journal of dentistry 42 (2014) 1509–1527

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/jden

Review

Periodontally compromised vs. periodontally healthy patients and dental implants: A systematic review and meta-analysis Bruno Ramos Chrcanovic a,*, Tomas Albrektsson a,b, Ann Wennerberg a a b

Department of Prosthodontics, Faculty of Odontology, Malmo¨ University, Malmo¨, Sweden Department of Biomaterials, Go¨teborg University, Go¨teborg, Sweden

article info

abstract

Article history:

Objectives: To test the null hypothesis of no difference in the implant failure rates, postop-

Received 22 April 2014

erative infection, and marginal bone loss for the insertion of dental implants in periodon-

Received in revised form

tally compromised patients (PCPs) compared to the insertion in periodontally healthy

28 August 2014

patients (PHPs), against the alternative hypothesis of a difference.

Accepted 25 September 2014

Methods: An electronic search without time or language restrictions was undertaken in March 2014. Eligibility criteria included clinical human studies, either randomized or not. Results: 2768 studies were identified in the search strategy and 22 studies were included. The

Keywords:

estimates of relative effect were expressed in risk ratio (RR) and mean difference (MD) in

Dental implants

millimetres. All studies were judged to be at high risk of bias, none were randomized. A total

Periodontal disease

of 10,927 dental implants were inserted in PCPs (587 failures; 5.37%), and 5881 implants in

Periodontitis

PHPs (226 failures; 3.84%). The difference between the patients significantly affected the

Implant failure rate

implant failure rates (RR 1.78, 95% CI 1.50–2.11; P < 0.00001), also observed when only

Postoperative infection

the controlled clinical trials were pooled (RR 1.97, 95% CI 1.38–2.80; P = 0.0002). There were

Marginal bone loss

significant effects of dental implants inserted in PCPs on the occurrence of postoperative

Meta-analysis

infections (RR 3.24, 95% CI 1.69–6.21; P = 0.0004) and in marginal bone loss (MD 0.60, 95% CI 0.33–0.87; P < 0.0001) when compared to PHPs. Conclusions: The present study suggests that an increased susceptibility for periodontitis may also translate to an increased susceptibility for implant loss, loss of supporting bone, and postoperative infection. The results should be interpreted with caution due to the presence of uncontrolled confounding factors in the included studies, none of them randomized. Clinical Significance: There is some evidence that patients treated for periodontitis may experience more implant loss and complications around implants including higher bone loss and peri-implantitis than non-periodontitis patients. As the philosophies of treatment may alter over time, a periodic review of the different concepts is necessary to refine techniques and eliminate unnecessary procedures. This would form a basis for optimum treatment. # 2014 Elsevier Ltd. All rights reserved.

* Corresponding author at: Department of Prosthodontics, Faculty of Odontology, Malmo¨ University, Carl Gustafs va¨g 34, SE-205 06, Malmo¨, Sweden. Tel.: +46 725 541 545; fax: +46 40 6658503. E-mail addresses: [email protected], [email protected] (B.R. Chrcanovic). http://dx.doi.org/10.1016/j.jdent.2014.09.013 0300-5712/# 2014 Elsevier Ltd. All rights reserved.

1510

1.

journal of dentistry 42 (2014) 1509–1527

Introduction

In an attempt to decrease implant failure rates, more attention is being placed on understanding the etiologic and risk factors that lead to the failure of dental implants.1 The question if patients with a history of periodontitis are more at risk for peri-implant disease has received increasing attention in the last years.2 There is some evidence that patients treated for periodontitis may experience more implant loss and complications around implants including higher bone loss and peri-implantitis than non-periodontitis patients.3 A history of treated periodontitis does not seem to adversely affect implant survival rates over short times of follow-up.4 A small number of periodontal maintenance patients seem to be refractory to treatment and go on to experience continued and significant tooth loss. These subjects also have a high level of implant complications and failure.5 However, the finding that titanium implants are but foreign bodies have resulted in a general questioning whether periodontitis and peri-implantitis are at all related forms of disease.6 Some clinicians assume that periodontally compromised patients (PCPs) present a potentially higher risk for implant failure than healthy individuals. The reason for this assumption is that a similar pathological bacterial flora forms around diseased teeth and diseased implants, though with some differences in partially and completely edentulous patients.7 Implants are rapidly colonized by indigenous periodontal pathogens in partially dentate patients harbouring periodontal lesions.7 Moreover, long-term outcomes demonstrated that implants in nonsmoking PCPs previously treated for periodontitis were more prone to developing marginal bone loss compared with those in PHPs.8 These results were obtained despite the fact that all patients were regularly enrolled in and were compliant with a supporting periodontal therapy (SPT) programme over 10 years.8 Fardal and Linden5 observed that smoking, stress and a family history of periodontal disease were identified as factors associated with a refractory outcome, and these variables remained significant after multivariate analysis. Another study showed that marginal bone level at 10 years was significantly associated with smoking, implant location, fullmouth probing attachment levels, and change, over time, in full-mouth probing pocket depths.9 Having said this, recent investigation demonstrated significantly a different mRNA signatures between periodontitis and peri-implantitis.10 Therefore, a pertinent question in relation to implant therapy in patients susceptible to periodontitis is whether these patients may also show an elevated risk for periimplant tissue destruction. Thus, the aim of this metaanalysis is to compare the survival rate of dental implants, postoperative infection, and marginal bone loss of dental implants inserted in PCPs and in periodontally healthy patients (PHPs). The present study presents a more detailed analysis of the influence of periodontal disease on the implant failure rates, previously assessed in a published systematic review.1

2.

Materials and methods

This study followed the PRISMA Statement guidelines.11 A review protocol does not exist.

2.1.

Objective

The purpose of the present review was to test the null hypothesis of no difference in the implant failure rates, postoperative infection, and marginal bone loss for the insertion of dental implants in PCPs compared to the insertion in PHPs, against the alternative hypothesis of a difference.

2.2.

Search strategies

An electronic search without time or language restrictions was undertaken in March 2014 in the following databases: PubMed, Web of Science, and the Cochrane Oral Health Group Trials Register. The following terms were used in the search strategy on PubMed: (dental implant [Text Word]) AND periodontal disease [Text Word] (dental implant [Text Word]) AND periodontitis [Text Word] The following terms were used in the search strategy on Web of Science, in all databases, refined by selecting the term ‘dentistry oral surgery medicine’ in the filter ‘research area’: (dental implant [Topic]) AND periodontal disease [Topic] (dental implant [Topic]) AND periodontitis [Topic] The following terms were used in the search strategy on the Cochrane Oral Health Group Trials Register: (dental implant OR dental implant failure OR dental implant survival OR dental implant success AND (periodontal disease OR periodontitis)) A manual search of dental implants-related journals, including British Journal of Oral and Maxillofacial Surgery, Clinical Implant Dentistry and Related Research, Clinical Oral Implants Research, European Journal of Oral Implantology, Implant Dentistry, International Journal of Oral and Maxillofacial Implants, International Journal of Oral and Maxillofacial Surgery, International Journal of Periodontics and Restorative Dentistry, International Journal of Prosthodontics, Journal of Clinical Periodontology, Journal of Dental Research, Journal of Dentistry, Journal of Oral Implantology, Journal of Craniofacial Surgery, Journal of Cranio-Maxillofacial Surgery, and Journal of Maxillofacial and Oral Surgery, Journal of Oral and Maxillofacial Surgery, Journal of Oral Rehabilitation, Journal of Periodontology, and Oral Surgery Oral Medicine Oral Pathology Oral Radiology and Endodontology was also performed. The reference list of the identified studies and the relevant reviews on the subject were also scanned for possible additional studies. Moreover, online databases providing information about clinical trials in progress were

journal of dentistry 42 (2014) 1509–1527

checked (https://clinicaltrials.gov/; www.centerwatch.com/ clinicaltrials; www.clinicalconnection.com).

2.3.

Inclusion and exclusion criteria

Eligibility criteria included clinical human studies, either randomized or not, comparing implant failure rates in any group of patients receiving dental implants that are being inserted in PCPs compared to their insertion in PHPs. For this review, implant failure represents the complete loss of the implant. Exclusion criteria were case reports, technical reports, animal studies, in vitro studies, and review papers.

2.4.

Study selection

The titles and abstracts of all reports identified through the electronic searches were read independently by the three authors. For studies appearing to meet the inclusion criteria, or for which there were insufficient data in the title and abstract to make a clear decision, the full report was obtained. Disagreements were resolved by discussion between the authors.

2.5.

Quality assessment

Quality assessment of the studies was executed according to the Newcastle–Ottawa scale (NOS).12 The NOS calculates the study quality on the basis of 3 major components: selection, comparability, and outcome for cohort studies. It assigns a maximum of 4 stars for selection, a maximum of 2 stars for comparability, and a maximum of 3 stars for outcome. According to that quality scale, a maximum of 9 stars/points can be given to an observational study, and this score represents the highest quality, where six or more points were considered high quality.

2.6.

Data extraction and meta-analysis

From the studies included in the final analysis, the following data was extracted (when available): year of publication, study design, unicenter or multicenter study, number of patients, patients’ age, follow-up, days of antibiotic prophylaxis, mouth rinse, implant healing period, failed and placed implants, postoperative infection (reported incidence of peri-implantitis), marginal bone loss, implant surface modification, periodontal disease definitions, periodontal therapy adopted, use of grafting procedures, and presence of smokers among the patients. Contact with authors for possible missing data was performed. Implant failure and postoperative infection were the dichotomous outcome measures evaluated. Weighted mean differences were used to construct forest plots of marginal bone loss, a continuous outcome. The statistical unit for the outcomes was the implant. Whenever outcomes of interest were not clearly stated, the data were not used for analysis. The I2 statistic was used to express the percentage of the total variation across studies due to heterogeneity, with 25% corresponding to low heterogeneity, 50% to moderate and 75% to high. The inverse variance method was used for

1511

random-effects or fixed-effects model. Where statistically significant (P < 0.10) heterogeneity is detected, a randomeffects model was used to assess the significance of treatment effects. Where no statistically significant heterogeneity is found, analysis was performed using a fixedeffects model.13 The estimates of relative effect for dichotomous outcomes were expressed in risk ratio (RR) and in mean difference (MD) in millimetres for continuous outcomes, both with a 95% confidence interval (CI). The degree of statistical significance was considered P < 0.05. Only if there were studies with similar comparisons reporting the same outcome measures was meta-analysis to be attempted. In the case where no events (or all events) are observed in both groups the study provides no information about relative probability of the event and is automatically omitted from the meta-analysis. In this (these) case(s), the term ‘not estimable’ is shown under the column of RR of the forest plot table. The software used here automatically checks for problematic zero counts, and adds a fixed value of 0.5 to all cells of study results tables where the problems occur. A funnel plot (plot of effect size vs. standard error) will be drawn. Asymmetry of the funnel plot may indicate publication bias and other biases related to sample size, although the asymmetry may also represent a true relationship between trial size and effect size. The data were analyzed using the statistical software Review Manager (version 5.2.11, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark, 2014).

3.

Results

3.1.

Literature search

The study selection process is summarized in Fig. 1. The search strategy resulted in 2768 papers. Two combinations of terms were used for PubMed and Web of Science, which resulted in a number of 360 duplicates. The three reviewers independently screened the abstracts for those articles related to the focus question. The initial screening of titles and abstracts resulted in 34 full-text papers; 2374 were excluded for not being related to the topic. The full-text reports of the remaining 34 articles led to the exclusion of 12 because they did not meet the inclusion criteria (6 did not inform of the number of implants per group, 2 were not evaluating implant failures, 2 had earlier follow-up of the same study, 1 had the same study published in another journal, 1 compared between patients with different types of periodontitis and not between PCPs and PHPs). Additional hand-searching of the reference lists of selected studies did not yield additional papers. Thus, a total of 22 publications were included in the review.

3.2.

Description of the studies

Detailed data of the 22 included studies are listed in Tables 1 and 2. Ten CCTs,2,14–22 and twelve retrospective analyses5,8,23–32 were included in the meta-analysis.

Study

Published

Study design

Patients (n) (number per group)

Patients’ age range (average) (years)

1996

RA (unicenter)

51 (31, G1; 20, G2)

16–72 (32.9)

2000

RA (multicenter)

440 (147, G1; 293, G2)

16–90 (53)

Polizzi et al. [14]

2000

CCT (multicenter)

143 (NM)

Watson et al. [15]

2000

CCT (unicenter)

26 (7, G1; 19, G2)

Hardt et al. [25]

2002

RA (unicenter)

50 (25, G1; 25, G2)

Karoussis et al. [16]

2003

CCT (unicenter)

Evian et al. [26]

2004

Rosenberg et al. [27] Mengel and

Healing period/ loading

Failed/placed implants (n)

Implant failure rate (%)

6 months (maxilla) 4 months (mandible) 6 months

5/62 (G1) 2/47 (G2) 39/375 (G1) 29/647 (G2)

8.06 (G1) 4.26 (G2) 10.40 (G1) 4.48 (G2)

14/98 (G1) 3/166 (G2) 0/7 (G1) 0/26 (G2) 8/100 (G1) 3/92 (G2) 2/21 (G1)

14.29 (G1) 1.81 (G2) 0 (G1) 0 (G2) 8 (G1) 3.26 (G2) 9.52 (G1)

3/91 (G2) 16/77 (G1) 6/72 (G2) 86/923 (G1) 37/588 (G2) 2/120 (G1a)

3.30 (G2) 20.78 (G1) 8.33 (G2) 9.32 (G1) 6.29 (G2) 1.67 (G1)

0/30 (G2) 10/122 (G1) 67/1803 (G2)

0 (G2) 8.20 (G1) 3.72 (G2)

1/41 (G1) 0/13 (G2) 17/68 (G1)

2.44 (G1) 0 (G2) 25 (G1)

0/2 (G2) 2/155 (G1b) 0/72 (G2) 16/252 (G1a) 8/261 (G2) 61/1171 (G1)

0 (G2) 1.29 (G1) 0 (G2) 6.35 (G1) 3.07 (G2) 5.21 (G1)

16/455 (G2) 204/5346 (G1c) 15/497 (G2) 4/40 (G1) 2/40 (G2) 5/34 (G1) 9/97 (G2)

3.52 (G2) 3.82 (G1) 3.02 (G2) 10 (G1) 5 (G2) 14.71 (G1) 9.28 (G2)

3/20 (G1) 1/20 (G2) 70/1512 (G1d) 23/747 (G2) 16/185 (G1b) 2/61 (G2)

15 (G1) 5 (G2) 4.63 (G1) 3.08 (G2) 8.65 (G1) 3.28 (G2)

6/198 (G1b) 0/54 (G2)

3.03 (G1) 0 (G2)

Antibiotics/ mouth rinse (days)

Mean 30.5 months (range 1–67) 7 years

10/NM

NM (47, females) NM (40, males) 22–63 (NM)

1, 3, and 5 years

NM

4 years

NM

6 months (maxilla) 3–4 months (mandible) ‘‘delayed’’

5 years

NM

NM

53 (8, G1; 45, G2)

NM (53.5, G1) NM (57.3, G2) NM

10 years

NM

4–6 months

RA (unicenter)

149 (77, G1; 72, G2)

NM

18–4030 days

NM

4–6 months

2004

RA (unicenter)

334 (151, G1; 183, G2)

13 years

7–10/28 days

5–9 months

2005

CCT (unicenter)

39 (27, G1a; 12, G2)

NM (61.1, G1) NM (49.5, G2) 19–59 (32–34, G1; 31, G2)

3 years

NM

NM

Flores-de-Jacoby [17] Wagenberg and Froum [28]

2006

RA (unicenter)

891 (NM)

14–94 (57.9)

Mean 71 months (range 12–193)

12/NM

Mengel et al. [18]

2007

CCT (unicenter)

17 (9, G1; 8, G2)

19–59 (34, G1; 31, G2)

NM

Fardal and Linden [5]

2008

RA (multicenter)

16 (14, G1; 2, G2)

NM (48)

Every 3 months over a 3-year period Mean 13.4 years

Gatti et al. [19]

2008

CCT (multicenter)

62 (33, G1b; 29, G2)

De Boever et al. [20]

2009

CCT (unicenter)

194 (84, G1a; 110, G2)

35–85 (56, G1) 18–61 (40, G2) 20–80 (53.8)

Anner et al. [29]

2010

RA (unicenter)

475 (311, G1; 164, G2)

NM (52)

(range 8–19) 5 years Mean 46.8 (G1) and 48.1 months (G2) Mean 30 months

NM NM/14

6 months (maxilla) 3 months (mandible) 6 months (maxilla) 3 months (mandible) (64 immediately loaded) 6 months (maxilla) 3 months (mandible) NM

0/7

Ranging from immediate to 11 months 5.9  2.1 months

NM

NM

(range 1–114) 5 years

3–5/14

10 years

NM

Ranging from immediate to 9 months 4–6 months

Gianserra et al. [30]

2010

RA (multicenter)

1477 (1281, G1c; 196, G2)

Matarasso et al. [8]

2010

RA (multicenter)

80 (40, G1; 40, G2)

18–85 (50–54, G1) 17–85 (29.9, G2) NM (46.5–48.1)

Simonis et al. [31]

2010

RA (unicenter)

55 (NM)

29–88 (68.7)

10–16 years

7–10/NM

3–4 months

Aglietta et al. [32]

2011

RA (multicenter)

40 (20, G1; 20, G2)

NM (51)

10 years

NM

4–6 months

Levin et al. [21]

2011

CCT (unicenter)

717 (434, G1d; 283, G2)

NM

2012

CCT (unicenter)

101 (73, G1b; 28, G2)

Mean 54 months (up to 144 months) 10 years

NM

Roccuzzo et al. [2]

NM (54, G1) NM (46, G2) NM

NM

3–6 months

Roccuzzo et al. [22]

2014

CCT (unicenter)

123 (91, G1b; 32, G2)

NM (53, G1) NM (43, G2)

10 years

NM

6–12 weeks

journal of dentistry 42 (2014) 1509–1527

Rosenquist and Grenthe [23] Brocard et al. [24]

Follow-up visits (or range)

1512

Table 1 – Detailed data of the included studies.

Study

Published

Study design

P value (for failure rate)

Postoperative infection

P value (for postoperative infection)

Marginal bone loss (mean  SD) (mm)

Periodontal therapy

Periodontal disease definitions

Implant surface modification (brand)

Rosenquist and Grenthe [23]

1996

RA (unicenter)

NM

4 (G1) 1 (G2)

NM

NM

Turned (Nobelpharma, Nobelpharma AB, Go¨teborg, Sweden

‘‘Extraction indication had been periodontitis’’

NM

Brocard et al. [24]

2000

RA (multicenter)

NM

NM

NM

NM

TPS (hollow screws, n = 464; solid screws, n = 251; hollow cylinders, n = 307; ITI, Straumann,

‘‘Prior to implant placement, some patients were treated for periodontal disease’’

Prior to implant placement, the G1 patients were treated for periodontal disease. This involved a

Waldenburg, Switzerland)

Observations

All implants in fresh extraction sockets, use of membranes in 5 patients 132 smokers, 66 bruxers, 177 sites with GBR

followed in some cases by periodontal surgery. All patients were enrolled in a periodontal maintenance programme with

Polizzi et al. [14]

2000

CCT (multicenter)

NM

NM

NM

NM

Turned (Bra˚nemark, Nobel Biocare AB, Go¨teborg, Sweden)

‘‘Periodontitis cited as a reason for tooth extraction’’, ‘‘history of periodontitis before

regular professional plaque control NM

tooth extraction’’

Watson et al. [15]

2000

CCT (unicenter)

NM

NM

NM

NM

Hydroxyapatitiecoated (Calcitek omniloc, Carlsbad,

Chronic periodontitis: pockets 4 mm and radiographic bone

USA)

loss

NM

146 implants in fresh extraction sockets, membranes used in 64 implants, 8 grafts Smokers were included, but the exact number

journal of dentistry 42 (2014) 1509–1527

hygienic phase consisting of scaling, root planning, and oral hygiene instructions,

was not informed, no grafts, only single-tooth restorations

1513

1514

Table 1 (Continued ) Study

Hardt et al. [25]

2002

2003

Study design

RA (unicenter)

CCT (unicenter)

P value (for failure rate)

Postoperative infection

P value (for postoperative infection)

NM

NM

NM

NM

NM

NM

et al. [16]

Evian et al. [26]

2004

RA (unicenter)

NM

NM

NM

Marginal bone loss (mean  SD) (mm) 2.2  0.8 (G1) 1.7  0.8 (G2)

Implant surface modification (brand)

Periodontal disease definitions

Turned (Bra˚nemark, Nobel Biocare AB, Go¨teborg, Sweden)

An overall descriptor of the patient’s experience of periodontal destruction before the time of implant therapy was generated through the calculation of an age-related periodontal bone loss score ‘‘Patients having lost

Mesial, 1.00  1.38 (G1)

TPS (hollow screws,

0.48  1.10 (G2) Distal, 0.94  0.73 (G1) 0.50  1.08 (G2)

ITI, Straumann, Waldenburg, Switzerland)

their teeth due to chronic periodontitis’’

NM

? (Paragon, Zimmer Dental, Carlsbad, USA)

‘‘Periodontal disease was diagnosed if probing depths were 5 mm or greater and associated with radiographic signs of bone loss. Patients who exhibited 1 or more teeth with periodontal disease, or who originally lost their teeth as a result of periodontitis, were considered to have periodontal disease’’

Periodontal therapy

Observations

NM

Fixed partial dentures in the maxillary posterior segments, no grafts

The patients had

28 implants

been treated for periodontal disease according to a comprehensive treatment strategy prior to the installation of

placed in 12 smokers (10 in G1, 18 in G2) and in 41 nonsmokers (11 in G1, 73 in G2)

implants Periodontal treatment was performed prior to or in conjunction with implant placement

Only patients who received a single implant

journal of dentistry 42 (2014) 1509–1527

Karoussis

Published

Rosenberg et al. [27]

2005

RA (unicenter)

CCT (unicenter)

NM

NM

NM

NM

NM

NM

NM

Turned (Bra˚nemark, Nobel Biocare AB, Go¨teborg, Sweden), TPS and SLA (ITI, Straumann, Waldenburg, Switzerland), TPS

‘‘Patients were classified as periodontally compromised if they had a history of periodontal disease that resulted in tooth

Prior to implant placement, all necessary periodontal, restorative, and endodontic treatment was

(IMZ, Biomet, Irvine, USA) acid-etched (Osseotite, 3i, Palm Beach Gardens, USA), hydroxyapatite-

loss. Patients were classified as periodontally healthy if tooth loss was not caused by periodontal disease and if no loss

completed, including extraction of hopeless teeth

coated (Swede-vent, Screw-vent, Corevent, Paragon, Encino, USA)

of attachment (with the exception of facial or lingual recession) or probing depth greater than 3–4 mm was present at the time of implant

1 year, 0.83  0.71 (G1 GA) 0.68  0.54 (G1 GC) 0.58  0.45 (G2) 3 years, 0.31  0.22

Turned (Mk II Bra˚nemark, Nobel Biocare AB, Go¨teborg, Sweden; n = 83), acid-etched

(G1 GA) 0.18  0.11 (G1 GC) 0.12  0.08 (G2)

(Osseotite, 3i Implant Innovations, Palm Beach Gardens, USA; n = 67)

placement’’ The diagnosis of generalized chronic and aggressive periodontitis was based on the American Academy of Periodontology criteria

All patients underwent periodontal surgery and were entered into a 3-month

–

No smokers

recall system, with an oral hygiene control with motivation and instruction where necessary. Subgingival scaling with root planing was performed at tooth surfaces with probing depths >4 mm and bleeding on probing

Wagenberg and Froum [28]

2006

RA (unicenter)

0.02

NM

NM

NM

Turned (Bra˚nemark, Nobel Biocare AB, Go¨teborg, Sweden, n = 1398), acidetched (Osseotite, 3i Implant Innovations, Palm Beach Gardens, USA; n = 527)

‘‘Teeth were lost because of periodontal disease’’

All patients were treated for their periodontal disease prior to or in conjunction with their implant treatment

journal of dentistry 42 (2014) 1509–1527

Mengel and Flores-deJacoby [17]

2004

All implants placed in fresh extraction sockets, bone grafts were utilized in all cases in which there was a residual space around the implant, 13 implants in sinus-lifts, 323

1515

implants in smokers

Study

Mengel et al. [18]

Published

2007

Study design

CCT (unicenter)

P value (for failure rate)

Postoperative infection

P value (for postoperative infection)

NM

NM

NM

Periodontal therapy

Observations

Marginal bone loss (mean  SD) (mm)

Implant surface modification (brand)

Periodontal disease definitions

1 year, 1.02  0.89 (G1) 0.52  0.23 (G2) 3 years, 0.27  0.22 (G1) 0.19  0.23 (G2)

Acid-etched (Osseotite, BIOMET/ 3i, Palm Beach Gardens, USA)

‘‘The diagnosis of generalized aggressive periodontitis was based on the

All generalized aggressive periodontitis patients underwent periodontal

Edentulous patients. G1 patients were fitted with removable

American Academy of Periodontology criteria’’

treatment and were entered into a 3month recall system

implant-tooth– supported superstructures, G2 patients received either fixed cemented

Fardal and Linden [5]

Gatti et al. [19]

2008

2008

RA (multicenter)

CCT (multicenter)

NM

NM

5 (G1) 0 (G2)

4 (G1) 0 (G2)

NM

NM

NM

2.57  1.06 (G1 GA) 2.72  0.44 (G1 GC) 1.24  1.09 (G2)

NM

‘‘The term refractory periodontal disease has been applied to such individuals who are characterized by continued

Patients received initial periodontal therapy, followed by at least 8 years of maintenance treatment in the

Even though the number of smokers was informed for the whole sample of the study, only

degeneration of the periodontium despite ongoing sanative, surgical and/or pharmacological therapy’’

specialist practice

the patients who received implants were considered here, and the number of smokers among these

Several (Nobel Biocare, Gothenburg, Sweden; Zimmer

‘‘The periodontal conditions were assessed using a modification of the

At the first visit, the periodontal conditions were assessed using a

Dental, Carlsbad, USA; Mathys, Bettlach, Switzerland; Straumann, Waldenburg, Switzerland;

Periodontal Screening and Recording index’’

modification of the Periodontal Screening and Recording index, and subsequently periodontal therapy

Dentsply Friadent, Mannheim, Germany)

patients was not informed 39 implants in grafts, 14 smokers (8 in G1, 6 in G2)

journal of dentistry 42 (2014) 1509–1527

implantsupported dentures in the maxilla or single-tooth implants. No smokers.

1516

Table 1 (Continued )

De Boever et al. [20]

Anner et al.

2009

2010

CCT (unicenter)

RA (unicenter)

NM

0.1498

NM

NM

NM

NM

Mesial, 0.28  0.7 (G1 + G2) Distal, 0.24  0.6 (G1 + G2)

NM

TPS and SLA (ITI, Straumann, Waldenburg, Switzerland; n = 259, TPS; n = 254, SLA)

NM

‘‘Periodontally susceptible patients with tooth loss due to periodontal disease and patients with periodontal disease’’

Before implant placement, all patients received, if necessary, periodontal nonsurgical and/or

11.4% of the patients were smokers (13 smokers in G1, 9 in G2), 10.3% former smokers, 134 ridge augmentations (75, G1; 59, G2)

NM

surgical therapy. Periodontally susceptible patients were enrolled in a strict maintenance programme 246 patients (51.7%) participated of a structured supportive periodontal programme Periodontal therapy

smokers

[29]

2010

RA (multicenter)

NM

NM

NM

NM

Several (3i Biomet,

‘‘Periodontal conditions were assessed using a modification of the Periodontal Screening and Recording (PSR) index’’ ‘‘The classification of

(non-surgical and surgical) was administered as required

in G1, 63 in G2)

Turned, 2.78  0.48 (G1)

Astra Tech, Camlog, Friadent-Dentsply, Nobel Biocare, Straumann, Sweden & Martina, Zimmer Dental) Turned (Bra˚nemark,

The patients

No smokers,

Nobel Biocare AB, Go¨teborg, Sweden; n = 40), TPS (ITI, Straumann, Waldenburg, Switzerland; n = 40)

the patients in the two groups was carried out on the basis of the diagnosis reported in the patient’s chart.’’

received individualized periodontal treatment before implant surgery. On the basis of the

only dental implants in a single-unit gap

et al. [30]

Matarasso

2010

RA (multicenter)

NM

NM

NM

et al. [8]

1.95  0.42 (G2) TPS, 2.32  0.41 (G1) 1.43  0.38 (G2)

549 smokers (486

results achieved after the periodontal treatment, the patients were placed on an

Simonis et al. [31]

2010

RA (unicenter)

0.327

13 (G1) 10 (G2)

0.006

Mesial, 2.2  3.4 (G1 + G2) Distal, 2.3  3.4 (G1 + G2)

TPS (ITI, Straumann, Waldenburg, Switzerland; solid screw, n = 116; hollow screw, n = 15)

NM

individually tailored maintenance care programme ‘‘All patients were instructed on how to maintain appropriate oral

journal of dentistry 42 (2014) 1509–1527

Gianserra

49 diabetics, 63

9 smokers, only implantsupported fixed restorations

hygiene around the implants and remaining teeth.’’

1517

1518

Table 1 (Continued ) Study

Aglietta et al. [32]

Roccuzzo et al. [2]

2011

2011

2012

Study design

RA (multicenter)

CCT (unicenter)

CCT (unicenter)

P value (for failure rate)

Postoperative infection

P value (for postoperative infection)

NM

NM

NM

NM

NM

NM

NM

NM

NM

Marginal bone loss (mean  SD) (mm) Turned, 3.47  1.09 (G1) 2.65  0.31 (G2) TPS, 3.77  1.43 (G1) 2.51  0.31 (G2)

NM

0.98  1.22 (G1 severe) 1.14  1.11 (G1 moderate) 0.75  0.88 (G2)

Implant surface modification (brand) Turned (Bra˚nemark, Nobel Biocare AB, Go¨teborg, Sweden; n = 20), TPS (ITI, Straumann, Waldenburg, Switzerland; n = 20) NM

TPS (ITI, Straumann, Waldenburg, Switzerland)

Periodontal disease definitions

Periodontal therapy

Observations

‘‘Patients treated for generalized chronic periodontitis’’

Patients were ‘‘treated for periodontitis’’

All patients were smokers, only dental implants in a single-unit gap

Patients were divided into different periodontal groups according to their

All periodontally involved patients had undergone cause-related as

81 diabetics, 103 smokers (71 in G1, 32 in G2)

periodontal diagnosis that was based on a classification of periodontal diseases

well as correctivephase periodontal interventions (if indicated) before dental implant placement

‘‘PCP received a score (S) on the basis of the number and depth of periodontal pockets according to the following formula: S = Number of

All patients received appropriate initial therapy, consisting, depending on the cases, in motivation, oral hygiene instruction and

pockets (5–7 mm) + 2 Number of pockets (8 mm)’’

scaling and root planing, with the aim to reduce to a minimal level periodontal pathogens. Hopeless teeth were extracted. Periodontal surgery was performed as needed. Guided tissue regeneration was pursued, when feasible

18 smokers (15 in G1, 3 in G2)

journal of dentistry 42 (2014) 1509–1527

Levin et al. [21]

Published

Roccuzzo et al. [22]

2014

CCT (unicenter)

NM

NM

NM

NM

Sandblasted and acid-etched (SLA, Straumann, Waldenburg, Switzerland)

‘‘PCP received a score (S) on the basis of the number and depth of periodontal pockets according to the following formula:

All patients received appropriate initial therapy, consisting, depending on the cases, in motivation, oral

S = Number of pockets (5–7 mm) + 2 Number of pockets (8 mm)’’

hygiene instruction and scaling and root planing, with the aim to reduce to a minimal level periodontal pathogens.

21 smokers (16 in G1, 5 in G2)

was pursued, when feasible NM – not mentioned; CCT – controlled clinical trial; RCT – randomized controlled trial; RA – retrospective analysis; G1 – group periodontitis; G2 – group non- periodontitis; NP – not performed; TPS – titanium plasma-sprayed; GBR – guided bone regeneration; GA – aggressive periodontitis group; GC – chronic periodontitis group. a Here the implants and the patients with chronic adult periodontitis and generalized aggressive periodontitis were put together in G1. b

Here the implants and the patients with severe chronic periodontitis and moderate chronic periodontitis were put together in G1. Here the number of patients and implants were considered for the patients followed for 5 years. The implants and the patients with severe periodontitis (569 patients, 2938 implants, 130 failures) and moderate periodontitis (712 patients, 2408 implants, 74 failures) at the 5-year follow-up were put together in G1. The numbers at baseline were different (1727 patients; 1469, G1, 258, G2), (severe periodontitis: 630 patients, 3260 implants, 130 failures; moderate periodontitis: 839 patients, 2813 implants, 74 failures). d Here the implants and the patients with severe chronic periodontitis and moderate chronic periodontitis were put together in G1. c

journal of dentistry 42 (2014) 1509–1527

Hopeless teeth were extracted. Periodontal surgery was performed as needed. Guided tissue regeneration

1519

1520

journal of dentistry 42 (2014) 1509–1527

Fig. 1 – Study screening process.

Three studies17,18,23 had a follow-up up to 3 years, one15 of 4 years, whereas 18 studies2,5,8,14,16,19–22,24–32 had a maximum follow-up of at least 5 years. From the studies with available data of patients’ age, four23,24,28,30 included non-adults patients. Three studies2,16,26 did not inform of the patients’ age. Only four studies5,19,23,31 provided information about postoperative infection, with 37 occurrences in a total of 184 patients receiving 537 implants. Some patients in thirteen studies2,5,15,16,19–22,24,28–31 were smokers, whereas in one study32 all patients were smokers, and three studies8,17,18 excluded smokers. One study14 inserted part of the implants in fresh extraction sockets, whereas in the other two23,28 all implants were inserted in fresh extraction sockets. Three studies8,26,32 included only patients who received a single implant. Patients were submitted to grafting procedures at the implant site in 5 studies.14,19,20,24,28 One study18 included only edentulous patients, and one25 inserted implants in the maxillary posterior segments only. Any kind of periodontal treatment previous to the implants insertion or a SPT was not mentioned to be performed in four studies.14,15,23,25 Seven studies2,17,19–22,30 included a comparison between periodontitis of different severities. From the 22 studies comparing PHPs and PCPs, a total of 10,927 dental implants were inserted in PCPs, with 587 failures (5.37%), and 5881 implants were inserted in PHPs, with 226 failures (3.84%). There were no implant failures in one study.15 From the 7 studies2,17,19–22,30 comparing periodontitis of different severities, a total of 4460 dental implants were inserted in patients with a more aggressive type of periodontitis, with

210 failures (4.71%), and 3308 implants were inserted in patients with a less aggressive type of periodontitis, with 106 failures (3.20%). The inherent problem in a meta-analysis such as in the present paper is that although some authors see a difference between ‘‘periodontitis’’ and ‘‘aggressive’’ or ‘‘chronic’’ periodontitis, others have not split the material in this manner. Therefore, comparisons are difficult and we will treat periodontitis as one entity in the remaining part of the paper. The most commonly used implants were the titanium plasma-sprayed from Straumann (Straumann, Waldenburg, Switzerland), in nine studies,2,8,16,19,20,24,27,31,32 and the Bra˚nemark (Nobel Biocare AB, Go¨teborg, Sweden), in eight studies.8,14,17,19,25,27,28,32 The latter was not exclusively used in six studies.8,17,19,27,28,32 Three studies5,21,29 did not inform what kind of implants were used. Three studies28,29,31 informed whether there was a statistically significant difference or not between the implant failure rates between the PCPs and PHPs, and only one28 found a statistical significance favouring PHPs. Six studies20,23,27,28,30,31 provided information about the use of prophylactic antibiotics. In one of them,20 it was informed that antibiotics were not prescribed to any patient. Four studies19,20,27,30 provided information about the use of chlorhexidine mouth rinse by the patients.

3.3.

Quality assessment

All studies except one23 were high quality. The scores are summarized in Table 2.

Table 2 – Quality assessment of the studies by the Newcastle-Ottawa scale. Study

Published

Selection Representativeness of the exposed cohort

Selection of external control

Comparability

Ascertainment of exposure

Outcome of interest not present at start

Comparability of cohorts

Rosenquist and Grenthe [23] Brocard et al. [24] Polizzi et al. [14] Watson et al. [15] Hardt et al. [25] Karoussis et al. [16] Evian et al. [26] Rosenberg et al. [27] Mengel and Flores-de-Jacoby [17] Wagenberg and Froum [28] Mengel et al. [18] Fardal and Linden [5] Gatti et al. [19] De Boever et al. [20] Anner et al. [29] Gianserra et al. [30] Matarasso et al. [8] Simonis et al. [31] Aglietta et al. [32] Levin et al. [21] Roccuzzo et al. [2] Roccuzzo et al. [22]

Assessment of outcome

Follow-up long enougha

Total (9/9) Adequacy of follow-up

Additional factor

1996

0

$

$

$

$

0

$

0

0

5/9

2000 2000 2000 2002 2003 2004 2004 2005

$ 0 $ 0 $ 0 $ $

$ $ $ $ $ $ $ $

$ $ $ $ $ $ $ $

$ $ $ $ $ $ $ $

$ $ $ $ $ $ $ $

$ $ 0 0 0 $ $ 0

$ $ $ $ $ $ $ $

$ $ 0 $ $ $ $ 0

0 0 $ $ 0 0 0 0

8/9 7/9 7/9 7/9 7/9 7/9 8/9 6/9

2006

0

$

$

$

$

$

$

$

0

7/9

2007 2008 2008 2009 2010 2010 2010 2010 2011 2011 2012 2014

0 $ $ 0 $ $ $ $ $ $ 0 0

$ $ $ $ $ $ $ $ $ $ $ $

$ $ $ $ $ $ $ $ $ $ $ $

$ $ $ $ $ $ $ $ $ $ $ $

$ $ $ $ $ $ $ $ $ $ $ $

0 0 0 $ $ 0 $ 0 $ $ 0 0

$ $ $ $ $ $ $ $ $ $ $ $

0 $ $ 0 0 $ $ $ $ 0 $ $

$ 0 $ 0 0 $ $ 0 $ 0 $ 0

6/9 7/9 8/9 6/9 7/9 8/9 9/9 7/9 9/9 7/9 7/9 6/9

journal of dentistry 42 (2014) 1509–1527

Main factor

Outcome

a Five years was chosen to be enough for the outcome ‘implant failure’ to occur. This time point was chosen due to the fact that Roccuzzo et al.2 showed that the difference between PHP and PCP is negligible during the first 5 years, but becomes more pronounced later on, being in accordance with the findings of Karoussis et al.,16 who first demonstrated that a 5-year follow-up is usually not sufficient to evaluate the differences in the clinical outcomes of the various groups of patients.

1521

1522

journal of dentistry 42 (2014) 1509–1527

Fig. 2 – Forest plot for the event ‘implant failure’ in the comparison between periodontally compromised vs. periodontally healthy patients.

3.4.

Meta-analysis

In this study, a fixed-effects model was used to evaluate the implant failure in the comparison between PCPs vs. PHPs, since statistically significant heterogeneity was not found (P = 0.87; I2 = 0%). The insertion of dental implants in PCPs or PHPs statistically affected the implant failure rates (P < 0.00001; Fig. 2), in favour of PHPs. A RR of 1.78 (95% CI 1.50–2.11) implies that failures when implants are inserted in PCPs are 1.78 times likely to happen than failures when implants are inserted in PHPs. Since the effect size could differ depending on the research methodology of the studies, a sensitivity analysis was performed. When only the CCTs were pooled, a RR of 1.97 resulted (95% CI 1.38–2.80; heterogeneity: P = 0.53; I2 = 0%; Fig. 3), also statistically affecting the implant failure rates (P = 0.0002). Given the variability of the included studies (varying lengths of follow-up, patient ages, number of implants, classification of severity of periodontitis etc.), the

analysis was also performed with a random-effects model. When all studies were evaluated or when only the CCTs were pooled, the significance of the treatment effect was the same as when the fixed-effects model was used, with the exact same values for the RR and the 95% CI. Only four studies5,19,23,31 provided information about postoperative infection. A fixed-effects model was used, due to lack of statistically significant heterogeneity (P = 0.54; I2 = 0%). The insertion of dental implants in PCPs or PHPs statistically affected the incidence of postoperative infections (P = 0.0004; Fig. 4), in favour of PHPs. A RR of 3.24 (95% CI 1.69– 6.21) was observed. When the analysis was performed with a random-effects model, the values for RR and 95% CI remained the same, as well as the significance of the treatment effect. Five studies8,16,18,25,32 provided information about the marginal bone loss with standard deviation, necessary for the calculation of comparisons in continuous outcomes, comparing PCPs and PHPs (212 implants in PCPs and 269 implants in PHPs; Fig. 5). A random-effects model was used to

Fig. 3 – Forest plot for the event ‘implant failure’ in the comparison between periodontally compromised vs. periodontally healthy patients, when only the CCTs were pooled.

journal of dentistry 42 (2014) 1509–1527

1523

Fig. 4 – Forest plot for the event ‘postoperative infection’ in the comparison between periodontally compromised vs. periodontally healthy patients.

Fig. 5 – Forest plot for the event ‘marginal bone loss’ comparing PCPs and PHPs. 1y – 1 year; 3y – 3 years; turned – turned implants; TPS – titanium plasma-sprayed implants.

evaluate the marginal bone loss, since statistically significant heterogeneity was found (P < 0.00001; I2 = 88%). There was statistically significant difference (MD 0.60, 95% CI 0.33–0.87; P < 0.0001) between the groups concerning the marginal bone loss, favouring PHPs.

3.5.

Publication bias

The funnel plot showed asymmetry when the studies reporting the outcome ‘implant failure’ in the comparison between PCPs vs. PHPs are analyzed, indicating possible presence of publication bias. Seven studies2,8,14,16,23,25,32 collaborated with the asymmetry (Fig. 6), and showed a wide CI range for RR. The study of Polizzi et al.14 was the only one outside the triangular 95% confidence region, showing a very

Fig. 6 – Funnel plot for the studies reporting the outcome event ‘implant failure’.

high value of RR together with a wide CI range (7.90, CI 95%, 2.33–26.82), showing heterogeneity in comparison with the other studies. When only the CCTs were pooled (Fig. 7), a possible presence of publication bias is still indicated.

4.

Discussion

Narrowing the inclusion criteria of studies increases homogeneity but also excludes the results of more trials and thus risks the exclusion of significant data.33 The issue is important because meta-analyses are frequently conducted on a limited number of RCTs. In meta-analyses such as these, adding more information from observational studies may aid in clinical reasoning and establish a more solid foundation for causal inferences.33 However, potential biases are likely to be greater for non-randomized studies compared with RCTs, so results should always be interpreted with caution when they are included in reviews and meta-analyses.34 The search strategy adopted here did not find any randomized study on the subject. Thus, the results must be interpreted carefully. The statistical heterogeneity stands for the variability in the intervention effects being evaluated in the different studies, and is a consequence of clinical or methodological diversity, or both, among the studies. The low level of heterogeneity observed when the outcomes ‘implant failure’ and ‘postoperative infection’ were analyzed is surprising, given the variability of the included studies (varying lengths of follow-up, patient ages, number of implants, classification of severity of periodontitis etc.). For this reason, a randomeffects model was also used to incorporate heterogeneity among studies, resulting in the same significance of the treatment effects. However, it is important to stress that care

1524

journal of dentistry 42 (2014) 1509–1527

Fig. 7 – Funnel plot for the studies reporting the outcome event ‘implant failure’, when only the CCTs were pooled.

must be taken in the interpretation of the chi-squared test, since it has low power in the (common) situation of a metaanalysis when studies have small sample size or are few in number. This means that while a statistically significant result may indicate a problem with heterogeneity, a non-significant result must not be taken as evidence of no heterogeneity.34 Some argue that, since clinical and methodological diversity always occur in a meta-analysis, statistical heterogeneity is inevitable.35 Thus, the test for heterogeneity is irrelevant to the choice of analysis; heterogeneity will always exist whether or not we happen to be able to detect it using a statistical test.35 The present meta-analysis showed that, regardless of how the studies were pooled, either when the retrospective studies and the CCTs were considered together, or when only the CCTs were considered, there was a statistically significant difference between PCPs and PHPs for the outcome ‘implant failure’, in favour of PHPs. In patients in whom teeth were lost for periodontal reasons, the disease may have decreased the available bone following tooth extraction or resulted in the necessity to place the implant with a more exposed surface to achieve ideal prosthetic position. Both of these situations may have resulted in a greater implant failure rate.28 Here it is important to mention the possible high influence of smoking habits on the implants failure rates. Smokers were included in 14 studies here reviewed, and the smoking habit is considered a confounding factor in the present meta-analysis. Tobacco smoking is considered the principal environmental risk factor affecting the pathogenesis of periodontitis but it is also responsible for a great number of other types of disease. Karoussis et al.16 showed in their study that there was a tendency for a poorer survival rate of implants in smokers vs. nonsmokers in patients with a history of periodontitis, indicating that the smoking patient susceptible to periodontitis yields a documented higher risk for implant loss than the non-smoking periodontitis patient or the patient not susceptible to periodontitis at all. Aglietta et al.32 showed that tobacco smokers with a history of treated periodontitis displayed lower implant survival rates and higher marginal bone loss rates compared with smokers PHPs, independent of other factors such as implant type, healing modality and operator, and despite the fact that smokers PCPs were treated for their

periodontal conditions before implant placement and were regularly enrolled in a SPT. When marginal bone loss was analyzed, there was a statistically significant difference between PCPs and PHP, favouring PHPs. Hardt et al.25 observed that the amount of longitudinal peri-implant bone loss is related to pretreatment experience of loss of periodontal bone support. The exact relationship between periodontitis and peri-implantitis remains unknown; we really do not know whether these two types of disease are similar in origin at all since a demonstrated positive correlation between PCP and implant failure in itself does not prove such a connection. In some studies PCPs demonstrated higher implant failure rates but this difference did not reach statistical significance. Obviously, treatment of periodontal infections before implant placement would seem important to avoid placing an implant in an infected bed. It is important to stress here that heterogeneity was identified among the group of studies reporting marginal bone loss. Thus, conclusions should be interpreted with caution and the analysis can at best lead to the generation of hypotheses. The fact that some of these studies reviewed here have a short follow-up is a confounding factor, even though it is hard to define what would be considered a short follow-up period to evaluate implant failures in PCPs. A longer follow-up period can lead to an increase in the failure rate, especially if it is extended beyond functional loading, because other prosthetic factors can influence implant failure from that point onward. This might have led to an underestimation of actual failures in some studies. Analysis of the data disclosed different patterns regarding the distribution of the implant losses over time in the two categories of patients. Rosenberg et al.27 hypothesized that one possible explanation for the difference between the two groups in this pattern of failure is the influence of the host, which plays an important role in the variable inflammatory process and may be significant in patients with a history of periodontal disease. Another possible explanation could be related to local factors. A reduced quantity of hard tissue in the PCP group may be related to periodontal loss prior to tooth extraction.27 Another confounding factor is the fact that the studies used different definitions for the presence of periodontal disease, depending on the threshold chosen for the definition of periodontitis, or which conjunct of characteristics may be considered a periodontal disease, i.e., the diagnostic criteria are less clear. Thus, a clear classification system needs to be implemented with clinical evaluation related to a more specific pathology. Moreover, the outcome measures were not related to the type of periodontitis in every study. When it was reported, there was a statistically significant difference concerning the implant failure rates, favouring the less aggressive type of periodontitis in comparison with the more aggressive. Numerous implants in some studies26 were placed at the same time as periodontal surgical procedures were being carried out. The influence of this co-therapy on implant contamination during the procedure has not been investigated. At the immediate and early implant placement it can be speculated that periodontitis-affected tissues might have had a negative local influence due to the presence of infrabony

journal of dentistry 42 (2014) 1509–1527

defects; this could increase the gap between bone and implant36 or jeopardize achievement of primary stability.37 It is unknown whether textured implant surfaces may be more vulnerable to infection than machined implant surfaces in patients with past or present periodontal disease.26 Some studies presented higher failure rates in PCPs when using TPS implants. The moderate micro-roughness of most modern implants did not seem coupled to more than 1–2% of periimplantitis when followed up for 10 years or more as indicated in a recent review of ten different long term clinical reports of Tioblast, SLA and TiUnite implants.38 Titanium with different surface modifications shows a wide range of chemical, physical properties, and surface topographies or morphologies, depending on how they are prepared and handled,39–41 and it is not clear whether, in general, one surface modification is better than another.42 Another possible limitation of some studies is that implants were not tested for stability during some of the late follow-up visits since many of the prostheses could not be removed because they were permanently cemented. This might have led to an underestimation of actual failures.43 Differences in prosthetic suprastructures including completely or partially edentulous patients in the same study are variables that must also be taken into account. The small number of patients in some studies5,15,17,18,32 also counts as a limitation. Moreover, groups were not completely comparable at baseline in some cases.5,16,17,29,30 The potentially most relevant differences were in terms of age and number of implants/prosthesis. The differences in age and number of implants could have plausible explanations, such as different patterns of tooth loss among groups.19 Patients with a previous history of periodontitis are likely to have lost more teeth because of periodontal disease, and therefore require more implants, whereas the healthy group usually includes patients who had lost teeth through trauma or were affected by tooth agenesia, and therefore they were likely to be younger and require fewer implants.19 The results of the present study have to be interpreted with caution because of its limitations. First of all, all confounding factors may have affected the long-term outcomes and not just the presence or not of periodontal disease, and the impact of these variables on the implant survival rate, postoperative infection and marginal bone loss is difficult to estimate if these factors are not identified separately between the two different procedures in order to perform a meta-regression analysis. The lack of control of the confounding factors limited the potential to draw robust conclusions. Second, due to the retrospective design of some studies the classification of the patients with respect to their experience of periodontal disease could be based only on preoperative radiographic data describing the amount of bone support at remaining teeth, since clinical data regarding the periodontal conditions at time for implant therapy or at subsequent follow-ups are not retrievable.44–46 Third, there are no RCTs in the analysis, and potential biases are likely to be greater for non-randomized studies compared with RCTs. The authors suggest that more research is needed on the history of bone tissue loss prior to implant placement in patients classified as ‘‘periodontally compromised’’ to evaluate the local factors affecting implant failure in these patients.

1525

Due to the multifaceted aspects of any infectious disease such as periodontitis, any correlations between this disease and peri-implantitis need not necessarily indicate that bone loss around teeth and implants is dependent on the same type of disease.

5.

Conclusion

The results of the present systematic review should be interpreted with caution due to the presence of uncontrolled confounding factors in the included studies, none of them randomized. Within the limitations of the existing investigations, the present study suggests that an increased susceptibility for periodontitis may also translate to an increased susceptibility for implant loss, loss of supporting bone, and postoperative infection.

Acknowledgements This work was supported by CNPq, Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico – Brazil. We would like to thank Dr. Ricardo Trindade.

references

1. Chrcanovic BR, Albrektsson T, Wennerberg A. Reasons for failures of oral implants. Journal of Oral Rehabilitation 2014;41:443–76. 2. Roccuzzo M, Bonino F, Aglietta M, Dalmasso P. Ten-year results of a three arms prospective cohort study on implants in periodontally compromised patients. Part 2: clinical results. Clinical Oral Implants Research 2012;23:389–95. 3. Ong CTT, Ivanovski S, Needleman IG, Retzepi M, Moles DR, Tonetti MS, et al. Systematic review of implant outcomes in treated periodontitis subjects. Journal of Clinical Periodontology 2008;35:438–62. 4. Klokkevold PR, Han TJ. How do smoking, diabetes, and periodontitis affect outcomes of implant treatment? International Journal of Oral and Maxillofacial Implants 2007;22(Suppl.):173–202. 5. Fardal O, Linden GJ. Tooth loss and implant outcomes in patients refractory to treatment in a periodontal practice. Journal of Clinical Periodontology 2008;35:733–8. 6. Albrektsson T, Dahlin C, Jemt T, Sennerby L, Turri A, Wennerberg A. Is marginal bone loss around oral implants the result of a provoked foreign body reaction? Clinical Implant Dentistry and Related Research 2014;16:155–65. 7. Apse P, Ellen RP, Overall CM, Zarb GA. Microbiota and crevicular fluid collagenase activity in the osseointegrated dental implant sulcus: a comparison of sites in edentulous and partially edentulous patients. Journal of Periodontal Research 1989;24:96–105. 8. Matarasso S, Rasperini G, Siciliano VI, Salvi GE, Lang NP, Aglietta M. A 10-year retrospective analysis of radiographic bone-level changes of implants supporting single-unit crowns in periodontally compromised vs. periodontally healthy patients. Clinical Oral Implants Research 2010;21:898–903. 9. Karoussis IK, Muller S, Salvi GE, Heitz-Mayfield LJA, Bragger U, Lang NP. Association between periodontal and periimplant conditions: a 10-year prospective study. Clinical Oral Implants Research 2004;15:1–7.

1526

journal of dentistry 42 (2014) 1509–1527

10. Becker ST, Beck-Broichsitter BE, Graetz C, Dorfer CE, Wiltfang J, Hasler R. Peri-implantitis versus periodontitis: functional differences indicated by transcriptome profiling. Clinical Implant Dentistry and Related Research 2014;16:401–11. 11. Moher D, Liberati A, Tetzlaff J, Altman DG, Grp P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of Internal Medicine 2009;151:264–9. W64. 12. Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 2000. Available from: http://www.ohri.ca/programs/ clinical_epidemiology/oxford.asp (accessed on 27.09.14). 13. Egger M, Smith GD. Principles of and procedures for systematic reviews. In: Egger M, Smith GD, Altman DG, editors. Systematic reviews in health care: meta-analysis in context. London: BMJ Books; 2003. p. 23–42. 14. Polizzi G, Grunder U, Goene R, Hatano N, Henry P, Jackson WJ, et al. Immediate and delayed implant placement into extraction sockets: a 5-year report. Clinical Implant Dentistry and Related Research 2000;2:93–9. 15. Watson CJ, Tinsley D, Sharma S. Implant complications and failures: the single-tooth restoration. Dental Update 2000;27:35–8. 40, 42. 16. Karoussis IK, Salvi GE, Heitz-Mayfield LJ, Bragger U, Hammerle CH, Lang NP. Long-term implant prognosis in patients with and without a history of chronic periodontitis: a 10-year prospective cohort study of the ITI Dental Implant System. Clinical Oral Implants Research 2003;14:329–39. 17. Mengel R, Flores-de-Jacoby LF. Implants in patients treated for generalized aggressive and chronic periodontitis: a 3year prospective longitudinal study. Journal of Periodontology 2005;76:534–43. 18. Mengel R, Kreuzer G, Lehmann KM, Flores-de-Jacoby L. A telescopic crown concept for the restoration of partially edentulous patients with aggressive generalized periodontitis: a 3-year prospective longitudinal study. International Journal of Periodontics and Restorative Dentistry 2007;27:231–9. 19. Gatti C, Gatti F, Chiapasco M, Esposito M. Outcome of dental implants in partially edentulous patients with and without a history of periodontitis: a 5-year interim analysis of a cohort study. European Journal of Oral Implantology 2008;1:45–51. 20. De Boever AL, Quirynen M, Coucke W, Theuniers G, De Boever JA. Clinical and radiographic study of implant treatment outcome in periodontally susceptible and nonsusceptible patients: a prospective long-term study. Clinical Oral Implants Research 2009;20:1341–50. 21. Levin L, Ofec R, Grossmann Y, Anner R. Periodontal disease as a risk for dental implant failure over time: a long-term historical cohort study. Journal of Clinical Periodontology 2011;38:732–7. 22. Roccuzzo M, Bonino L, Dalmasso P, Aglietta M. Long-term results of a three arms prospective cohort study on implants in periodontally compromised patients: 10-year data around sandblasted and acid-etched (SLA) surface. Clinical Oral Implants Research 2014;25:1105–12. 23. Rosenquist B, Grenthe B. Immediate placement of implants into extraction sockets: implant survival. International Journal of Oral and Maxillofacial Implants 1996;11:205–9. 24. Brocard D, Barthet P, Baysse E, Duffort JF, Eller P, Justumus P, et al. A multicenter report on 1,022 consecutively placed ITI implants: a 7-year longitudinal study. International Journal of Oral and Maxillofacial Implants 2000;15:691–700. 25. Hardt CRE, Grondahl K, Lekholm U, Wennstrom JL. Outcome of implant therapy in relation to experienced loss of periodontal bone support – a retrospective 5-year study. Clinical Oral Implants Research 2002;13:488–94.

26. Evian CI, Emling R, Rosenberg ES, Waasdorp JA, Halpern W, Shah S, et al. Retrospective analysis of implant survival and the influence of periodontal disease and immediate placement on long-term results. International Journal of Oral and Maxillofacial Implants 2004;19:393–8. 27. Rosenberg ES, Cho SC, Elian N, Jalbout ZN, Froum S, Evian CI. A comparison of characteristics of implant failure and survival in periodontally compromised and periodontally healthy patients: a clinical report. International Journal of Oral and Maxillofacial Implants 2004;19:873–9. 28. Wagenberg B, Froum SJ. A retrospective study of 1925 consecutively placed immediate implants from 1988 to 2004. International Journal of Oral and Maxillofacial Implants 2006;21:71–80. 29. Anner R, Grossmann Y, Anner Y, Levin L. Smoking, diabetes mellitus, periodontitis, and supportive periodontal treatment as factors associated with dental implant survival: a long-term retrospective evaluation of patients followed for up to 10 years. Implant Dentistry 2010;19:57–64. 30. Gianserra R, Cavalcanti R, Oreglia F, Manfredonia MF, Esposito M. Outcome of dental implants in patients with and without a history of periodontitis: a 5-year pragmatic multicentre retrospective cohort study of 1727 patients. European Journal of Oral Implantology 2010;3:307–14. 31. Simonis P, Dufour T, Tenenbaum H. Long-term implant survival and success: a 10–16-year follow-up of nonsubmerged dental implants. Clinical Oral Implants Research 2010;21:772–7. 32. Aglietta M, Siciliano VI, Rasperini G, Cafiero C, Lang NP, Salvi GE. A 10-year retrospective analysis of marginal bone-level changes around implants in periodontally healthy and periodontally compromised tobacco smokers. Clinical Oral Implants Research 2011;22:47–53. 33. Shrier I, Boivin JF, Steele RJ, Platt RW, Furlan A, Kakuma R, et al. Should meta-analyses of interventions include observational studies in addition to randomized controlled trials? A critical examination of underlying principles. American Journal of Epidemiology 2007;166:1203–9. 34. Higgins JPT, Green S, Cochrane Collaboration.Cochrane handbook for systematic reviews of interventions. Chichester, England/Hoboken, NJ: Wiley-Blackwell; 2008. 35. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. British Medical Journal 2003;327:557–60. 36. Carlsson L, Rostlund T, Albrektsson B, Albrektsson T. Implant fixation improved by close fit. Cylindrical implantbone interface studied in rabbits. Acta Orthopaedica Scandinavica 1988;59:272–5. 37. Ivanoff CJ, Sennerby L, Lekholm U. Influence of initial implant mobility on the integration of titanium implants – an experimental study in rabbits. Clinical Oral Implants Research 1996;7:120–7. 38. Albrektsson T, Buser D, Sennerby L. Crestal bone loss and oral implants. Clinical Implant Dentistry and Related Research 2012;14:783–91. 39. Chrcanovic BR, Pedrosa AR, Martins MD. Chemical and topographic analysis of treated surfaces of five different commercial dental titanium implants. Materials Research 2012;15:372–82. 40. Chrcanovic BR, Lea˜o NLC, Martins MD. Influence of different acid etchings on the superficial characteristics of Ti sandblasted with Al2O3. Materials Research 2013;16:1006–14. 41. Chrcanovic BR, Martins MD. Study of the influence of acid etching treatments on the superficial characteristics of Ti. Materials Research 2014;17:373–80. 42. Wennerberg A, Albrektsson T. Effects of titanium surface topography on bone integration: a systematic review. Clinical Oral Implants Research 2009;20:172–84.

journal of dentistry 42 (2014) 1509–1527

43. Grondahl K, Lekholm U. The predictive value of radiographic diagnosis of implant instability. International Journal of Oral and Maxillofacial Implants 1997;12:59–64. 44. Chrcanovic BR, Abreu MH, Freire-Maia B, Souza LN. Facial fractures in children and adolescents: a retrospective study of 3 years in a hospital in Belo Horizonte, Brazil. Dental Traumatology 2010;26:262–70.

1527

45. Chrcanovic BR, Souza LN, Freire-Maia B, Abreu MH. Facial fractures in the elderly: a retrospective study in a hospital in Belo Horizonte, Brazil. Journal of Trauma 2010;69:E73–8. 46. Chrcanovic BR, Abreu MH, Freire-Maia B, Souza LN. 1,454 mandibular fractures: a 3-year study in a hospital in Belo Horizonte, Brazil. Journal of Cranio-Maxillofacial Surgery 2012;40:116–23.