Nom original: hopp2017.pdf
Titre: Comparison of marginal bone loss and implant success between axial and tilted implants in maxillary All‐on‐4 treatment concept rehabilitations after 5 years of follow‐up
Ce document au format PDF 1.5 a été généré par Arbortext Advanced Print Publisher 9.1.580/W Unicode / Adobe LiveCycle PDF Generator, et a été envoyé sur fichier-pdf.fr le 25/11/2018 à 21:07, depuis l'adresse IP 86.227.x.x.
La présente page de téléchargement du fichier a été vue 219 fois.
Taille du document: 447 Ko (11 pages).
Confidentialité: fichier public
Aperçu du document
Received: 1 June 2017
Revised: 17 July 2017
Accepted: 18 July 2017
Comparison of marginal bone loss and implant success between
axial and tilted implants in maxillary All-on-4 treatment concept
rehabilitations after 5 years of follow-up
jo Nobre, RDH, MSc2 | Paulo Malo
Milena Hopp, DMD1 | Miguel de Arau
Dr. Bernd Quantius & Milena Hopp, Private
Practice of Implantology and
Periodontology, Giesenkirchener Str. 40,
€ nchengladbach 41238, Germany
Research and Development Department,
Clinic, Lisbon, Portugal
Oral Surgery Department, Malo
Background: There is need for more scientific and clinical information on longer-term outcomes
of tilted implants compared to implants inserted in an axial position.
Purpose: Comparison of marginal bone loss and implant success after a 5-year follow-up between
axial and tilted implants inserted for full-arch maxillary rehabilitation.
Material and Methods: The retrospective clinical study included 891 patients with 3564 maxillary
jo Nobre, Malo
Miguel de Arau
Avenida dos Combatentes, 43, piso 11,
1600-042 Lisboa, Portugal.
implants rehabilitated according to the All-on-4 treatment concept. The follow-up time was 5
years. Linear mixed-effect models were performed to analyze the influence of implant orientation
(axial/tilted) on marginal bone loss and binary logistic regression to assess the effect of patient
characteristics on occurrence of marginal bone loss >2.8 mm. Only those patients with measurements of at least one axial and one tilted implant available were analyzed. This resulted in a data
Nobel Biocare Services AG, Grant/Award
set of 2379 implants (1201 axial, 1178 tilted) in 626 patients (5reduced data set).
Results: Axial and tilted implants showed comparable mean marginal bone losses of 1.14 6 0.71
and 1.19 6 0.82 mm, respectively. Mixed model analysis indicated that marginal bone loss levels at
5 years follow up was not significantly affected by the orientation (axial/tilted) of the implants in
the maxillary bone. Smoking and female gender were associated with marginal bone loss >2.8 mm
in a logistic regression analysis. Five-year implant success rates were 96%. The occurrence of
implant failure showed to be statistically independent from orientation.
Conclusions: Within the limitations of this study and considering a follow-up time of 5 years, it
can be concluded that tilted implants behave similarly with regards to marginal bone loss and
implant success in comparison to axial implants in full-arch rehabilitation of the maxilla. Longerterm outcomes (10 years 1) are needed to verify this result.
All-on-4, biological complication, edentulous maxilla, immediate function, longer-term outcome,
marginal bone loss, tilted implants
1 | INTRODUCTION
edentulous jaw in one operation without nerve transposition and/or
bone grafting. This is achieved by strategic positioning of four implants
Immediate function implant-supported rehabilitations have proven to
and maximization of the existing bone volume.6,15–18
be safe and reliable in the short and midterm outcomes,1–10 while pro-
The All-on-4 treatment concept provides that two implants are
viding the patients with the psychological advantage of immediate
placed anteriorly and two posteriorly. The anterior implants are placed
fixed restorations.11–14 The All-on-4 treatment concept is one of these
axial, whereas the posterior implants are tilted between 308 and 458.
immediate-function protocols, allowing the rehabilitation of an
The use of tilted implants posteriorly allows for use of longer implants.
Clin Implant Dent Relat Res. 2017;19:849–859.
C 2017 Wiley Periodicals, Inc.
This in turn enhances primary stability, enables cantilever reduction
surgery and daily for 6 days thereafter. Cortisone medication (predni-
with optimized prosthetic support as well as reduces the need for bone
m, Portusone [Meticorten Schering-Plough Farma Lda, Agualva-Cace
augmentation by optimized use of available bone.6,15–19 Although pro-
gal], 5 mg) was given daily in a regression mode (15–5 mg) from the
viding these benefits, the tilting of posterior implants may—from a bio-
day of surgery until 4 days postoperatively. Anti-inflammatory medica-
mechanical point of view—have the disadvantage of promoting
tion (ibuprofen, 600 mg, Ratiopharm Lda, Carnaxide, Portugal) was
marginal bone loss due to increased stress to surrounding bone.20–22
administered for 4 days postoperatively starting on day 4. Analgesics
However, abundantly available clinical data did not support this hypo-
(clonixine [Clonix, Janssen-Cilag Farmaceutica Lda, Barcarena, Portu-
thetical concern. Previous systematic reviews23 and meta-analyses24,25
gal], 300 mg) were given on the day of surgery and postoperatively for
did not reveal differences in marginal bone loss and implant failure
the first 3 days if needed. Antacid medication (Omeprazole, 20 mg, Lis-
rates between axial and tilted implants in short and medium term out-
bon, Portugal) was given on the day of surgery and daily for 6 days
comes. The few available longer-term outcomes support these results
for the mandible26,27 and the maxilla28,29 showing no significant differ-
A mucoperiosteal flap was raised along the top of the ridge with
ences in marginal bone level change and implant survival rates between
relieving incisions on the buccal aspect in the molar area. The insertion
axial and tilted implants over 5 years of follow-up assuming no addi-
of the implants followed standard procedures, except that we per-
tional risk of implant failure by using tilted implants posteriorly. How-
formed under-preparation to achieve an insertion torque between 30
ever, due to the limited data available for the maxilla, additional
scientific, and clinical information on longer-term outcomes of axial and
tilted implant is needed. Thus, the aims of this study were to investigate the effect of implant orientation (axial, tilted) on marginal bone
loss 5 years after insertion for full-arch maxillary rehabilitations using
the All-on-4 treatment concept and to examine possible risk indicators
for advanced marginal bone loss (>2.8 mm). The cutoff value of
2.8 mm at 5 years was chosen based on universally accepted initial
bone remodeling of up to 2.0 mm in the first year when considering
immediate function, together with a marginal bone loss 0.2 mm annually thereafter.30 The null hypothesis was that marginal bone loss levels
are not significantly influenced by implant orientation (axial, tilted).
and 50 Ncm before final seating of the implants. The implant platform
was positioned 0.8 mm above the crest (corresponding to the lower
corner of the implant neck for Brånemark system Mk III or Mk IV
implants) or flush with the bone level (NobelSpeedy implants). In the
case of tilted implants, we sometimes performed leveling of the bone
crest to ensure that the implant platform was situated at the level of
the bone crest, and to create space for the angulated abutment. Bicortical anchorage was established whenever needed. A surgical guide (Allon-4 Guide, Nobel Biocare AB) was used for implant placement assistance to facilitate the insertion of the tilted implants and the precise
positioning of the implants in relation to the opposing jaw. The
implants were inserted between the anterior wall of the maxillary sinus
2 | MATERIAL AND METHODS
This retrospective study was performed in a private clinic, Malo
Lisbon, Portugal. Patients were treated between November 2002 and
June 2010. Patients were included provided the need for fixed prosthetic full-arch maxillary rehabilitations supported by immediate function implants due to edentulism or the presence of teeth in very poor
periodontal conditions. The same team performed surgery and prosthetic restoration. Exclusion criteria were active chemotherapy and
radiotherapy, and patients who were not further followed at the private practice. An Ethical Committee approved the study (Ethics Com-
with an angulation between 308 and 458 relative to the occlusal plane.
The posterior implants were tilted distally along the anterior sinus wall,
emerging typically at the second premolar position. The two anterior
implants were positioned axially; however, in cases with severe bone
resorption and/or buccal concavities, vestibular tilting was employed.
Multiunit abutments (Nobel Biocare) were connected to the implants,
with tilted implants provided with angulated Multiunit abutments (308)
and anterior axial implants provided with straight (08) or angulated Multiunit abutments (178). After closing and suturing the flap with 3–0
nonresorbable sutures (B Braun Silkam, Aesculap Inc., Center Valley,
PA), the abutments were accessed by means of a punch.
mittee for Health, Lisbon, Portugal; authorization no. 001/2016).
2.2 | Prosthetic protocol
2.1 | Surgical protocol
High-density acrylic resin (PalaXpress Ultra, Heraeus Kulzer GmbH,
The medical history of each patient was reviewed, and each patient
Hanau, Germany) and acrylic resin crowns (Premium teeth, Heraeus
received a clinical examination complemented with radiographic exami-
Kulzer GmbH) provisional prostheses with Temporary Coping Multi-
nations (an orthopantomography to assess bone height and computer-
Unit Titanium (Nobel Biocare AB) were delivered on the day of surgery.
ized tomography scan to assess bone volume and anatomical
Considering the patients’ preference, definitive acrylic-resin prostheses
structures). The surgical procedures were performed under local anes-
of the same type, metal-ceramic prostheses with a titanium framework
thesia with mepivacaine chlorhydrate with epinephrine 1:100 000
(Procera, Nobel Biocare AB) and all-ceramic Alumina crowns (Procera
(Scandinibsa 2%, Inibsa Laboratory, Barcelona, Spain). All patients were
crowns; NobelRondo Ceramics, Nobel Biocare AB) or metal-acrylic
sedated with diazepam (Valium 10 mg, Roche, Amadora, Portugal) prior
prostheses with a titanium framework (Procera) high-density acrylic
to surgery. Antibiotics (amoxicillin 875 mg 1 clavulanic acid 125 mg,
resin (PalaXpress Ultra,) and acrylic resin crowns (Premium teeth), were
Labesfal, Campo de Besteiros, Portugal) were given 1 hour prior to
delivered, at the earliest, 6 months’ postsurgery.
Orthopantomography at 5 years after maxillary
rehabilitation with All-on-4 treatment and definitive prosthesis
2.3 | Follow up visits and maintenance protocol
toward the implant axis. The radiographs were calibrated using the disThe authors instructed the patients to follow a strict protocol for postoperative maintenance of the oral hygiene. Follow up visits were performed at 10 days, 2, 4, and 6 months, 1 year and every 6 months
thereafter. The routine checkups consisted in the assessment of clinical
parameters, followed by prophylaxis and dental hygiene instructions.
tance between implant threads. A representative radiographic followup is illustrated in Figures 1–6.
Secondary clinical endpoints included prosthetic survival (on
patient level) and implant success as well as incidence of biological
complications (at implant level) at 5 years of function. Biological complications encompassed infection (in the first 12 months postsurgery), fis-
2.4 | Clinical endpoints
tula, abscess, mucositis (mucosal inflammation with or without probing
Primary clinical endpoint was marginal bone loss at 5 years of follow-
implant pathology (probing pocket depths >4 mm with marginal bone
up. Marginal bone levels were assessed through periapical radiographs
loss with or without suppuration) measured using a plastic periodontal
taken at baseline (day of surgery) and 5 years of function. Periapical
probe calibrated to 0.25 N (Hawe Click Probe, Hawe Neos, Bioggio,
radiographs were made using the parallel technique with a film holder
Switzerland). Prosthetic survival was based on function, with prosthesis
(Super-bite, Hawe-Neos), and its position was adjusted manually for an
removal classified as failure. Implant success was based on the Malo
estimated orthogonal film position. A reader blinded to the patient
Clinic success criteria31: (1) implant fulfilled its intended function as
information examined all radiographs of the implants for marginal bone
support for reconstruction (the potential existence of a sleeping
level. Each periapical radiograph was scanned at 300 dpi with a scanner
implant was considered a failure); (2) implant was stable when individu-
(HP Scanjet 4890, HP Portugal, Paço de Arcos, Portugal), and the mar-
ally and manually tested (the prosthesis was removed at each follow-
ginal bone level was assessed with image analysis software (Image J
up appointment); (3) no signs of persistent infection; (4) no radiolucent
version 1.40g for Windows, National Institutes of Health). The refer-
areas around the implants; (5) good aesthetic outcome in the rehabilita-
ence point for the reading was the implant platform (the horizontal
tion (classified as the absence of aesthetic complaints from the patient
interface between the implant and the abutment), and marginal bone
or prosthodontist); and (6) construction of an implant-supported fixed
level was measured up to the first contact between implant and bone.
prosthesis that provided patient comfort and good hygienic
pocket depths >4 mm and absence of marginal bone loss), or peri-
The radiographs were calibrated digitally using the distance between
implant threads as reference. The measurements were performed on
the mesial and distal sites, and average values were calculated. The
radiographs were accepted (74%) or rejected (26%) for evaluation
based on the clarity of the implant threads; a clear thread guarantees
both sharpness and an orthogonal direction of the radiographic beam
F I G U R E 2 Postoperative orthopantomography after maxillary Allon-4 treatment
Periapical radiograph of the posterior first quadrant
tilted implant at 5-year follow-up
TA BL E 1
Descriptive patient data
(years; mean, range)
56 (20 – 85)
Mix of implant-supported
and fixed prosthesis
Systemic comorbidity (total)
F I G U R E 5 Periapical radiograph of the axial implants in the
anterior second sextant at 5-year follow-up
maintenance (classified as the absence of comfort and hygiene complaints from the patient). Implants not complying with the criteria as
well as implant removal were considered failures.
In addition to the clinical endpoints, patient and implant related
AIDS (n 5 3); syndrome (n 5 1); kidney (n 5 2), autoimmune (n 5 3), neurological (n 5 7), or hemorrhagic disease (n 5 1).
Patients with more than one comorbidity.
data were compiled to be used as explanatory fixed and random factors for bone loss in a generalized linear mixed model and potential risk
predictors for marginal bone loss >2.8 mm in a logistic regression
model. The factors are given consideration in Tables 1–3 and were
2.5 | Data sets for analyses
chosen based on assumed relevance as explanatory variable and risk
A total of 891 patients with 3564 maxillary implants were initially
predictor for bone loss as well as their data availability.
included in this study for the analysis of implant success and prosthesis survival as well as biological complications (5complete data
set). The final data set for marginal bone loss analysis was lower
because: (1) 91 implants failed before 5 years follow up; (2) the
treatment of 29 implants was modified within the 5 years period
under consideration limiting the validity of the All-on-4 treatment
concept and (3) 26% of the periapical radiographs were not readable. The remaining data set of 2539 implants in 715 patients was
then used to evaluate the effect of patient characteristics on marginal bone loss >2.8 mm using a binary logistic regression model. To
avoid bias due to unbalanced data, the data set was further adapted
for statistics (descriptive statistics and GLMM analysis) of marginal
bone loss levels focusing on comparisons between axial and tilted
implants: Only those patients with measurements of at least one
axial and one tilted implant available were analyzed. This resulted in
a data set of 2379 implants (1201 axial, 1178 tilted) in 626 patients
(5reduced data set). Different data sets were used and not consis-
F I G U R E 6 Periapical radiograph of the posterior second quadrant
tilted implant at 5-year follow-up
tently lowered to the reduced data set, in order to retain the most
comprehensive data set possible for each analysis.
TA BL E 2
to ascertain the effect of patient characteristics on the likelihood of
Descriptive implant data
% of total
patients have peri-implant bone loss >2.8 mm on at least one
implant. Odds ratios and corresponding 95% confidence intervals
Brånemark system Mk III
Brånemark system Mk IV
Implant width (mm)
(CIs) were estimated for the explanatory variables. The regression
model was performed on patient level. Univariate analyses were
used to identify covariates associated with marginal bone loss
>2.8 mm. Covariates with P < .20 in univariate analyses were
included in the regression model. The level of significance was 5%,
statistical power was 80%. All data, except for the GLMM, was
analyzed using the software SPSS version 20 (IBM SPSS, NY).
3 | RESULTS
Implant length (mm)
Total data set (n 5 3564)
Reduced data set (n 5 2379)
3.1 | Patient and implant characteristics
A total of 891 patients with 3564 maxillary implants supporting 891
prostheses were included in this study. Three-hundred-sixty-four were
males, 527 were females. The average age was 56 years (range 20–83).
About 287 patients (32.2%) presented systemic comorbidities, namely:
a cardiovascular (n 5 189), a thyroid (n 5 24), a rheumatologic (n 5 37),
an oncological disease (n 5 17), diabetes (n 5 36), hepatitis (n 5 16), a
hemorrhagic (n 5 1), a neurological (n 5 7), a kidney (n 5 2), or an autoimmune disease (n 5 3), HIV positive (n 5 3) or a syndrome (von Willebrand disease, n 5 1). Forty-nine patients presented more than one
systemic condition. Two-hundred-eleven of the included patients
Number of failed implants is given in parenthesis.
(23.7%) were smokers. Implant diameter differed between 3.3 and
5 mm and implant length differed between 7 and 18 mm. The opposing
2.6 | Statistics
mandibular occlusal surfaces consisted of removable prostheses (3.1%
of patients), natural teeth (16.9% of patients), implant-supported fixed
Descriptive statistics were calculated for implant success, incidence of
biological complications (complete data set) as well as marginal bone
loss (reduced data set). In addition, mean marginal bone loss was calcu-
Combinations of type, width and length of implants (all of
external connection) assessed in this study
TA BL E 3
lated for 20 randomly composed subsets (100 implants each) of the
data set, to assess the impact of the data set size on calculated mean
values. Subsamples of 100 implants were chosen, since this is the scale
of implants analyzed in one of the few available and comparable
system Mk III
% of total
Implant success was analyzed using life table analysis. McNemar
tests were performed to assess differences in marginal bone loss
>2.8 mm, biological complications and implant failure between axial
system Mk IV
and tilted implants. Generalized linear mixed models (GLMM) were
adopted to analyze the influence of implant orientation (axial/tilted)
on marginal bone loss at 5 years of follow-up. Implant length and
orientation (axial/tilted), gender (m/f), smoking (y/n), cardiovascular
condition (CVD; y/n) as well as opposing dentition (4 categories
with increasing strength of resistance) were evaluated as fixed
effects, and age and patient as random factors. Interaction terms
between CVD and age, CVD and smoker and CVD, age and smoker
were checked for improvement of the model. A GLMM without consideration of implant orientation was compared with the respective
model including implant orientation using a likelihood ratio test. The
GLMMs were computed in R v3.3.2 using the glmer function from
the lme4 package. A binary logistic regression model was performed
130 axial [10.8%] and 157 tilted [13.3%]) lost more than 2 mm of periimplant bone. Advanced bone loss (>2.8 mm) occurred with 97
implants [4.0%; 39 axial (3.9%) and 58 tilted (4.9%)]. Seventy-nine of
the 97 implants had either biological complications (infection, fistula, or
peri-implant pathology; n 5 28) previously and/or were connected in a
prosthesis that presented mechanical complications (n 5 68). The difference of the proportion of advanced marginal bone loss (>2.8 mm) at
axial and tilted implants on patient level was statistically significant
(McNemar, X2 5 8.82, P 5 .003; n 5 626 patients). A binary logistic
regression model performed to ascertain the effects of patient characteristics on the likelihood of patients have marginal bone loss > 2.8 mm
was statistically significant (X2 5 28.174, P < .001; Table 5). Of the five
predictor variables, only three were statistically significant: smoking,
gender, and cardiovascular condition. Smokers had 2.032 times higher
odds, female gender 1.925 times higher odds and patients without carBox plots of bone loss at axial and tilted implants after
5 years. The dashed line marks the mean and the central horizontal
line the median of the values; the box edges the first and third
quartile. The interquartile range within the box includes the central
50% of the values. The whiskers show the 5th and 95th percentile.
Outliers are indicated as single points
diovascular condition 2.42 times higher odds to exhibit advanced marginal bone loss (>2.8 mm) than nonsmoker, males and patients with
cardiovascular disease, respectively (Table 5).
3.3 | Implant success and biological complications
After 5 years 91 implants failed, giving a 5-years success rate of 96%
prostheses (33.9% of patients), or a mixture of implant-supported and
at implant level. Failure of tilted implants (69 implants failed; 96.1%
fixed prostheses over natural teeth (46.1% of patients). Patient and
success rate) was numerically nearly equal to failure of axial implants
implant related data are given consideration in Tables 1–3.
(76 implants lost; 95.7% success rate). The occurrence of implant failure on patient level showed to be statistically independent from
3.2 | Marginal bone loss after 5 years and risk factors
implant orientation (McNemar, X2 (1, n 5 891 patients) 5 .65, P 5 .42).
Life table analysis revealed implant failure occurring predominantly
Regarding marginal bone loss axial and tilted implants showed an iden-
within the first 2 years of follow-up for both axial and tilted implants
tical median value of 1.05 mm and mean values of 1.14 6 0.71 mm
(Table 6). Prosthesis survival was 99.8% (889/891). Three-hundred-
(range 0–6.9 mm) and 1.19 6 0.82 mm (range 0–8.4 mm), respectively,
thirteen implants (8.8%) in 209 patients (23.5%) showed biological
at 5 years of follow-up (Figure 7). Calculated marginal bone loss from
complications, namely infection (n 5 24), fistula (n 5 4), mucositis
randomly composed subsets (100 implants) of the present data set
(n 5 189) peri-implant pathology (n 5 95), or abscess (n 5 1). Biological
resulted in mean values between 0.97 and 1.5 mm. Generalized linear
complications were observed with 131 axial and 182 tilted implants.
mixed models were computed with and without implant orientation as
The difference of the proportion of biological complications at axial
fixed factor (Table 4). Inclusion of implant orientation (axial/tilted) as
and tilted implants on patient level was statistically significant (McNe-
fixed factor did not result in a significantly improved model (P 5 .503),
mar, X2 (1, n 5 891 patients) 5 14.82, P < .01). The biological complica-
which indicates that the marginal bone loss at 5 years follow up was
tions were resolved (absence of inflammation and probing pocket
not significantly affected by the implant orientation (axial/tilted) in the
depths </5 4 mm without reoccurrence during the study follow-up)
maxillary bone. Seventy-one implants [2.9% of all implants; 36 axial
for 242 implants through nonsurgical intervention by scaling with an
(3% of axial implants), 35 tilted (3% of tilted implants)] revealed mar-
ultrasonic device and application of 0.2% chlorhexidine gel and for 10
ginal bone loss <0.2 mm. Two-hundred-eighty-seven implants (12%;
implants through administration of systemic antibiotics and open flap
Likelihood ratio test for statistical difference between generalized linear mixed models (GLMM) fitted for bone loss 5 years followup (a) including and (b) excluding implant orientation (axial/tilted) as fixed factor
TA BL E 4
GLMM without implant orientationb
GLMM with implant orientationc
AIC, Akaike information criterion; BIC, Bayesian information criterion; logL, log likelihood.
Implant length (7 categories), gender (m/f), smoker (y/n), cardiovascular condition (y/n), opposing dentition (4 categories) were entered as fixed effects
and age (cont.) and patient as random factors.
Formula: glmer (boneloss implength 1 (1 | age) 1 gender 1 smoker 1 cvd 1 opposite dent 1 (1 | patient).
Formula: glmer (boneloss implength 1 orientation 1 (1 | age) 1 gender 1 smoker 1 cvd 1 opposite dent 1 (1 | patient).
TA BL E 5
Binary logistic regression model assessing the effect of potential risk factors for marginal bone loss >2.8 mm
n 5 715
Likelihood ratio test
P 5 .75
surgical intervention, in order to mechanically clean the implant surface
implants placed using the All-on-4 treatment concept. Linear mixed
in conjunction with 0.2% chlorhexidine. The biological complications
model analysis showed that implant orientation (axial/tilted) does not
were not solved for 61 implants with 7 implants ending up failed and
significantly contribute to the explanation of the peri-implant bone loss
54 implants without compromise of the survival but included as failures
pattern after 5 years’ follow-up. Hence, the null hypothesis of nonexis-
considering the success criteria adopted.
tence of a significant influence of implant orientation on marginal bone
loss after 5 years could not be rejected. This result is congruent with
4 | DISCUSSION
findings based on reviews and meta-analysis of short-term outcomes.25,33,34 Chrcanovic and colleagues performed a meta-analysis
Implant tilting in immediate function implant-supported rehabilitations
focusing on the difference in implant failure rate, marginal bone loss,
has provided the possibility of rehabilitating patients with anatomic lim-
and postoperative infections for patients being rehabilitated by tilted
itations that preclude the insertion of implants in an axial position.
or by axially placed implants.25 The authors included data from 44 pub-
However, it is still under debate if tilted implants bear a higher risk of
lications in their analysis, resulting in a total of 5029 tilted and 5732
implant failure compared to axial implants by peri-implant bone loss or
axial implants. No significant effects of tilted implants on the occur-
pathology over time.32 Long-term data are scarce. In this retrospective
rence of marginal bone loss and implant failure rates was observed
cohort study, the authors investigated the 5-year follow-up results of
using a random-effects and a fixed-effects model, respectively. Further-
marginal bone loss and success rates of axial and tilted maxillary
more, another meta-analysis from Ata-Ali and colleagues investigated
TA BL E 6
Five-years life table analysis of (a) axial and (b) tilted implants
on interval (%)
success rate (%)
(a) 5-years life table analysis of axial implants
(b) 5-years life table analysis of tilted implants
Implants at risk
n 5 5 implants with unresolved biological complications.
n 5 9 implants with unresolved biological complications.
n 5 7 implants with unresolved biological complications.
n 5 4 implants with unresolved biological complications.
n 5 7 implants with unresolved biological complications.
n 5 6 implants with unresolved biological complications.
n 5 3 implants with unresolved biological complications.
the outcome of oral rehabilitations with tilted dental implants.35 The
designed to examine to which extent marginal bone loss of the dif-
authors included 13 publications (seven retrospective studies and six
ferently angled implants increases over time, comparing our marginal
prospective studies) and performed three meta-analysis resulting in:
bone loss measurements from rehabilitation of completely edentu-
One-hundred-sixty-six tilted and 241 axial implants for the meta-
lous maxillae with short-term and long-term results available in the
analysis evaluating the success rates in the retrospective studies; 379
literature it can be concluded that (1) our results do not indicate that
tilted and 425 axial implants for the meta-analysis evaluating the success
tilted implants placed using the All-on-4 treatment concept promote
rates in the prospective studies; and 230 tilted and 270 axial implants
marginal bone loss to a higher extent than axial implants over time
for the meta-analysis evaluating the marginal bone loss in the prospec-
and that (2) mean bone loss over time might be generally lower than
tive studies. No evidence of differences in success rate were found in
expected based on short-term data obtained from small data sets.
prospective or retrospective studies, nor differences in marginal bone
Advanced marginal bone loss (>2.8 mm) occurred with about 4%
loss was registered between tilted and axial implants. However, in con-
of the implants (39 axial, 58 tilted). Comparing the occurrence of mar-
trast to the present study, limitations, such as variability in the techni-
ginal bone loss >2.8 mm with biological and mechanical complication
ques used, lack of analysis of confounding factors, no differentiation
observed at these implants assume that advanced marginal bone loss
between mandibular and maxillary implants and assessment of primarily
might at least partly be interpreted as being secondary to biological or
short-term data hamper definitive conclusions on long-term effects.
mechanical complications. Seventy-nine of the 97 implants had previ-
In this study, the mean marginal bone loss observed with axial
ous infection, fistula, or peri-implant pathology and/or were connected
and tilted implants was 1.14 6 0.71 and 1.19 6 0.82 mm, respec-
in a prosthesis that presented mechanical complications. The propor-
tively, retrieved from the reduced data set with 2379 implants
tion of tilted implants with advanced marginal bone loss (>2.8 mm)
(1201 axial, 1178 tilted) in 626 patients. These values compare
was significantly higher than of axial implants; however, the overall
favorably with other longer-term outcomes (5 and 7 years) for maxil-
number of affected implants is low and in many cases (59% of patients,
lary implants placed using the All-on-4 treatment concept or similar
n 5 53 implants) only a single implant per patient showed marginal
immediate loading protocols.6,36,37 Niedermaier and colleagues
bone loss of this severity. This suggests a low potential impact on pros-
measured mean marginal bone losses of 1.30 6 0.35 mm for axial
thesis survival and thus very low clinical relevance of this statistical
and 1.30 6 0.42 mm for tilted implants 5 years after treatment of
finding. Thus, the authors do not rate this statistical result to be contra-
severely atrophied and edentulous jaws by means of implant-
dictory to the GLMM analysis of the whole bone loss data set resulting
supported immediately loaded fixed full-arch dentures using four to
in no relevant influence of implant orientation on marginal bone loss
six implants per jaw.36 Results might not be comparable directly,
after 5 years of function.
because the authors did not differentiate between treatment proto-
Logistic regression analysis revealed smoking and female gen-
cols and mandibular and maxillary implants. Vervaeke and colleagues
der to be associated with, and thus potential risk indicators for,
for instance, found jaw of treatment to be a predictor of peri-
marginal bone loss >2.8 mm at 5 years when controlled for other
implant bone loss with more marginal bone loss observed in the
variables of interest. Smokers exhibited a more than 2-fold risk
maxilla;38 however, other studies did not find differences.23,34
than nonsmokers. The impact of smoking on implant success is still
Nevertheless, in accordance with our findings and the literature,
controversially debated.34 Some studies observed a negative
Niedermaier and colleagues did not observe significant differences
impact of smoking on implant failure or marginal bone loss,38,39
in bone loss between tilted and axial implants during a follow-up of
others did not.27,36,38 Since smoking habit is rarely precisely
and colleagues reported a 3 and 5 years mean marMalo
defined and the impact assessment varies greatly between descrip-
ginal bone loss of 1.52 6 0.3 mm and 1.95 60.4 mm, respectively,
tive statistics, univariate and multivariate analysis it is difficult to
after rehabilitation of completely edentulous maxillae using the All-
draw definitive conclusions. Our study support assumptions of
on-4 treatment concept.6 Readable radiographs were available for
higher risk of bone loss in smokers as suggested by various meta-
621 implants (196 patients) at 3 years and 106 implants (33 patients)
analysis and reviews.38–41 Female gender showed a 2-fold risk than
at 5 years of follow-up. Treatment as well as patient and implant
males for advanced marginal bone loss. Higher marginal bone loss
characteristics were very similar to the present study. Since mean
was previously reported in female patients in long-term follow-up
values of small data sets are in general more prone to outliers the
studies.42–47 The higher risk of patients without CVD compared to
authors assume that the lower marginal bone loss values obtained in
patients with CVD is counterintuitive and the statistical outcome
our study may be due to the larger data set analyzed. Calculated
might be influenced by the fact that the patients with CVD-group
marginal bone loss from randomly composed subsets (100 implants)
was much smaller than the patients without CVD-group and the
of the present data set resulted in mean values of 0.97 to up to
patients without CVD-group consisted of healthy as well as patients
1.5 mm. The lower mean marginal bone loss for axial and tilted
with other comorbidities, such as diabetes. Hence, further research
implants obtained in this study and the larger data set confirm that
is needed focusing on the impact of patient and implant character-
the proportion of implants with low marginal bone loss after 5 years
istics on the long-term marginal bone loss outcome in immediate
is high. This is further supported by even lower median values for
function full-arch rehabilitations when adjusted for other relevant
axial and tilted implants of 1.05 mm. Even if this study was not
Survival rates of 97.5% and 93.8% were obtained after 5 years at
implant and patient level, respectively. Survival rates were similar for
factors using regression and mixed model analysis and the less than
20% rate of patients lost to follow-up represent study strengths.
axial and tilted implants. These findings are consistent with results of
previous short-term and long-term studies using similar treatment protocols. These studies report survivals of axial and tilted implants in the
range of 95%-100%.6,34,36,48 Life table analysis revealed implant failure
occurring predominantly within the first 2 years of follow-up for both
axial and tilted implants. About 71% of the failed implants in our study
were lost in the first year, which is in line with previous research on
immediate function implant supported restorations.6,14,48 A very low
number of implant failures occurred in the subsequent years and no
significant association between implant orientation (axial/tilted) and
implant failure was found. Thus, the authors assume that marginal
bone loss over time does not markedly provoke failure of either axial
5 | CONCLUSION
Within the limitations of this study and considering a follow-up time of
5 years, it can be concluded that tilted implants do not promote marginal bone loss and implant failure in comparison to axial implants in
full-arch rehabilitation of the maxilla. Future research on the impact of
implant orientation on implant survival or success should focus on the
long-term follow-up of these rehabilitations (10 years 1) and should
consider implant success as well as marginal bone loss and soft tissue
or tilted implants, at least up to 5 years of follow-up.
Consideration of biological complications after implant placement
AC KNOWLEDG MENT S
is important for assessing the long-term success of immediate function
The authors thank Mr. Sandro Catarino for all the help in data man-
full-arch rehabilitations, since more severe biological complication, such
agement and acknowledge the statistical support of Dr. Richard
as peri-implant pathology may indicate an increased risk for failure
Ottermanns, RWTH Aachen University.
over time. However, if properly recognized and treated, a biological
complication might be saved. The biological complications observed in
the present study were infection, mucositis and peri-implant pathology
as well as a few events of fistulae or abscesses. The study shows a
slight but statistically significant higher proportion of biological complications occurring at tilted compared to axial implants on patient level.
Low levels of oral hygiene, history of periodontitis and mechanical
complications were frequently suggested to be potential risk factors
for biological complications;42,43 however, the authors are not aware of
any other study examining the association of biological complications
with implant orientation. Due to several reasons it is assumed that the
statistical finding in the present study is of low clinical relevance: (1)
The overall rate of biological complications after 5 years follow-up was
low (8.8% of implants); (2) the majority of biological complications
observed were well amenable for periodontal treatment; (3) in many
affected patients (65%) only single implants exhibited biological complication, which is considered to bear a low risk of prosthesis failure in
full-arch implant-supported rehabilitations irrespective of the location
of the biological complication. Nevertheless, an effective recall program
is crucial to individuate complications in the beginning avoiding the
evolution of these in major complication that may lead to implant
This study exhibits certain limitations, which should be taken into
consideration when interpreting the results. The single center retrospective design used has drawbacks, such as lack of external validation and
data collection not adjusted to the focus of this study. A further limitation is the loss of data due to the 26% of nonreadable radiographs,
thereby providing a potential bias by unbalancing the data set and
underestimating the marginal bone loss evaluation. To accommodate
for these limitations, the data set used for marginal bone loss comparisons was adjusted as described in Material and Methods, for example,
by ensuring that data for at least one axial and one tilted implant was
available per patient. The large sample size, consideration of various
RE FE RE NCE S
~arrocha-Diago M, Carrillo-Garcîa C, Boronat-Lopez A, García Pen
Mira B. Comparative study of wide-diameter implants placed after
dental extraction and implants positioned in mature bone for molar
replacement. Int J Oral Maxillofac Impl. 2008;23:497–501.
 Chen ST, Wilson TG, Hämmerle CHF. Immediate or early placement
of implants following tooth extraction: review of biological basis,
clinical procedures, and outcomes. Int J Oral Maxillofac Impl. 2004;
 Esposito M, Grusovin MG, Polyzos IP, Felice P, Worthington HV.
Timing of implant placement after tooth extraction: immediate,
immediate-delayed or delayed implants? A Cochrane systematic
review. Eur J Oral Implantol. 2010;3:189–205.
 Fugazzotto PA. Implant placement at the time of mandibular molar
extraction: description of technique and preliminary results of 341
cases. J Periodontol. 2008;79:737–747.
 Kohen J, Matalon S, Block J, Ormanier Z. Effect of implant insertion
and loading protocol on long-term stability and crestal bone loss: a
comparative study. J Prosthet Dent. 2016;115:697–702.
P, Nobre M, Lopes A, Francischone C, Rigolizzo M. All-on-4”
immediate-function concept for completely edentulous maxillae: a
clinical report on the medium (3 years) and long-term (5 years) outcomes. Clin Imp Dent Rel Res. 2012;14:e139–e150.
 Grunder U. Immediate functional loading of immediate implants in
edentulous arches: two-years results. Int J Periodont Rest. 2001;21:
 Altintas NY, Taskesen F, Bagis B, Baltacioglu E, Cezairli B, Senel FC.
Immediate implant placement in fresh sockets versus implant placement in healed bone for full-arch fixed prostheses with conventional loading. Int J Oral Maxillofac Surg. 2016;45:226–231.
 Zhang S, Wang S, Song Y. Immediate loading for implant restoration
compared with early or conventional loading: a meta-analysis.
J Cranio Maxillofac Surg. In Press.
 Moraschini V, Porto Barboza E. Immediate versus conventional loaded
single implants in the posterior mandible: a meta-analysis of randomized controlled trials. Int J Oral Maxillofac Surg. 2016;45:85–92.
 Menassa M, de Grandmont P, Audy N, Durand R, Rompre P, Emami
E. Patients’ expectations, satisfaction, and quality of life with immediate loading protocol. Clin Oral Impl Res. 2016;27:83–89.
 Avila G, Galindo P, Rios H, Wang HL. Immediate implant loading: current status from available literature. Impl Dent. 2007;16:235–245.
 Vogel R, Smith-Palmer J, Valentine W. Evaluating the health economic implications and cost-effectiveness of dental implants: a literature review. Int J Oral and Maxillofac Impl. 2013;28:343–356.
P, Rangert B, Eng M, Dvärsäter L. Immediate function of Bra Malo
nemark implants in the esthetic zone: a retrospective clinical study
with 6 months to 4 years of follow-up. Clin Impl Dent Rel Res. 2000;
P, Rangert B, Nobre M. All-on-4 immediate function concept
with Brånemark system implants for completely edentulous mandibles: a retrospective clinical study. Clin Impl Dent Rel Res. 2003;5:2–9.
P, Rangert B, Nobre M. All-on-4 immediate-function concept
with Brånemark system implants for completely edentulous maxillae: a 1-year retrospective clinical study. Clin Impl Dent Rel Res.
 Capelli M, Zuffetti F, Testori T, Romeo D, Taschieri S, Del Fabbro
M. Immediate rehabilitation of the completely edentulous jaws with
fixed prostheses supported by upright and tilted implants. A multicenter clinical study. Int J Oral Maxillofac Impl. 2007;22:639–644.
 Weinstein R, Agliardi E, Fabbro MD, Romeo D, Francetti L. Immediate rehabilitation of the extremely atrophic mandible with fixed fullprosthesis supported by four implants. Clin Impl Dent Rel Res. 2012;
€ m H. Tilting of posterior
 Krekmanov L, Kahn M, Rangert B, Lindstro
mandibular and maxillary implants for improved prosthesis support.
Int J Oral Maxillofac Impl. 2000;15:405–414.
 Clelland N, Gilat A, McGlumphy E, Brantley W. A photoelastic and
strain gauge analysis of angle abutments for an implant system. Int
J Oral Maxillofac Impl. 1993;8:541–548.
 Watanabe F, Yoshiaki H, Komatsu S, Ramos T, Fukuda H. Finite element analysis of the influence of implant inclination, loading position,
and load direction on stress distribution. Odontology. 2003;91:31–36.
 Almeida E, Rocha EP, Freitas AC, et al. Tilted and short implants
supporting fixed prosthesis in an atrophic maxilla: a 3D-FEA biomechanical evaluation. Clin Impl Dent Rel Res. 2015;1(17):e332–e342.
 Del Fabbro M, Bellini CM, Romeo D, Francetti L. Tilted implants for
the rehabilitation of edentulous jaws: a systematic review. Clin Impl
Dent Rel Res. 2012;14:612–621.
 Monje A, Chan HL, Suarez F, Galindo-Moreno P, Wang HL.
Marginal bone loss around tilted implants in comparison to
straight implants: a meta-analysis. Int J Oral Maxillofac Impl. 2012;
retrospective longitudinal study. Clin Oral Impl Res. 2007;18:
 Albrektsson T, Zarb G. Determinants of correct clinical reporting. Int
J Prosthodont. 1998;11:517–521.
P, De Arau
jo Nobre M, Lopes A. The rehabilitation of com Malo
pletely edentulous maxilla with different degrees of resorption with
four or more immediately loaded implants: a 5-years retrospective
study and a new classification. Eur J Oral Implantol. 2011;4:227–243.
 Browaeys H, Dierens M, Ruyffelaert C, Matthijs C, De Bruyn H,
Vandeweghe S. Ongoing crestal bone loss around implants subjected to computer-guided flapless surgery and immediate loading
using the All-on-4® concept. Clin Imp Dent Rel Res. 2014;17:
 Menini M, Signori A, Tealdo T, et al. Tilted implants in the immediate loading rehabilitation of the maxilla: a systematic review. J Dent
 Patzelt SBM, Bahat O, Reynolds MA, Strub JR. The All-on-four
treatment concept: a systematic review. Clin Imp Dent Rel Res.
~arrocha-Oltra D, Candel-Marti E, Pen
 Ata-Ali J, Pen
Oral rehabilitation with tilted dental implants: a metaanalysis. Med
Oral Patol Oral Cir Bucal. 2012;17:e582–e587.
 Niedermaier R, Stelzle F, Riemann M, Bolz W, Schuh P, Wachtel H.
Implant-supported immediately loaded fixed full-arch dentures:
evaluation of implant survival rates in a case cohort of up to 7
years. Clin Imp Dent Rel Res. 2017;19(1):4–19.
 Rosen A, Gynther G. Implant treatment without bone grafting in
edentulous severely resorbed maxillas: a long-term follow-up study.
J Oral Maxillofac Surg. 2007;65:1010–1016.
 Vervaeke S, Collaert B, Cosyn J, Deschepper E, De Bruyn H. A
multifactorial analysis to identify predictors of implant failure
and peri-implant bone loss. Clin Imp Dent Rel Res. 2015;1(17):
 Clementini M, Rossetti PHO, Penarrocha D, Micarelli C, Bonachela
WC, Canullo L. Systemic risk factor for peri-implant bone loss: a
systematic review and meta-analysis. Int J Oral Maxillofac Surg.
 Strietzel FP, Reichart PA, Kale A, Kulkarni M, Wegner B, Ku
Smoking interferes with the prognosis of dental implant treatment:
a systematic review and meta-analysis. J Clin Periodontol. 2007;34:
 Hinode D, Tanabe S, Yokoyama M, Fujisawa K, Yamauchi E, Miyamoto Y. Influence of smoking on osseointegrated implant failure: a
meta-analysis. Clin Oral Implants Res. 2006;17:473–478.
 Chrcanovic BR, Albrektsson T, Wennerberg A. Tilted versus axially
placed dental implants: a meta-analysis. J Dent. 2015;43:149–170.
P, Salvado F.
jo Nobre M, Mano Azul A, Rocha E, Malo
 de Arau
Attributable fractions, modifiable risk factors and risk stratification
using a risk score for peri-implant pathology. J Prosthodont Res.
P, Nobre M, Lopes A, Moss SM, Molina GJ. A longitudinal
study of the survival of All-on-4 implants in the mandible with up
to 10 years of follow-up. J Am Dent Assoc. 2011;142:310–320.
 Monje A, Aranda L, Diaz KT, et al. Impact of maintenance therapy
for the prevention of peri-implant diseases: a systematic review and
meta-analysis. J Dent Res. 2016;95(4):372–379.
 Francetti L, Romeo D, Corbella S, Taschieri S, Del Fabbro M. Bone
level changes around axial and tilted implants in full-arch fixed
immediate restorations. Interim results of a prospective study. Clin
Imp Dent Rel Res. 2012;14:646–654.
n E, Gay-Escoda C.
 Mir-Mari J, Mir-Orfila P, Valmaseda-Castello
Long-term marginal bone loss in 217 maschined-surface implants
placed in 68 patients with 5 to 9 years of follow-up: a retrospective
study. Int J Oral Maxillofac Imp. 2012;27:1163–1169.
 Aparicio C, Perales P, Rangert B, Eng M. Tilted implants as an alternative to maxillary sinus grafting: a clinical, radiologic, and periotest
study. Clin Imp Dent Rel Res. 2001;3:39–49.
 Wagenberg BD, Froum SJ, Eckert SE. Long-term bone stability
assessment around 1187 immediately placed implants with 1- to
22-year follow-up. Int J Oral Maxillofac Impl. 2013;28:605–612.
€ m JL. Bone level changes at axial- and non Koutouzis T, Wennstro
axial-positioned implants supporting fixed partial dentures. A 5-year
P, Gonçalves Y, Sabas A, Salvado F. Den jo Nobre M, Malo
 de Arau
tal implants in diabetic patients: retrospective cohort study
reporting on implant survival and risk indicators for excessive marginal bone loss at 5 years. J Oral Rehabil. 2016;43(11):863–870.
 Babbush CA, Kutsko GT, Brokloff J. The all-on-four immediate function treatment concept with Nobel Active implants: a retrospective
study. J Oral Implantol. 2011;37:431–445.
 Balshi TJ, Wolfinger GJ, Slauch RW, Balshi SF. A retrospective analysis of 800 Brånemark system implants following the all-on-FourTM
protocol. J Prosthodont. 2014;23(2):83–88.
jo Nobre M, Malo
How to cite this article: Hopp M, de Arau
Comparison of marginal bone loss and implant success between
axial and tilted implants in maxillary All-on-4 treatment concept
rehabilitations after 5 years of follow-up. Clin Implant Dent Relat
Res. 2017;19:849–859. https://doi.org/10.1111/cid.12526