CLINICAL

RESEARCH
Immediate Versus Delayed Loading of
Dental Implants Supporting Fixed
Full-Arch Maxillary Prostheses:
A 10-year Follow-up Report
Paolo Pera, MD, DDS, PhD
Maria Menini, DDS, PhD
Paolo Pesce, DDS, PhD
Marco Bevilacqua, DDS
Francesco Pera, DDS, PhD
Tiziano Tealdo, DDS, MS, CDT

Division of Implant and Prosthetic Dentistry, Department of Surgical Sciences (DISC),

University of Genoa, Genoa, Italy.

Purpose: To compare clinical outcomes of immediate vs delayed implant loading in edentulous maxillae
with full-arch fixed prostheses. Materials and Methods: Two patient groups were identified for this study:
(1) the test group (TG), which included 34 patients (19 women, 15 men; mean age 56.7 years) treated with
the Columbus Bridge Protocol with 4 to 6 postextractive implants loaded within 24 hours (163 implants total);
and (2) the control group (CG), which included 15 patients (6 women, 9 men; mean age 59.96 years) treated
with a traditional two-stage delayed loading rehabilitation using 6 to 9 implants inserted in healed sites (97
implants total). All patients were rehabilitated with full-arch fixed prostheses in the maxilla. Results: At the
10-year follow-up, no difference in the implant cumulative survival rate between the TG (93.25%) and CG
(94.85%) was found. Mean bone loss was significantly lower in the TG (mean: 2.11 mm) compared to the
CG (mean: 2.65 mm). All original prostheses were maintained and functioning satisfactorily. Conclusion:
Maxillary full-arch immediate loading represents a valid alternative to the traditional delayed loading
rehabilitation. Int J Prosthodont 2019;32:27–31. doi: 10.11607/ijp.5804

A
reduced number of implants, including tilted distal implants, are used to
support immediately loaded prostheses in the edentulous maxilla. However,
long-term controlled studies of this particular clinical protocol are lacking.1,2
The present authors have reported 3- and 6-year outcome experiences3,4 using a simi-
lar approach, specifically described as the Columbus Bridge Protocol (CBP), preceding
this 10-year follow-up report comparing CBP outcomes to those observed with a
delayed implant loading protocol.

MATERIALS AND METHODS

Correspondence to:
Between September 2005 and January 2006, a consecutive cohort of 49 patients Dr Maria Menini
comprising 25 women (mean age 54.8 years) and 24 men (mean age 61.5 years) Division of Implant and Prosthetic
with edentulous maxillae or compromised maxillary dentitions referred to the Division Dentistry, Department of Surgical
Sciences (DISC - Pad. 4)
of Implant and Prosthetic Dentistry (Department of Surgical Sciences) of Genoa Ospedale S. Martino
University was selected for this study. Enrollment inclusion criteria included a strong L. Rosanna Benzi 10
desire to be treated with full-arch implant-supported fixed prostheses, presence of 16132 Genova (Italy)
Fax: 0039 0103537402.
good general health, and no contraindications for undergoing oral surgery. Past his- Email: maria.menini@unige.it
tory of smoking or parafunctional habits did not disqualify any patient. Smokers were
counseled to give up smoking, but a smoking-cessation protocol was not demanded Submitted January 17, 2018;
accepted July 23, 2018.
(none of the included patients were heavy smokers [> 20 cigarettes/day]). Additional ©2019 by Quintessence
patient characteristics are listed in Table 1. Publishing Co Inc.

Volume 32, Number 1, 2019 27

© 2019 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Clinical Research

Table 1   Population Characteristics at Baseline periodontal probe) were evaluated

Test group Control group after removal of the fixed screw-
Population characteristics retained prosthesis. A mixed-effects
Total patients at baseline, n 34 15 logistic regression model was used
Patients followed up through- 31 11 to statistically compare the PI and
out the study period (10 y), n BoP between the two groups, with
Mean age at baseline (y) 56.7 59.96 these two characteristics used as in-
Women, n (%) 19 (55.8) 6 (40) dependent variables and the group
Smokers, n (%) 12 (35.3) 6 (40) representing the binary dependent
Edentulous patients, n (%) 0 (0) 15 (100) variable. For MBL and PD, two dis-
Dental implant characteristics tinct mixed-effects linear models
Implant type Osseotite and Osseotite NT Osseotite and Osseotite NT were used, with the group as the in-
Implant surface DAE (Osseotite) DAE (Osseotite) dependent variable and mean MBL
Implant length (mm) 13–18 10–15 and PD as the dependent variables.
Implant diameter (mm) 4 4 Log-rank test was used to com-
Placement protocol Immediate or delayed Delayed pare implant CSRs between the two
(postextractive) (healed bone)
groups. A P value < .05 was consid-
Total implants at baseline, n 163 97
ered statistically significant. Stata
Mean no. of implants per 4.8 (4–6) 6.5 (6–9)
patient (range)
(v. 14) was used for the computation.
Prosthodontic characteristics
Loading time Within 24 h Mean 8.75 mo
RESULTS
Framework Cast palladium alloy Cast palladium alloy
Veneering material Acrylic/composite resina Composite resin
All patients were followed for at least
Retention Screw-retained Screw-retained
10 years after surgery for a mean
Opposing dentition Natural dentition (n = 14), Natural dentition (n = 1),
follow-up interval of 10.2 years.
natural + fixed implant natural + fixed implant Three TG patients (12 implants)
restoration (n = 5), restoration (n = 4), and four CG patients (26 implants)
FFIP (n = 8,) RPD (n = 5), FFIP (n = 3), RPD (n = 3),
implant overdenture (n = 2) implant overdenture (n = 4)
were unavailable for follow-up and
dropped out (three patients died,
Definitive prostheses (delivered after a mean healing period of 4.5 months) presented with composite
resin veneering material. FFIP = full-arch fixed implant prosthesis; RPD = removable partial denture. two relocated, one was not able to
aImmediate provisional prostheses were provided with acrylic resin occlusal surfaces.
attend the scheduled follow-up ap-
pointment due to motor disability,
and one was untraceable).
Two treatment protocols were employed, and the patients were divided As previously described,3,4 10
into two groups. The test group (TG) was treated with the CBP3,4 (Fig 1), implants in the TG and 4 in the CG
with 4 to 6 implants loaded within 24 hours. This treatment was prescribed failed during the first 12 months
to 34 patients using 163 implants. In the control group (CG), patients were following implant placement. Two
treated with a traditional two-stage delayed loading protocol. The 15 pa- other implants (one TG and one
tients in this group received 6 to 9 implants each for a total of 97 implants CG, respectively) failed in the period
that were loaded following a mean postsurgical interval of 8.75 months. between 6 and 10 years as a result
Patients in the CG were made edentulous at least 3 months prior to implant of progressive bone loss and were
surgery, while the TG patients underwent teeth extraction followed by im- replaced. New implants were im-
mediate implant placement and subsequent prosthetic loading. mediately inserted and loaded, and
All of the patients were treated by three of the present authors (P.P., T.T., these implants were not included
and M.B.), who scrupulously followed identical treatment protocols.3,4 in the survival calculations or the
Recall appointments for assessment of oral hygiene, monitoring of out- peri-implant bone level evaluations.
comes, and recording observations were carried out at least once a year Consequently, the statistical analysis
and per individual patient needs.5 All observed or reported technical and was conducted for 206 implants (66
biologic complications were recorded. in the CG and 140 in the TG).
Parameters evaluated at the 10-year follow-up included implant cumula- The implant CSR was 93.25% for
tive survival rate (CSR), plaque index (PI), bleeding on probing (BoP), prob- the TG and 94.85% for the CG, with
ing depth (PD), and marginal bone loss (MBL) evaluated with periapical no statistical difference between
radiographs.3,4 the two groups (P = .58) (Fig 2). At
PI (score of 0 to 4 for each implant, evaluated using an erythrosine gel), BoP the 10-year follow-up appointment,
(score of 0 to 4 for each implant, evaluated using a nonmetallic periodontal all original fixed prostheses were
probe), and PD (recorded at 4 points for each implant using a nonmetallic functioning and did not need to be

28 The International Journal of Prosthodontics

© 2019 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Pera et al

a b

Fig 1   (a) Panoramic radiograph and (b, c) intraoral photographs of

a test group patient after 10 years of function.

100 3.0
90
P = .58 P = .58
Cumulative survival rate (%)

80 2.5

70
2.0
Mean (mm)

60
50 1.5
40
30 Control 1.0 Control
20 Test Test
.5
10
0 0
0 12 23 34 48 60 72 84 96 108 120 132 T0 T12 T24 T36 T72 T120
Follow-up (mo) Time (mo)

Fig 2   Life table analysis for implants in the test and control groups. Fig 3   Comparison of bone level between baseline (T0) and 10-year
follow-up appointment (T120) for the test and control groups.

replaced except for the prostheses in the two patients information regarding clinical complications in the two
who lost implants and needed modification. groups is reported in Table 3.
Implants in the TG had significantly lower MBL at the
10-year follow-up (P = .006) (Fig 3, Table 2). PI and BoP DISCUSSION
values were significantly higher in the TG, but there was
no significant difference in PD (Table 2). The applied immediate functional loading protocol
All patients reported satisfaction with their overall demonstrated good outcomes without significant
implant-prosthodontic treatment outcome. Additional differences in implant or prosthodontic CSRs when

Volume 32, Number 1, 2019 29

© 2019 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Clinical Research

Table 2   Bone Loss, Plaque Index (PI), Bleeding on Probing (BoP), and Probing Depth (PD) Data at 10 Years
Bone loss (mm), PI, mean (SD); median BoP, mean (SD); median PD, mean (SD); median
Group mean (SD) (25th–75th percentile) (25th–75th percentile) (25th–75th percentile)
Control 2.65 (1.34) 1.96 (1.53); 2 (1–4) 1.09 (1.38); 1 (0–1) 2.20 (0.90); 2.25 (1.5–2.5)
Test 2.11 (1.32) 2.49 (1.54); 3 (1–4) 1.93 (1.65); 2 (0–4) 2.56 (1.17); 2.25 (1.75–3)
P value .003 .023 .024 .23
SD = standard deviation.

Table 3   Biologic and Technical Complications at considered clinically significant, either (less than 2% dif-
10-Year Follow-up ference after 10 years). Indeed, in a larger sample size a
Test Control statistically significant difference might be reached, but
group group from a clinical point of view, a difference would not be
Biologic complications reached if the proportion of failures is maintained.
Failed implants at 1-y follow-up, n (%) 10 (6.13) 4 (4.12) It must be pointed out that the difference between
Failed implants at 10-y follow-up, n (%) 11 (6.75) 5 (5.15) the TG and CG was not only the time of implant load-
Patients with peri-implantitisa, n (%)
ing, since no differences in clinical outcomes are to be
11 (7.9) 8 (12)
expected between immediate and delayed loading pro-
Patients experiencing pain/discomfortb, n 2 1
tocols due to the loading time after the first healing pe-
Patients with inflammation under 2 2 riod. Differences between the CBP and the traditional
prosthesis, n
delayed loading protocol are also influenced by other
Patients experiencing TMJ pain, n 0 0 diverse considerations, including time of teeth extrac-
Technical complications tion, number of implants, implant inclination, and im-
Failed prostheses, n 0 0 plant length. Therefore, factors other than loading time
Events of implant fracture, n 0 0 might have affected the results.
Events of implant component fracture, n 0 0
CG patients had at least six implants each; therefore,
when one implant failed, it was generally not neces-
Events of abutment screw loosening, n 5 4
sary to replace it in order to maintain the full-arch fixed
Events of prosthetic screw loosening, n 7 6 prosthesis. Seven TG patients underwent re-entry sur-
Events of veneering material chipping, n 6 4 gery due to implant failure, while there was only one
Events of major fracture of veneering 3 3 (in the 6- to 10-year period) in the CG. However, the
material, n majority of implant failures in the TG (10 out of 11) oc-
aPeri-implantitis
defined as implants with presence of bleeding on probing curred during the first 3 months after surgery. Patients
and bone loss greater than 3 mm.
bSoreness
due to great transmucal height. in the TG received immediate implants in postextraction
sites, and full-arch fixed prostheses were placed within
24 hours. When failed implants were replaced, the new
compared to the traditional delayed loading protocol. implants were immediately inserted and immediately
The TG showed significantly lower mean MBL at all time loaded in the same day. As a consequence, despite re-
points (the CG showed 0.6 mm, 0.7 mm, 0.8 mm, and entry surgery, patients in the TG always wore full-arch
0.5 mm more MBL after 12, 36, 72, and 120 months, re- fixed prostheses and were not subjected to a lot of extra
spectively), with a statistically significant difference be- treatment.
tween the two groups. However, the major difference In contrast, patients in the CG were made edentulous
in peri-implant bone level between the two groups was before receiving implants 3 months later, and the fixed
established during the first healing period (baseline to 1 prosthesis was placed after a mean healing period of
year). This was followed by a steady-state condition that 8.75 months. Overall, patients in the CG wore remov-
was maintained between 12 and 72 months, while the able complete dentures for 11.75 months. Although this
TG exhibited a slight increase in bone loss in the subse- was not specifically investigated in the present study,
quent 6- to 10-year follow-up period (Fig 3). morbidity and discomfort for missing a fixed rehabilita-
Although a greater number of implants failed in the tion was significantly lower in the TG compared to the
TG (6.75%) compared to the CG (5.15%), the majority CG.
of these implants failed during the first months after Achievement of the reported favorable outcomes
implant insertion (at 12 months, the CSR was 93.9% respected the key determinants of osseointegration
in the TG and 95.9% in the CG), and differences in irrespective of the time of implant loading and relied
CSR between the two groups were not statistically sig- on the use of commercially pure titanium implants to
nificant at any time point at the implant level. It seems guarantee biologic inertia at the bone-implant inter-
reasonable to suggest that this difference cannot be face by avoidance of noxious ion release and on the

30 The International Journal of Prosthodontics

© 2019 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Pera et al

maintenance of the biomechanical inertia (ie, avoidance CONCLUSIONS

of implant micromotion). The latter was assured in the
TG patients by:
•• Primary stability (determined by bone quality,
implant macrodesign, and underpreparation of the
implant site)
•• Widespread polygonal support6
•• Choice of material (shock-absorbing veneering
material and rigid framework)6
•• Occlusal scheme7
•• Passively fitting screw-retained prostheses
Several authors consider biologic factors (ie, plaque
accumulation) to be a possible cause of MBL and implant
failure.8 In the present study, although plaque accumu-
lation was quite high at the 10-year follow-up (Table
2), MBL and implant failures were mainly concentrated
in the first months after implant insertion and loading.
When sufficient bone volume is present all around the
implants and biomechanical aspects are under control,
it appears that plaque accumulation alone may not be
the obvious cause of marginal bone resorption or lead
to the breakdown of the ankylotic-like bone-implant
interface.9
It is not possible to compare these observations with
similar reports, since small studies with low statistical
power frequently lead the reader to interpret pub-
lished reports as suggesting no differences between
outcomes. However, published systematic reviews1,10
underscore the effectiveness of simplified implant pro-
tocols that provide successful outcomes for full-arch
immediate loading when using a reduced number of
implants and distal tilted implants. Moreover, reported
clinical outcomes may be sensitive to clinician experi-
ence, and a learning curve is required to achieve com-
parable favorable results.
A 10-year follow-up report on the application of the
CBP for immediate rehabilitation of the edentulous
maxilla led to similar implant and prosthodontic clini-
cal outcomes and less bone loss when compared to the
traditional two-stage protocol.
ACKNOWLEDGMENTS
The authors thank dental technicians Luca Scaglione, Piercarlo
Seghesio, Arcangelo Traversi, and Alessandro Manca for producing
the dental prostheses. A special thanks to Dr Francesco Bagnasco
(Division of Implant and Prosthetic Dentistry, Department of Surgical
Sciences-DISC, University of Genoa, Italy) for his help in data col-
lection and to Dr Alessio Signori (Department of Health Sciences-
DISSAL, University of Genoa) for the statistical analysis of the data.
The authors report no conflicts of interest.
REFERENCES
  1. Chrcanovic BR, Albrektsson T, Wennerberg A. Tilted versus axially placed
dental implants: A meta-analysis. J Dent 2015;43:149–170.
  2. Balshi TJ, Wolfinger GJ, Slauch RW, Balshi SF. A retrospective analysis of
800 Brånemark System implants following the All-on-Four protocol.
J Prosthodont 2014;23:83–88.
  3. Tealdo T, Bevilacqua M, Menini M, et al. Immediate versus delayed load-
ing of dental implants in edentulous maxillae: A 36-month prospective
study. Int J Prosthodont 2011;24:294–302.
  4. Tealdo T, Menini M, Bevilacqua M, et al. Immediate versus delayed
loading of dental implants in edentulous patients’ maxillae: A 6-year
prospective study. Int J Prosthodont 2014;27:207–214.
  5. Menini M, Dellepiane E, Pesce P, et al. Hygienic and dietetic guidelines
for implant-supported full-arch immediate loading prostheses. Int J Oral
Dent Health 2015;1:1–5.
  6. Ogawa T, Dhaliwal S, Naert I, et al. Impact of implant number, distribu-
tion and prosthesis material on loading on implants supporting fixed
prostheses. J Oral Rehabil 2010;37:525–531.
  7. Beyron H. Occlusion: Point of significance in planning restorative proce-
dures. J Prosthet Dent 1973;30:641–652.
  8. Pesce P, Menini M, Tealdo T, Bevilacqua M, Pera F, Pera P. Peri-implantitis:
A systematic 
review of recently published papers. Int J Prosthodont
2014;27:15–25.
  9. Menini M, Setti P, Pera P, Pera F, Pesce P. Peri-implant tissue health and
bone resorption in patients with immediately loaded, implant-supported,
full-arch prostheses. Int J Prosthodont 2018;31:327–333.
10. Papaspyridakos P, Chen CJ, Chuang SK, Weber HP. Implant loading
protocols for edentulous patients with fixed prostheses: A systematic
review and meta-analysis. Int J Oral Maxillofac Implants 2014;
29(suppl):s256–s270.

Literature Abstract

Evidence-Based Dentistry Update on Silver Diamine Fluoride

Silver diamine fluoride incorporates the antibacterial effects of silver and the remineralizing actions of a high-concentration fluoride. It
effectively arrests the disease process on most lesions treated. Systematic reviews of clinical trials confirm the effectiveness of silver diamine
fluoride as a caries-arresting agent for primary teeth and root caries, as well as its ease of use, low cost, and relative safety. No caries
removal is necessary to arrest the caries process, so the use of silver diamine fluoride is appropriate when other forms of caries control
are not available or feasible. A sign of arrest is the dark staining of the lesions and affected tooth structures. That could be a deterrent for
patients who have esthetic concerns. A thorough informed consent is recommended to ensure high patient satisfaction. Silver diamine
fluoride use for caries control is recommended as part of a comprehensive caries management program in which individual needs and risks
are considered.
Crystal YO, Niederman R. Dent Clin North Am 2019;63:45–68. References: 68. Reprints: Y.O Crystal, yoc1@nyu.edu —Steven Sadowsky, USA

Volume 32, Number 1, 2019 31

© 2019 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.