A. Zembić Immediate vs.

early loading of dental

R. Glauser
A. Khraisat
implants: 3-year results of a randomized
C.H.F. Hämmerle controlled clinical trial

Authors’ affiliations:
A. Zembić, Department of Fixed and Removable
Prosthodontics and Dental Material Science, School
of Dental Medicine, University of Zurich, Zurich,
Switzerland
R. Glauser, Private Practice, Zürich, Switzerland
A. Khraisat, Department of Conservative Dentistry
and Fixed Prosthodontics, Faculty of Dentistry,
University of Jordan, Amman, Jordan
C.H.F. Hämmerle, Department of Fixed and
Removable Prosthodontics and Dental Material
Science, School of Dental Medicine, University of
Zurich, Zurich, Switzerland
Corresponding author:
Dr Anja Zembić
Clinic for Fixed and Removable Prosthodontics and
Dental Material Science
University of Zurich
Plattenstrasse 11
CH-8032 Zurich, Switzerland
Tel.: þ 41 44 634 32 57
Fax: þ 41 44 634 43 05
e-mail: anja.zembic@zzmk.uzh.ch
Key words: bone level, dental implants, early, immediate, partially edentulous
Abstract
Objectives: The aim of the present study was to test whether or not immediately loaded
implants exhibit the same survival rates as early loaded implants.
Material and methods: Eleven patients with bilateral free end mandibles were randomly
assigned to treatment either with immediately (test) or early loaded implants (control). Test
implants received provisionals in occlusion on the day of surgery, control implants 6 weeks
later. Parameters assessed included implant stability quotient (ISQ), plaque, prosthesis
stability and radiographs at baseline (implant insertion), 1 and 3 years. The statistical
analysis was performed by means of Student’s paired t-test and Wilcoxon’s signed-rank test.
The level of significance was set at Po0.05.
Results: After a mean observation period of 39.8 months (36.7–53.1), three test implants
were lost in two patients resulting in a survival rate of 85% compared with 100% for control
implants. At baseline, the mean marginal bone level was significantly higher at test
implants (mean ¼ 0.36 mm, SD  0.5) compared with control implants (1.08  0.37 mm).
For both test and control implants, the bone level significantly decreased from baseline
to 3 years (test: 1.51  0.79 mm; control: 0.89  0.94 mm). The bone loss until 3 years was
not significantly different between test and control group. There was no significant
difference for ISQ both at test and control implants between baseline (test: 63.59  4.62 mm,
control: 65.35  7.43 mm) and 3 years (test: 66.47  7.47 mm, control 68.80  8.75 mm).
Conclusions: Immediate loading was associated with a lower implant survival rate.
Although the test implants were placed with increased sink depth compared with the
control implants, the marginal bone levels were not different between test and control at
3 years.

Date:
Accepted 20 October 2009
To cite this article:
Zembić A, Glauser R, Khraisat A, Hämmerle CHF.
Immediate vs. early loading of dental implants: 3-year
results of a randomized controlled clinical trial.
Clin. Oral Impl. Res. 21, 2010; 481–489.
doi: 10.1111/j.1600-0501.2009.01898.x
Among the initially identified prerequisites
for osseointegration and successful implant
treatment were the primary implant stabi-
lity at the time of placement and the lack of
micromotion during the phase of secondary
bone integration (Albrektsson et al. 1981).
Absence of micromotion is crucial for pre-
venting soft-tissue encapsulation of the
implant (Brunski et al. 1979).
For a long time, it was presumed that
premature loading would inhibit direct
bone apposition onto the implant surface
(osseointegration), thus compromising nor-
mal osseointegration. It was therefore re-
commended to include a 3- to 6-month
healing period before implants were loaded
(Brånemark et al. 1977). This approach is
termed the conventional loading protocol.
Several experimental and clinical studies
have shown that immediate loading does
not necessarily lead to fibrous tissue heal-
ing, but a bone-to-implant contact may be

c 2010 John Wiley & Sons A/S

 481
Zembić et al  Immediate vs. early loading
achieved, which is similar to that of con-
ventionally loaded implants (Piattelli et al.
1997; Ledermann et al. 1999; Romanos
et al. 2001; Nkenke et al. 2003, 2005a,
2005b; Siar et al. 2003; Rocci et al. 2003a).
Analysing the reactions of the surrounding
soft tissues, a systematic review could not
find a deleterious effect resulting from the
immediate loading of implants (Glauser
et al. 2006).
More recently, efforts have therefore
been made to develop immediate loading
protocols. The potential benefits include
reduced treatment times, increase of mas-
ticatory function and improvement of pa-
tient comfort and psychological conditions
(Kinsel et al. 2000; Ganeles et al. 2001;
Roccuzzo et al. 2001; Ganeles & Wismeijer
2004).
In the third ITI Consensus Conference,
immediate loading was defined as a restora-
tion placed in occlusion with the opposing
dentition within 48 h of implant placement
(Cochran et al. 2004). Early loading was
defined as a restoration in contact with the
opposing dentition and placed at least 48 h
after implant placement but not later than
3 months afterward. Delayed loading was
defined as a prosthesis attached in a second
procedure, which occurs some time later
than the conventional healing period of 3–6
months (Cochran et al. 2004).
Immediate loading procedures were first
applied to rehabilitate completely edentu-
lous mandibles with successful results
(Ledermann 1979; Schnitman et al. 1990,
1997; Tarnow et al. 1997; Ganeles et al.
2001; Grunder 2001; Testori et al. 2001).
Subsequently, reports were published on
the treatment of partially edentulous pa-
tients with immediately loaded implants
(Jaffin et al. 2000; Chatzistavrou et al.
2003; Rocci et al. 2003a, 2003b; Ganeles
& Wismeijer 2004; Glauser et al. 2005;
Machtei et al. 2007; Schincaglia et al.
2007). A recent study on immediately
loaded fixed dental prosthesis in the mand-
ible reported promising survival rates of
98.4% after 4 years of loading (Ostman
et al. 2008). In this study, no control group
was included.
At present, four randomized controlled
clinical trials (RCTs) have been published
comparing immediate and early loading in
partially edentulous patients (Testori et al.
2007; Galli et al. 2008; Ganeles et al. 2008;
Merli et al. 2008). Out of these studies, one
482 | Clin. Oral Impl. Res. 21, 2010 / 481–489
study included implant-supported single
crowns and fixed partial dentures (Ganeles
et al. 2008). Another study did not specify,
whether single teeth or multiple teeth were
to be replaced (Merli et al. 2008). At
observation periods of up to 24 months,
no significant differences were found be-
tween test and control implants regarding
implant survival, biologic and prosthetic
complications. Although the results were
encouraging for both procedures, it has to
be kept in mind that in all these studies the
prostheses were not in occlusal contact for
the first 2–6 months (Testori et al. 2007;
Galli et al. 2008; Ganeles et al. 2008; Merli
et al. 2008).
One RCT compared immediately and
early loaded implants in partially edentu-
lous patients with occlusal contacts
(Fischer et al. 2008). At 1 year, immedi-
ately loaded implants exhibited a lower
survival rate (93.8% and 96%) than early
loaded implants (98% and 100%; Fischer
et al. 2008).
Except for the present 3-year study, there
are thus no clinical trials reporting on
immediately compared with early loaded
implants with an observation period longer
than 2 years.
Hence, evidence on the outcome of im-
mediately compared with early loaded im-
plants in partial edentulism is scarce.
The aim of this randomized controlled
clinical study was to test whether or not
immediately loaded implants exhibit the
same survival and complication rates as
early loaded implants.
Material and methods
Patients
The present clinical trial was carried out in
accordance with the Declaration of Hel-
sinki. Each patient received oral and writ-
ten information about the study and
provided informed consent. The study pro-
tocol was approved by the local ethical
committee.
Patients, who exhibited bilateral free end
situations in the mandible desiring fixed
reconstructions, were recruited and conse-
cutively included in this study. All patients
met the following inclusion criteria:
(1) Bilateral posterior edentulism with suf-
ficient bone volume for the placement
of two to three implants posterior to
the canine.
(2) Good physical and mental condition.
(3) Opposing dentition present in the area
intended for implant placement.
Patients were not included into the
study, if they met any of the following
exclusion criteria:
(1) Systemic condition jeopardizing suc-
cessful implant therapy.
(2) Lack of compliance.
(3) Disorders in the area of the planned
implant placement such as chronic
bone diseases, present or previous tu-
mors or irradiation.
(4) Previous bone grafting in the area
planned for implantation.
(5) Clinical signs of bruxism or other
severe functional disorders.
Smoking and local bone regenerative
procedures associated with the implant
placement were allowed within the study.
Patient history was recorded and the
necessary clinical and radiographic exam-
inations were performed and a treatment
plan was developed.
The two sides of the mandible were
randomly assigned to either immediate
loading (test side) or early loading (control
side). The randomization was carried out
using a lot just before the surgery for im-
plant placement.
Surgical procedure
All patients were treated at the University
of Zurich, Clinic for Fixed & Removable
Prosthodontics and Dental Material
Science by one clinician.
The patients were instructed in oral
hygiene procedures and, if needed, scaling
and root planing were performed. The sur-
geries on both sides of the mandible were
performed during the same visit using the
same clinical protocols. Crestal incisions
were made under local anesthesia (Ultra-
cain D-S forte; Aventis Pharma Deutsch-
land, Frankfurt, Germany). If teeth were to
be extracted, this was carried out with
minimal trauma in order to preserve the
buccal bone plate. The extraction sockets
were carefully cleaned and all granulation
tissues were removed. Vertical releasing
incisions were performed when necessary.
Mucoperiosteal flaps were raised, the im-
plant beds prepared and the sites were
c 2010 John Wiley & Sons A/S

 Zembić et al  Immediate vs. early loading

classified with respect to bone quantity and
quality (Lekholm & Zarb 1985).
Tapping was performed when necessary.
Primary implant stability was obtained at
all sites. Depending on the number of teeth
to be replaced, two to three implants were
s
inserted (Brånemark MK IV, TiUnite ;
Nobel Biocare AB, Gothenburg, Sweden).
All implant diameters measured 4 mm
whereas implant lengths ranged from 7 to
15 mm depending on the bone height avail-
able (Table 1). The implants were placed in
the prosthetically correct position irrespec-
tive of the bone width. In cases of exposed
implant threads, regenerative procedures
were performed with autogenous bone
and a xenograft material (Geistlich Bio-
s
Oss ; Geistlich Pharma AG, Wolhusen,
Switzerland) covered by a resorbable col-
s
lagen membrane (Geistlich Bio-Gide ;
Geistlich Pharma AG). When necessary,
resorbable pins made of polylactic acid
s
(Resorpin ; Geistlich Pharma AG) were
used for membrane fixation. Tension-free
flap adaptation allowing for transmucosal
healing in both groups was obtained and
the flaps were sutured with e-PTFE sutures
s
(GoreTex ; W.L. Gore & Associates Inc.,
Flagstaff, AZ, USA).
The patients received antibiotics
s
(Clamoxyl 750 mg, three times a day
[Glaxosmithkline AG, Münchenbuchsee,
Switzerland]) and analgetics (Mefenacid
s
500 mg, three times a day [Streuli Pharma
AG, Uznach, Switzerland]) for 4–5 days.
Mouth rinsing was prescribed twice daily
for 2 weeks with a solution of 0.2% chlor-
hexidine digluconate. Sutures were re-
moved 7–10 days postoperatively.
Prosthetic procedures
The implants of the test group were recon-
structed with screw-retained temporary
FDP at the day of surgery. Based on the
diagnostic wax-up, acrylic provisionals
were fabricated in the dental laboratory.
Table 1. Distribution (n) of implant length in test (immediately loaded) and control (early
loaded) implants
Test (immediate) (n ¼ 22)
Implant length (mm)
7 1
8.5 0 2
10 8 8
11.5 4 2
13 5 9
15 4 1
Utilizing the manufacturer’s cylinders spe-
cifically made for temporaries, these provi-
sionals were directly relined in the
patient’s mouth using a hybride composite
and all the implants were thus splinted
s
(Tetric ; Ivoclar Vivadent AG, Schaan,
Liechtenstein). The reconstructions were
designed to exhibit full centric occlusal
contacts whereas excentric contacts were
avoided.
In the control group, abutments (Multi-
s
Unit-Abutments ; Nobel Biocare AB)
were placed on the day of surgery and
impressions were taken at the abutment
level. The acrylic provisionals were fabri-
cated in the laboratory and inserted
6 weeks after implant placement. All
patients were enrolled in an individually
designed maintenance care program with
visit of one to four times per year.
Clinical and radiographic evaluations
Follow-up examinations were performed at
1, 2, 4, and 6 weeks and at 2, 3, 6, 12, 13.5
and 36 months. For the control group, the
1-year follow-up was set at 13.5 months, as
the prosthesis was inserted 6 weeks after
implant placement.
The occurrence of adverse events was
recorded at each follow-up examination.
Radiographs (Kodak Ultra Speed; East-
man Kodak Co., Rochester, NY, USA)
were taken perpendicularly to the implant
axis with the long-cone paralleling techni-
que (Updegrave 1951) at the time of pros-
thesis insertion, at 3, 6, 12 and 36 months.
For an evaluation of the marginal bone
level, the radiographs were digitized and
the level of the crestal bone was assessed.
Two independent examiners measured the
distance in 0.1 mm increments between
the implant shoulder as the reference point
and the most coronal bone-to-implant con-
tact mesially and distally of the implants.
The known distance between three im-
plant threads (1.8 mm) was used for
Control (early) (n ¼ 22)
0
purposes of calibration and determination
of the exact magnification of the images.
When a difference between the two mea-
surements of 0.5 mm was recorded, the
mean value of the two measurements was
used. In cases of differences 40.5 mm, the
radiographs were reexamined by both ex-
aminers and the results were discussed
until a consensus was found (bone level
measurements in accordance to Rodoni
et al. 2005).
Implant stability was assessed by reso-
nance frequency measurements (Osstellt;
Integration Diagnostics AB, Sävedalen,
Sweden) at implant insertion and each
follow-up examination. All measurements
were recorded at the abutment levels. A
first-generation resonance frequency trans-
ducer was used.
In addition, at all examinations follow-
ing the insertion of the provisionals, the
presence of technical complications (screw
loosening, fracture of materials) was as-
sessed.
The plaque control record (PCR) was
assessed at the buccal and lingual aspect
of the implants at 12 and 36 months
(O’Leary et al. 1972).
Survival rate
A ‘‘surviving implant’’ was considered, if
the implant remained in the jaw following
prosthesis removal.
Statistical analysis
Descriptive statistics were applied to the
data by a biostatistician masked with re-
spect to the clinical protocols. The patient
was considered as the statistical unit. In
situations, where more than two implants
were present on one or both sides of the
mandible, only two were chosen for analy-
sis. The implant in the middle position was
excluded from the analysis. Thus, every
patient contributed two test and two con-
trol implants. The values measured for the
two test or two control implants in each
patient were averaged, as well as for mesial
and distal bone levels. The statistical ana-
lysis of implant stability quotient (ISQ)
values for both groups was performed at
baseline, 12 and 36 months. Missing data
at any of the three time points was not
considered for the analysis. The signifi-
cance of differences between the test and
the control group was assessed by means of
Student’s paired t-test and Wilcoxon’s

c 2010 John Wiley & Sons A/S

 483 | Clin. Oral Impl. Res. 21, 2010 / 481–489
Zembić et al  Immediate vs. early loading
signed-rank test. The paired t-test was
applied only for the analysis of the variables
for which the Kolmogorov–Smirnov test
was not significant. In the other cases,
the non-parametric Wilcoxon’s signed-
rank test was applied. The level of signifi-
cance was set at Po0.05. A Bonferroni
correction could be applied by considering
the results of the tests with a P-value
smaller than 0.05/k as significant. In this
formula, k is equal to the number of
different tests. For k equal to 5, the cor-
rected a niveau is equal to 0.01.
Results
Patients and implants
Eleven patients (three women, eight men)
with 51 implants were included in the
study. The mean age of the patients was
54.8 years (range 37.8–68.6 years) at the
time of surgery. Five patients were smo-
kers. Ten patients were available for the
3-year examination. The mean follow-up
time after implant placement was 39.8
months (36.7–53.1).
In the test group, 25 implants and in the
control group, 26 implants were placed.
According to the protocol, all middle im-
plants were excluded from the analysis
leaving 22 implants under investigation in
each of the two groups. In two patients,
three implants were placed on the test side.
In another, three patients the control side
included three implants. Finally, in one
patient on both test and control sides, three
implants were placed. Hence, a total of
seven implants, which were standing be-
tween two study implants, were excluded
from the analysis.
Thirty-six implants (81.8%) were placed
in healed sites. Eight implants (18.2%)
were placed into the alveoli immediately
after tooth extraction. Out of these, three
(6.8%) were test and five (11.4%) were
control implants (Table 2).
In total, 31 implants (test: 68.2%, con-
trol: 72.7%) had bone volume deficiencies
and were treated with GBR procedures.
The remaining 13 implants could be placed
according to standard procedures.
Adverse events and implant failures
One week after surgery, in one patient
soft-tissue dehiscencies were observed at
both test and control implants. At the
484 | Clin. Oral Impl. Res. 21, 2010 / 481–489
Table 2. Distribution (n) of patients with implants that were placed in extraction sockets
and healed sites
Test (immediate) n ¼ 11
Implants placed in extraction sockets 2 patients (3 implants)
Implants placed in healed sites 9 patients (19 implants)
Table 3. Mean marginal bone level in millimeters (mean, SD) of test and control implants at
baseline (implant placement), 12 and 36 months, brackets indicate significant differences
at po0.05
Baseline
Bone level
Test 0.36  0.50
Control 1.08  0.37
most distal test implant, a painful soft-
tissue ulceration was present. Two years
after implant insertion, all the three
splinted test implants in this patient had
to be removed due to loss of osseointegra-
tion. The patient was a smoker. The im-
plant lengths were 10, 10 and 7 mm. One
test implant was an immediate implant.
Minor local bone regeneration was per-
formed at test sites at implant placement
to regenerate infrabony defects. ISQ values
decreased from baseline (63, 69, 67) to 2
months (51, 53, 55). No further ISQ mea-
surements were performed. This patient
also exhibited impaired wound healing at
various areas of the skin. In spite of general
medical examinations, the cause for the
disturbed wound healing could not be
identified. In this patient, statistical ana-
lysis was only performed up to the 1-year
examination.
In another patient, one test implant had
to be removed due to lack of osseointegra-
tion at the 1-year examination. This im-
plant was the most mesial of three splinted
implants, 13 mm long and was placed in
conjunction with minor bone grafting. The
patient was a smoker. The ISQ value of
this implant decreased from 70 at implant
placement to 52 at the 6-week follow-up.
No further ISQ measurements were per-
formed due to uncomfortable sensations at
this implant.
All other patients exhibited uneventful
wound healing and normal hard and soft-
tissue integration at both test and control
implants.
This resulted in a cumulative implant
survival rate of 85% for immediately
loaded and 100% for early loaded implants
at 3 years.
Control (early) (n ¼ 11)
4 patients (5 implants)
7 patients (17 implants)
12 months 36 months
1.63  0.63 1.91  0.55
1.50  0.54 1.95  0.88
Clinical and radiographic evaluations
The bone level at the test implants mea-
sured 0.36  0.5 mm (mean and standard
deviation) at implant insertion, 1.63 
0.63 mm at 1 year and 1.91  0.55 mm
at 3 years (Table 3, Fig. 1). The mean bone
resorption at test implants was significant
both from baseline to 1 year (1.23 
0.89 mm; P ¼ 0.0017) and baseline to 3
years (1.51  0.79 mm; P ¼ 0.0004)
(Table 4).
The bone level of the control implants
amounted to 1.08  0.37 mm at implant
insertion, 1.50  0.54 mm at 1 year and
1.95  0.88 mm at 3 years (Table 3,
Fig. 2). The decrease of the bone level was
0.44  0.66 mm (P ¼ 0.0645) from base-
line to 1 year. There was a significant
decrease of bone from baseline to 3 years
(0.89  0.94; P ¼ 0.0151) (Table 4).
The bone levels for test and control
implants at baseline differed significantly
with higher values for the test implants
(P ¼ 0.0017) (Table 3). The bone loss until
3 years was not significantly different be-
tween test and control group (P ¼ 0.0690).
The distribution of test and control
implants with marginal bone resorption
in 0.5 mm measures is demonstrated in
Fig. 3.
ISQ values assessed at all follow-up ex-
aminations, including the ones used for
statistical analysis, are shown in Table 6,
Fig. 4. There was no significant difference
for ISQ both at test and control implants
between baseline (test: 63.59  4.62,
control: 65.35  7.43) and 1 year (test:
66.58  6.35, P ¼ 0.2087; control:
69.77  6.80, P ¼ 0.3197) (Table 5). In
addition, there was no significant differ-
ence between baseline and 3 years (test:
c 2010 John Wiley & Sons A/S

 Zembić et al  Immediate vs. early loading

66.47  7.47, P ¼ 0.2738; control: 68.80 

8.75, P ¼ 0.2157). Comparing the change
in ISQ from baseline to 3 years for test
and control implants, there was no signifi-
cant difference between the two groups
(P ¼ 0.3661).
Regarding prosthesis stability at 3 years,
30% of the provisionals exhibited fractures
of the veneering composite in both groups.
In three patients, metal–ceramic crowns
were inserted to replace the provisionals,
which showed extensive fractures. No
screw loosenings occurred in either group.
At the 1-year follow-up examination,
45% of the test and 35% of the control
implants exhibited plaque. At the 3-year
follow-up examination, 73.7% of the test
and 60% of the control implants exhibited
plaque.

Fig. 1. Scatter plot of mean marginal bone levels from test implants at baseline (implant placement), 1 and 3
years. Discussion

This RCT revealed a lower survival rate for

Table 4. Marginal bone resorption in millimeters (mean, SD) from the time of baseline immediately loaded implants (85%) com-
(implant placement) to 1 and 3 years pared with early loaded implants (100%)
Baseline to 1 year n Baseline to 3 years n
after an observation period of 3 years. Bone
Bone resorption levels decreased significantly in both
Test 1.23  0.89n 10 1.51  0.79n 9
groups from baseline to 3 years. In both
Control 0.44  0.66 10 0.89  0.94n 10
groups, the implant stability measure-
n
Significant differences during the observation period at Po0.05. ments revealed an increase in ISQ values
from baseline to 1 year and virtually no
further changes until 3 years.
Previous prospective studies on immedi-
ate loading without control groups reported
survival rates ranging from 88.5% to 100%
after observation periods ranging from 1 to
7 years in partially dentate mandibles
(Nikellis et al. 2004; Degidi & Piattelli
2005; Glauser et al. 2005). Regarding im-
mediate vs. conventional loading, the only
two studies existing with a control group
showed 2-year survival rates of 95.7% and
100% for immediately loaded implants
compared with 100% for conventionally
loaded implants in partially dentate pa-
tients (Cannizzaro & Leone 2003; Roma-
nos & Nentwig 2006).
The lower survival rate for immediately
loaded implants in the present study is in
agreement with previous clinical studies of
similar design, comparing immediate and
early loading in partially edentulous pa-
tients (Testori et al. 2007; Galli et al.
2008). Although in these studies, occlusal
Fig. 2. Scatter plot of mean marginal bone levels from control implants at baseline (implant placement), 1 and contacts were avoided for the first 2 months
3 years. unlike in the present study, where occlusal

c 2010 John Wiley & Sons A/S

 485 | Clin. Oral Impl. Res. 21, 2010 / 481–489
Zembić et al  Immediate vs. early loading
contacts were incorporated in both test
and control groups. In another con-
trolled human trial, immediately and de-
layed loaded implants with full occlusal
contacts in partially edentulous patients
were compared (Cannizzaro & Leone
2003). Interestingly, after a mean loading
period of 24 months, the immediately
loaded implants showed a higher survival
rate (100%) than the delayed loaded im-
plants (93%; Cannizzaro & Leone 2003).
A RCT evaluated the influence of immedi-
ately loaded single implants with and with-
out occlusal contacts on survival, bone loss
and esthetic outcome (Lindeboom et al.
2006). At 1 year of loading, the survival
rate for implants with occlusal contacts
was even higher than for implants without
Fig. 3. Distribution of test and control implants (n, y-axis) with marginal bone resorption in 0.5 mm measures
from baseline to 3 years – frequency analysis.
Fig. 4. Distribution of mean implant stability quotient (ISQ) values and standard deviations of test and control implants at the various visits.
486 | Clin. Oral Impl. Res. 21, 2010 / 481–489
occlusal contacts (92% vs. 88%). No sig-
nificant differences were found with regard
to bone loss and gingival esthetics. The
patients were prescribed a soft diet and
informed to avoid chewing in the area of
the implants.
It thus remains to be determined to what
extent occlusal contacts play a role for the
survival of immediately loaded implants.
The bone levels in this study decreased
from baseline to 3 years in both groups
(1.5  0.8 mm for test and 0.9  0.9 mm
for control implants). The bone levels at test
and control implants differed significantly at
baseline with more deeply placed test im-
plants. As a result, the bone resorption
between baseline and the 3-year examina-
tion was more excessive for test implants
and resulted in similar bone levels at test and
control implants at 3 years. One possible
reason for this initial difference may be that
the randomization was performed before
placement of the implants. Hence, the sur-
geon may have been biased during implant
placement. In a similarly designed recent
study, the randomization was also per-
formed before implant placement resulting
in immediately loaded implants placed sig-
nificantly deeper than early loaded implants
(Ganeles et al. 2008). The bone-level change
was significantly greater for immediately
loaded implants (0.9  0.9 mm) compared
with early loaded implants (0.6  1 mm) at
1 year of loading. Adjusting for the difference
in the initial insertion depth, however,
statistical analysis showed no significant
difference of bone-level changes between
the loading groups (Ganeles et al. 2008).
In a recent randomized clinical trial of
similar design mentioned above, a signifi-
cant marginal bone loss was found both for
immediately (1.1  0.58 mm) and early
loaded implants (1.1  0.54 mm) during
14 months of observation (Galli et al.
2008).
In the present study, efforts were made
to obtain a favorable distribution of the
loading forces to the bone-to-implant inter-
face. All implants in the same quadrant
Table 5. Implant stability quotient ISQ
(mean, SD) for test and control implants
at baseline (implant placement), 12 and 36
months, no significant differences
Baseline 12 36
months months
ISQ
Immediate 63.5  4.7 66.7  6.2 66.5  7.3
Early 65.3  7.4 69.8  6.8 68.8  8.8
c 2010 John Wiley & Sons A/S


Table 6. Distribution of mean ISQ values and standard deviations at all time points for both groups (no test for statistical significance was
performed)
Baseline 1 week 2 weeks
Immediate 63.3 63.1 62.9
Standard deviation 3.3 6.5 3.0
Early 65.5 64.6 64.4
Standard deviation 3.3 3.5 3.7
were splinted. This was carried out in order
to reduce the maximum loading force to
one single implant. In addition, working
side contacts were restricted to the canines
in order to reduce the lateral forces on the
implants. These measures were taken with
the intention to keep micromotion below
the physiological threshold (Brunski et al.
1979; Brunski 1999).
In the current study, eight implants were
placed into extraction sockets and 36 im-
plants in healed site. Even though more
control (n ¼ 5) than test (n ¼ 3) implants
were placed in extraction sockets, there
were no failures in the control group,
whereas one test implant placed in an
extraction socket failed. The results of
one retrospective study showed a higher
risk for failure for immediately loaded sin-
gle implants, when placed into fresh ex-
traction sockets compared with placement
into healed sites (Chaushu et al. 2001).
The survival rates were 82.4% at 13
months for immediately loaded implants
in extraction sockets (n ¼ 19) in compari-
son with 100% for immediately loaded
implants in healed sites (n ¼ 9) at 16
months.
Depending on the clinical situation,
some of the patients in the present study
received two and others three implants in
one mandibular quadrant. For reasons of
uniformity, only the two externally located
implants were chosen for analysis. The
magnitude of effect of this incongruency
between the two groups remains unclear,
but as it was not systematically performed
in one group or the other, it is expected to
be minimal.
With respect to implant stability, the
results indicated an increase in ISQ in
References
Albrektsson, T., Brånemark, P.-I., Hanson, H.A. &
Lindström, J. (1981) Osseointegrated titanium
implants: requirements for ensuring a long-
lasting, direct bone to implant anchorage in
c 2010 John Wiley & Sons A/S
 487 |
4 weeks 6 weeks 2 months 3 months
62.1 64.1 64.7 63.9
3.6 5.3 5.5 6.6
64.4 67.1 64.1 65.3
5.1 5.8 4.7 3.6
both groups from implant placement to
1 year. From 1 to 3 years, the ISQ values
remained stable. This ISQ increase during
the first year of loading was also observed
in other studies on immediately loaded
implants (Glauser et al. 2005; Fischer
et al. 2008). Obviously, the implant stability,
which is a reflection of the bone-to-im-
plant interface, is not negatively influenced
by the immediate loading protocol. Con-
cerning the implant stability in conven-
tionally loaded implants, there also were
no significant differences in previous stu-
dies compared with immediately loaded
implants in partially edentulous patients
(measured with Periotest; Cannizzaro &
Leone 2003; Romanos & Nentwig 2006).
The high plaque scores in this study
might be attributed to the acrylic material
of the provisionals. A recent review found
an influence of surface roughness, surface-
free energy and the chemical composition
of biomaterials on the biofilm formation on
restorative materials (Teughels et al. 2006).
Based on these data, it is expected that the
rather rough surfaces of the acrylic were the
reason for the high plaque scores.
Several factors have earlier been pro-
posed, which may decrease the risks for
failure of immediately loaded implants,
including: under-preparation of the bone
at the implant sites in order to condensate
bone and achieve higher primary stability,
non-occluding provisional restorations
within the first 2 months of healing, sta-
bilization of implants by splinting and the
use of biocompatible prosthetic material
(for a review, see Morton et al. 2004). In
the present study, only splinting was in-
cluded from the list above. It would be
interesting to know, if the inclusion of
man. Acta Orthopaedica Scandinavica 52:
155–170.
Brånemark, P.-I., Hansson, B.O., Adell, R., Breine,
U., Lindtrom, J., Hallen, O. & Ohman, A. (1977)
Zembić et al  Immediate vs. early loading
6 months 12 months 13.5 months 36 months
65.1 67.1 67.2 66.1
4.0 2.0 3.5 3.2
68.9 68.9 69.8 70.4
2.0 2.8 2.8 4.0
the other factors would have increased the
survival rate of the implants in the present
study. Hence, RCTs are necessary to test
whether or not and to what extent these
suggested factors affect the survival rates of
immediately loaded implants.
Owing to the small number of patients
examined and the other limitations listed,
the results of the present study should be
interpreted with caution. Nevertheless, the
results indicate that immediate loading
represents a greater risk for implant failure
in the intermediate term compared with
early loading.
Conclusions
The present study demonstrated a higher
failure rate for immediately loaded im-
plants in occlusion in the posterior mand-
ible compared with early loaded implants.
Although the test implants were placed
with increased sink depth compared with
the control implants, the marginal bone
levels were not different between test and
control at 3 years.
Acknowledgements The authors
thank Dr Malgorzata Roos (Institute for
Social and Preventive Medicine,
Department for Biostatistics, University
of Zurich, Switzerland) for the
statistical analysis of the data. The
study was partially supported by the
Department of Fixed and Removable
Prosthodontics and Dental Material
Science, School of Dental Medicine,
University of Zurich, Switzerland and a
research grant from Nobel Biocare AB,
Gothenburg, Sweden.
Osseointegrated implants in the treatment of
edentulous jaw. Experience from a 10-year period.
Scandinavian Journal of Plastic and Reconstruc-
tive Surgery 16 (Suppl.): 1–132.
Clin. Oral Impl. Res. 21, 2010 / 481–489
Zembić et al  Immediate vs. early loading
Brunski, J.B. (1999) In vivo bone response to bio-
mechanical loading at the bone-dental-implant
interface. Advances in Dental Research 13:
99–119.
Brunski, J.B., Moccia, A.F. Jr, Pollack, S.R., Korost-
off, E. & Trachtenberg, D.I. (1979) The influence
of functional use of endosseous dental implants on
the tissue–implant interface. II. Clinical aspects.
Journal of Dental Research 58: 1970–1980.
Cannizzaro, G. & Leone, M. (2003) Restoration of
partially edentulous patients using dental im-
plants with a microtextured surface: a prospective
comparison of delayed and immediate full occlu-
sal loading. The International Journal of Oral &
Maxillofacial Implants 18: 512–522.
Chatzistavrou, M., Felton, D.A. & Cooper, L.F.
(2003) Immediate loading of dental implants in
partially edentulous patients: a clinical report.
Journal of Prosthodontics 12: 26–29.
Chaushu, G., Chaushu, S., Tzohar, A. & Dayan, D.
(2001) Immediate loading of single-tooth im-
plants: immediate versus non-immediate implan-
tation. A clinical report. The International
Journal of Oral & Maxillofacial Implants 16:
267–272.
Cochran, D.L., Morton, D. & Weber, H.P. (2004)
Consensus statements and recommended clinical
procedures regarding loading protocols for endoss-
eous dental implants. The International Journal
of Oral & Maxillofacial Implants 19 (Suppl.):
109–113.
Degidi, M. & Piattelli, A. (2005) 7-year follow-up of
93 immediately loaded titanium dental implants.
Journal of Oral Implantology 31: 25–31.
Fischer, K., Backstrom, M. & Sennerby, L. (2008)
Immediate and early loading of oxidized tapered
implants in the partially edentulous maxilla:
a 1-year prospective clinical, radiographic, and
resonance frequency analysis study. Clinical Im-
plant Dentistry & Related Research 11: 69–80.
Galli, F., Capelli, M., Zuffetti, F., Testori, T. &
Esposito, M. (2008) Immediate non-occlusal vs.
early loading of dental implants in partially
edentulous patients: a multicentre randomized
clinical trial. Peri-implant bone and soft-tissue
levels. Clinical Oral Implants Research 19:
546–552.
Ganeles, J., Rosenberg, M.M., Holt, R.L. & Reich-
man, L.H. (2001) Immediate loading of implants
with fixed restorations in the completely edentu-
lous mandible: report of 27 patients from a private
practice. The International Journal of Oral &
Maxillofacial Implants 16: 418–426.
Ganeles, J. & Wismeijer, D. (2004) Early and im-
mediately restored and loaded dental implants for
single-tooth and partial-arch applications. The
International Journal of Oral & Maxillofacial
Implants 19 (Suppl.): 92–102.
Ganeles, J., Zollner, A., Jackowski, J., ten Bruggen-
kate, C., Beagle, J. & Guerra, F. (2008) Immediate
and early loading of straumann implants with a
chemically modified surface (slactive) in the pos-
terior mandible and maxilla: 1-year results from a
prospective multicenter study. Clinical Oral Im-
plants Research 19: 1119–1128.
Glauser, R., Ruhstaller, P., Windisch, S., Zembic,
A., Lundgren, A., Gottlow, J. & Hammerle, C.H.
(2005) Immediate occlusal loading of Branemark
488 | Clin. Oral Impl. Res. 21, 2010 / 481–489
system tiunite implants placed predominantly in
soft bone: 4-year results of a prospective clinical
study. Clinical Implant Dentistry & Related
Research 7 (Suppl. 1): S52–59.
Glauser, R., Zembic, A. & Hammerle, C.H. (2006)
A systematic review of marginal soft tissue at
implants subjected to immediate loading or im-
mediate restoration. Clinical Oral Implants Re-
search 17 (Suppl. 2): 82–92.
Grunder, U. (2001) Immediate functional loading of
immediate implants in edentulous arches: two-
year results. International Journal of Periodontics
and Restorative Dentistry 21: 545–551.
Jaffin, R.A., Kumar, A. & Berman, C.L. (2000)
Immediate loading of implants in partially and
fully edentulous jaws: a series of 27 case reports.
Journal of Periodontology 71: 833–838.
Kinsel, R.P., Lamb, R.E. & Moneim, A. (2000)
Development of gingival esthetics in the edentulous
patient with immediately loaded, single-stage,
implant-supported fixed prostheses: a clinical report.
The International Journal of Oral & Maxillofacial
Implants 15: 711–721.
Ledermann, P. (1979) Bar-prosthetic management
of the edentulous mandible by means of
plasma-coated implantation with titanium
screws. Deutsche Zahnärztliche Zeitschrift 34:
907–911.
Ledermann, P.D., Schenk, R.K. & Buser, D. (1999)
Long-lasting osseointegration of immediately
loaded bar-connected TPS screws after 12 years
of function: a histologic case report of a 95-year-
old patient. International Journal of Periodontics
and Restorative Dentistry 18: 553–556.
Lekholm, U. & Zarb, G.A. (1985) Patient selection
and preparation. In: Brånemark, P.-I., Zarb, G.A.
& Albrektsson, T., eds. Tissue-Integrated Pros-
theses: Osseointegration in Clinical Dentistry,
199–210. Chicago: Quintessence.
Lindeboom, J.A., Frenken, J.W., Dubois, L., Frank,
M., Abbink, I. & Kroon, F.H. (2006) Immediate
loading versus immediate provisionalization of
maxillary single-tooth replacements: a prospec-
tive randomized study with biocomp implants.
The International Journal of Oral and Maxillofa-
cial Surgery 64: 936–942.
Machtei, E.E., Frankenthal, S., Blumenfeld, I., Gut-
macher, Z. & Horwitz, J. (2007) Dental implants
for immediate fixed restoration of partially eden-
tulous patients: a 1-year prospective pilot clinical
trial in periodontally susceptible patients. Journal
of Periodontology 78: 1188–1194.
Merli, M., Bernardelli, F. & Esposito, M. (2008)
Immediate versus early nonocclusal loading of
dental implants placed with a flapless procedure
in partially edentulous patients: preliminary re-
sults from a randomized controlled clinical trial.
International Journal of Periodontics and Re-
storative Dentistry 28: 453–459.
Morton, D., Jaffin, R. & Weber, H.P. (2004) Im-
mediate restoration and loading of dental im-
plants: clinical considerations and protocols. The
International Journal of Oral & Maxillofacial
Implants 19 (Suppl.): 103–108.
Nikellis, I., Levi, A. & Nicolopoulos, C. (2004)
Immediate loading of 190 endosseous dental im-
plants: a prospective observational study of 40
patient treatments with up to 2-year data. The
International Journal of Oral & Maxillofacial
Implants 19: 116–123.
Nkenke, E., Fenner, M., Vairaktaris, E.G., Neukam,
F.W. & Radespiel-Troger, M. (2005a) Immediate
versus delayed loading of dental implants in the
maxillae of minipigs. Part II: histomorphometric
analysis. The International Journal of Oral &
Maxillofacial Implants 20: 540–546.
Nkenke, E., Lehner, B., Fenner, M., Roman, F.S.,
Thams, U., Neukam, F.W. & Radespiel-Troger,
M. (2005b) Immediate versus delayed loading of
dental implants in the maxillae of minipigs: fol-
low-up of implant stability and implant failures.
The International Journal of Oral & Maxillofa-
cial Implants 20: 39–47.
Nkenke, E., Lehner, B., Weinzierl, K., Thams, U.,
Neugebauer, J., Steveling, H., Radespiel-Troger,
M. & Neukam, F.W. (2003) Bone contact,
growth, and density around immediately loaded
implants in the mandible of mini pigs. Clinical
Oral Implants Research 14: 312–321.
O’Leary, T.J., Drake, R.B. & Naylor, J.E. (1972) The
plaque control record. Journal of Periodontology
43: 38.
Ostman, P.O., Hellman, M. & Sennerby, L. (2008)
Immediate occlusal loading of implants in the
partially edentate mandible: a prospective 1-year
radiographic and 4-year clinical study. The Inter-
national Journal of Oral & Maxillofacial Im-
plants 23: 315–322.
Piattelli, A., Paolantonio, M., Corigliano, M. &
Scarano, A. (1997) Immediate loading of titanium
plasma-sprayed screw-shaped implants in man: a
clinical and histological report of two cases. Jour-
nal of Periodontology 68: 591–597.
Rocci, A., Martignoni, M., Burgos, P.M., Gottlow,
J. & Sennerby, L. (2003a) Histology of retrieved
immediately and early loaded oxidized implants:
light microscopic observations after 5 to 9 months
of loading in the posterior mandible. Clinical
Implant Dentistry & Related Research 5 (Suppl.
1): 88–98.
Rocci, A., Martignoni, M. & Gottlow, J. (2003b)
Immediate loading of Branemark system tiunite
and machined-surface implants in the posterior
mandible: a randomized open-ended clinical trial.
Clinical Implant Dentistry & Related Research 5
(Suppl. 1): 57–63.
Roccuzzo, M., Bunino, M., Prioglio, F. & Bianchi,
S.D. (2001) Early loading of sandblasted and acid-
etched (SLA) implants: a prospective split-mouth
comparative study. Clinical Oral Implants Re-
search 12: 572–578.
Rodoni, L.R., Glauser, R., Feloutzis, A. & Ham-
merle, C.H. (2005) Implants in the posterior
maxilla: a comparative clinical and radiologic
study. The International Journal of Oral & Max-
illofacial Implants 20: 231–237.
Romanos, G., Toh, C.G., Siar, C.H., Swaminathan,
D., Ong, A.H., Donath, K., Yaacob, H. & Nent-
wig, G.H. (2001) Peri-implant bone reactions to
immediately loaded implants. An experimental
study in monkeys. Journal of Periodontology 72:
506–511.
Romanos, G.E. & Nentwig, G.H. (2006) Immediate
versus delayed functional loading of implants in
the posterior mandible: a 2-year prospective clin-
ical study of 12 consecutive cases. International
c 2010 John Wiley & Sons A/S


Journal of Periodontics and Restorative Dentistry
26: 459–469.
Schincaglia, G.P., Marzola, R., Scapoli, C. & Scotti, R.
(2007) Immediate loading of dental implants sup-
porting fixed partial dentures in the posterior
mandible: a randomized controlled split-mouth
study – machined versus titanium oxide implant
surface. The International Journal of Oral &
Maxillofacial Implants 22: 35–46.
Schnitman, P.A., Wohrle, P.S. & Rubenstein, J.E.
(1990) Immediate fixed interim prostheses supported
by two-stage threaded implants: methodology and
results. Journal of Oral Implantology 16: 96–105.
Schnitman, P.A., Wöhrle, P.S., Rubenstein, J.E., Da
Silva, J.D. & Wang, N.H. (1997) Ten-year results
for Brånemark implants immediately loaded with
fixed prostheses at implant placement. The Inter-
national Journal of Oral & Maxillofacial Im-
plants 12: 495–503.
Siar, C.H., Toh, C.G., Romanos, G., Swaminathan,
D., Ong, A.H., Yaacob, H. & Nentwig, G.H.
(2003) Peri-implant soft tissue integration of im-
mediately loaded implants in the posterior maca-
que mandible: a histomorphometric study.
Journal of Periodontology 74: 571–578.
c 2010 John Wiley & Sons A/S
 489 |
Tarnow, D.P., Emtiaz, S. & Classi, A. (1997) Im-
mediate loading of threaded implants at stage 1
surgery in edentulous arches: ten consecutive case
reports with 1- to 5-year data. The International
Journal of Oral & Maxillofacial Implants 12:
319–324.
Testori, T., Galli, F., Capelli, M., Zuffetti, F. &
Esposito, M. (2007) Immediate nonocclusal ver-
sus early loading of dental implants in partially
edentulous patients: 1-year results from a multi-
center, randomized controlled clinical trial. The
International Journal of Oral & Maxillofacial
Implants 22: 815–822.
Testori, T., Szmukler-Moncler, S., Francetti, L., Del
Fabbro, M., Scarano, A., Piattelli, A. & Wein-
stein, R.L. (2001) Immediate loading of osseotite
implants: a case report and histologic analysis
after 4 months of occlusal loading. International
Journal of Periodontics and Restorative Dentistry
21: 451–459.
Teughels, W., Van Assche, N., Sliepen, I. & Quir-
ynen, M. (2006) Effect of material characteristics
and/or surface topography on biofilm develop-
ment. Clinical Oral Implants Research 17
(Suppl. 2): 68–81.
Zembić et al  Immediate vs. early loading
Updegrave, W.J. (1951) The paralleling extension-
cone technique in intraoral dental radiography.
Oral Surgery, Oral Medicine, Oral Pathology,
Oral Radiology, and Endodontology 4:
1250–1261.
Supporting Information
Additional Supporting Information
may be found in the online version of
this article:
Table S1. Supporting information in
accordance with the CONSORT
Statement 2001 checklist used in
reporting randomized trials.
Please note: Wiley-Blackwell is not
responsible for the content or
functionality of any supporting
materials supplied by the authors. Any
queries (other than missing material)
should be directed to the
corresponding author for the article.
Clin. Oral Impl. Res. 21, 2010 / 481–489