doi:10.1111/j.1365-2591.2009.01561.

REVIEW

Endodontics or implants? A review of decisive

criteria and guidelines for single tooth restorations
and full arch reconstructions

N. U. Zitzmann, G. Krastl, H. Hecker, C. Walter & R. Weiger

Clinic for Periodontology, Endodontology and Cariology, University of Basel, Hebelstrasse, Basel, Switzerland

Abstract implant diseases are established and detected. Good

long-term success rates and greater flexibility in clinical
Zitzmann NU, Krastl G, Hecker H, Walter C, Weiger R.
management indicate that RCT or retreatment should
Endodontics or implants? A review of decisive criteria and
be performed first in most instances unless the tooth is
guidelines for single tooth restorations and full arch recon-
judged to be unrestorable. When deciding if a compro-
structions. International Endodontic Journal, 42, 757–774, 2009.
mised tooth of questionable prognosis should be
This review describes practical criteria and a systematic maintained or replaced by an implant, both local,
process to aid the treatment planning decision of site-specific and more general patient-related factors
whether to preserve teeth by root canal treatment should be considered. Following systematic evaluation
(RCT) or extract and provide an implant. Recommen- and consideration of the best treatment option in a
dations presented are based on best available evidence particular case, a treatment recommendation may then
from the literature and the expert views of specialists in be given in favour or against tooth retention. Whilst
endodontics and restorative dentistry, including dental single risks are possibly accepted for single tooth
implantology. A MEDLINE search was conducted using restorations, teeth with questionable prognosis and
the terms ‘root canal therapy’, ‘dental implants’, multiple pre-treatment requirements are better not
‘decision making’, ‘treatment planning’, ‘outcome’ included as abutments in fixed dental prostheses to
and ‘human’, and supplemented by hand-searching. reduce the risk to survival of the entire restoration.
When evaluating the outcome of root canal treatment,
Keywords: decision making, dental implants,
an observation period of 4–5 years is required for
endodontics, long-term results, review, tooth prognosis.
complete healing of periapical lesions. Dental implants,
however, present a de novo situation and a functional Received 23 September 2008; accepted 4 February 2009
period of at least 5 years is often required before peri-

implant. Dentists appear to make the decision for

Introduction
extracting a tooth on the basis of multiple risk factors
Clinicians frequently face the dilemma of whether to including endodontic and periodontal criteria, remain-
endodontically treat and retain a questionable tooth or ing tooth structure, restorability with core build-ups
to extract and potentially replace it with a dental and post and core, extent of previous restorations as
well as the perceived strategic value of a tooth within
the dentition. Whilst single identifiable risks may be
Correspondence: Prof Dr Nicola U. Zitzmann, PhD, Associate easy to manage clinically, the presence of multiple risks
Professor, Clinic for Periodontology, Endodontology and Cari-
ology, University of Basel, Hebelstrasse 3, CH-4056 Basel,
appears to jeopardize the survival of a compromised
Switzerland (Tel.: +41 61 267 2613; fax: +41 61 267 2659; tooth (Pothukuchi 2006, Wolcott & Meyers 2006).
e-mail: n.zitzmann@unibas.ch). Evidence-based data from the literature should be the

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 757
 Endodontics or implants? Zitzmann et al.

foundation for the individual risk assessment and In implant studies, the reader must also be aware of the
determination of the long-term prognosis of the respec- differences in outcome data either based on implant
tive tooth requiring root canal treatment (RCT) or level or on the restoration level, which involves both
extraction and replacement with an implant. The implants and superstructures (Pjetursson et al. 2004a).
literature, however, contains inconsistencies in terms In response to the differences in success criteria for
of the definitions of success and survival of endodon- RCT and implants, Iqbal & Kim (2007) restricted their
tically treated teeth and implants (Iqbal & Kim 2007). outcome measure to ‘survival’, which was defined as
Equally, the reported success rates do not necessarily the clinically observed presence of the root canal
equate to the probability of a favoured outcome treated tooth or implant in the mouth. The authors
(prognosis) when applied to a particular case or clinical included 13 studies involving RCT and 55 with
scenario (John et al. 2007). In a systematic review, implants in their meta-analysis, with only one study
Iqbal & Kim (2007) observed that much more stringent (Doyle et al. 2006), involving a comparison of both.
outcome criteria were normally applied to the assess- With proportion estimates for survival of 94% for RCT
ment of ‘successful’ RCT, including the absence of a and 96% for implant-supported single crowns (ISC) at
periapical radiolucency. On the other hand, the use of 5 years, and 97% (RCT) and 94% (ISC) at 6 years and
less stringent criteria in implant studies (generally overlapping confidence intervals at any time-point, the
simple survival) may translate inherently to higher review did not reveal any differences between the two
success rates. This is even more obvious when early treatment modalities. Comparing initial nonsurgical
implant losses that occur during the initial healing RCT and single tooth implants (STI) in a retrospective
period are not accounted for. According to a recent cross-sectional analysis, similar failure rates (6%) were
review, the survival of sound and even compromised reported for both treatments, but significantly more
and treated teeth surpassed that of oral implants, implants required some type of post-treatment inter-
provided that implant loss before loading was added to vention and were classified as ‘surviving’ instead of
that during function over 10 years (Holm-Pedersen ‘successful’ (Doyle et al. 2006). Hence, clinical compli-
et al. 2007). Further misunderstanding is provoked cations were observed in 18% of the restored implant
because, in some studies, survival or retention rates cases and 4% amongst the RCT teeth. In RCT teeth,
include both successful teeth/implants and those clas- these complications were mainly related to endodontic
sified as surviving (instead of reporting successful, retreatment, or persistent apical periodontitis (AP) as
surviving and failed teeth/implants separately, Fig. 1). assessed from radiographs, whilst in implants, several

RCT tooth in function implant in function

clinical & no clinical signs no clinical signs symptomatic situation absence of bleeding bleeding on probing mobility
and absence and persisting and radiolucency, on probing, suppuration, with/without
radiographic or decreasing radiolucency no further bone loss bone loss
measures radiolucency treatment feasible

healthy diseased loss of

diagnosis & healed or healing continued disease state vertical fracture etc.
osseointegration
“success“ “survival“1 “failure“ “success“ “survival“1 “failure“
prognosis
good questionable hopeless/untreatable good questionable hopeless/untreatable

treatment no treatment, optional, no treatment,

requirements supportive in case of progression extraction necessary removal
supportive
maintenance maintenance
1 In some studies survival data are presented as sum of surviving and successful RCT teeth/ implants

Figure 1 Success criteria for root canal treated (RCT) teeth and implants.

758 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

technical problems occurred or surgical interventions Table 1 Search strategy and two-step selection procedure
were required to treat peri-implantitis (Doyle et al. Keyword combination in Number of
2006). Medline retrieved articles
In numerous publications discussion occurred on Root canal therapy AND dental 153
whether tooth preservation by nonsurgical and surgi- implants AND human
cal endodontic means, or extraction and replacement Decision making AND root 20 (plus 13 already
with an implant is more valuable in the long-term, i.e., canal therapy AND outcome listed)
Decision making AND dental 22 (plus 13 already
whether ‘the implant is better than a tooth’ or ‘the
implants AND outcome listed)
implant is a more reliable abutment’ (Lewis 1996, Treatment planning AND dental 10 (plus 17 already
Bader 2002, Cohn 2005, Felton 2005, Ruskin et al. implants AND root canal therapy listed)
2005, Trope 2005, Dawson & Cardaci 2006, Spang- Total 205
berg 2006, Thomas & Beagle 2006, Torabinejad & Manual search 30

Goodacre 2006, White et al. 2006, Mordohai et al. 1st step: screening of 235 titles and Number of
2007, Iqbal & Kim 2008, Kao 2008). These publica- abstracts, reasons for elimination: eliminated
articles
tions focus mainly on single anterior or posterior teeth
with compromised prognosis and their possible replace- Non-english publication 13
No abstract available 17
ment by an implant. In most clinical situations,
Single case report or conference report 6
however, the conditions of the adjacent teeth and the Not related to the current subject 97
entire dentition must be considered when deciding
2nd step full text analysis of 102 articles Number of
upon adequate treatment.
(72 from PubMed, 30 from eliminated
Due to the similar outcomes of implant treatment hand-searching), reasons articles
and RCT, the decision to treat a tooth endodontically or for elimination:
replace it with an implant, must be based on factors Not related to the current subject 51
other than anticipated treatment outcome alone. It was Identical issue discussed by the same 2
the aim of this review to describe the decisive criteria author in another journal (more
and a systematic procedure for deciding upon end- pertinent publication was selected)

odontic treatment or the implant alternative, based on Included articles (49):

best evidence from the literature. Regarding those From PubMed 34
From manual search 15
treatment considerations which lack distinct evidence-
based guidelines, a consensus was accomplished
amongst the authors specialized in endodontics and The combinations of search terms resulted in a list of
restorative dentistry, including dental implantology. 205 publications from PubMed, and an additional 30
papers were retrieved by hand-searching. In the first
step, titles and abstracts were screened. In the second
Search strategy and inclusion
step, full text analysis was performed from 102 possibly
of publications in the review
relevant publications, out of which 49 were included
A MEDLINE search (PubMed) up to July 2008 (data- (Table 1). Publications from the same author discuss-
base 1966–2008 July, week 4) was conducted using ing identical issues were identified and the more
different keyword combinations including the terms relevant publication was selected for this review.
‘root canal therapy’, ‘dental implants’, ‘decision mak-
ing’, ‘treatment planning’, ‘outcome’ and ‘human’
Longevity of root canal treated teeth
(Table 1). In addition, bibliographies of all relevant
and implants
papers and previous review articles were hand-
searched. Any relevant work published in the English When comparing outcome data for root canal treated
language and presenting pertinent information related teeth and dental implants, clinicians must be aware
to single-tooth and full arch reconstructions was that several differences exist, associated with the origin
considered for inclusion in the review. Titles were of the tooth and the implant, the definition and
excluded, if no abstract was available, single case interpretation of success and survival, the study design
reports or conference reports were presented, or the and samples, operators conducting the treatment, and
topic was not related to the subject of the current changes in treatment modalities overtime (Fig. 1).
review. Several preoperative, intraoperative and postoperative

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 759
 Endodontics or implants? Zitzmann et al.

Table 2 Factors influencing endodontic and implant treatment outcome

Initial RCT Endodontic retreatment Apical surgery Implant treatment

Preoperative + Vital pulp tissue + Root canal filling + Orthograde retreatment ) Insufficient bone
) Periapical lesion >2 mm short of the apex feasible volume
+ No periapical lesion + Significant overfill or root ) Specific anatomic
) Large periapical lesion canal filling >2 mm short of findings
) Altered root-canal the apex ) History of periodontitis
morphology or ) Lesion ‡5 mm ) Previous implant
perforation ) Persisting lesion despite failure
) Adequate existing root satisfactory root canal ) Insufficient oral
canal filling filling hygiene and smoking
) Combined endo-perio (see also Table 3)
lesion
) Previous surgical
treatment

Intraoperative + Root canal filling with + Addressing previous + Root-end filling +/) Type of implant and
no voids extending to technical shortcomings ) Poor accessibility surface
2 mm within apex + Adequate root canal +/) Type of bone
(radiographically) filling feasible ) Fenestration, bone
+ Sufficient coronal defects
restoration ) Specific anatomic
) Missed canals and findings
inadequate cleaning ) Bone augmentation
) Errors such as ledging, ) Immediate implant
instrument fracture, placement
root perforations
) Inadequate obturation
) Root canal filling
>2 mm short of the apex
or overfill

Postoperative ) Restoration failure ) Restoration failure +/) No obvious influence ) Wound healing
(coronal leakage) (coronal leakage, no by antibiotics problems
cuspal coverage) ) Iatrogenic factors
(e.g., excess cement)
) Insufficient oral
hygiene and smoking
) Peri-implantitis

+ positively influencing factors; ) negatively influencing factors.

factors influence the prognosis of root canal treatment, of infection. Although clinical symptoms regularly
and have also been identified for the implant treatment diminish within several hours or days of initiating root
outcome (Table 2). canal treatment, complete healing of the periapical
bony lesion may require several months or even years
(Friedman 2002). The absence of clinical symptoms
Success and survival of RCT teeth
and a radiograph with an intact periodontal ligament
A tooth considered for primary RCT or endodontic space in the apical region are indications of healing,
retreatment may has been in function for many years whilst the persistence of AP is a sign of a continued
or even decades. Reasons for treatment may include disease state. If the radiolucency decreases overtime
irreversible pulpitis due to microbial infection originat- (within 4–5 years), the pathosis is also considered to be
ing from a carious lesion, trauma or periodontal ‘healing’. This healing pattern, particularly in teeth
involvement, or AP in teeth with nonvital pulp. The with AP at the time of initial treatment, indicates that
starting point for any longevity assessment is thus a success rates of RCT (in terms of periapical health) start
disease state, involving the pulp tissues and/or the at 0% and increase overtime. Fristad et al. (2004)
periapical bone and the primary goal is the eradication found a 95.5% radiographic success rate with retreated

760 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

teeth recalled 20–27 years postoperatively, whilst the vertical root fractures, presence of true cysts, or foreign
same sample had a 85.7% success 10 years previously. body reactions, e.g., to overfilled root canals (Sjögren
The teeth deemed to be failures radiographically at 10– et al. 1990, Kojima et al. 2004, Stoll et al. 2005).
17 years were still functioning after another 10 years (2) Nonendodontic reasons for RCT failure are related
and healing was observed after the extended observa- to pre-existing factors such as severe periodontal
tion time. This study not only shows the potential for disease, or to post-endodontic factors such as recurrent
late healing, but also the inadequacy of a ‘radiographic caries, improper reconstructions with coronal leakage
only’ assessment (Fristad et al. 2004, Wolcott & Meyers and subsequent reinfection or fracture (Ray & Trope
2006). Applying only clinical measures (no signs and 1995, Aquilino & Caplan 2002, Iqbal et al. 2003). RCT
symptoms), however, led to an overestimation of teeth not restored with crowns were extracted at a rate
favourable outcomes, whilst the radiographic measure 6.0 times greater than teeth crowned after root filling
(with/without periapical radiolucency) was found to be (Aquilino & Caplan 2002).
a better predictor for the outcome of RCT (Farzaneh In a study evaluating the reasons for failure of RCT
et al. 2004b). The use of cone beam computed tomog- teeth, prosthetic reasons (crown fracture, root fracture
raphy with three-dimensional images, has the potential at the level of a post, traumatic fracture) dominated
to add further information about the periapical status of and explained almost 60% of the failures; 32% failed
endodontically treated teeth (Walter et al. 2009). The due to periodontal reasons, whilst pure endodontic
awareness that pulpal and periradicular disease may be failures (vertical root fracture, instrumentation failure,
managed, but not always entirely eliminated led to an root resorption) were rare and accounted for less than
important change in evaluating outcomes (Fig. 1). 10% (Vire 1991). Whilst prosthetic and periodontal
Hence, RCT outcome is better evaluated in terms of failures occurred following 5–5.5 years on average,
‘healed or healing/success’, ‘diseased/survival’ and endodontic failures were recognized within a 2-year
‘failure’ rather than just ‘success’ and ‘failure’ (Fried- period after RCT had been completed (Vire 1991).
man 2002, Farzaneh et al. 2004b). Similarly, Chen et al. (2008) reported from an epidem-
According to a recent meta-analysis, the pooled iologic study that extensively decayed or unrestorable
outcome of primary RCT was 75% when strict teeth were the main reason for tooth extractions (40%).
success criteria (absence of periapical radiolucency) Other causes were tooth fracture (28%), and periodon-
were applied, and reached 85% based on loose tal disease (23%), whilst endodontic reasons were rare
criteria (reduction in size of radiolucency) (Ng et al. at 9% (Chen et al. 2008).
2007). Preoperative absence of a periapical radiolu- Clinical studies investigating the long-term survival
cency, root filling with no voids, root filling extending of fixed dental prostheses (FDP) showed that as soon as
to 2 mm within the radiographic apex and satisfac- 1 or more RCT abutments were involved, the survival
tory coronal restoration were found to improve the rate of all restoration at 20 years was reduced to 57%
outcome of primary RCT significantly (Table 2) (Ng compared with 69% when the FDP comprised abut-
et al. 2007, 2008b). In teeth without a periapical ments with healthy pulps only (De Backer et al. 2006,
radiolucency, initial RCT secured a success rate of 2008). According to a multivariate analysis of abut-
96% after 8–10 years, whilst healing was reduced to ment failures (365 teeth with vital pulps, 122 root filled
86% in cases with pulp necrosis and periapical teeth), additional influencing factors other than RCT
radiolucency (Sjögren et al. 1990). Highest success were distal terminal position in the FDP, and advanced
rates exceeding 90% (with periapical health as marginal bone loss as initially assessed from radio-
outcome measure) have been achieved following graphs. Several variables were stronger multivariately
RCT in teeth with vital pulps (Friedman 2002, than bivariately and this indicated that a combination
Hørsted-Bindslev & Løvschall 2002, Gesi & Bergen- of risk factors is the most detrimental for the longevity
holtz 2003). of the restorations (Palmqvist & Söderfeldt 1994).
Reasons for persistent or emerging disease associ- In epidemiological studies investigating the retention
ated with root filled teeth are either endodontic in of RCT teeth based on data from insurance companies,
nature, or, more frequently, related to nonendodontic so called ‘untoward events’ yielding further insurance
factors: claims such as extraction, retreatment, or apical
(1) Endodontic causes include residual intracanal surgery were evaluated (Lazarski et al. 2001, Salehrabi
infection in nonaccessible regions of the canal system & Rotstein 2004, Chen et al. 2008). Eight years after
or periapical infections due to persisting microbiota, initial nonsurgical root canal treatment, 96% of all

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 761
 Endodontics or implants? Zitzmann et al.

teeth (almost 1.5 million) were retained without any was described as inflammatory reactions associated
untoward event; 0.4% required nonsurgical retreat- with loss of supporting bone around an implant in
ment, in 0.6% apical surgery was performed, and 2.9% function. Hence, peri-implantitis is clinically diagnosed
were extracted. Extractions occurred mainly within by bleeding on probing (and/or suppuration) in com-
3 years from completion of the RCT, and affected bination with radiographic bone loss (Heitz-Mayfield
primarily those teeth without full coronal coverage 2008). During the first year following implant place-
(Salehrabi & Rotstein 2004). Chen et al. (2007, 2008) ment, bone remodelling may cause bone resorption in
also reported a high 5-year tooth retention rate of 93% the marginal area (average 1.3–1.5 mm around
following nonsurgical RCT in more than 1.5 million implants placed at the bone level). Any further bone
teeth. In all, almost 10% were affected by untoward loss, particularly reaching ‡2.5 mm, is considered as
events (6.9% of the teeth were extracted, 2.3% required disease manifestation (Berglundh et al. 2002), and
nonsurgical retreatment, and in 0.5% apical surgery affects at least 28% of subjects (Zitzmann & Berglundh
was performed) (Chen et al. 2007, 2008). 2008a, Zitzmann et al. 2008b). Despite disease pro-
gression, the implant remains nonmobile until the
apical portion of implant osseointegration is affected
Success and survival of dental implants
(Listgarten 1997). In the implant literature, the
A functioning dental implant represents a de novo majority of studies report implant survival rates defined
situation, in which neither caries nor endodontic as simple retention (Berglundh et al. 2002, Zitzmann &
problems exist. In contrast to root canal treatment, Berglundh 2008a, Zitzmann et al. 2008b). If success
implants are placed into relatively healthy surround- criteria are applied, the absence of clinical symptoms,
ings. Complications and failures, however, occur either no signs of inflammation and a limited marginal bone
prior to implant osseointegration (early implant loss) or loss (e.g., not exceeding 0.2 mm after the first year in
after initially successful osseointegration (late implant function) (Smith & Zarb 1989) are frequently men-
loss) and disease manifestation may necessitate several tioned. In several studies, however, disease symptoms
years or even decades of function (Quirynen et al. are not consistently investigated, i.e., probing is not
2007). Osseointegration is considered to be a phenom- applied, and bone level assessments are made from
enon of direct apposition of bone substance on the panoramic radiographs with limited accuracy (Zitz-
implant surface followed by structural adaptation in mann & Berglundh 2008a, Zitzmann et al. 2008b).
response to mechanical load (Schenk & Buser 1998). Implant success can be further compromised by
Whilst initial implant fixation following placement is several biological and technical complications, which
simply derived from mechanical stabilization, osseoin- occur in association with dental implants and implant
tegration with an intimate contact between living bone restorations. Whilst biological complications comprise
and the titanium surface requires several weeks any type of peri-implant diseases, a large variety of
(Berglundh et al. 2003, Abrahamsson et al. 2004). technical complications exist related to mechanical
Early implant failures occur mainly during the first damage of the implant, implant components and/or the
weeks or months after implant placement and are superstructure (e.g., implant fracture, abutment or
frequently related to surgical trauma, complicated occlusal screw loosening or fracture, fracture of
wound healing, insufficient primary stability and/or veneering or framework, loss of retention in cemented
initial overload (Listgarten 1997, Esposito et al. 1998). restorations). According to a review analyzing long-
Late implant losses are caused by microbial infection, term results of fixed implant restorations, 39% of all
overload or toxic reactions from implant surface patients were affected by complications or failures
contamination (e.g., acid remnants). Whilst overload during a 5-year observation period (Pjetursson et al.
leads to a sudden loss of osseointegration with implant 2004a). The 10-year survival rates reached 93%
mobility, microbial infection initiates peri-implant (implant-FDP) and 94% (single tooth implants, STI)
mucositis that corresponds to gingivitis and may on an implant level, whilst survival of the implant
progress into peri-implantitis that corresponds to peri- restorations varied between 87% (implant-FDP) and
odontitis. According to the consensus report from the 90% for the ISC (Pjetursson et al. 2004a). It should be
1st European Workshop on Periodontology (EWOP, noted that implant reconstructions exposed to biolog-
Albrektsson & Isidor 1994), peri-implant mucositis was ical or technical complications were at greater risk of
defined as a reversible inflammatory reaction in the soft recurrent problems or failures (Brägger et al. 2005).
tissues surrounding an implant, and peri-implantitis Similarly, patients who had experienced an implant

762 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

failure, had a 30% increased risk of further failures et al. 2001, Salehrabi & Rotstein 2004, Iqbal & Kim
(Weyant & Burt 1993). 2007, 2008). Hence, comparing data from RCT teeth
Risk factors for developing peri-implant diseases are and implant studies in meta-analyses mixes efficacy
patient-related (e.g., susceptibility to periodontitis, dia- and effectivenss.
betes), environmental (e.g., cigarette smoking, alcohol
consumption), technological (e.g., exposed rough
Root canal treatment changes
implant surface), or local. These local factors comprise
poor personal plaque control, or iatrogenic factors such During the last decade, RCT has benefited from
as insufficient access for oral hygiene due to implant improvements in techniques and equipment such as
position and/or restoration contour, or excess cement nickel–titanium rotary instruments, electronic apex
(Table 2). Limited evidence is available for an associ- locators and microscopic magnification for nonsurgical
ation between peri-implant disease and rough implant and surgical therapies (Manning 2000, John et al.
surfaces or genetic traits (Quirynen et al. 2007, Heitz- 2007). Improvements in long-term success of surgical
Mayfield 2008). Late implant failures due to occlusal or nonsurgical RCT applying new technical develop-
overload occur when the load bearing threshold set by ments have, however, not yet been proven on the basis
the biological phenomenon of osseointegration has of outcome of treatment provided by general dental
been exceeded. Very little is known about this individ- practitioners (effectiveness) (Ng et al. 2007).
ual threshold and possible influencing factors such as
bone quality, implant surface modifications and the
Changes in implant therapy
type and direction of forces. Whilst clenching exerts
mainly vertical forces, bruxism creates excessive lateral When implant treatment was introduced in the 1970s,
forces, which are suggested to be less well tolerated several restrictions were defined in order to minimize
(Meffert 1997). the risk of implant failure or complications. Hence,
implant therapy was not recommended in patients with
xerostomia, osteoporosis, aggressive forms of periodon-
History and recent changes in endodontic
titis and heavy smokers (Brånemark 1985). Today, it is
and implant therapies
evident that the peri-implant tissues are not affected by
For both RCT and dental implants, efficacy i.e., the hyposalivation and/or the symptoms of xerostomia.
successful maintenance under optimal conditions has Further, a reduced bone mineral density in osteoporotic
been proven mainly in controlled longitudinal studies patients entails a reduced bone-to-implant contact, but
in university settings. Both treatment options have also does not appear to inhibit osseointegration (Table 3).
been evaluated for effectiveness, i.e., the effect has been Implant indications have been extended to patients
verified under ordinary conditions as shown in retro- with a history of periodontitis and also to smokers
spective studies and community-based trials. RCT has accepting an increased risk for complications and
always been a part of general dental practice and failures (See ‘General endodontic and implant contra-
specialist recognition was granted in some parts of the indications’). In an initial attempt at cautious restraint,
world in the 1960s or later. However, in the 1970s and any type of potential risk for implant failures was
1980s, dental implants, were mainly placed at a excluded, whilst current implant treatment modalities
specialist level, whilst today it is a common treatment consciously include further risk factors such as imme-
modality amongst general dentists. Although some diate implant loading, even combined with immediate
evidence suggests that general practitioners achieve implant placement (Aparicio et al. 2003, Schropp &
implant survival rates similar to those of specialists Isidor 2008).
(Andersson et al. 1998), it is assumed that results differ
particularly for demanding procedures requiring addi-
General endodontic and implant
tional bone augmentation and aesthetic management
contraindications
in the anterior region.
In the literature, data about implant survival and In patients with high caries activity, possibly related to
success are still dominated by studies from university dry mouth as a common side effect of several medica-
clinics and/or specialists documenting its efficacy, tions (e.g., antihypertensives, diuretics, antidepres-
whilst several studies investigating RCT include work sants, atropine, anticonvulsants, spasmolysants and
from undergraduates and general practice (Lazarski appetite suppressants) or associated with syndromes

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 763
 Endodontics or implants? Zitzmann et al.

Table 3 Contraindications and increased risk for implant failures

Disease Assessment

Medical contraindications Acute infectious diseases Absolute, but temporarily; wait for recovery
Cancer chemotherapy Absolute, but temporarily; reduced immune status
Systemic bisphosphonate Risk of bisphosphonate-induced osteonecrosis (BON)
medication (‡2 year)
Renal osteodystrophia Increased risk for infection, reduced bone density
Severe psychosis Absolute; risk of regarding the implant as foreign body
and requesting removal despite of successful
osseointegration
Depression Relative
Pregnancy Absolute, but temporarily; to avoid additional stress
and radiation exposure
Unfinished cranial growth with Relative, but temporarily; to avoid any harm to the
incomplete tooth eruption growth plates, to avoid inadequate implant position
in relation to the residual dentition; utilize hand-wrist
radiograph to evaluate end of skeletal growth; single
tooth implants in the anterior region not before 25th
year of age
Intraoral contraindications Pathologic findings at the oral soft- Temporarily; increased risk for infection, wait until
and/or hard tissues healing is completed
Increased risk for implant History of (aggressive) periodontitis Relative, requires supportive periodontal care;
failure or complications increased risk to develop peri-implantitis
Heavy smoking ‡10 pack-years Relative or absolute, indicates cessation protocol;
(particularly in combination with wound healing problems, locally reduced
HRT/oestrogen), alcohol and drug abuse vascularization, impaired immunity, reduced
bone turn over
Insufficient oral hygiene Absolute; wound healing problems, infection
Uncontrolled parafunctions Relative; increased risk for technical complications
Post head and neck radiation therapy Absolute, but temporarily; reduced bone remodelling,
risk of osteoradionecrosis, implant placement
6–8 weeks before or ‡1 year after radiotherapy
Osteoporosis Relative; reduced bone-to-implant contact; consider
calcium substitution, prolong healing period and
avoid high torque levels for abutment screw fixation
Uncontrolled diabetes Relative, requires medical treatment; wound healing
problems (impaired immunity, microvascular
diseases)
Status post chemotherapy, Absolute, but temporarily; wound healing problems,
immuno-suppressants medical advice required (consider corticosteroid
or steroid long-term medication, cover)
uncontrolled HIV infection

(e.g., Sjögren), less effort will be made to maintain a (Marending et al. 2005). Other authors suggested a
questionable tooth, and implant treatment may be possible negative influence of smoking on the prognosis
favoured. Further, patients with diabetes seem to have of RCT teeth, but this was mainly attributed to delayed
a somewhat increased likelihood of endodontic compli- bone healing, and to an increased prevalence of
cations (symptomatic periapical diseases and flare-ups) periodontal disease and root caries in smokers (Duncan
following nonsurgical RCT, particularly in cases with & Pitt Ford 2006).
preoperative periradicular lesions (Fouad & Burleson There are few absolute and permanent implant
2003). Impaired integrity of the patient’s nonspecific contraindications, but several temporary restrictions
immune system was found to be a significant predictor such as incomplete cranial growth (Table 3) (Zitzmann
for a negative outcome of initial nonsurgical RCT or & Berglundh 2008a, Zitzmann et al. 2008b). In young
retreatment, whilst other patient-related factors such as adults requiring single tooth replacement in the max-
age and smoking had no impact on the healing rate illary anterior region, implant placement should be

764 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

postponed until after the age of 25 due to the changes Further treatment modalities in case
in anterior face height and posterior rotation of the of primary endodontic failure
mandible, particularly in women (Jemt et al. 2007).
In cases of endodontic failure following primary RCT,
This continuous alveolar bone development entails a
nonsurgical retreatment is generally indicated provided
vertical infraposition of the implant with the mucosal
that the root canals are accessible (Fig. 2). White et al.
margin too far apical and significant aesthetic implica-
(2006) stated that endodontic surgery has been largely
tions may occur. Patients under intravenous bis-
replaced by endodontic retreatment in specialist end-
phosphonate medication for more than 2 years and a
odontic practice over the past decade. According to a
history of complicated wound healing, e.g., following
recent systematic review, the pooled success rate for
tooth extraction, are not a candidate for implant
secondary RCT (judged by complete or incomplete
treatment due to the risk of bisphosphonate-induced
healing) was 77% each (Ng et al. 2008a). Defining
osteonecrosis (BON) (Edwards et al. 2008). Considering
success as the absence of AP and any associated signs
early and late implant losses as well as biological and
and symptoms, the 4- to 6-year overall success of
technical complications, several factors were identified
orthograde retreatment was reported to be 81% (Far-
to be associated with an increased risk for implant
zaneh et al. 2004a). In cases with preoperative AP, the
failure or complications (Table 3). According to a
success rate of retreatment was lower (78%) than if
recent review, smoking is a significant risk factor for
such radiolucency was absent at the time of retreat-
implant treatment and augmentation procedures
ment (97%, Table 2). These differences in outcome
accompanying implant therapies (Strietzel et al.
reflect the divergent indications either to improve a
2007). In these situations with an enhanced risk for
RCT in a tooth with no AP (e.g., before inclusion as
implant failure, preference to tooth preservation, and
abutment in an FDP), or to retreat a symptomatic tooth
avoidance of extraction and of further implant surgery
with AP (Bergenholtz et al. 1979, Farzaneh et al.
should be considered also in teeth with a questionable
2004a). Further, the success of endodontic retreatment
prognosis.

star ting point irreversible pulpitis or periapical periodontitis,

RCT required (non-surgical)

assessment of - perio: periodontal health, sufficient residual attachment ?

tooth prognosis - endo: RCT feasible, root canals accessible ?
- reconstructive: sufficient residual tooth substance ? (crown lengthening or or thodontic extrusion
possibly required), adequate restoration feasible to avoid bacterial leakage

outcome of + -
non-surgical RCT

diagnosis intracanal infection isolated periapical root canal not accessible tooth untreatable
infection

treatment non-surgical retreatment surgical treatment periradicular surger y tooth extraction

option (periapical resection (hemi-, tooth-sectioning)
and retrograde obturation)
outcome + - + - + -

fur ther treatment

no replacement,
implant treatment,
prosthetic restoration

Figure 2 Treatment considerations for root canal treated (RCT) teeth.

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 765
 Endodontics or implants? Zitzmann et al.

depends on whether alterations in the natural course of According to the agreement of specialists, surgical
the root canals were caused by previous root canal intervention is probably not worth whilst in cases of
treatment (i.e., technical errors such as transportation, limited prognosis of periapical surgery, e.g., when
stripping or perforation, Table 2). Hence, teeth in nonsurgical retreatment is not feasible due to inhib-
which no significant anatomical changes were made ited access by an adhesively cemented zirconium or
by the previous RCT (‘root-canal-morphology- metal post reaching close to the apex. The buccal
respected’) had 87% success, whilst only 47% of the fenestration created to gain access to the periapical
teeth with damaged anatomy from previous treatment area may not heal with an intact bony plate
(‘root-canal-morphology-altered’) were successful after resulting in a compromised site and precipitate the
2 years (Gorni & Gagliani 2004). need for additional bone grafting if further implant
Surgical treatment is a valuable alternative if treatment is planned (Greenstein et al. 2008). In
nonsurgical retreatment is not successful, not indi- molar teeth with sufficient root separation, periradic-
cated (e.g., primary treatment was performed under ular surgery in terms of root amputation or root
best possible conditions), or not feasible (e.g., in teeth resection can be considered, particularly when a root
with adhesively cemented overextended zirconium or canal is not accessible and there is concomitant
metal posts, teeth with alterations of the natural periodontal involvement. Divergent failure rates have
course of the root canal such as ledge formation from been documented for root-resected molars and vary
previous treatment, abutment teeth in existing FDP between 7% and 38% after 10 years follow-up
with radiographic root canal obturation; Fig. 2). (Langer et al. 1981, Carnevale et al. 1998). Extrac-
Apical resection eliminates the periapical lesion (e.g., tion of periodontally involved molars with advanced
in lesions refractory to conventional treatment) or attachment loss, however, frequently entails complex
other irritants from the periapical tissues, allows bone augmentation procedures to prepare an ade-
healing and is best combined with a root-end filling quate implant site.
(John et al. 2007). The large variety of healing rates
(37–91%) reported in a review (Friedman 2005) may
Further treatment modalities in case
entail some restraint in considering resected teeth as
of implant failure
abutments for an FDP (See ‘Restorative aspects’).
However, 80–94% of resected teeth remained in Early implant failures and late losses due to overload
symptom-free function, even if they were not healed are, in most instances, first recognized by implant
(Friedman 2005). Detailed analysis of the data mobility and there is no treatment available to save a
revealed that the prognosis for apical resection is less mobile implant (Albrektsson & Isidor 1994). In sites
favourable, when no nonsurgical retreatment was affected by peri-implantitis, applied therapies aim in a
performed in advance and an infection possibly resolution of the infection, but these measures are not
persisted in the root canal system. Additional factors predictably successful in achieving reosseointegration
for a reduced prognosis of periapical surgery are: poor in the previously contaminated region (Claffey et al.
accessibility in the molar region, persisting lesion 2008, Renvert et al. 2008). Depending on disease
despite apparently satisfactory filling, size of the lesion progression, implant loss occurs sooner or later and is
‡5 mm, coronal leakage and surgical retreatment frequently accompanied by substantial alveolar ridge
(Table 2) (Kvist & Reit 1999, Wang et al. 2004, defects (Lindhe & Meyle 2008). Further reimplantation
Friedman 2005). Whilst first surgical interventions may then entail additional bone augmentation in a
resulted in 74% success, surgical retreatment had an staged approach.
outcome of 62% (Kvist & Reit 1999, Wang et al. It has been mentioned that particularly in younger
2004, Friedman 2005). Repeated periapical surgery is patients, where a significantly long-term prognosis is
only useful when primary surgery was performed required, a more aggressive approach in replacing
under poor conditions such as inadequate equipment. questionable teeth with implants would be justified
Another factor for impaired prognosis of periapical (Mordohai et al. 2005, 2007). The continuous alveolar
surgery is periodontal involvement of the respective bone growth, aesthetic concerns particularly in single
tooth: whilst isolated endodontic lesions revealed a tooth restorations, gingival recession over years, a
95% successful outcome following endondontic micro- possible susceptibility to periodontal and peri-implant
surgery, combined endodontic-periodontal lesions had diseases are important aspects that rather imply a more
a reduced healing rate of 77.5% (Kim et al. 2008). restrained approach, facilitating tooth maintenance for

766 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

several years or even decades before tooth extraction case is when tissue loss is minimal and the coronal
may become inevitable. restoration is a simple composite filling. It is much more
These aspects related to the longevity of root canal complicated if a single crown is involved, possibly
treated teeth and implants, indicate that: (i) most requiring crown-lengthening through surgery or
endodontic failures are related to nonendodontic fac- orthodontic extrusion to facilitate sufficient cervical
tors and RCT teeth survive better if properly restored ferrule with the definitive crown engaging at least
(with single crowns being more favourable than FDPs), 1.5 mm tooth structure (Libman & Nicholls 1995, Tan
(ii) failures due to endodontic reasons can frequently be et al. 2005, Türp et al. 2007). Surgical crown-length-
resolved by any type of retreatment, (iii) most implant ening for a tooth already compromised by a large post
failures are directly related to the implant itself and channel, and a poor crown to root ratio, does, however,
entail implant removal. place the respective tooth at high risk and extraction
may be more appropriate (Bader 2002). Any additional
pre-treatment requirement adds complexity, may pres-
Restorative aspects
ent further complications and risks, increases treatment
According to the view of the specialists, good long-term costs, and probably reduces the patient’s willingness to
prognosis and greater flexibility in clinical management accept RCT rather than implant placement (Torabine-
indicate that RCT and even retreatment should be jad & Goodacre 2006).
performed first in most instances unless the tooth is If the degree of difficulty of the planned therapy is
judged to be untreatable when implants are considered assessed, it seems that any type of endodontic or
(Fig. 2). As soon as other compromising factors or risks periodontal treatment is less time-consuming, less
exist, such as insufficient coronal tooth structure and/ expensive, and easier to perform in anterior teeth than
or moderate to severe periodontal involvement, the in multi-rooted premolars and molars, due to the
time and cost efforts engaged with the RCT may be simpler root morphology and root canal anatomy, and
questionable. When deciding if an impaired tooth with better accessibility and visibility particularly for peri-
a questionable prognosis is maintained or extracted and apical surgery (Fig. 3). After RCT in the anterior
possibly replaced by an implant, several different region, however, greyness of the clinical crown possibly
aspects have to be taken into account. These aspects impairs the aesthetic outcome and indicates bleaching
comprise site-specific factors, the entire oral situation and/or crown restorations. In implant treatment, the
and patient-related factors (Messer 1999). clinical crown can be designed so as to ideally mimic
the symmetric situation on the contralateral site
provided that the implant position is appropriate. The
Site-specific aspects
aesthetic outcome is, however, often compromised due
In order to evaluate the prognosis of a specific tooth, all to soft tissue recession from unpredictable healing
required treatment measures should be listed initially following tooth extraction and implant surgery. Inci-
and their degree of difficulty assessed. These treatment sors have a marked undulation of the cemento-enamel-
needs comprise not only nonsurgical and/or surgical junction as well as of the gingival margin with long
endodontics, but post and core build-ups, periodontal interproximal papillae, which are specific for anterior
treatment, restorations or crowning. Crown-lengthen- teeth and are bound to a sound periodontium. In the
ing or orthodontic extrusion are possibly needed in gingiva surrounding teeth, the collagen fibres are
addition (Palmer & Howe 1999, Greenstein et al. 2007, attached to the root cementum and are arranged in
Mordohai et al. 2007). Particularly in periodontics, an groups or bundles with distinct orientations such as
initial phase of pre-treatment followed by a re-evalua- dentogingival, dentoperiosteal, circular and transseptal
tion is required to facilitate a complete estimation of the fibres. Around implants, however, there is no peri-
site-specific response and the patient’s compliance. odontal ligament and the implant lacks a lining
After successful periodontal treatment, however, teeth cementum with inserting collagen fibres (Berglundh
with reduced periodontal support are also capable of et al. 1991). Particularly in patients with high aesthetic
serving as foundations for single crowns or as abut- demands and a thin mucosal biotype, greater efforts
ments for FDPs (Nyman & Lindhe 1979). One of the should be made to save a questionable anterior tooth in
most decisive site-specific factors is the remaining order to ensure preservation of the soft tissue architec-
coronal tooth substance, which determines the dimen- ture (Kan et al. 2003, Greenstein et al. 2008). Posterior
sion and extent of the coronal restoration. The easiest teeth with questionable prognosis, however, are

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 767
 Endodontics or implants? Zitzmann et al.

pre-/treatment anterior posterior

periodontal single-rooted multi-rooted,

(root morphology, furcation involvement,
accessibility) length of root trunk

endodontics simpler anatomy difficult curves,

(root canal anatomy) accessory canals

restorative/ reconstructive
aesthetics (gingival undulation, marked undulation, small undulation,
papilla preservation, aesthetics significant aesthetics marginal
contralateral symmetry)

tooth preservation implant placement1

1sufficient bone volume provided

Figure 3 Local factors influencing the predictability of treatment outcomes.

replaced by an implant with less restraint, than in the conventional FDP may be favourable over tooth main-
aesthetic zone where concerns about the risks of tenance at high costs and increased risk for failure.
gingival recession and a possible lack of interproximal The same is true, if implant placement is needed in
mucosal tissues are of greater importance. the adjacent tooth positions (i.e., anterior and posterior
of the questionable tooth). Hence, a three-unit implant-
FDP supported by two implants and tooth removal is a
Oral situation
more reasonable treatment plan as compared to three
As soon as any restorative treatment requirements of single crowns with the questionable tooth maintained
an RCT tooth have been defined, the situation of the between the two implants (Fig. 4).
adjacent teeth and the entire remaining dentition is If the RCT tooth is planned to serve as an abutment
included in the treatment planning (Fig. 4) (Palmer & and is located in a strategic position within a long-span
Howe 1999, Bader 2002). For a questionable tooth in tooth-supported FDP, its prognosis has to be good in
an intact arch, which can be kept as a free-standing order to ensure a noncompromised long-term success of
unit, a greater latitude for therapy can be implemented the entire reconstruction (Davarpanah et al. 2000,
for retention, whilst a complex prosthetic plan possibly Bader 2002). Having in mind that the potential risk for
indicates extraction of a compromised abutment tooth. failure from endodontic, periodontal or prothodontic
In accordance with the view of specialists, the following reasons after a 10-year observation period is 10% each,
clinical scenarios are common in clinical practice and these multiple risk factors may theoretically accumu-
have to include assessments of potential risks and late and entail a reduced long-term success rate of 73%
evaluation of the prognosis of the RCT tooth as well as by multiplying 0.93.
of the entire restoration: On the other hand, with a questionable RCT abut-
If maintenance or extraction of a questionable tooth is ment located in a strategic position of an existing and
considered and the adjacent teeth obviously require full otherwise sufficient reconstruction, all efforts are made
crown restorations, extraction and replacement by a to save the tooth and the restoration.

768 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

extraction
t ti off a single
i l RCT ttooth
th

no treatment implant treatment not feasible and

implant-supported single crown (ISC) involvement of adjacent teeth required
suitable as abutment
fixed dental prosthesis (FDP)
prognosis of adjacent teeth?
not suitable as abutment in
a reconstruction
maintain as single unit
extraction

extraction of an RCT tooth in a full-arch rehabilitation

sufficient periodontal support available additional implant support required

fixed dental prosthesis (FDP) implant-FDP
potential implant positions?
resulting in which type of restorations (extent of tooth- or implant-supported FDP)?
strategic/ elective extraction of maintainable teeth with questionable prognosis ?
prognosis of a single tooth versus treatment-related prognosis of the entire restoration

tooth with poor prognosis, extraction indicated

tooth with questionable prognosis, maintainable
tooth with good prognosis, pretreatment indicated

Figure 4 Reconstructive aspects in treatment planning.

In full arch reconstructions and few potential adjacent teeth are sound and implant placement would
abutments maintained in a spread position, long-span require additional bone and/or soft tissue augmenta-
tooth-supported FDPs, particularly those with high tions, it may not be included as abutment in a long-
numbers of pontics and few abutments, can be span FDP. The situation of the remaining dentition and
avoided by adding implants supporting either single the full-mouth treatment planning decides, at least in
crowns or short-span implant-FDPs. After introducing part, whether or not to maintain a questionable tooth.
implant-supported restorations as a treatment option Hence, a tooth with a relatively good prognosis, but
in a specialist practice, the number of long-span FDP requiring tremendous pre-treatment efforts may be
was reduced and the overall failure rate of tooth- intended for extraction, as treatment requirements in
supported FDPs decreased from 4% to 2% at 5– the adjacent tooth positions (either tooth- or implant-
10 years observation (Walton 2009). Using less supported) overrule the decision made for the single
compromised teeth as abutments, not necessarily tooth (Fig. 4).
extracting and replacing them, but placing implants
in addition, facilitates an improved outcome of tooth-
Patient-related factors
supported FDP.
Replacing missing posterior teeth in free-end situa- The patient’s expectations, medical contraindications
tions is a clear indication for implant placement in (See ‘General endodontic and implant contraindica-
order to reduce the risk of FDP with distal cantilevers tions’ and Table 3) and his/her financial position are
and to avoid combined tooth-implant-supported FDPs, further aspects taken into account during treatment
as both treatment modalities are associated with an planning (Palmer & Howe 1999, Dawson & Cardaci
increased risk of failure (Lang et al. 2004, Pjetursson 2006, Zitzmann & Berglundh 2008a, Zitzmann et al.
et al. 2004b). 2008b). In general, RCT including a restoration with a
Whilst a questionable tooth is probably planned for a single crown is less expensive, and entails fewer dental
single crown without restraints, particularly when the visits in a shorter time period than an ISC (Moiseiwitsch

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 769
 Endodontics or implants? Zitzmann et al.

& Caplan 2001). According to a cross-sectional study, 2005, Pothukuchi 2006, Wolcott & Meyers 2006).
even ISC performed as one- or two-stage procedure or Patient’s values and expectations may lead to a more
as immediate placement had a longer time-to-function value-based dentistry, where the patients’ perceived
than RCT teeth (median 250 days for ISC vs. 67 days benefit form the treatment outweighs the clinical
for RCT) (Doyle et al. 2006). Although recent protocols decision-making procedure (Eckert 2005). It must be
encourage immediate loading, most implant situations noted, however, that as soon as fixed or removable
require several months for completion of treatment to dental prostheses are part of the treatment require-
ensure undisturbed osseointegration and maturation of ments, the practitioner has the full responsibility for the
the soft tissues. In addition to lower initial costs with reconstruction comprising also the laboratory work
RCT compared with ISC, a greater cost-benefit ratio is and he/she will possibly hesitate to include a question-
assumed, since RCT retains a natural tooth provided able tooth in order to reduce the risk of the entire
that no residual pathology of clinical significance restoration due to economic considerations.
persists, the tooth is in function and causes neither
discomfort for the patient nor any aesthetic impairment
Conclusions
(Torabinejad et al. 2007).
According to the agreement of specialists, all site- A simple comparison of long-term survival or success
specific, oral and patient-related factors should be rates of root filled teeth and implants does not fulfil the
evaluated systematically, the strategic value of the demand for a comprehensive decision-making process,
tooth is determined and a risk analysis is performed which includes multiple factors to evaluate, individual
before any definitive decision is taken. The existing case evaluation and a thorough treatment planning.
evidence on the best treatment option in this particular Several retrieved publications implied that the decision
case should be taken into account, and a treatment for extraction of a natural tooth depends less on the
recommendation is then given in favour or against health of that individual tooth, but rather on the
tooth maintenance. After case presentation and thor- overall rehabilitation planned and that sacrificing a
ough objective information about the risk assessment, tooth can be preferable for a ‘better, more predictable,
prognosis, possible complications and treatment alter- more economic long-term rehabilitation on implants’.
natives, the final choice rests with the patient, who Applying this opinion without critical appraisal of site-
either accepts or refuses the treatment proposal. Patient specific and patient-related factors may fail to recognize
attitudes, e.g., opposed to or approving of implants and/ risks for complications and failures possibly associated
or bone augmentation procedures, seeking final solu- with implant treatment.
tions or not, may set a questionable tooth on a higher For single tooth restorations, an increased risk in
or lower strategic value. For instance, when molar restoring a tooth with a questionable prognosis is
extraction would result in sinus lifting procedures, acceptable in a particular case. The respective tooth,
which the patient wishes to avoid, the high strategic however, should not be included as an abutment in a
value justifies multiple extensive procedures for tooth long-span FDP. Multiple risk factors may indicate tooth
retention. On the contrary, with multiple issues asso- extraction and possible replacement by an implant,
ciated with endodontic therapy, minimal coronal tooth particularly in the posterior region and when aesthetics
structure with decayed root dentine, and/or high caries is not paramount. Although priority should be given to
index, which prohibits crown lengthening, replacement preservation of the natural dentition, implant place-
of the tooth with lower value may be prudent to avoid ment enhances treatment planning options, thereby
potential complications (Palmer & Howe 1999, Mordo- facilitating short-span reconstructions or single units
hai et al. 2005, 2007). Treatment alternatives should with reduced risk of failure for the patient and the
be estimated as true supplementing therapeutic options practitioner. Hence, using implants for replacement of
rather than as competing treatments, particularly single missing teeth may facilitate retention of a
when these are not part of the practitioners own neighbouring compromised tooth, which otherwise
repertoire. Especially for more demanding endodontic would have been extracted.
or implant therapies, referral to an experienced spe- In case of full-mouth rehabilitation, single tooth
cialist (endodontist and implantologist), who has the prognosis and the site-specific treatment recommenda-
clinical skills, the necessary equipment and resources is tion is possibly overruled by the overall treatment
in the best interest of the patient and should be planning and a therapy-related decision for a strategic
encouraged when appropriate (Messer 1999, Cohn extraction may be required to perform reasonable

770 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

reconstructions with uncompromised long-term prog- Brägger U, Karoussis I, Persson R, Pjetursson B, Salvi G, Lang
nosis. Particularly in patients with a history of previous N (2005) Technical and biological complications/failures
implant loss, and in young patients, in whom the final with single crowns and fixed partial dentures on implants: a
tooth position is not settled, and susceptibility to 10-year prospective cohort study. Clinical Oral Implants
Research 16, 326–34.
periodontal and/or peri-implant diseases are not yet
Brånemark P-I (1985) Introduction to Osseointegration. In:
predictable, the threshold for tooth extraction should be
Brånemark P-I, Zarb G, Albrektsson T, eds. Tissue-Integrated
at its greatest. Irrespective of the type of the selected Prostheses: Osseointegration in Clinical Dentistry, 1st edn.
treatment option involving teeth and/or implants, Chicago: Quintessence, pp. 11–76.
ongoing maintenance is required to assure sufficient Carnevale G, Pontoriero R, di Febo G (1998) Long-term effects
periodontal and peri-implant care, and to detect and of root-resective therapy in furcation-involved molars. A 10-
treat any type of biological or technical complication at year longitudinal study. Journal of Clinical Periodontology 25,
an early stage in order to reduce the risk of compro- 209–14.
mising the longevity of the reconstruction. Chen SC, Chueh LH, Hsiao CK, Tsai MY, Ho SC, Chiang CP
(2007) An epidemiologic study of tooth retention after
nonsurgical endodontic treatment in a large population in
References Taiwan. Journal of Endodontics 33, 226–9.
Chen SC, Chueh LH, Hsiao CK, Wu HP, Chiang CP (2008)
Abrahamsson I, Berglundh T, Linder E, Lang NP, Lindhe J
First untoward events and reasons for tooth extraction after
(2004) Early bone formation adjacent to rough and turned
nonsurgical endodontic treatment in Taiwan. Journal of
endosseous implant surfaces. An experimental study in the
Endodontics 34, 671–4.
dog. Clinical Oral Implants Research 15, 381–92.
Claffey N, Clarke E, Polyzois I, Renvert S (2008) Surgical
Albrektsson T, Isidor F (1994) Consensus report of session IV.
treatment of peri-implantitis. Journal of Clinical Periodontol-
In: Lang NP, Karring T, eds. Proceedings of The First European
ogy 35(Suppl. 8), 316–32.
Workshop on Periodontolog. London: Quintessence, pp. 365–
Cohn SA (2005) Treatment choices for negative outcomes
9.
with non-surgical root canal treatment: non-surical retreat-
Andersson B, Ödman P, Lindvall AM, Brånemark PI (1998)
ment vs surgical retreatment vs implants. Endodontic Topics
Five-year prospective study of prosthodontic and surgical
11, 4–24.
single-tooth implant treatment in general practices and at a
Davarpanah M, Martinez H, Tecucianu JF, Fromentin O,
specialist clinic. International Journal of Prosthodontics 11,
Celletti R (2000) To conserve or implant: which choice of
351–5.
therapy? International Journal of Periodontics & Restorative
Aparicio C, Rangert B, Sennerby L (2003) Immediate/early
Dentistry 20, 412–22.
loading of dental implants: a report from the Sociedad
Dawson AS, Cardaci SC (2006) Endodontics versus implan-
Espanola de Implantes World Congress consensus meeting
tology: to extirpate or integrate? Australian Endodontic
in Barcelona, Spain, 2002. Clinical Implant Dentistry and
Journal 32, 57–63.
Related Research 5, 57–60.
De Backer H, Van Maele G, De Moor N, Van den Berghe L, De
Aquilino SA, Caplan DJ (2002) Relationship between crown
Boever J (2006) A 20-year retrospective survival study of
placement and the survival of endodontically treated teeth.
fixed partial dentures. International Journal of Prosthodontics
Journal of Prosthetic Dentistry 87, 256–63.
19, 143–53.
Bader HI (2002) Treatment planning for implants versus root
De Backer H, Van Maele G, De Moor N, Van den Berghe L
canal therapy: a contemporary dilemma. Implant Dentistry
(2008) Long-term results of short-span versus long-span
11, 217–23.
fixed dental prostheses: an up to 20-year retrospective
Bergenholtz G, Lekholm U, Milthon R, Heden G, Odesjo B,
study. International Journal of Prosthodontics 21, 75–85.
Engstrom B (1979) Retreatment of endodontic fillings.
Doyle SL, Hodges JS, Pesun IJ, Law AS, Bowles WR (2006)
Scandinavian Journal of Dental Research 87, 217–24.
Retrospective cross sectional comparison of initial nonsur-
Berglundh T, Lindhe J, Ericsson I, Marinello CP, Liljenberg B,
gical endodontic treatment and single-tooth implants.
Thomsen P (1991) The soft tissue barrier at implants and
Journal of Endodontics 32, 822–7.
teeth. Clinical Oral Implants Research 2, 81–90.
Duncan HF, Pitt Ford TR (2006) The potential association
Berglundh T, Persson L, Klinge B (2002) A systematic review
between smoking and endodontic disease. International
of the incidence of biological and technical complications in
Endodontic Journal 39, 843–54.
implant dentistry reported in prospective longitudinal stud-
Eckert SE (2005) Value in dentistry. International Journal of
ies of at least 5 years. Journal of Clinical Periodontology
Oral Maxillofacial Implants 20, 341–2.
29(Suppl. 3), 197–212. discussion 32-3
Edwards BJ, Hellstein JW, Jacobsen PL, Kaltman S, Mariotti A,
Berglundh T, Abrahamsson I, Lang NP, Lindhe J (2003) De
Migliorati CA (2008) Updated recommendations for mana-
novo alveolar bone formation adjacent to endosseous
ging the care of patients receiving oral biphosphonate
implants. Clinical Oral Implants Research 14, 251–62.

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 771
 Endodontics or implants? Zitzmann et al.

therapy: an advisory statement from the American Dental preserving natural teeth with nonsurgical endodontic ther-
Association Council on Scientific Affairs. Journal of American apy. Journal of Endodontics 34, 519–29.
Dental Association 139, 1674–7. Iqbal MK, Johansson AA, Akeel RF, Bergenholtz A, Omar R
Esposito M, Hirsch JM, Lekholm U, Thomsen P (1998) (2003) A retrospective analysis of factors associated with
Biological factors contributing to failures of osseointegrated the periapical status of restored, endodontically treated
oral implants. (II). Etiopathogenesis. European Journal of Oral teeth. International Journal of Prosthodontics 16, 31–8.
Sciences 106, 721–64. Jemt T, Ahlberg G, Henriksson K, Bondevik O (2007) Tooth
Farzaneh M, Abitbol S, Friedman S (2004a) Treatment outcome movements adjacent to single-implant restorations after
in endodontics: the Toronto study. Phases I and II: Ortho- more than 15 years of follow-up. International Journal of
grade retreatment. Journal of Endodontics 30, 627–33. Prosthodontics 20, 626–32.
Farzaneh M, Abitbol S, Lawrence HP, Friedman S (2004b) John V, Chen S, Parashos P (2007) Implant or the natural
Treatment outcome in endodontics-the Toronto Study. tooth - a contemporary treatment planning dilemma?
Phase II: initial treatment. Journal of Endodontics 30, 302–9. Australian Dental Journal 52(Suppl.), S138–50.
Felton DA (2005) Implant or root canal therapy: a prostho- Kan JYK, Rungcharassaeng K, Umezu K, Kois JC (2003)
dontist’s view. Journal of Esthetic Restorative Dentistry 17, Dimensions of peri-implant mucosa: an evaluation of
197–9. maxillary anterior single implants in humans. Journal of
Fouad AF, Burleson J (2003) The effect of diabetes mellitus on Periodontology 74, 557–62.
endodontic treatment outcome: data from an electronic Kao RT (2008) Strategic extraction: a paradigm shift that is
patient record. Journal of the American Dental Association changing our profession. Journal of Periodontology 79, 971–
134, 43–51. quiz 117-8. 7.
Friedman S (2002) Prognosis of initial endodontic therapy. Kim E, Song JS, Jung IY, Lee SJ, Kim S (2008) Prospective
Endodontic Topics 2, 59–88. clinical study evaluating endodontic microsurgery outcomes
Friedman S (2005) The prognosis and expected outcome of for cases with lesions of endodontic origin compared with
apical surgery. Endodontic Topics 11, 219–62. cases with lesions of combined periodontal-endodontic
Fristad I, Molven O, Halse A (2004) Nonsurgically retreated origin. Journal of Endodontics 34, 546–51.
root-filled teeth – radiographic findings after 20–27 years. Kojima K, Inamoto K, Nagamatsu K et al. (2004) Success rate
International Endodontic Journal 37, 12–8. of endodontic treatment of teeth with vital and nonvital
Gesi A, Bergenholtz G (2003) Pulpectomy – studies on pulps. A meta-analysis. Oral Surgery, Oral Medicine, Oral
outcome. Endodontic Topics 5, 57–70. Pathology, Oral Radiology and Endodontics 97, 95–9.
Gorni FG, Gagliani MM (2004) The outcome of endodontic Kvist T, Reit C (1999) Results of endodontic retreatment: a
retreatment: a 2-yr follow-up. Journal of Endodontics 30, randomized clinical study comparing surgical and nonsur-
1–4. gical procedures. Journal of Endodontics 25, 814–7.
Greenstein G, Greenstein B, Cavallaro J (2007) Prerequisite for Lang NP, Pjetursson BE, Tan K, Brägger U, Egger M, Zwahlen
treatment planning implant dentistry: periodontal prognos- M (2004) A systematic review of the survival and compli-
tication of compromised teeth. Compendium of Continuing cation rates of fixed partial dentures (FPDs) after an
Education in Dentistry 28, 436–46, quiz 47, 70. observation period of at least 5 years. II. Combined tooth-
Greenstein G, Cavallaro J, Tarnow D (2008) When to save or implant-supported FPDs. Clinical Oral Implants Research 15,
extract a tooth in the esthetic zone: a commentary. 643–53.
Compendium of Continuing Education in Dentistry 29, 136– Langer B, Stein SD, Wagenberg B (1981) An evaluation of
45. quiz 46, 58. root resections. A ten-year study. Journal of Periodontology
Heitz-Mayfield LJ (2008) Peri-implant dieases: diagnosis and 52, 719–22.
risk factors. Journal of Clinical Periodontology 35(Suppl. 8), Lazarski MP, Walker WA III, Flores CM, Schindler WG,
292–304. Hargreaves KM (2001) Epidemiological evaluation of the
Holm-Pedersen P, Lang NP, Müller F (2007) What are the outcomes of nonsurgical root canal treatment in a large
longevities of teeth and oral implants? Clinical Oral Implants cohort of insured dental patients. Journal of Endodontics 27,
Research 18(Suppl. 3), 15–9. 791–6.
Hørsted-Bindslev P, Løvschall H (2002) Treatment outcome of Lewis S (1996) Treatment planning: teeth versus implants.
vital pulp treatment. Endodontic Topics 2, 24–34. International Journal of Periodontics & Restorative Dentistry
Iqbal MK, Kim S (2007) For teeth requiring endodontic 16, 366–77.
treatment, what are the differences in outcomes of restored Libman WJ, Nicholls JI (1995) Load fatigue of teeth restored
endodontically treated teeth compared to implant-supported with cast posts and cores and complete crowns. International
restorations? International Journal of Oral Maxillofacial Journal of Prosthodontics 8, 155–61.
Implants 22(Suppl.), 96–116. Lindhe J, Meyle J (2008) Peri-implant diseases: Consensus
Iqbal MK, Kim S (2008) A review of factors influencing Report of the Sixth European Workshop on Periodontology.
treatment planning decisions of single-tooth implants versus Journal of Clinical Periodontology 35, 282–5.

772 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal
 Zitzmann et al. Endodontics or implants?

Listgarten MA (1997) Clinical trials of endosseous implants: Pothukuchi K (2006) Case assessment and treatment plan-
issues in analysis and interpretation. Annals of Periodontol- ning: what governs your decision to treat, refer or replace a
ogy 2, 299–313. tooth that potentially requires endodontic treatment? Aus-
Manning S (2000) Creating a space: when to extract. Annals of tralian Endodontic Journal 32, 79–84.
the Royal Australasian College of Dental Surgeons 15, 240–2. Quirynen M, Abarca M, Van Assche N, Nevins M, van
Marending M, Peters OA, Zehnder M (2005) Factors affecting Steenberghe D (2007) Impact of supportive periodontal
the outcome of orthograde root canal therapy in a general therapy and implant surface roughness on implant outcome
dentistry hospital practice. Oral Surgery, Oral Medicine, Oral in patients with a history of periodontitis. Journal of Clinical
Pathology, Oral Radiology and Endodontics 99, 119–24. Periodontology 34, 805–15.
Meffert RM (1997) Issues related to single-tooth implants. Ray HA, Trope M (1995) Periapical status of endodontically
Journal of the American Dental Association 128, 1383–90. treated teeth in relation to the technical quality of the root
Messer HH (1999) Clinical judgement and decision making in filling and the coronal restoration. International Endodontic
endodontics. Australian Endodontic Journal 25, 124–32. Journal 28, 12–8.
Moiseiwitsch J, Caplan D (2001) A cost-benefit comparison Renvert S, Roos-Jansåker A-M, Claffey N (2008) Non-surgical
between single tooth implant and endodontics. Journal of treatment of peri-implant mucositis and peri-implantitis: a
Endodontics 27, 235. literature review. Journal of Clinical Periodontology 35(Suppl.
Mordohai N, Reshad M, Jivraj SA (2005) To extract or not to 8), 305–15.
extract? Factors that affect individual tooth prognosis Ruskin JD, Morton D, Karayazgan B, Amir J (2005) Failed root
Journal of the California Dental Association 33, 319–28. canals: the case for extraction and immediate implant
Mordohai N, Reshad M, Jivraj S, Chee W (2007) Factors that placement. Journal of Oral and Maxillofacial Surgery 63, 829–
affect individual tooth prognosis and choices in contempo- 31.
rary treatment planning. British Dental Journal 202, 63–72. Salehrabi R, Rotstein I (2004) Endodontic treatment outcomes
Ng YL, Mann V, Rahbaran S, Lewsey J, Gulabivala K (2007) in a large patient population in the USA: an epidemiological
Outcome of primary root canal treatment: systematic review study. Journal of Endodontics 30, 846–50.
of the literature - part 1. Effects of study characteristics on Schenk RK, Buser D (1998) Osseointegration: a reality.
probability of success. International Endodontic Journal 40, Periodontology 2000(17), 22–35.
921–39. Schropp L, Isidor F (2008) Timing of implant placement
Ng YL, Mann V, Gulabivala K (2008a) Outcome of secondary relative to tooth extraction. Journal of Oral Rehabilitation
root canal treatment: a systematic review of the literature. 35(Suppl 1), 33–43.
International Endodontic Journal 41, 1026–46. Sjögren U, Hagglund B, Sundqvist G, Wing K (1990) Factors
Ng YL, Mann V, Rahbaran S, Lewsey J, Gulabivala K (2008b) affecting the long-term results of endodontic treatment.
Outcome of primary root canal treatment: systematic review Journal of Endodontics 16, 498–504.
of the literature - Part 2. Influence of clinical factors. Smith DE, Zarb GA (1989) Criteria for success of osseointe-
International Endodontic Journal 41, 6–31. grated endosseous implants. Journal of Prosthetic Dentistry
Nyman S, Lindhe J (1979) A longitudinal study of combined 62, 567–72.
periodontal and prosthetic treatment of patients with Spangberg LS (2006) To implant, or not to implant: that is the
advanced periodontal disease. Journal of Periodontology 50, question. Oral Surgery, Oral Medicine, Oral Pathology, Oral
163–9. Radiology and Endodontics 101, 695–6.
Palmer R, Howe L (1999) Dental implants. 3. Assessment of Stoll R, Betke K, Stachniss V (2005) The influence of different
the dentition and treatment options for the replacement of factors on the survival of root canal fillings: a 10-year
missing teeth. British Dental Journal 187, 247–55. retrospective study. Journal of Endodontics 31, 783–90.
Palmqvist S, Söderfeldt B (1994) Multivariate analyses of Strietzel FP, Reichart PA, Kale A, Kulkarni M, Wegner B,
factors influencing the longevity of fixed partial dentures, Küchler I (2007) Smoking interferes with the prognosis of
retainers, and abutments. Journal of Prosthetic Dentistry 71, dental implant treatment: a systemic review and
245–50. meta-analysis. Journal of Clincal Periodontology 34, 523–
Pjetursson BE, Tan K, Lang NP, Brägger U, Egger M, Zwahlen 44.
M (2004a) A systematic review of the survival and Tan PL, Aquilino SA, Gratton DG et al. (2005) In vitro fracture
complication rates of fixed partial dentures (FPDs) after an resistance of endodontically treated central incisors with
observation period of at least 5 years. I. Implant-supported varying ferrule heights and configurations. Journal of
FDPs. Clinical Oral Implants Research 15, 625–42. Prosthetic Dentistry 93, 331–6.
Pjetursson BE, Tan K, Lang NP, Brägger U, Egger M, Zwahlen Thomas MV, Beagle JR (2006) Evidence-based decision-
M (2004b) A systematic review of the survival and making: implants versus natural teeth. Dental Clinics of
complication rates of fixed partial dentures (FPDs) after an North America 50, 451–61, viii.
observation period of at least 5 years. IV. Cantilever or Torabinejad M, Goodacre CJ (2006) Endodontic or dental
extension FDPs. Clinical Oral Implants Research 15, 667–76. implant therapy: the factors affecting treatment planning.

ª 2009 International Endodontic Journal International Endodontic Journal, 42, 757–774, 2009 773
 Endodontics or implants? Zitzmann et al.

Journal of the American Dental Association 137, 973–7, quiz dentistry in a prosthodontic practice. International Journal
1027-8. of Prosthodontics 22, 127–35.
Torabinejad M, Anderson P, Bader J et al. (2007) Outcomes of Wang Q, Cheung GS, Ng RP (2004) Survival of surgical
root canal treatment and restoration, implant-supported endodontic treatment performed in a dental teaching
single crowns, fixed partial dentures, and extraction without hospital: a cohort study. International Endodontic Journal
replacement: a systematic review. Journal of Prosthetic 37, 764–75.
Dentistry 98, 285–311. Weyant RJ, Burt BA (1993) An assessment of survival rates
Trope M (2005) Implant or root canal therapy: an endodon- and within-patient clustering of failures for endosseous oral
tist’s view. Journal of Esthetic Restorative Dentistry 17, 139– implants. Journal of Dental Research 72, 2–8.
40. White SN, Miklus VG, Potter KS, Cho J, Ngan AY (2006)
Türp JC, Heydecke G, Krastl G, Pontius O, Antes G, Zitzmann Endodontics and implants, a catalog of therapeutic con-
NU (2007) Restoring the fractured root-canal-treated max- trasts. Journal of Evidence Based Dental Practice 6, 101–9.
illary lateral incisor: in search of an evidence-based Wolcott J, Meyers J (2006) Endodontic re-treatment or
approach. Quintessence International 38, 179–91. implants: a contemporary conundrum. Compendium of
Vire DE (1991) Failure of endodontically treated teeth: Continuing Education in Dentistry 27, 104–10, quiz 11-2.
classification and evaluation. Journal of Endodontics 17, Zitzmann NU, Berglundh T (2008a) Definition and prevalence
338–42. of peri-implant diseases. Journal of Clinical Periodontology
Walter C, Kaner D, Berndt DC, Weiger R, Zitzmann NU (2009) 35(Suppl. 8), 286–91.
3D imaging as a preoperative tool in decision making for Zitzmann NU, Margolin MD, Filippi A, Weiger R, Krastl G
furcation surgery. Journal of Clinical Periodontology 36, 250– (2008b) Patient assessment and diagnosis in implant
7. treatment. Australian Dental Journal 53(Suppl. 1), S3–10.
Walton TR (2009) Changes in patient and FDP profiles
following the introduction of osseointegrated implant

774 International Endodontic Journal, 42, 757–774, 2009 ª 2009 International Endodontic Journal