Version 20/02/18

RANDOMISED CONTROLLED CLINICAL TRIAL

Cannizzaro et al Fixed-on-3 versus fixed-on-4

Immediate loading of three (fixed-on-3) versus four (fixed-on-4) implants

supporting cross-arch fixed prostheses: One-year results from a multicenter

randomised controlled trial

Gioacchino Cannizzaro, Marco Cavallari, Matteo Lazzarini, Giuseppe Purello D'ambrosio,

Gianluigi Scialpi, Salvatore Audino, Eugenio Velasco-Ortega, Daniela Rita Ippolito, Marco

Esposito

Gioacchino Cannizzaro, MD, DDS, Adjunct Professor, Department of Dentistry, Vita-Salute

University, IRCCS San Raffaele, Milan, and private practice, Pavia, Italy

Marco Cavallari, DDS, private practice, Saluzzo, Italy

Matteo Lazzarini, DDS, private practice, Forlì, Italy

Giuseppe Purello D'Ambrosio, DDS, private practice, Rivarolo Canavese, Italy

Gianluigi Scialpi, DDS, private practice, Fasano, Italy

Salvatore Audino, DDS, private practice, Soresina, Italy

Eugenio Velasco-Ortega, MD, DDS, Professor of General Dentistry and Gerodontology,

University of Seville, Seville, Spain

Daniela Rita Ippolito, DDS, Resident, Department of Biomedical and Neuromotor Sciences,

Unit of Orthodontics, University of Bologna, Bologna, Italy

Marco Esposito, DDS, PhD, Free lance researcher and Associated Professor, Department of

Biomaterials, The Sahlgrenska Academy at Göteborg University, Göteborg, Sweden

Correspondence to: Dr Marco Esposito, Casella Postale 34, 20862 Arcore (MB), Italy. E-mail:

espositomarco@hotmail.com

Keywords: dental implants, full edentulism, immediate loading, randomised controlled trial

Purpose: To evaluate the outcome of three (fixed-on-3 = Fo3) versus four (fixed-on-4 = Fo4)

implants immediately restored with metal-resin screw-retained cross-arch prostheses in fully

edentulous jaws. Materials and Methods: Forty-eight edentulous or to be rendered edentulous

patients were randomised in 6 centres (8 patients per centre) to the Fo3 group (24 patients: 12

upper and 12 lower jaws) and to the Fo4 group (24 patients: 12 upper and 12 lower jaws)

according to a parallel group design. To be immediately loaded implants had to be inserted with a

minimum torque of 40 Ncm. Outcome measures were prosthesis and implant failures,

complications and peri-implant marginal bone level changes evaluated up to 1-year post-loading.

Results: One maxillary prosthesis per group was delayed loaded because implants could not be

placed with a torque of at least 40 Ncm. Ten patients in the Fo3 group and 4 in the Fo4 group had

implants placed flapless. One-year after loading no drop-out occurred. One patient of the Fo3

group lost 3 implants versus 3 patients of the Fo4 group who lost 4 implants, the difference being

no statistically significant (risk difference = -0.0833; 95% CI: -0.2715 to 0.0969; Fisher’s exact

test P = 0.609). One mandibular Fo3 and one maxillary Fo4 prosthesis failed. Six Fo3 patients

were affected by complications versus 3 Fo4 patients (risk difference = 0.125; 95% CI: -0.1011

to 0.3401; Fisher’s exact test; P = 0.461). Both groups lost marginal bone in a statistically

significant way (0.22±0.20 mm for Fo3 patients and 0.40±0.22 mm for Fo4 patients), with Fo3

group showing less marginal peri-implant bone loss than Fo4 group (estimate of the difference =

-0.18±0.06 mm; 95% CI: -0.31 to -0.06; P=0.006). There were no differences in clinical

outcomes between the six operators. Conclusions: These preliminary results suggest that

immediately loaded cross-arch prostheses of both jaws can be supported by only three dental

implants at least up to 1-year post-loading, though longer follow-ups are definitively needed.

Conflict of interest statement: Sweden & Martina, the manufacturer of the implants used in this

investigation, supported this trial, however data belonged to the authors and by no means did the

sponsor interfere with the conduct of the trial or the publication of its results.

INTRODUCTION

When rehabilitating edentulous jaws with fixed implant-supported prostheses usually at least four

implants are placed(1) though there are several alternative options in terms of number of implants

than can be used and the position they can be placed in. The more implants are placed, the more

increase costs and difficulties to fabricate precise metal frameworks to be passively fixed on the

implants. To reduce costs, often overdentures are made on four, two and sometimes even on

single implants(2, 3). Results from a Cochrane review(4) which also evaluated the number of

implants needed to support different type of prostheses, suggested that two implants supporting

an overdenture may be sufficient to have good clinical results, however these results were based

only on two randomised controlled trials (RCTs)(5-8). These observations were reaffirmed in a

more recent systematic review(9) and in two other RCTs(10, 11). While overdentures are

considered a cheaper alternative to implant-supported fixed dental prostheses, they may require a

higher prosthetic maintenance for replacing worn-out components which in the long term might

make this treatment option not so cheap as it appeared initially(12).

From the patient perspective, it would ideal to obtain a functional fixed prosthesis within

a couple of days from implant placement, with a minimal surgical intervention, reducing

discomfort, treatment time and costs, providing the risk of implant failure is not increased.

Therefore it would be interesting to know whether it could be possible to rehabilitate fully

edentulous patients using just three, immediately loaded, implants to support a cross-arch fixed

prosthesis.

It is possible to load immediately two to four implants with overdentures without

increasing the risk for implant failures, especially in fully edentulous mandibles(13-17). More

recent evidence suggests that it is also possible to load immediately two mandibular implants

with total fixed prostheses(18-22). However all the previous mentioned trials but one(22) were

conducted by a single and the same operator only, therefore the generalization of the results to

other settings could be hazardous. It would be interesting to know whether these results could be

obtained by other operators as well.

The aim of this multicenter RCT was to compare the clinical outcome of three (test group)

versus four (control group) implants immediately restored with metal-resin screw-retained cross-

arch prostheses in edentulous jaws. The null hypothesis was that there would be no difference in

the outcomes between the two procedures, against the alternative hypothesis of a difference. This

article is written following the CONSORT statement for improving the quality of RCTs reports

(http://www.consort-statement.org/) and presents the preliminary 1-year post-loading results.

This study was designed to have a 10-year post-loading follow-up.

MATERIALS AND METHODS

Trial design

This was a multicenter-centre (6 centres) RCT of parallel group design including with two

arms, balanced randomisation and blind assessment, when possible. Patients edentulous in both

upper and lower jaws were randomised in equal numbers into an immediately loaded prosthesis to

be hold either by three (Fixed-on-3; Fig 1a-h and Fig 2a-c) or 4 (Fixed-on-4; Fig 3a-c and 4a-c)

implants.

Eligibility criteria for participants

Any edentulous patient requiring an implant-supported prosthesis, who was 18 years old

or older and able to understand and sign an informed consent, was eligible for inclusion in this

trial. Eligible patients needed to have bone volumes allowing the placement of four implants with

a diameter of at least 3.8 mm and 8.5 mm long. Immediate post-extractive implants were

included.

Patients were not accepted into the study if any of the following exclusion criteria was

present:

• general contraindications to implant surgery at discretion of the surgeon

• uncontrolled diabetes

• pregnancy or lactation

• addiction to alcohol or drugs

• psychiatric problems

• unrealistic expectations

• irradiated in the head and/or neck

• treated or under treatment with intravenous amino-bisphosphonates

• poor oral hygiene and motivation

• active infection or severe inflammation in the area intended for implant placement

• need of bone-augmentation procedures

• participation to other trials, if it could interfere with the conduct of the current trial

• unable to commit to a 10-year follow-up

• lack of opposing occluding dentition/prosthesis/dentures at the time of loading.

Patients were categorised into three groups according to what they declared: non smokers;

moderate smokers (up to 10 cigarettes per day); heavy smokers (more than 10 cigarettes per day).

Patients were also categorised according to their skeletal class into 1, 2 and 3 class; according to

the clinical appearance of their masseter muscles, into normal or hyper-developed, and according

to the type of dentition of the opposite jaw (natural/fixed or full dentures).

All patients received thorough explanations and signed a written informed consent form

prior to be enrolled in the trial. Ninety-six patients (48 with fully edentulous mandibles and 48

with fully edentulous maxillae) were to be recruited and treated in 12 Italian private practices by

12 different operators who had extensive experience in immediate loading procedures, however

only 6 centres actually provided the requested data.

Preliminary screening was performed on cone-bean computed tomography scans. In case

of metal-reinforced dentures, transparent resin denture replicas were made to be used as

individual impression trays. Patients received professional oral hygiene prophylaxis or

periodontal treatment when required prior to implantation.

All patients received prophylactic antibiotic therapy: amoxicillin 2 g orally one hour prior

to the intervention. Patients allergic to penicillin were given clindamycin 600 mg 1 hour prior to

the intervention. Patients rinsed with a chlorhexidine mouthwash 0.2% for 1 minute just prior to

any intervention. Local anaesthesia was administered using Articain with adrenaline 1:100.000.

After local anaesthesia administration, surgeons were informed whether to place 3 or 4 implants,

opening a sequentially numbered opaque sealed envelope. Flaps could be elevated at discretion of

the surgeons. Three or four implant sites were prepared according to what dictated by the random

allocation. In case of three mandibular implants one implant was placed centrally in order to

create a triangle with two sides of similar length. In maxillae, the central implants were placed

taking into consideration the patient's aesthetic demands. Position and angulation of the implants

were determined by bone anatomy but it was attempted to minimize the need and the length of

cantilevers, whenever possible. Implants were placed as straight and parallel as possible. Sites

were underprepared according to bone quality which was quantified tactically at drilling into soft,

medium and hard bone. In principle, in the presence of soft bone, sites were underprepared with

drills of two diameter size smaller than the implants to be placed. In the presence of medium

bone quality, sites were underprepared with one drill diameter size smaller. In case of hard bone,

the final drill, corresponding to the implant diameter, was used. Syra tapered, transmucosal

titanium grade 4 implants (Sweden & Martina, Due Carrare, PD, Italy) with a zirconium sand-

blasted acid etched surface and an external hexagon were used. Operators were free to choose

implant lengths (8.50, 10.00, 11.50, 13.00 and 15.00 mm) and diameters (3.80, 4.25 and 5.00

mm) according to clinical indications and their preferences. In maxillae the implant version with

the larger threads was used (Syra SL). It was attempted place implant slightly supracrestally

leaving the 1 mm machined collar in contact with soft tissues. Implants were inserted in the

prepared sites with a torque 40 N/cm and, once the motor stopped, manually with a ratchet until

seated in the proper position. At the protocol formulation phase, it was decided that implants

inserted with 40 Ncm or less torque, could not to be loaded immediately. Surgeons, at their

discretion, could either load these implants anyway, replace them with larger diameter or longer

ones, prepare alternative implant sites, or leave the implants to heal submerged for 3 months.

Single sutures were given when needed. Prosthetic procedures were initiated immediately after

implant placement and were identical for both groups. The patient’s own denture or a transparent

denture replica was trimmed and used as individual impression tray. Temporary titanium

abutments were placed on the implants and impressions were taken with Impregum F (Espe

Dental AG, Seefeld, Germany). Dentures were blocked directly to the titanium temporary

abutments with Pattern Resin (GC America Inc., Alsip, IL, USA). Once the abutment/transfers

were removed with the impression tray, healing screws were placed onto the implants. The

provisional titanium abutments were prepared on models in the dental laboratory, and metal

frameworks were cast with a sagittal section of at least 3 mm in height and 1.5 mm in width. The

frameworks were cemented onto the prepared abutments using an anaerobic self-curing cement

for metals (NOBIL-FIX, Nobil-Metal, Villafranca d'Asti, Italy). Ten to 12 composite teeth were

added and, when present, first molars were not in static or dynamic occlusion. After surgery

patients were instructed to avoid brushing at surgical sites and to rinse twice a day with 0.12%

chlorhexidine for 2 weeks. A cold and soft diet was recommended for one week. Analgesics

(ibuprofen 400 mg or in case of allergy to ibuprofen, paracetamol 1 g was be given instead) were

provided to be taken twice a day during meals on patient’s demand.

Patients were seen after two days when the definitive cross-arch prostheses were screwed

onto the implants with a standard torque of 25 Ncm. Periapical radiographs of the study implants

were according to the parallel technique. If the peri-implant marginal bone levels were

unreadable, radiographs were taken again. Patients were seen again after 2 days and again after

about one week, when sutures, if placed were removed, occlusion was rechecked and oral

hygiene instructions were given.

Patients were recalled for maintenance every 6 months and occlusion was carefully

checked at each maintenance visit. One-year after loading, periapical radiographs were taken and

the screw-retained prostheses were removed to check implant stability.

The following outcome measures were evaluated:

• Prosthesis failure: loss or replacement of the prosthesis for any reasons.

• Implant failure: presence of any mobility of the individual implant and/or any infection

dictating implant removal. Implant facture or deformation of the connections rendering

the implant unusable were also considered as failures. Implant stability was measured 1-

year post-loading, after having removed the prosthesis by tightening the implant abutment

screw with a 20 Ncm force.

• Any biological or mechanical complications.

• Peri-implant marginal bone level changes assessed on periapical radiographs taken with

the paralleling technique at implant placement and 1 year after loading. Non-digital

radiographs were scanned in TIFF format with a 600 dpi resolution, and stored in a

personal computer. Peri-implant marginal bone levels were measured using the

UTHSCSA Image Tool 3.0 (The University of Texas Health Science Center, San

Antonio, USA) software. The software was calibrated for every single image using the

known implant diameter. Measurements of the mesial and distal bone crest level adjacent

to each implant were made to the nearest 0.01 mm and averaged at patient level and then

at group level. The measurements were taken parallel to the implant axis. Reference

points for the linear measurements were: the most coronal portion of the implant neck

and the most coronal point of bone-to-implant contact.

All maintenance procedures and implant stability assessments were performed by local

independent outcome assessors, who were not aware of the nature of the study, whereas

complications were addressed by the treating clinicians. One dentist (Dr Cinzia Torchio) not

aware of the nature of the study, performed all radiographic assessments without knowing group

allocation, therefore the outcome assessor was blind, however, the difference of the number of

implants was obvious when looking at the radiographs.

A real sample size calculation was not made, however if we would expect to have 10% of

prosthesis failures in the FO4 group and 20% in FO3 group, a two group Fisher exact test with a

0.050 two-sided significance level would have 90% power to detect the difference a proportion of

0.200 and a proportion of 0.100 (odds ratio of 0.444) when the sample size in each group is 286.

In reality the sample size should have been by far larger but this was definitively above our

recruitment capabilities so it was decided to include 48 patients in each group, thought only 24

patients per group were actually treated.

Twenty-four computer generated restricted random lists (one for upper and one for lower

jaw of patients at each centre) were used to create two groups with equal numbers of patients by

Dr Marco Esposito who was not involved in patient recruitment or treatment and was the only

one to have access to the random list stored in a password-protected portable computer. The

random codes were enclosed in sequentially numbered, identical, opaque, sealed envelopes.

Envelopes were to be opened sequentially only after local anaesthesia was delivered; therefore

treatment allocation was concealed to the investigators in charge of enrolling and treating the

patients of this trial.

All data analysis was carried out according to a pre-established analysis plan. The patient

was the statistical unit of the analyses. A dentist (Dr Daniela Rita Ippolito) with expertise in

biomedical statistics analysed the data, without knowing group allocation. Fisher’s exact

probability tests were used to compare the relative numbers of patients who had a prosthesis

failure, an implant failure or a complication. With regard to continuous outcomes (bone levels)

independent-samples t-tests were used to investigate the differences in baseline measurements

and bone loss at 1 year while an analysis of covariance (ANCOVA) was used to compare the

mean radiographic values at 1 year in order to account for the baseline value (covariate). To

detect any changes in marginal peri-implant bone levels between time points (baseline and 1 year

after loading), a dependent t test was used. Differences in the proportion of patients with

prosthesis failures, implant failures and complications (dichotomous outcomes) among the 6

centres were evaluated through chi-square tests, whereas a One-way ANOVA test was used to

compare peri-implant bone loss among the centres. All statistical comparisons were conducted at

the 0.05 level of significance.

RESULTS

Seventy-seven patients were originally screened, however 29 patients were not enrolled for the

following reasons: 18 patients did not want to participate in study; 5 patients because under

treatment with oral anticoagulants for severe conditions; 4 patients did not really understood what

the study was about; 1 patient did not wanted to have an immediate loading procedure; 1 patient

planned to emigrate after 6 months.

Forty-eight patients (24 with edentulous mandibles and 24 with edentulous maxillae)

were consecutively recruited and treated from February 2015 to February 2016 and were

randomised just after delivery of local anaesthesia: 24 to Fo3 group (12 with edentulous

mandibles and 12 with edentulous maxillae) and 24 to the Fo4 group. All patients were treated

according to the allocated interventions. No patient dropped out and the data of all patients were

evaluated in the statistical analyses.

The following deviations from the operative protocol occurred:

• In the Fo3 group, 2 patients had 1 implant each placed with a torque inferior to 40 Ncm

but were immediately loaded anyhow.

• In the Fo4 group, 7 patients had one or 2 implants (9 implants in total) placed with a

torque inferior to 40 Ncm but were immediately loaded anyhow.

The follow-up focused on the time between implant placement and 1-year post-loading.

The main baseline patient characteristics are presented in Table 1. There were no apparent

differences in baseline characteristics between the two group except for implant length which was

higher in the Fo3 group (11.65±1.16 mm) than in the Fo4 group (10.85±1.18 mm). At placement

16 implants in 11 patients (5 implants in 3 Fo3 patients and 11 implants in 8 Fo4 patients) did not

achieve the minimal insertion torque required (more than 40 Ncm). In particular, 4 implants in 2

Fo4 patients (2 in lower and 2 in upper jaws) were replaced by larger diameters ones without

obtaining yet the requested 40 Ncm insertion court. Of these implants: 11 implants in 9 patients

(2 implants in 2 Fo3 patients and 9 implants in 7 Fo4 patients) were loaded anyhow; finally the

loading of two maxillary prostheses in patient of each group who had 5 implants inserted with

less than 40 Ncm was delayed of about 3 months. Bone quality, implant lengths and diameters

are presented in Table 2.

Prosthesis failures: One prosthesis per group failed. There were no differences for

prosthesis failures between the two groups (risk difference = 0.0; 95% CI: -0.1644 to 0.1644;

Fisher’s exact test P = 1.000). One Fo3 patient lost all 3 implants and his mandibular prosthesis

60 days after loading, whereas one Fo4 patient lost 2 implants and her maxillary prosthesis 60

days after loading.

Implant failures: One patient of the Fo3 group lost 3 implants versus 3 patients of the Fo4

group who lost 4 implants, the difference being no statistically significant ((risk difference = -

0.0833; 95% CI: -0.2715 to 0.0969; Fisher’s exact test P = 0.609). The patient of the Fo3 group

lost all 3 mandibular implants 2 months after loading. He referred that all implants were mobile

soon after their placement but he never attended the planned controls. Wide bony craters with pus

were noticed at implant removal. In the Fo4 group, one patient lost 2 maxillary implants after 2

months. The patient was a heavy smoker with hypertrophic masseters and skeletal class 3 and one

of the 2 implants was placed in soft bone with a torque inferior to 40 Ncm. after 3 weeks the

patient did not feel real pain, but he felt the prosthesis bending when biting bread but she did not

go to her dentist until she felt a stronger pain. Another Fo4 patient lost one mandibular post-

extractive implant 6 months after loading. She just noticed a moderated pain at mastication. The

prosthesis was transformed into a Fo3. The last Fo4 patient lost one distal maxillary implants 10

months after loading. The prosthesis was shortened removing the 2 premolars from the side of the

failed implant and one on the controlateral side and a replacement implant was placed.

Complications: Six Fo3 patients were affected by complications versus 3 Fo4 patients, the

difference being not statistically significant (risk difference = 0.125; 95% CI: -0.1011 to 0.3401;

Fisher’s exact test; P = 0.461). Complications are described in Table 3 and were all successfully

treated.

Peri-implant marginal bone levels: There was no statistically significant difference

between the two groups for peri-implant bone levels at implant placement (difference = -

0.02±0.05 mm; 95% CI: -0.130 to 0.083 mm; P = 0.655). Both groups gradually lost marginal

peri-implant bone in a statistically significant way (P<0.001; Table 4). One year post loading,

Fo3 implants lost an average of 0.22±0.20 mm peri-implant bone versus 0.40±0.22 mm of the

Fo4. The difference in the bone loss at 1 year between the groups was statistically significant

(difference: -0.18±0.06 mm; 95% CI: -0.31 to -0.06; P=0.006; Table 4).

There were no statistically significant difference by comparing the outcomes of the 6

different centres (Table 5).

DISCUSSION

This RCT was designed to evaluate whether 3 immediately loaded implants supporting

cross-arch fixed prostheses could be reliable treatment option to prostheses supported

conventionally by 4 implants in both upper and lower jaws. The 1-year findings of the present

trial are encouraging since only one prosthesis per group was lost. The present results closely

match those of other two studies where cross-arch prostheses were loaded on 3 implants only(19,

20). Even mandibular cross-arch prostheses supported by only 2 immediately loaded implants

were shown to be successful up to 5 years if fabricated with robust metal frameworks(18, 22).

These preliminary results, taken all together, suggest that the clinical outcome up to 1-year of Fo3

prostheses appears not to be dramatically different from those of prostheses supported by four or

more implants, thought longer follow-up are definitively needed to avoid unpleasant surprises

most likely of biomechanical nature.

The only difference observed at baseline between the 2 groups was related to implant

length. On average, Fo3 implants were 0.8 mm shorter than Fo4 implants. This difference, which

may not have relevant clinical implications, might be easily explained by the tendency of

operators to place slightly longer implants when their number was reduced.

The only statistically significant difference observed comparing the 2 groups was that Fo3

implants lost 0.18 mm less bone that Fo4 implants one year after loading. While it is difficult to

provide a convincing explanation for this observation, it could be argue than a difference o less

than 0.2 mm in bone loss may not have a major clinical impact.

At protocol level, it was decided to load immediately only those implants which could be

placed with an insertion torque greater than 40 Ncm, even though 8 implants that achieved an

insertion torque around 35 Ncm were immediately loaded anyway. To achieve a high insertion

torque, implant sites were underprepared to a various degree according to bone quality,

nevertheless in poor bone quality, especially in maxillae, it was not possible to achieve

consistently torque values above 40 Ncm. However, only for 2 patients, one per group, the

loading of their maxillary prostheses was delayed since the requested minimum torque of 40 Ncm

could not be obtained. A high insertion torque is likely be one of the most relevant factors which

may explain the good results of the present trial(23, 24).

The major limitations of the present study are the low number of included patients, too

low to be able to detect possible statistically significant differences in prosthesis/implant failures

and the short duration of the follow-up. Regarding the duration of the follow-up it was planned to

follow these patients for 10 years so in some years we should have some additional useful

information. This study was designed to investigate whether the Fo3 treatment concept was a

viable treatment option. Unfortunately, half of the centres who accepted to join the trial did not

recruit any patients. At trial initiation (and still now) we were unable to predict with reasonable

certainty the long-term outcome of the Fo3 rehabilitations therefore it seemed prudent not to

recruit a larger patient population into the trial.

Both procedures were tested in real clinical conditions, using relatively broad patient

inclusion criteria and several experience operators, therefore the generalisation (external validity)

of the present findings to other settings can be done with more confidence since the operators

obtained similar results, however it should also be considered that the follow-up is yet too short

and longer follow-ups are needed.

CONCLUSIONS

These preliminary 1-year results suggest that both immediately loaded mandibular and maxillary

cross-arch fixed prostheses can be successfully supported by only three dental implants, however

longer follow-ups are needed.

Table 1: Patients’ and interventions’ characteristics (n = patients).

Fo3 (n=24) Fo4 (n=24)

Females 9 (37.5%) 11 (45.8%)

Mean age at implant insertion (range) 57.5 years (39-78) 59.3 years (36-78)

Skeletal class 1 18 (75.0%) 19 (79.2%)

Skeletal class 2 3 (12.5%) 3 (12.5%)

Skeletal class 3 3 (12.5%) 2 (8.3%)

Non-smokers 18 (75.0%) 13 (54.2%)

Smoking up to 10 cigarettes/day 4 (16.7%) 9 (37.5%)

Smoking more than 10 cigarettes/day 2 (8.3%) 2 (8.3%)

Patient with hypertrophic masseters 5 (20.8%) 7 (29.2%)

Patients wearing dentures in the opposing jaw at loading 12 (50.0%) 14 (58.3%)

Total number of inserted implants 72 96 Implants inserted in fresh extraction sockets 15 (20.8%) 21 (21.9%)

Patients who had implants placed flapless 10 (41.7%) 4 (16.7%)

Patients who had implants inserted with <40 Ncm 3 (12.5%) 8 (33.3%)

Patients who had implants inserted with <40 Ncm but immediately loaded anyhow

2 (8.3%) 7 (29.2%)

Patients who had implants delayed loaded 1 (4.2%) 1 (4.2%)

Distal cantilever mean length in mm 10.9 mm 12.5 mm

Table 2: Bone quality, implant length and diameters (n = implants).

Bone quality Fo3 (n=72) Fo4 (n=96)

Soft 5 (6.9%) 9 (9.4%) Medium 44 (61.1%) 56 (58.3%)

Hard 23 (31.9%) 31 (32.3%) Implant length

8.5 mm 1 (1.4%) 10 (10.4%) 10,0 mm 15 (20.8%) 29 (30.2%)

11.5 mm 0 (0.0%) 1 (1.0%)

13.0 mm 32 (44.4%) 48 (50.0%) 15.0 mm 24 (33.3%) 8 (8.3%)

Implant diameter

3.80 mm 47 (65.3%) 65 (67.7%)

4.25 mm 19 (26.4%) 26 (27.1%) 5.00 mm 6 (8.3%) 5 (5.2%)

Table 3: Complications and their treatment up to 1 year after implant loading by study group and

jaw location. All complications successfully treated (n = patients).

Complication type Fo3

(n=24)

Fo4

(n=24) Fenestration at implant placement - not treatment 1 mand 0 Post-implantation haemorrhage - 1 suture 1 max 0

Pain at chewing few days post-loading - removed cantilevers & byte for 4 months 1 max 0

Complain for too wide hygienic interdental spaces - closed on patient's request 0 1 mand Distal screw loosening twice in 3 weeks 1 month after loading - screw replaced 1 max 0

Teeth detached 4 times 3 to 4 months after loading - remade framework lining in composite 1 mand 0

Pain due to failure of 2 implants 2 months after loading - implants removed 0 1 max Peri-implant mucositis 3 months after loading - prosthesis recontouring 0 1 max

Prosthesis tooth detachment 7 months after loading - reattached 1 max 0

Totals 6 3

Table 4: Mean radiographic peri-implant marginal bone levels between groups and time periods

and mean changes in peri-implant marginal bone levels at different times.

Implant placement 1 year postloading Baseline-1 year postloading P-value*

N Mean (SD) N Mean (SD) N Mean (SD) 95% CI

Fo3 23 0.24 (0.20) 23 0.46 (0.30) 23 0.22 (0.20) 0.13 to 0.30 <0.001

Fo4 22 0.25 (0.17) 22 0.64 (0.27) 22 0.40 (0.22) 0.30 to 0.49 <0.001

Difference (SE); 95% CI -0.02 (0.05); -0.130 to 0.083 -0.19 (0.08); -0.36 to -0.02 -0.18 (0.06) -0.31 to -0.06

P-value for groups 0.655** 0.006*** 0.006**

*Dependent t-test; **Independent-samples t-test; ***Analysis of covariance at 1-year postloading with baseline value as a covariate.

Table 5: Comparison of outcomes between the 6 different centres at 1 year after loading

Cannizzaro Cavallari Lazzarini Purello Scialpi Audino P value

Patients with prosthesis failures 0/8 0/8 0/8 0/8 1/8 1/8 0.525*

Patients with implant failures 0/8 0/8 1/8 1/8 1/8 1/8 0.823* Patients with complications 3/8 1/8 1/8 2/8 0/8 2/8 0.478*

Peri-implant bone loss (mm) 0.39±0.28 0.27±0.19 0.30±0.22 0.39±0.29 0.21±0.10 0.25±0.22 0.565** * chi-square test; ** One-Way ANOVA

FIGURE LEGENDS

Fig 1a-h: Sequence of treatment of one of the mandibular patients treated by Dr Cannizzaro

randomly allocated to the Fo3 group: a) preoperative view; b) initial preparation with the 3.8 mm

drill, directly into the mucosa; c) in the presence of soft bone the implant site was underprepared

of two diameter drill sizes than the implant to be placed; c) view of the first prepared site; d)

implant placement of the implant in central position; e) all the 3 implants are positioned flapless;

f) within 8 hours a metal-resin screw-retained prosthesis is fabricated and delivered; g) control

ortopantomograph at prosthesis delivery; h) ortopantomograph at 1-year after loading.

Fig 1a

Fig 1b

Fig 1c

Fig 1d

Fig 1e

Fig 1f

Fig 1g

Fig 1h

Fig 2a-c: Panoramic orthopantomographs of one of the maxillary patients treated by Dr

Cannizzaro randomly allocated to the Fo4 group: a) preoperative view; b) at delivery of the

immediate prosthesis; c) at 1-year after loading.

Fig 2a

Fig 2b

Fig 2c

Fig 3a-c: Panoramic orthopantomographs of one of the mandibular patients treated by Dr

Cannizzaro randomly allocated to the Fo4 group: a) preoperative view; b) at delivery of the

immediate prosthesis; c) at 1-year after loading.

Fig 3a

Fig 3b

Fig 3c

Fig 4a-c: Panoramic orthopantomographs of one of the maxillary patients treated by Dr

Cannizzaro randomly allocated to the Fo4 group: a) preoperative view; b) at delivery of

the immediate prosthesis; c) at 1-year after loading.

Fig 4a

Fig 4b

Fig 4c

REFERENCES

1. Brånemark P-I, Hansson BO, Adell R, Breine U, Lindström J, Hallén O, et al. Osseointegrated implants in the treatment of the edentulous jaw. Experience from a 10-year period. Stockholm: Almqvist & Wiksell International, 1977. 2. Cordioli G, Majzoub Z, Castagna S. Mandibular overdentures anchored to single implants: a five-year prospective study. The Journal of Prosthetic Dentistry 1997;78:159-165. 3. Alsabeeha NH, Payne AG, De Silva RK, Thomson WM. Mandibular single-implant overdentures: preliminary results of a randomised-control trial on early loading with different implant diameters and attachment systems. Clinical Oral Implants Research 2011;22:330-337. 4. Coulthard P, Esposito M, Jokstad A, Worthington HV. Interventions for replacing missing teeth: surgical techniques for placing dental implants. Cochrane Database of Systematic Reviews 2003: Chichester, UK: John Wiley & Sons, Ltd.

5. Batenburg RHK, Raghoebar GM, Van Oort RP, Heijdenrijk K, Boering G. Mandibular overdentures supported by two or four endosteal implants. A prospective, comparative study. International Journal of Oral and Maxillofacial Surgery 1998;27:435-439. 6. Wismeijer D, van Waas MAJ, Mulder J, Vermeeren JIJF, Kalk W. Clinical and radiological results of patients treated with three treatment modalities for overdentures on implants of the ITI Dental Implant System. A randomized controlled clinical trial. Clinical Oral Implants Research 1999;10:297-306. 7. Wismeijer D, van Waas MAJ, Vermeeren JIJF, Kalk W. Patients' perception of sensory disturbances of the mental nerve before and after implant surgery: a prospective study of 110 patients. The British journal of oral & maxillofacial surgery 1997;35:254-259. 8. Wismeijer D, Van Waas MAJ, Vermeeren JIJF, Mulder J, Kalk W. Patient satisfaction with implant-supported mandibular overdentures. A comparison of three treatment strategies with ITI-dental implants. International Journal of Oral and Maxillofacial Surgery 1997;26:263-267. 9. Klemetti E. Is there a certain number of implants needed to retain an overdenture? Journal of Oral Rehabilitation 2008;35 Suppl 1:80-84. 10. Visser A, Raghoebar GM, Meijer HJ, Batenburg RH, Vissink A. Mandibular overdentures supported by two or four endosseous implants. A 5-year prospective study. Clinical Oral Implants Research 2005;16:19-25. 11. Meijer HJ, Raghoebar GM, Batenburg RH, Visser A, Vissink A. Mandibular overdentures supported by two or four endosseous implants: a 10-year clinical trial. Clinical Oral Implants Research 2009;20:722-728. 12. Palmqvist S, Sondell K, Swartz B. Implant-supported maxillary overdentures: Outcome in planned and emergency cases. The International Journal of Oral and Maxillofacial Implants 1994;9:184-190. 13. Chiapasco M, Abati S, Romeo E, Vogel G. Implant-retained mandibular overdentures with Brånemark System MKII implants: a prospective comparative study between delayed and immediate loading. The International Journal of Oral and Maxillofacial Implants 2001;16:537-546. 14. Romeo E, Chiapasco M, Lazza A, Casentini P, Ghisolfi M, Iorio M, et al. Implant-retained mandibular overdentures with ITI implants. Clinical Oral Implants Research 2002;13:495-501. 15. Turkyilmaz I, Tumer C. Early versus late loading of unsplinted TiUnite surface implants supporting mandibular overdentures: a 2-year report from a prospective study. Journal of Oral Rehabilitation 2007;34:773-780. 16. Cannizzaro G, Leone M, Esposito M. Immediate versus early loading of two implants placed with a flapless technique supporting mandibular bar-retained overdentures: a single-blinded, randomised controlled clinical trial. European Journal of Oral Implantology 2008;1:33-43. 17. Esposito M, Grusovin MG, Maghaireh H, Worthington H. Interventions for replacing missing teeth: different times for loading dental implants. Cochrane Database of Systematic Reviews 2013;2013: Chichester, UK: John Wiley & Sons, Ltd. 18. Cannizzaro G, Felice P, Lazzarini M, Ferri V, Leone M, Trullenque-Eriksson A, et al. Immediate loading of two flapless placed mandibular implants supporting cross-arch fixed prostheses: A 5-year follow-up prospective single cohort study. European Journal of Oral Implantology 2016;9:165-177. 19. Cannizzaro G, Felice P, Loi I, Viola P, Ferri V, Leone M, et al. Immediate loading of bimaxillary total fixed prostheses supported by five flapless-placed implants with machined

surfaces: A 6-month follow-up prospective single cohort study. European Journal of Oral Implantology 2016;9:67-74. 20. Cannizzaro G, Gastaldi G, Gherlone E, Vinci R, Loi I, Trullenque-Eriksson A, et al. Two or three machined vs roughened surface dental implants loaded immediately supporting total fixed prostheses: 1-year results from a randomised controlled trial. Eur J Oral Implantol 2017;10:279-291. 21. Cannizzaro G, Loi I, Viola P, Ferri V, Leone M, Trullenque-Eriksson A, et al. Immediate loading of two (fixed-on-2) versus three (fixed-on-3) implants placed flapless supporting cross-arch fixed prostheses: One-year results from a randomised controlled trial. European Journal of Oral Implantology 2016;9:143-153. 22. Cannizzaro G, Felice P, Gherlone E, Barausse C, Ferri V, Leone M, et al. Immediate loading of 2 (fixed-on-2) versus 4 (fixed-on-4) implants placed with a flapless technique supporting mandibular cross-arch fixed prostheses: 3-year results from a pilot randomised controlled trial. European Journal of Oral Implantology 2017;10:133-145. 23. Ottoni JM, Oliveira ZF, Mansini R, Cabral AM. Correlation between placement torque and survival of single-tooth implants. International Journal of Oral and Maxillofacial Implants 2005;20:769-776. 24. Cannizzaro G, Leone M, Ferri V, Viola P, Gelpi F, Esposito M. Immediate loading of single implants inserted flapless with medium or high insertion torque: a 6-month follow-up of a split-mouth randomised controlled trial. European Journal of Oral Implantology 2012;5:333-342.