 95.6% rated the result excellent. Still, only three of 24 studies measured satisfaction formally with a questionnaire or VAS.

Practical takeaways

All-on-4 is a predictable route for the atrophic jaw in a patient who declines regeneration, and the suitable patient is systemically healthy with enough interforaminal bone. Because most implant failures occur in year one, care should focus on primary stability and early load control.

What patients mainly face long term is prosthetic complication, not implant loss, and they should hear from the start that the prosthesis needs periodic maintenance. If the opposing arch is natural teeth or ceramic, fracture risk is higher and nightguard use plus occlusion control become more serious. Occlusion control is the core of maintenance: keep excursive contacts inside the intercanine zone and keep the cantilever out of contact in the early months.

A patient with a periodontitis history needs a regular peri-implant recall plan, because peri-implantitis is the second most common biologic complication and the available data are incomplete and poorly defined. Finally, do not quote the 99.8% figure in counseling; a 95–98% range is the more cautious and accurate statement.