Disclosure. Dr. Maloney has no commercial relationship with any clinic, marketplace, implant manufacturer, or zirconia milling laboratory in Turkey, Vietnam, Mexico, Hungary, or elsewhere. She did not receive payment, travel, accommodation, equipment, or any other consideration in connection with this piece. Standing disclosures are at /disclosures/. Last reviewed: 2026-05-09.
The headline product of high-volume dental tourism in 2026 is not the single implant. It is the full-arch monolithic zirconia bridge on four to six implants, quoted at $7,500–9,500 USD per arch in Antalya, $9,000–12,000 in Cancún, $11,000–14,000 in Ho Chi Minh City, and somewhere between $24,000 and $42,000 AUD per arch in Sydney for the same scope of work. The price differential is real, and it is the largest single-procedure differential in international dentistry. It is also the procedure I have retreated more international failures from in the last twenty-four months than any other.
I will not tell you the procedure is wrong. The procedure is right for a defined patient population, and Maló’s original 2003 case series on the immediate-function four-implant concept in the edentulous mandible — the procedure most patients now know as “All-on-4” — is one of the genuinely paradigm-shifting clinical papers of modern implant dentistry. A patient with a long-edentulous lower jaw, adequate anterior bone, no parafunction, no untreated periodontal-disease history, and a realistic understanding of what a fixed full-arch prosthesis can and cannot do is a patient for whom the procedure can be the right answer.
What I am telling you is that the patient who is offered this procedure on a one-week trip to Antalya for $8,000, after a thirty-minute video consult, on the basis of a panoramic radiograph and a holiday photograph, is being offered the right procedure for a different patient. The cases I retreat are almost never failures of the implant. They are failures of the prosthetic geometry. The arithmetic of the prosthesis — cantilever length, occlusal scheme, vertical dimension, framework rigidity, screw-access alignment, milled-fit tolerance — is what determines whether a four-implant full-arch survives at five years or fractures at eighteen months. The arithmetic does not change in Antalya. The price is what changes. The procedure that the price is buying is, in a measurable fraction of the cases, no longer the procedure the Maló paper described.
This piece is not a country review. It is a procedure review. The country in front of the procedure is incidental to the question of whether the prosthetic geometry has been worked out. I have seen the failure pattern in cases originating from Turkey, Mexico, Hungary, Thailand, Vietnam, and — less frequently but not zero — Australia. The failure pattern is procedure-shaped, not country-shaped. The reason it concentrates at the $8,000 Antalya price point is structural rather than geographical: at that price point, the time and the prosthetic infrastructure required to do the procedure correctly do not survive the throughput economics. I will say what the procedure is, when it is the right answer, when it is the wrong answer, and what the patient considering the headline price needs to ask before she pays the deposit.
What zirconia full-arch actually is
A monolithic zirconia full-arch prosthesis is a single CAD/CAM-milled bridge, made from a sintered block of yttria-stabilised tetragonal zirconia polycrystal — the material that Wikipedia describes under zirconium dioxide — that replaces every tooth in one jaw and is screw-retained to four, five, or six titanium implants. The implants are placed in the available bone of the maxilla or mandible; the bridge is screwed into the implants through small access holes drilled vertically through the prosthesis from the occlusal surface; the access holes are then sealed with composite. Zirconia is hard, dense, biologically inert, optically opaque enough to mask underlying titanium, and mechanically suited to monolithic milling at a thickness that resists fracture under occlusal load.
The procedure family is sometimes called All-on-4 (four implants per arch), All-on-6, or — generically — fixed-detachable full-arch implant-supported prosthesis. The Wikipedia entry on All-on-4 describes the original Maló protocol accurately. The variant that dominates the dental-tourism market in 2026 is the four-implant version with two anterior implants placed axially and two posterior implants tilted distally to engage the maxillary tuberosity or the area mesial to the mental foramen, supporting a monolithic zirconia bridge with a posterior cantilever extending the prosthesis to the second-molar position.
Each part of that sentence is doing clinical work. The four-implant version reduces fixture count and surgical time. The tilting reduces the need for sinus augmentation in the posterior maxilla. The cantilever extends the occlusal table without requiring more implants. The monolithic zirconia eliminates the porcelain-veneered ceramic layer that historically chipped on framework-supported prostheses. Each of these is, taken individually, a real engineering improvement over the older protocols. Taken together, they produce a procedure that is faster, cheaper, and more reproducible than the older alternatives — provided the prosthetic geometry is worked out correctly for the specific patient.
The prosthetic geometry is the question.
When the procedure is the right answer
There is a defined patient profile for whom the published evidence supports zirconia full-arch as a defensible, well-evidenced choice. I am giving the strongest version of the case before I give the cautious version, because that is the order in which a patient should hear it.
Long-term edentulous patient, lower jaw, adequate anterior bone. The original Maló series was specifically the edentulous mandible. The bone in the anterior mandible between the mental foramina is among the most reliable bone in the human jaw for implant integration, and the four-implant configuration in this region has the longest follow-up evidence behind it. Survival rates above 95% at five years are routine in published series in this configuration in this jaw.
Patient who has rejected, failed, or cannot tolerate a removable lower denture. The fixed-prosthesis pathway is not the first-line answer to edentulism. Conventional dentures are. The All-on-X procedure is one of three escalation pathways from conventional dentures (the others being implant-retained overdentures and bar-retained overdentures), and it is specifically the right escalation for the patient who needs the prosthesis fully fixed and is prepared to accept the cleaning regime, the cost, and the maintenance load that a fixed prosthesis brings.
Patient with no untreated periodontal disease, no untreated bruxism, no uncontrolled diabetes, no active smoking habit. Each of these is an independent risk factor for implant failure, and the published survival data on All-on-X is overwhelmingly drawn from populations in which these factors were controlled or excluded. The patient who is told “your gum disease has been treated, you are stable, your sugars are under 7%, you have stopped smoking, and you do not grind” is the patient the data describes. The patient who is offered the same procedure with active periodontal pockets, an A1c of 9, a pack-a-day habit, and bruxism that has flattened her remaining teeth is being offered a procedure in a population the data does not describe.
Patient who has had a treatment-planning conversation that included the alternatives. A defensible plan considers and documents why the patient is not having a removable overdenture, why the patient is not having implant-retained anchors with a removable prosthesis, why the patient is not having a longer-span fixed bridge on more implants. If the plan is the same plan every patient walks out with, the plan is not a plan. It is a product.
For this patient, in this configuration, the procedure works, has long-term evidence, and produces functional, aesthetic, and biological outcomes the patient is generally satisfied with. The Pjetursson 2014 systematic review of long-term outcomes of full-arch implant-supported fixed prostheses, and the Sulaiman 2020 review of monolithic zirconia performance, are the two bodies of evidence I lean on for the strong version of this case. I am not arguing against the procedure. I am arguing about who it is for.
When the procedure is the wrong answer
The wrong-answer version is not always a wrong-procedure version. Sometimes the procedure is right and the execution is wrong. Sometimes the procedure is wrong on the patient and the execution would have been wrong regardless. Both produce failures, and the patient cannot tell, from the outside, which kind she is going to get.
There are five wrong-answer patterns I see repeatedly in returned-from-overseas cases. I will name each, name what the published evidence says about it, and name what the prosthetic geometry would have looked like in the right-answer version.
Wrong answer 1 — The patient who could have kept teeth
The single most common wrong-answer pattern is the patient who arrived at the clinic with restorable teeth and left without them. A full-arch reconstruction begins with full-arch extraction. Once the teeth are out, the decision is irreversible. The patient who arrived with eight to twelve teeth that had been written off elsewhere — periodontally compromised but not hopeless, restoratively neglected but not unsalvageable — and who leaves with all of those teeth in a kidney dish has been recruited into a procedure that the underlying dentition did not require.
I see this in roughly a third of the returned-from-overseas full-arch cases I review. The patient was told the teeth were beyond saving. The treatment-planning radiograph, when I look at it later, shows otherwise. The published case-selection guidance for All-on-X is that the procedure is for the edentulous or near-edentulous arch, with appropriate consideration of the alternatives. A patient with eight restorable teeth has not been triaged into a near-edentulous category by clinical evidence; she has been triaged into it by a treatment planner whose product was the bridge.
The right-answer alternative for this patient is a stratified plan: periodontal treatment first, restorative work where indicated, single-tooth or shorter-span bridge replacement of unrestorable teeth, removable partial as a reversible interim, and reassessment over twelve to twenty-four months. If, after that, the dentition has continued to fail, the All-on-X conversation can be had honestly, with a patient who has seen the alternatives in her own mouth.
Wrong answer 2 — The cantilever was too long
The single most common technical failure pattern in returned cases is excessive distal cantilever. The cantilever is the portion of the bridge that extends posteriorly past the most distal implant, replacing molar teeth without an implant directly underneath. Cantilever length is the variable that most strongly predicts mechanical complication rates in the published literature. The Storelli 2018 systematic review of cantilever length and prosthetic complications found that prosthetic complications increased materially with cantilever lengths above 15 mm, and that even within “biomechanically acceptable” cantilever ranges, longer cantilevers produced more screw loosening, more framework fractures, and more peri-implant bone loss adjacent to the most distal implant.
What I see in returned cases is cantilevers extended to the second-molar position to deliver the patient a prosthesis with twelve to fourteen teeth per arch — the “movie-star smile” length — when the underlying anatomy and implant configuration support a defensible cantilever to the first-molar or second-premolar position only. The patient receives the longer prosthesis because the longer prosthesis photographs better and feels, on day one, more “complete.” At month eighteen, the most distal implant is loose, the framework has fractured, or both.
The right-answer alternative is a shorter cantilever, a fifth implant in the posterior, or a six-implant configuration with two posterior cantilevers each shorter than 10 mm. Each is biomechanically defensible. The fourteen-tooth-per-arch monolithic zirconia on a four-implant configuration is, in the cases I have seen, mechanically defensible only sometimes.
Wrong answer 3 — The occlusion was registered to the holiday, not to the bite
A full-arch reconstruction requires a re-registration of the patient’s bite. The vertical dimension — the height of the face when the teeth are together — has to be reset. The position of the lower jaw relative to the upper has to be captured. The occlusal scheme — which teeth contact in which positions, in centric and in excursion — has to be designed. Each of these decisions is a documented, deliberate clinical step in a defensible plan.
The cases I retreat are generally cases in which one of the three was skipped. Vertical dimension was set by the technician at the lab, not by the clinician at the chair. Centric relation was approximated rather than registered. The occlusal scheme is an out-of-the-box milled occlusion with no consideration of the patient’s specific muscular pattern, parafunction, or pre-treatment guidance system.
The clinical signature of skipped occlusal work is recognisable. The patient reports a feeling that her bite is “off” but cannot localise it. The posterior teeth show wear facets that should not be present at six months. The TMJ shows tenderness on palpation that the patient did not have before. The framework, on radiograph, shows screw-access misalignments that betray a fit forced on day one rather than fitted to a registered occlusion. Sometimes a fixture has fractured. Sometimes the bridge has fractured at the cantilever. Sometimes a posterior implant has a 3-mm marginal bone loss around it after eighteen months — a finding the Schwarz peri-implantitis consensus paper describes as the early imaging signature of an over-loaded fixture.
The right-answer alternative requires time. Centric relation registration takes a session of its own; vertical dimension takes another; the wax-up takes another; the verification jig takes another. A patient who is in and out of a chair in three days has had none of these properly performed.
Wrong answer 4 — The bone was not what the panoramic showed
A panoramic radiograph (orthopantomogram) is a planar projection of a three-dimensional anatomy. It shows bone height in two dimensions. It does not show bone width, sinus pneumatisation in the third dimension, the buccolingual position of the inferior alveolar nerve, or the cortication of the residual ridge. A defensible plan for a posterior maxillary or posterior mandibular implant in a patient with significant atrophy requires a CBCT (cone-beam computed tomography), reviewed by the treating clinician, with the patient present, before the surgical plan is finalised.
I see returned cases where the implant has been placed adjacent to or through the inferior alveolar nerve, where the implant has missed the available bone width by 1–2 mm and is partially exposed buccally, where the implant has perforated the lingual cortex of the mandible, or where the implant in the posterior maxilla has — as I described in the Leanne Abeyance Antalya case — ended up in the sinus or the nasal cavity. These are imaging-decision failures, not surgical-skill failures. The clinician is operating on the imaging she has. If the imaging is a panoramic from a clinic that does not own a CBCT, the imaging does not contain the information the surgical decision required.
The right-answer alternative — short implants in atrophic posteriors, supported by the Lemos 2016 meta-analysis; tilted implants supported by the Del Fabbro review; or staged augmentation followed by delayed placement — is the same alternative pathway I describe in the bone-grafting decision piece. The procedure has alternatives. The plan has to consider them. If the plan is “four implants, immediate load, monolithic zirconia, three days” in every patient regardless of imaging, the plan is not considering them.
Wrong answer 5 — The patient was a bruxer
Parafunctional clenching and grinding apply repeated, off-axis loads to a prosthesis. Monolithic zirconia is harder than enamel and harder than the opposing dentition the patient had before the prosthesis was placed. The result is a prosthesis that does not wear, but transmits the load down the implant-bone interface. In a bruxer, a four-implant full-arch monolithic zirconia produces a measurable rate of fixture-level fracture, screw fracture, and peri-implant bone loss the same review evidence I have already cited describes.
A defensible plan for a bruxer either screens the patient out of monolithic zirconia (and substitutes a softer veneered framework, more implants, or a removable design) or accepts the increased risk and manages it with a hard occlusal nightguard, scheduled occlusal-adjustment recalls, and an explicit conversation with the patient about why her risk profile differs.
The cases I retreat in this category are bruxers who were never asked. The intake form did not include parafunction screening, the wear pattern on the pre-treatment dentition was not noted, the temporomandibular examination was a five-second jaw-open-and-close. The plan was monolithic zirconia regardless. At two years, two implants are loose, the prosthesis has fractured at the cantilever, and the patient has flown 9,000 km to a domestic specialist who sees the wear facets on her remaining canines and asks why the planning team did not.
What the right plan actually looks like
I am writing this section because Brand voice requires the right answer, not only the wrong answer. The patient considering a zirconia full-arch — in any country — should be able to recognise what a competent treatment plan contains, so she can recognise what she is being offered against the standard.
A competent zirconia full-arch plan, in 2026, contains the following elements. Each is a question the patient can ask. Each can be answered. None of them is exotic.
A pre-operative CBCT, reviewed in the patient’s presence, with the clinician naming the bone height in millimetres at each planned implant site, the buccolingual width, the sinus floor topography (in the maxilla) or the inferior alveolar nerve position (in the mandible), and the planned fixture length, diameter, and angle.
A documented consideration of alternatives. Why is this patient not having a removable overdenture? Why not implant-retained anchors with a removable prosthesis? Why are short or tilted implants not being substituted for sinus augmentation? Each “considered and ruled out” is a one-line answer in the plan. If they are not in the plan, the plan has been quoted.
A staged timeline. Surgical placement, four to six months of integration, prosthetic finalisation. Immediate-load protocols (bridge fitted on day one) are defensible in the right patient configuration, but immediate load to a final monolithic zirconia is not standard. The standard immediate-load pathway is a temporary acrylic or PMMA prosthesis worn during integration, replaced by the final zirconia prosthesis at four to six months. A plan that fits a final monolithic zirconia in three days is operating outside the standard pathway.
An occlusal scheme worked up and verified. Centric relation registered, vertical dimension determined, wax-up tried in, verification jig used to confirm passive fit, occlusal scheme reviewed before the final prosthesis is milled. Each step is a separate appointment.
A cantilever length within the published biomechanical envelope, documented in the plan. The number is a real number, in millimetres. The number is below 15 mm and is justified against the implant configuration.
A written aftercare protocol, naming a domestic contact and a complication pathway, signed before the deposit. The prosthesis will need maintenance. Screws will loosen. Composite occlusal-access seals will wear. The patient will need a hygienist who can clean the prosthesis. If something fractures, the patient needs to know what happens. The same conditions I described for the posterior-maxilla sinus-implant case apply here.
A clinician with documented case volume in full-arch reconstruction, supported by a prosthodontic-trained colleague or a prosthodontist directly. Surgical placement is not the same skill as full-arch prosthetic reconstruction. The clinics that do this procedure well, in any country, generally separate the two roles or have a single clinician with formal training in both.
The patient who is being offered this procedure on a one-week trip with a panoramic radiograph, an immediate-load monolithic zirconia, a cantilever-to-second-molar prosthesis, no parafunction screening, no occlusal registration, and no written aftercare contact is not being offered the procedure the published evidence describes. She is being offered a different procedure that shares a name with it.
What I would change my view on
I hold this position because of (a) the published mechanical-complication evidence on cantilever length, occlusal scheme, and prosthetic geometry, (b) the long-term survival data that holds in populations where parafunction, periodontal disease, and diabetes were screened, and (c) my own clinical caseload of returned-from-overseas full-arch cases over the last twenty-four months. The evidence that would update it:
A multi-country prospective cohort, N>500, of monolithic zirconia full-arch prostheses placed under the dental-tourism throughput model — three to seven days from extraction to final prosthesis — with five-year follow-up reporting implant survival, prosthetic complication rates, peri-implant bone loss, patient-reported function, and remediation cost, with patient subgroups stratified by parafunction status, periodontal-disease history, and pre-operative imaging completeness. No such cohort currently exists in the published literature, to my knowledge. The closest comparators are single-clinic case series at conventional throughput, which is not the population the question is about.
A demonstration that a major dental-tourism platform has begun publishing complication-rate data for the procedure, including the specific failure patterns described above (cantilever fracture, screw loosening, peri-implant bone loss adjacent to the distal implant, prosthesis-edge fracture under bruxism load), with a methodology a third party can audit.
Until then, my view is what I have written. The procedure is right for a specific patient. The $8,000 quote is the wrong answer to the right question for most of the patients who receive it, because the price is buying a different procedure than the published evidence describes.
What you can do, today, if you are the patient
If you are considering a zirconia full-arch — in any country, at any price — do five things before you pay the deposit.
First, read the bone-grafting decision piece and ask the questions in the framework it sets out. The same logic applies to the full-arch decision: the alternatives have to be considered before the headline procedure is the right answer.
Second, get a domestic specialist second opinion on the treatment plan and the imaging. Show the plan to a prosthodontist or implant surgeon in your home country. The consultation will cost $200–500 in your local currency and is the single highest-leverage protective step. The structural argument for why a domestic second opinion matters is in the dental tourism trust gap analysis.
Third, verify the cost figure against the publication’s implant cost-by-country reference, with currency, date, and what’s-included specified. A full-arch quote should specify, line by line: implant brand and number, abutment system, prosthetic material, occlusal scheme, immediate-load versus delayed-load, included follow-up appointments, and inclusions and exclusions for any required augmentation. If the quote is a single number, it is not a quote. It is a price.
Fourth, ask the seven questions above. Ask them in writing. If the answers come back as marketing copy, the plan has not been worked up. If the answers come back specific, in millimetres and in months, the plan has been worked up regardless of where the clinic is.
Fifth, read the worked example of what happens when these conditions are not met. The Leanne Abeyance case is the cleanest illustration in 2026 of a high-volume cosmetic clinic operating outside the conditions the procedure requires. The piece you are reading and the Abeyance piece are companions: one is the procedure-level account of when the price hides the geometry, the other is the case-level account of what the geometry, when ignored, looks like on the imaging eighteen months later.
I have written the same closing sentence in several pieces by now. It is the closest thing this publication has to a recommendation. The question is not where to go. The question is what the plan in front of you actually contains. A plan that contains the elements I have set out above produces, in the published evidence, a defensible outcome. A plan that does not contain them produces, in the cases I retreat, the failure pattern I have described. The country in front of the plan does not change either fact.
For the upstream reading on why the patient cannot, from a clinic’s website, tell which version of the procedure she is being offered, see the dental tourism trust gap. For the broader Turkey-teeth picture, including the prep-side failure mode that dominates the press coverage, see the honest account of the Turkey teeth phenomenon. For the posterior-maxilla case that became the BBC scan, see the Leanne Abeyance Antalya review. For the bone-grafting decision that often accompanies the full-arch plan, see why most dental implants do not need bone grafting. For a worked example of the cosmetic full-arch failure mode applied to a specific Vietnamese clinic, see the Metal Dental Clinic, Da Nang review.