Interventions for replacing missing teeth: attachment systems for implant overdentures in edentulous jaws
Abstract
Background
Implant overdentures are one of the most common treatment options used to rehabilitate edentulous patients. Attachment systems are used to anchor the overdentures to implants. The plethora of attachment systems available dictates a need for clinicians to understand their prosthodontic and patient‐related outcomes.
Objectives
To compare different attachment systems for maxillary and mandibular implant overdentures by assessing prosthodontic success, prosthodontic maintenance, patient preference, patient satisfaction/quality of life and costs.
Search methods
Cochrane Oral Health's Information Specialist searched the following databases: Cochrane Oral Health's Trials Register (to 24 January 2018); Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 12) in the Cochrane Library (searched 24 January 2018); MEDLINE Ovid (1946 to 24 January 2018); and Embase Ovid (1980 to 24 January 2018). The US National Institutes of Health Trials Registry (ClinicalTrials.gov) and the World Health Organization International Clinical Trials Registry Platform were searched for ongoing trials on 24 January 2018. No restrictions were placed on the language or date of publication when searching the electronic databases.
Selection criteria
All randomised controlled trials (RCTs), including cross‐over trials on maxillary or mandibular implant overdentures with different attachment systems with at least 1 year follow‐up.
Data collection and analysis
Four review authors extracted data independently and assessed risk of bias for each included trial. Several corresponding authors were subsequently contacted to obtain missing information. Fixed‐effect meta‐analysis was used to combine the outcomes with risk ratios (RR) for dichotomous outcomes and mean differences (MD) for continuous outcomes, with 95% confidence intervals (95% CI). We used the GRADE approach to assess the quality of evidence and create 'Summary of findings' tables.
Main results
We identified six RCTs with a total of 294 mandibular overdentures (including one cross‐over trial). No trials on maxillary overdentures were eligible. Due to the poor reporting of the outcomes across the included trials, only limited analyses between mandibular overdenture attachment systems were possible.
Comparing ball and bar attachments, upon pooling the data regarding short‐term prosthodontic success, we identified substantial heterogeneity (I2 = 97%) with inconsistency in the direction of effect, which was unexplained by clinical or methodological differences between the studies, and accordingly we did not perform meta‐analyses for this outcome. Short‐term re‐treatment (repair of attachment system) was higher with ball attachments (RR 3.11, 95% CI 1.68 to 5.75; 130 participants; 2 studies; very low‐quality evidence), and there was no difference between both attachment systems in short‐term re‐treatment (replacement of attachment system) (RR 1.18, 95% CI 0.38 to 3.71; 130 participants; 2 studies; very low‐quality evidence). It is uncertain whether there is a difference in short‐term prosthodontic success when ball attachments are compared with bar attachments.
Comparing ball and magnet attachments, there was no difference between them in medium‐term prosthodontic success (RR 0.84, 95% CI 0.64 to 1.10; 69 participants; 1 study; very low‐quality evidence), or in medium‐term re‐treatment (repair of attachment system) (RR 1.75, 95% CI 0.65 to 4.72; 69 participants; 1 study; very low‐quality evidence). However, after 5 years, prosthodontic maintenance costs were higher when magnet attachments were used (MD ‐247.37 EUR, 95% CI ‐346.32 to ‐148.42; 69 participants; 1 study; very low‐quality evidence). It is uncertain whether there is a difference in medium‐term prosthodontic success when ball attachments are compared with magnet attachments.
One trial provided data for ball versus telescopic attachments and reported no difference in prosthodontic maintenance between the two systems in short‐term patrix replacement (RR 6.00, 95% CI 0.86 to 41.96; 22 participants; 1 study; very low‐quality evidence), matrix activation (RR 11.00, 95% CI 0.68 to 177.72; 22 participants; 1 study; very low‐quality evidence), matrix replacement (RR 1.75, 95% CI 0.71 to 4.31; 22 participants; 1 study; very low‐quality evidence), or in relining of the implant overdenture (RR 2.33, 95% CI 0.81 to 6.76; 22 participants; 1 study; very low‐quality evidence). It is uncertain whether there is a difference in short‐term prosthodontic maintenance when ball attachments are compared with telescopic attachments.
In the only cross‐over trial included, patient preference between different attachment systems was assessed after only 3 months and not for the entire trial period of 10 years.
Authors' conclusions
For mandibular overdentures, there is insufficient evidence to determine the relative effectiveness of different attachment systems on prosthodontic success, prosthodontic maintenance, patient satisfaction, patient preference or costs. In the short term, there is some evidence that is insufficient to show a difference and where there was no evidence was reported. It was not possible to determine any preferred attachment system for mandibular overdentures.
For maxillary overdentures, there is no evidence (with no trials identified) to determine the relative effectiveness of different attachment systems on prosthodontic success, prosthodontic maintenance, patient satisfaction, patient preference or costs.
Further RCTs on edentulous cohorts must pay attention to trial design specifically using the same number of implants of the same implant system, but with different attachment systems clearly identified in control and test groups. Trials should also determine the longevity of different attachment systems and patient preferences. Trials on the current array of computer‐aided designed/computer‐assisted manufactured (CAD/CAM) bar attachment systems are encouraged.
PICOs
Plain language summary
Attachments for implant dentures
Review question
The aim of this review was to assess different attachments used for upper or lower jaw implant dentures with respect to their success, wear and tear, patient satisfaction, patient preference and cost.
Background
For adults with complete tooth loss, the modern approach is implant dentures with attachment systems connecting the implants to the undersurface of the dentures. It is important to do the review as the choice of the number of implants and the design of the attachments influences prosthesis success, the amount of wear and tear, patient satisfaction, preference and costs.
Study characteristics
Authors from Cochrane Oral Health carried out this review and the evidence is up to date to 24 January 2018. A total of six trials on adults with complete tooth loss were included with a total of 294 lower jaw dentures (anchored by one or more implants). The review looked at different attachment systems on the same implant systems. The six trials did not each evaluate the same attachment systems. There were no eligible trials with upper jaw implant dentures.
Key results
There is insufficient evidence to determine any significant differences between lower jaw implant denture attachment systems and an absence of evidence for upper jaw implant denture attachment systems. Further randomised controlled trials on people with complete tooth loss wearing dentures must pay specific attention to trial design using the same implant system and the same number of implants, but different attachment systems to determine their longevity and patient preferences.
Quality of the evidence
We judged the quality of the evidence to be very low. In all the included trials there were relatively few participants and few events, and there were serious limitations in the trial designs with data missing or not all outcomes reported.
Authors' conclusions
Summary of findings
| Interventions for replacing missing teeth: attachment systems for implant overdentures in edentulous jaws | ||||||
| Patient or population: edentulous adults receiving implant overdentures in one or both jaws to overcome problems with conventional complete dentures | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | Number of participants | Certainty of the evidence | Comments | |
| Risk with bar attachment system | Risk with ball attachment system | |||||
| Success (short term) | Study population | ‐ | 130 | ⊕⊝⊝⊝ | Considerable heterogeneity (I2 = 97%). Pooling of data was not done | |
| See comment | See comment | |||||
| Re‐treatment (repair) (short term) | Study population | RR 3.11 | 130 | ⊕⊝⊝⊝ | ||
| 141 per 1000 | 437 per 1000 | |||||
| Re‐treatment (replace) (short term) | Study population | RR 1.18 | 130 | ⊕⊝⊝⊝ | ||
| 78 per 1000 | 92 per 1000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Risk of bias (serious), inconsistency (very serious), imprecision (very serious); downgraded by 3 levels. | ||||||
| Interventions for replacing missing teeth: attachment systems for implant overdentures in edentulous jaws | ||||||
| Patient or population: edentulous adults receiving implant overdentures in one or both jaws to overcome problems with conventional complete dentures | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | Number of participants | Certainty of the evidence | Comments | |
| Risk with telescopic attachment system | Risk with ball attachment system | |||||
| Patrix replaced (short term) | Study population | RR 6.00 | 22 | ⊕⊝⊝⊝ | ||
| 91 per 1000 | 545 per 1000 | |||||
| Matrix activated (short term) | Study population | RR 11.00 | 22 | ⊕⊝⊝⊝ | ||
| 0 per 1000 | 0 per 1000 | |||||
| Matrix replaced (short term) | Study population | RR 1.75 | 22 | ⊕⊝⊝⊝ | ||
| 364 per 1000 | 636 per 1000 | |||||
| Reline of implant overdenture (short term) | Study population | RR 2.33 | 22 | ⊕⊝⊝⊝ | ||
| 273 per 1000 | 635 per 1000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Risk of bias (serious), imprecision (very serious); downgraded by 3 levels. | ||||||
| Interventions for replacing missing teeth: attachment systems for implant overdentures in edentulous jaws | ||||||
| Patient or population: edentulous adults receiving implant overdentures in one or both jaws to overcome problems with conventional complete dentures | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | Number of participants | Certainty of the evidence | Comments | |
| Risk with magnetic attachment system | Risk with ball attachment system | |||||
| Success (medium term) | Study population | RR 0.84 | 69 | ⊕⊝⊝⊝ | ||
| 826 per 1000 | 694 per 1000 | |||||
| Re‐treatment (repair) (medium term) | Study population | RR 1.75 | 69 | ⊕⊝⊝⊝ | ||
| 174 per 1000 | 304 per 1000 | |||||
| Costs (medium term) | The mean costs (medium term) was EUR 2286.34 | MD 247.37 EUR lower | ‐ | 69 | ⊕⊝⊝⊝ | |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Risk of bias (serious), imprecision (very serious); downgraded by 3 levels. | ||||||
Background
Description of the condition
Edentulism according to the Glossary of Prosthodontic Terms (Ferro 2017; van Blarcom 2005) is defined as the state of being without any natural teeth. Complete tooth loss is an irreversible condition.
From an epidemiological perspective, there are numerous related risk factors leading to edentulism, with caries and periodontal disease seen as the main causes according to the World Health Organization (WHO) Global Oral Health databank (Brown 2009; Felton 2009; Petersen 2005a; Petersen 2005b). It is considered a disability by the WHO. Edentulism is also markedly affected by several factors including access to care, attitude towards dental hygiene, dentist/population ratios, oral health knowledge, education level, socioeconomic status, and lifestyle. Accordingly, its prevalence is known and varies considerably between countries and between different regions within the same country (Douglass 2002; Enami 2013; Mojon 2004; Petersen 2005a). It is higher in populations of lower income and educational levels, in rural areas, and in females; with the later being restricted to certain countries (Enami 2013; Felton 2009). Edentulism is always accompanied by reduction in the quality of life, due to it adversely affecting both oral and general health (Kapur 1964; Locker 1988; MacEntee 2003; Petersen 2005b). Tooth loss is associated with an increased risk of early mortality (Gupta 2018). Historically it has been identified to be a major oral disease entity (Atwood 1971).
From a prosthodontic perspective, the complete loss of either maxillary or mandibular teeth or both leads to adverse aesthetic and biomechanical sequelae including residual ridge resorption, degenerative changes, impaired masticatory function and loss of neuromuscular control (Hobkirk 2013; Zarb 2004).
Rehabilitation of edentulism improves quality of life and reduces morbidity. A recent systematic review on the rehabilitation of edentulism and mortality showed most of the included studies indicated a higher proportion of deceased edentulous patients not using dentures as compared to denture wearers (Gupta 2018).
Edentulism is managed through prosthodontic rehabilitation, either implant‐ or tissue‐supported, with a fixed or removable prosthesis (Ferro 2017; Gupta 2018; van Blarcom 2005).
Description of the intervention
An overdenture is a removable dental prosthesis that covers and rests on one or more remaining natural teeth, the roots of natural teeth, and/or dental implants (Ferro 2017; van Blarcom 2005). Removable implant overdentures are one standard of care (Feine 2002; Fitzpatrick 2006) used to resolve the problems of edentulism and the limitations of conventional complete dentures (Zarb 2004). Removable overdentures are connected to the implants using different attachment systems (Payne 2013; Preiskel 1996). An attachment system is defined as a mechanical device for the fixation, retention and stabilization of an overdenture (Ferro 2017; van Blarcom 2005). All attachment systems are comprised of two parts: the matrix being the receptacle component, and the patrix which closely fits into the matrix either mechanically, magnetically or by friction fit (Ferro 2017; Laney 2007; van Blarcom 2005). One part of the attachment system is connected to the implants and the reciprocal part incorporated within the undersurface of the overdenture.
Ball (stud‐shaped), magnetic and telescopic attachment systems are most commonly extra‐radicular (in which the male patrix element projects from the implant abutment) but also intra‐radicular (where the male patrix element forms part of the denture base and engages a depression on the implant abutment). Bar attachment systems can be divided into two groups, those allowing rotational movement between the components (termed bar joints ‐ where a spacer between the patrix and matrix is used) and comparatively rigid ones allowing no movement (termed bar units ‐ where no spacer between the patrix and matrix is used) (Preiskel 1996). Overdenture bars have historically been soldered, cast with milled designs, made using spark‐erosion or milled precision bars (Sadowsky 2007). Aligning with the path of insertion of the overdenture, the patrix and matrix can be made of metal alloys or plastic of various diameters and configurations and range from O‐rings to pillar‐shaped projections (telescopic attachment systems) with varying amount of retention (Laney 2007; Preiskel 1985; Preiskel 1996). All ball (or stud), magnetic, telescopic and bar attachment systems are adjustable or replaceable or both (Ferro 2017; Laney 2007; Preiskel 1996; van Blarcom 2005). Today, it is more common to see the use of computer‐aided designed/computer‐assisted manufactured (CAD/CAM) technology for both bar attachment systems alone or including additional ball (stud) attachments.
Mandibular overdentures (most commonly opposing complete maxillary dentures) are usually assisted by either one or two unsplinted implants with free‐standing ball (stud‐shaped), magnetic or telescopic attachment systems, or alternatively by two or as many as four splinted implants connected by bar joints or bar units (Alsabeeha 2009; Burns 2000; Carlsson 2003; Payne 2000a; Payne 2013).
Maxillary overdentures (most commonly opposing mandibular dentitions) have historically been assisted by at least four splinted implants using one of the bar attachment systems (Mericske‐Stern 2003; Payne 2013; Sadowsky 2007), but today can be assisted by as few as unsplinted two implants with ball or stud attachment systems (Zembic 2014).
Implant overdentures depend on their attachment systems for prosthodontic treatment outcomes, but are independent of peri‐implant marginal/crestal bone loss traditionally used in determining implant success or survival (Kim 2012; Klemetti 2008). Prosthodontic success is influenced by post‐insertion maintenance or technical complications, adjustments/repairs or aftercare of the attachment systems (Andreiotelli 2010; Bragger 2015; Cehreli 2010a; Kim 2012). Terminology and methodology of reporting prosthodontic outcomes varies with attachment system design and implant systems (Payne 2001a). Patient preference, patient satisfaction (including quality of life) and costs may also be independent of the attachment system used (Kim 2012).
How the intervention might work
The intervention is by way of implants, the attachment system and the overdentures. The attachment systems provide anchorage, retention and stability for implant overdentures. This facilitates improved masticatory function (Geertman 1999; van Kampen 2004), bite forces (Fontijin‐Tekamp 1998), food selection (Allen 2002) and quality of life (Awad 2003; Feine 1998).
There are four broad groups of attachment systems.
-
Ball/stud attachment systems: these are the simplest and the most widely used. They have a considerable stress‐breaking/stress‐relieving effect, provide adequate amount of retention and stability, are available in several vertical heights, and can be used with non‐parallel implants. Although the size of ball/stud attachments currently used are small, the choice of historical larger ball/stud attachment systems is influenced by inter‐arch space (Preiskel 1996) (Figure 1).
-
Bar attachment systems: these are made of single or multiple bars attached to and splinting the implants together, with the clips (riders) positioned in the undersurface of the overdenture. They offer high retentive capacities, reduce loading forces over the implants, and aid in correcting misaligned implants. However, the bulk of the attachments limits its application where there is limited inter‐arch space and minimal residual ridge resorption. If used with tapered arches it may encroach on the tongue space and influence speech. Gingival hyperplasia or mucosal enlargement can develop under the bars (Payne 2001b), and the plaque control and hygiene measures are more complicated (Preiskel 1996) (Figure 2).
-
Magnet attachment systems: these offer the advantage of self‐seating the prosthesis, which is especially suitable for elderly patients with limited manual dexterity or arthritis. Their fabrication procedures are relatively simple, can be used with maligned implants, and the amount of lateral loads transmitted by the retainer are less. Intraoral corrosion remains a main drawback, since it leads to rapid loss of retention and the replacement of the attachments becomes inevitable. Plaque tends to accumulate more around magnets, thus meticulous hygiene measures are required (Preiskel 1996) (Figure 3).
-
Telescopic attachment systems: these are composed of primary copings attached to implants and secondary telescopic crowns embedded in the overdenture. Hygiene measures are much easier and accessible in this system, and the secondary telescopic crowns add to the high retention and stability of the overdenture. Metal display of the primary crowns when the overdenture is removed can influence aesthetics (Langer 1980; Preiskel 1985) (Figure 3).
Ball stud attachment systems.
Copyright© 2018 Payne A, Zarb G. Implant overdentures. In: Zarb G, Hobkirk J, Eckert S, Jacob R, editor(s). Prosthodontic Treatment for Edentulous Patients: Complete Dentures and Implant‐Supported Prostheses. 13th edition. St Louis, Missouri, USA: Mosby, 2013:330‐9 (Payne 2013): reproduced with permission.
Bar attachment systems.
Copyright© 2018 Payne A, Zarb G. Implant overdentures. In: Zarb G, Hobkirk J, Eckert S, Jacob R, editor(s). Prosthodontic Treatment for Edentulous Patients: Complete Dentures and Implant‐Supported Prostheses. 13th edition. St Louis, Missouri, USA: Mosby, 2013:330‐9 (Payne 2013): reproduced with permission.
Magnet telescopic attachment systems.
Copyright© 2018 Professor Ralf Kohal, Germany; Professor Yoshinobu Maeda, Japan: reproduced with permission.
Why it is important to do this review
To determine whether any implant overdenture attachment system is more successful than others in terms of prosthodontic success, prosthodontic maintenance, patient preference, patient satisfaction and costs. Implant overdentures are used widely, but there are significant differences between some countries in their use related to healthcare systems which can favour or hinder their choice as a treatment option by clinicians (Carlsson 2004). A recent global survey completed by 116 prosthodontists from 33 countries showed that implant overdenture treatment for their edentulous patients is common, but there was great variation regarding the number of implants and the attachment systems used (Kronstrom 2017). The majority (84%) reported using two implants, while 13% used four implants for overdenture retention. Only one respondent used a single implant for retention of the mandibular overdenture, and two reported using three implants. There were great variations in the use of attachment systems for mandibular implant overdentures with as many as 10 different types listed in terms of usage, with the most common being the Locator® attachments system (Zest Anchors LLC, Espandido, CA, USA).
Clinicians usually rely on their expertise, personal preferences, dental technician preferences or most commonly commercial influence in attachment system selection (Naert 2003). Attachment systems wear and deteriorate over time causing prosthodontic maintenance events for clinicians and additional costs to patients in terms of re‐treatment. The constant evolution and changes in overdenture attachment systems principally driven by commercial implant companies, has resulted in some older designs being superseded by newer designs or alternatively subsequently withdrawn as problems develop with them in clinical practice (Walton 2006). In addition, there are pertinent aspects of sequential post‐treatment costs related to the prosthodontic maintenance of overdenture attachment systems which is different to fixed implant bridges.
Objectives
To compare different attachment systems for maxillary and mandibular implant overdentures by assessing prosthodontic success, prosthodontic maintenance, patient preference, patient satisfaction/quality of life and costs.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) including cross‐over trials with at least 1 year follow‐up on attachment systems, and maxillary or mandibular overdentures or both, reporting prosthodontic and patient outcomes.
Types of participants
Edentulous adults receiving implant overdentures in one or both jaws to overcome problems with conventional complete dentures.
Types of interventions
-
The same number of implants of the same implant system comparing different attachment systems (more than one).
-
Ball/stud, magnetic, telescopic or bar attachment systems with mandibular 1‐, 2‐, 3‐, or 4‐implant overdentures using splinted or unsplinted prosthodontic designs.
-
Ball/stud, magnetic, telescopic or bar attachment systems with maxillary 1‐, 2‐, 3‐, 4‐, 5‐, or 6‐implant overdentures using splinted or unsplinted prosthodontic designs.
Types of outcome measures
Primary outcomes
-
Prosthodontic success by specific categorization (Appendix 1) and six‐field protocol (Appendix 2).
-
Prosthodontic maintenance by general categorization.
Secondary outcomes
-
Patient preference in cross‐over trials.
-
Patient satisfaction or quality of life assessed by a validated measure.
-
Cost (treatment time or material costs or both).
The outcomes would be assessed at the following time intervals:
-
Short term: 1 to 3 years.
-
Medium term: 3 to 5 years.
-
Long term: 5 to 10 years.
Search methods for identification of studies
Electronic searches
Cochrane Oral Health's Information Specialist conducted systematic searches in the following databases for randomised controlled trials and controlled clinical trials without language or publication status restrictions:
-
Cochrane Oral Health's Trials Register (searched 24 January 2018) (Appendix 3);
-
Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 12) in the Cochrane Library (searched 24 January 2018) (Appendix 4);
-
MEDLINE Ovid (1946 to 24 January 2018) (Appendix 5);
-
Embase Ovid (1980 to 24 January 2018) (Appendix 6).
Subject strategies were modelled on the search strategy designed for MEDLINE Ovid. Where appropriate, they were combined with subject strategy adaptations of the highly sensitive search strategy designed by Cochrane for identifying randomised controlled trials and controlled clinical trials as described in the Cochrane Handbook for Systematic Reviews of Interventions Chapter 6 (Lefebvre 2011).
Searching other resources
The following trial registries were searched for ongoing studies:
-
US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (clinicaltrials.gov; searched 24 January 2018) (Appendix 7);
-
World Health Organization International Clinical Trials Registry Platform (apps.who.int/trialsearch; searched 24 January 2018) (Appendix 7).
We wrote to authors of the identified RCTs, checked the bibliographies of all identified RCTs and used personal contacts in an attempt to identify unpublished or ongoing RCTs. We also contacted corresponding authors for further information to principally clarify aspects of the risk of bias tables, as well as unpublished data.
We did not perform a separate search for adverse effects of interventions used, we considered adverse effects described in included studies only.
Data collection and analysis
We used methods which were specified in the review protocol (Payne 2009) and updated where necessary to conform with the latest Methodological Expectations of Cochrane Intervention Reviews (MECIR) (MECIR 2016).
Selection of studies
Four review authors (Alan Payne (AP), Nabeel Alsabeeha (NA), Momen A Atieh (MAA), and Marwah Anas El‐Wegoud (MAEW)) independently assessed studies for eligibility by initially screening titles, abstracts and keywords of every record retrieved through the electronic searches. The search was designed to be sensitive and include controlled clinical trials, these were filtered out early in the selection process if they were not randomised. We retrieved full‐text articles for further assessment when studies met the inclusion criteria. In the presence of more than one publication of the same trial, we reviewed all the publications. We linked together studies with multiple publications under a single study ID. Any disagreements were resolved by discussion and consultation with a fifth author. Where resolution was not possible, a sixth review author was consulted. We excluded any studies that had insufficient data.
Data extraction and management
We initially designed a form to extract data. For eligible studies, four review authors (AP, MAEW, MAA, and NA) independently extracted the data using the agreed form. We resolved discrepancies through discussion or, if required, we consulted a fifth person. We entered the data into Review Manager software (Review Manager 2014) and checked them for accuracy. When information was unclear, we contacted authors of the original reports to provide further details. The review authors were not blinded to the study authors' names, institutional affiliations, or journal of publication.
Assessment of risk of bias in included studies
Three review authors (AP, MAEW, MAA) independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). There were no disagreements on the assessment of risk of bias in the included studies.
Random sequence generation (checking for possible selection bias)
We described for each included trial the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We assessed the method as:
-
low risk of bias (any truly random process, e.g. random number table; computer random number generator);
-
high risk of bias (any non‐random process, e.g. odd or even date of birth; hospital or clinic record number): these studies were excluded as per the pre‐specified eligibility criteria; or
-
unclear risk of bias.
Allocation concealment (checking for possible selection bias)
We described for each included trial the method used to conceal allocation to interventions prior to assignment and assess whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.
We assessed the methods as:
-
low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
-
high risk of bias (open random allocation; unsealed or non‐opaque envelopes);
-
unclear risk of bias.
Blinding of participants and personnel (checking for possible performance bias)
We described for each included trial the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered studies to be at low risk of bias if they were blinded, or if we judge that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed the methods as:
-
low, high or unclear risk of bias for participants and for personnel.
Blinding of outcome assessment (checking for possible detection bias)
We described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed the methods used to blind outcome assessment as:
-
low, high or unclear risk of bias.
Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)
We described for each included trial, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported and the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information is reported, or was supplied by the trial authors, we re‐included missing data in the analyses which we undertook.
We assessed methods as:
-
low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups);
-
high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; 'as treated' analysis done with substantial departure of intervention received from that assigned at randomisation);
-
unclear risk of bias.
We categorized greater than 20% missing data as 'high' risk of bias.
Selective reporting (checking for reporting bias)
We described for each included trial how we investigated the possibility of selective outcome reporting bias and what we found.
We assessed the methods as:
-
low risk of bias (where it is clear that all of the trial's pre‐specified outcomes and all expected outcomes of interest to the review have been reported);
-
high risk of bias (where not all the trial's pre‐specified outcomes have been reported; one or more reported primary outcomes were not pre‐specified; outcomes of interest are reported incompletely and so cannot be used; trial fails to include results of a key outcome that would have been expected to have been reported);
-
unclear risk of bias.
Other bias
We described for each included trial any important concerns we have about other possible sources of bias.
We assessed whether each trial was free of other problems that could put it at risk of bias:
-
low risk of other bias;
-
high risk of other bias;
-
unclear whether there is risk of other bias.
Overall risk of bias
We have made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). With reference to all domains above, we assessed the likely magnitude and direction of bias and whether we considered it likely to impact on the findings.
We assessed the methods as:
-
low risk of other bias (low risk of bias for all key domains);
-
high risk of other bias (high risk of bias for one or more key domains); or
-
unclear (unclear risk of bias for one or more key domains).
We planned to explore the impact of the level of bias through undertaking Sensitivity analysis.
Measures of treatment effect
We carried out statistical analysis using the Review Manager software (Review Manager 2014).
Dichotomous data
For dichotomous data, we presented results as summary risk ratio (RR) together with 95% confidence intervals (CIs).
Continuous data
For continuous data, we used the mean difference (MD) with 95% CI.
Unit of analysis issues
The statistical unit was the patient and the overdenture, not the number of implants, nor the type of implants used. In the analysis of cross‐over trials, we included data, if any, from the first period, if the duration of the period was at least 1 year.
Dealing with missing data
For included trials, we noted levels of attrition. We planned to explore the impact of including trials with high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis. For all outcomes, we carried out analyses, as far as possible, on an intention‐to‐treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and analyse all participants in the group to which they were allocated, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number randomised minus any participants whose outcomes were known to be missing.
Assessment of heterogeneity
We have assessed statistical heterogeneity in each meta‐analysis using the Tau2, I2 and Chi2 statistics. We regarded heterogeneity as substantial if I2 was greater than 30% and either Tau2 was greater than zero, or there was a low P value (less than 0.10) in the Chi2 test for heterogeneity.
Assessment of reporting biases
In future updates of this review, if there are 10 or more trials in the meta‐analysis we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.
Data synthesis
We carried out statistical analysis using the Review Manager software (Review Manager 2014). We have used fixed‐effect meta‐analysis for combining data where it is reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and we judged the trials' populations and methods sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if we detected substantial statistical heterogeneity, we explored this by sensitivity analysis followed by random‐effects if required.
Subgroup analysis and investigation of heterogeneity
Our planned subgroup analysis by the duration of follow‐up was not conducted as there were no data.
Sensitivity analysis
Our planned sensitivity analysis to explore the effect of trial quality assessed by omitting trials rated as high risk of bias and unclear when considering allocation concealment and incomplete outcome data was not conducted because the meta‐analysis of the primary outcomes included data from two trials only.
GRADE and 'Summary of findings' tables
We used the GRADE approach (Schünemann 2013) to create 'Summary of findings' tables (summary of findings Table for the main comparison; summary of findings Table 2; summary of findings Table 3).
GRADEpro 2015 was used to import data from Review Manager (RevMan 2014) in order to create the 'Summary of finding' tables. A summary of the intervention effect and a measure of quality for each of the above outcomes was produced using the GRADE approach. The GRADE approach uses five considerations (trial limitations, consistency of effect, indirectness, imprecision, and publication bias) to assess the quality of the body of evidence for each outcome. The evidence was downgraded from 'high quality' by one level for serious (or by two levels for very serious) limitations in the five mentioned considerations.
Results
Description of studies
See: Characteristics of included studies; Characteristics of excluded studies.
Results of the search
Our electronic searches identified a total of 876 references, coupled with 17 references detected by manual searches or monitoring e‐mail journal Table of Contents alerts. 421 references were assessed after duplicates were removed. Thereafter, 348 references were discarded after screening titles and abstracts leaving 73 references potentially eligible for inclusion. Assessment of full texts of these 73 references found a further 13 to be discarded as not meeting the criteria. The remaining 60 trials were scrutinized, leaving 23 (43 reports) being excluded studies, finally leaving six (17 reports) included studies (Figure 4).
Study flow diagram.
Authors of all six included studies were contacted for additional information and we received replies from five of them (Cristache 2014; Cune 2010; Kappel 2016; Naert 1999; Walton 2002).
Authors of 17 excluded studies were contacted for additional information and we received replies from 12 of them (Burns 2011; de Souza 2015; ELsyad 2012; Krennmair 2006; Krennmair 2008; Krennmair 2012; Krennmair 2012b; Kronstrom 2010; Slot 2013; Slot 2014; Walton 2009; Wismeijer 1997).
It is relevant that clinical experience suggests that prosthodontic maintenance of a mechanical nature usually take more than 1 year to manifest themselves.
Included studies
Six trials were eligible to be included (Cepa 2017; Cristache 2014; Cune 2010; Kappel 2016; Naert 1999; Walton 2002). All were on mandibular overdentures, with no trials on maxillary overdentures being eligible.
Characteristics of the trial settings and design
-
Two trials were conducted in Germany (Cepa 2017; Kappel 2016), one conducted in the Netherlands (Cune 2010), one in Belguim (Naert 1999), one in Romania (Cristache 2014), and one in Canada (Walton 2002).
-
Five trials had a parallel group study design (Cepa 2017; Cristache 2014; Kappel 2016; Naert 1999; Walton 2002).
-
One trial had a cross‐over study design (Cune 2010).
-
Five trials declared industry support (Cepa 2017; Cristache 2014; Kappel 2016; Naert 1999; Walton 2002).
-
All trials were conducted in university clinics or research centres.
Study duration ranged between 2 years (Kappel 2016) and 10 years (Cune 2010; Naert 1999).
Characteristics of the interventions
By prosthodontic design of mandibular or maxillary overdentures
-
Ball/stud or bar attachment systems for mandibular 2‐implant overdentures using splinted and unsplinted prosthodontic designs: two trials (Kappel 2016; Walton 2002).
-
Ball/stud or telescopic attachment systems for mandibular 2‐implant overdentures using an unsplinted prosthodontic design: one trial (Cepa 2017).
-
Ball/stud, magnetic or bar attachment systems for mandibular 2‐implant overdentures using splinted and unsplinted prosthodontic designs: three trials (Cristache 2014; Cune 2010; Naert 1999).
-
No included trials for maxillary implant overdentures.
By attachment system of mandibular overdentures
-
Locator® attachments (Zest Anchors, Escondido, CA, USA) or an egg‐shaped bar (Dolder®, Sub‐TecWirobond® MI bar abutment and Dolder bar; BEGO Implant GmbH & Co KG, Germany) (Kappel 2016).
-
Branemark® 2.25 mm ball patrices and titanium alloy spring matrices; Branemark® 2 mm single round gold bar joint patrices and single clip matrices (Walton 2002).
-
Ankylos® (Dentsply Implants, Mannheim, Germany) ball attachments or prefabricated conus (telescopic) attachments with connecting matrices were polymerized into spacers of the metal framework using a self‐curing resin (Pattern Resin LS, GC Europe, Leuven, Belgium) (Cepa 2017).
-
Straumann® 2.25 mm ball patrices and Dalla Bona type gold alloy matrices; titanium alloy spring matrices; Locator® (ZEST Anchors, Escondido, CA 92029 USA) stud matrices and plastic patrices; magnetic attachments (Titanmagnetics® Steco system‐technick, Hamburg, Germany) (Cristache 2014).
-
Frialit‐2 (Friadent, Dentsply, Mannheim, Germany) 2.25 mm ball patrices and Dalla Bona type metallic matrices; single bar joint matrices and metal clips (Friadent); magnetic attachments (Dyna, Bergen, the Netherlands) (Cune 2010).
-
Branemark® 3.25 mm ball patrices and O‐ring matrices; Branemark® abutments with single egg‐shaped Dolder bar joint patrices and clip matrices (Cendres et Metaux, Biel/Bienne, Switzerland); Branemark® abutments with open field magnetic attachments (Dyna, Bergen, the Netherlands) (Naert 1999).
By attachment system of maxillary overdentures
-
No included trials for maxillary implant overdentures.
Characteristics of the outcome measures
-
The primary outcome of prosthodontic success by specific categorization was recorded in two trials (Cristache 2014; Walton 2002). Unpublished data on this outcome was provided by email by authors of Kappel 2016.
-
The primary outcome of prosthodontic maintenance by general categorization was recorded in six trials (Cepa 2017; Cristache 2014; Cune 2010; Kappel 2016; Naert 1999; Walton 2002).
-
The secondary outcome of patient preference in cross‐over trials, was not recorded in the only included cross‐over trial (Cune 2010).
-
The secondary outcome of patient satisfaction assessed by a validated measure was recorded in three trials (Cune 2010; Naert 1999; Walton 2002). However, we could not use any data from these studies. Quality of life as assessed by a validated measure was not recorded in any of the included trials.
-
The secondary outcome of cost analysis (treatment time and/or material costs), was recorded in two trials (Cristache 2014; Walton 2002).
Participants
For more details, see the Characteristics of included studies table.
Inclusion criteria
For mandibular overdentures:
-
healthy participants with adequate bone height; at least 1 year of experience wearing conventional complete dentures (Naert 1999; Walton 2002); persistent dissatisfaction/problems/complaints with their conventional dentures (Cepa 2017; Cune 2010; Naert 1999; Walton 2002); agreement for a follow‐up period (Cristache 2014); willing to consent and commit to participation in the trial (Kappel 2016; Naert 1999; Walton 2002).
For maxillary overdentures:
-
no included studies for maxillary implant overdentures.
Exclusion criteria
For mandibular overdentures:
-
general contraindications for oral surgical procedures (Cepa 2017; Cristache 2014; Kappel 2016);
-
insufficient bone volume to harbour two implants with a minimum length of 10 mm, unrealistic expectations of the prosthodontic treatment outcome, gag reflex, absence of maxillary complete denture, insufficient inter‐arch space, patients in Class IV classification (McGarry 1999) or Angle Class II relationship (Cristache 2014; Naert 1999);
-
participants were not enrolled in the trial after implant surgery because the prosthodontist found that the implants diverged more than 15 degrees from each other, or that the implants were located less than 20 mm or more than 35 mm apart, because of evidence that such an orientation and location of implants could disturb the stability and maintenance of the implant overdenture (Walton 2002).
For maxillary overdentures:
-
no studies included for maxillary implant overdentures.
Sample size
The number of participants in the included studies ranged between 18 (Cune 2010) and 100 (Walton 2002).
Excluded studies
The reasons for exclusion of 23 trials were.
For mandibular overdentures:
-
either not randomised controlled trials (Akca 2013; Karabuda 2008; Krennmair 2006; Krennmair 2012b; Mericske‐Stern 2009; Payne 2000b);
-
more than one implant system was used on one implant comparing different attachment systems (Alsabeeha 2011);
-
one implant system, different numbers of implants were used (Burns 2011; Wismeijer 1997);
-
the same number of implants used, two different implant systems were used each with different attachment systems (Cehreli 2010b);
-
different implant systems were used, different numbers of implants were used with different attachment systems (de Souza 2015);
-
two subtypes of bar attachment, and not between the main attachment system types (ball/stud, magnetic, telescopic or bar attachment systems) (ELsyad 2012);
-
did not use the same number of implants in all included patients (Gotfredsen 2000);
-
two subtypes of the ball attachment, and not between the main attachment system types (ball/stud, magnetic, telescopic or bar attachment systems) (Kleis 2010; Krennmair 2012);
-
two subtypes of the bar attachment, and not between the main attachment system types (ball/stud, magnetic, telescopic or bar attachment systems) (Krennmair 2008);
-
one implant system was used, different numbers of implants were used with only one attachment system (Kronstrom 2010; Walton 2009);
-
it was on both maxillary and mandibular overdentures with one implant system, using the same number of implants, there were both titanium and zirconia implants used as well as only one attachment system used per jaw. Different attachment systems were not used in the maxilla and mandible for comparison (Osman 2014);
-
more than one implant system was used on the same number of implants comparing different attachment systems (Watson 2002).
For maxillary overdentures:
-
more than one implant system was used on the same number of implants comparing different attachment systems (Al‐Zubeidi 2012a);
-
on maxillary overdentures and one implant system was used, different numbers of implants were used with only one attachment system (Slot 2013; Slot 2014).
Further details for the reasons for their exclusion are in the Characteristics of excluded studies table.
Risk of bias in included studies
We have provided detailed descriptions of the risk of bias in the included trials in the 'Risk of bias' tables.
See Figure 5 and Figure 6 for a summary of 'Risk of bias' assessments.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Of the six included trials, only three noted adequate sequence generation (Cune 2010; Naert 1999; Walton 2002), while the other three trials did not provide any description of how the sequence was generated (Cepa 2017; Cristache 2014; Kappel 2016). All but one included trial (Kappel 2016 assessed as unclear) noted adequate allocation concealment and were at low risk of bias.
Blinding
Neither the participants nor the caregivers were blinded in the included trials due to the nature of the intervention, and since it is an operative procedure and the outcomes are not likely to be influenced by lack of blinding of participants and personnel, we considered the risk of performance bias to be low. Regarding detection bias, we assessed blinding separately for different classes of outcomes as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), and assessed the risk as low in objective outcomes (e.g. prosthodontic maintenance) and high in patient‐reported outcomes since lack of blinding can potentially introduce bias for this class of outcomes through multiple pathways (different expectations from the two groups and biased assessment of the effect).
Incomplete outcome data
For attrition bias, only two trials were at low risk (Cristache 2014; Kappel 2016), while the remaining four trials were at high risk due to reporting greater than 20% missing data (Cepa 2017; Cune 2010; Naert 1999), performing 'as treated' (Cune 2010) or 'per protocol' (Cepa 2017; Naert 1999) analyses, and presence of discrepancies between the reports of the same trial regarding the number of participants available for follow‐up and the reasons for dropouts (Walton 2002).
Selective reporting
Three of the included trials were at low risk of selective reporting of outcomes (Cepa 2017; Naert 1999; Walton 2002). The other three were at high risk either due to failure to report all the outcomes pre‐specified in their registered protocols (Cristache 2014; Kappel 2016), or due to failure to include results for key outcomes that would be expected to have been reported for such a study (Cune 2010).
Other potential sources of bias
The risk of other bias was low in four trials (Cepa 2017; Kappel 2016; Naert 1999; Walton 2002), and was high (Cristache 2014) and unclear (Cune 2010) in the other two trials.
Overall risk of bias
According to the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), overall risk of bias of all the included studies was high, due to being assessed at high risk of bias for one or more key domains.
Effects of interventions
See: Summary of findings for the main comparison Ball compared to bar attachment systems for implant overdentures in edentulous jaws; Summary of findings 2 Ball compared to telescopic attachment systems for implant overdentures in edentulous jaws; Summary of findings 3 Ball compared to magnetic attachment systems for implant overdentures in edentulous jaws
See summary of findings Table for the main comparison; summary of findings Table 2; summary of findings Table 3.
The objectives of this review were to assess different attachment systems for maxillary and mandibular implant overdentures with respect to the outcomes measures of prosthodontic success, prosthodontic maintenance, patient preference, patient satisfaction (including quality of life), and costs.
By attachment system of maxillary overdentures
No maxillary overdenture trials were identified.
By attachment system of mandibular overdentures
Ball versus bar attachment systems
Two studies (Kappel 2016; Walton 2002) compared ball and bar attachments and reported results of 3 years of follow‐up.
Prosthodontic success by specific categorization
(See Appendix 1 for categorization and Appendix 2 for six‐field protocol.)
Kappel 2016 and Walton 2002 provided data that could be used and categorized into the six‐field protocol. Upon pooling the data obtained from the two trials regarding the outcome prosthodontic success, we identified substantial heterogeneity (I2 = 97%) with inconsistency in the direction of effect, which was unexplained by clinical or methodological differences between the studies, and accordingly we did not perform meta‐analysis as this could produce misleading results (Analysis 1.1). However, re‐treatment (repair) was higher in the ball attachment group (risk ratio (RR) 3.11, 95% confidence interval (CI) 1.68 to 5.75; 2 studies; 130 participants) (Analysis 1.2), and there was no difference between both systems in re‐treatment (replace) (RR 1.18, 95% CI 0.38 to 3.71; 2 studies; 130 participants) (Analysis 1.3). It is uncertain whether there is a difference in short‐term prosthodontic success when ball attachments are compared with bar attachments.
Prosthodontic maintenance by general categorization
Although this outcome was recorded by Kappel 2016 and Walton 2002, they did not provide usable data for inclusion in the review.
Patient satisfaction /quality of life assessed by a validated measure
Only Walton 2002 assessed patient satisfaction but reported narrative results and failed to provide any usable numbers. Quality of life as assessed by a validated measure was not recorded in any of the trials.
Cost (treatment time and/or material costs)
Walton 2002 reported costs but the data could not be used due to failure to report means and standard deviations and only providing the median of the costs.
Ball versus telescopic attachment systems
One trial compared ball and telescopic attachments (Cepa 2017).
Prosthodontic success by specific categorization
(See Appendix 1 for categorization and Appendix 2 for six‐field protocol.)
Cepa 2017 did not assess this outcome.
Prosthodontic maintenance by general categorization
Among the six included studies, only Cepa 2017 provided usable data for this outcome. This trial compared ball and telescopic attachments, and after 3 years of follow‐up reported no difference between both systems in patrix replacement (RR 6.00, 95% CI 0.86 to 41.96) (Analysis 2.1), matrix activation (RR 11.00, 95% CI 0.68 to 177.72) (Analysis 2.2), matrix replacement (RR 1.75, 95% CI 0.71 to 4.31) (Analysis 2.3), or in relining of the implant overdenture (RR 2.33, 95% CI 0.81 to 6.76) (Analysis 2.4). It is uncertain whether there is a difference in short‐term prosthodontic maintenance when ball attachments are compared with telescopic attachments.
Patient satisfaction /quality of life assessed by a validated measure
Cepa 2017 assessed patient satisfaction but did not use validated measures and we could not use the data. Quality of life as assessed by a validated measure was not recorded.
Cost (treatment time and/or material costs)
Cepa 2017 did not assess this outcome.
Ball, bar or magnetic attachment systems
Three trials compared ball, bar and magnetic attachment systems (Cristache 2014; Cune 2010; Naert 1999). Cune 2010 was the only included trial with a cross‐over design. We could not use Cune 2010 results because outcomes were reported only after 3 months which is not the time frame of interest to the review, and at 10 years which were results of "as treated analysis", and accordingly could not be used.
Prosthodontic success by specific categorization
(See Appendix 1 for categorization and Appendix 2 for six‐field protocol.)
Only Cristache 2014 provided data that could be used and categorized into the six‐field protocol. Cristache 2014 compared ball and magnetic attachments and reported no difference between both systems in success (RR 0.84, 95% CI 0.64 to 1.10; 1 study; 69 participants) (Analysis 3.1) or re‐treatment (repair) (RR 1.75, 95% CI 0.65 to 4.72; 1 study; 69 participants) (Analysis 3.2) after a follow‐up of 5 years. It is uncertain whether there is a difference in medium‐term prosthodontic success when ball attachments are compared with magnet attachments.
Prosthodontic maintenance by general categorization
Although this outcome was recorded by Cristache 2014; Cune 2010; Naert 1999, they did not provide usable data for inclusion in the review.
Patient preference in cross‐over trials
Among the six included trials, only one had a cross‐over design (Cune 2010). It did not assess patient preference between the attachments used during the trial's observation period.
Patient satisfaction /quality of life assessed by a validated measure
Two trials (Cune 2010; Naert 1999) reported patient satisfaction. However, we could not use any data from these studies. Quality of life as assessed by a validated measure was not recorded in any of the trials.
Cost (treatment time and/or material costs)
Cristache 2014 compared between total aftercare and costs of prosthodontic maintenance after 5 years with ball and magnet attachments, and reported higher costs when magnets were used (mean difference (MD) ‐247.37 EUR, 95% CI ‐346.32 to ‐148.42; 1 study; 69 participants) (Analysis 3.3).
Discussion
Summary of main results
For mandibular overdentures, there is insufficient evidence to determine the relative effectiveness of different attachment systems on prosthodontic success, prosthodontic maintenance, patient satisfaction, patient preference or costs. In the short term, there is some evidence that is insufficient to show a difference and where there was no evidence was reported. It was not possible to determine any preferred attachment system for mandibular overdentures.
For maxillary overdentures, there is no evidence (with no trials identified) to determine the relative effectiveness of different attachment systems on prosthodontic success, prosthodontic maintenance, patient satisfaction, patient preference or costs.
There were six randomised controlled trials (RCTs) on mandibular overdentures (all opposing conventional complete maxillary dentures) included in this review evaluating ball/stud, magnetic, telescopic or bar attachment systems (Cepa 2017; Cristache 2014; Cune 2010; Kappel 2016; Naert 1999; Walton 2002). Two of the trials were over 2 years, one of 3 years, one over 5 years, and two over 10 years. There were no eligible trials on maxillary overdentures. Methodology of reporting the primary and secondary outcomes varied widely, but possibly less so in two trials (Cristache 2014; Walton 2002).
With regard to reporting the primary outcome of prosthodontic success by specific categorization (Appendix 1; Appendix 2), there is some evidence that there is no difference between ball and magnetic attachments and specifically re‐treatment (repair) up to 5 years. However, after 5 years, prosthodontic maintenance costs were higher when magnet attachments were used (Cristache 2014). In addition, with ball and bar attachments up to 3 years there appears to be no difference between both systems in re‐treatment (replace), however when looking at re‐treatment (repair) there is evidence that this is higher in the ball attachment group, which resulted in bar attachments having a higher prosthodontic success (Kappel 2016; Walton 2002). It is uncertain whether there is a difference in short‐term prosthodontic success when ball attachments are compared with bar attachments as well as when ball attachments are compared with magnet attachments.
Although prosthodontic maintenance by general categorization was evaluated in all six trials, there was also considerable variation in the way in which the prosthodontic maintenance or technical complications, adjustments/repairs or aftercare of the attachment systems were documented. Related to this prosthodontic maintenance by general categorization, there is some evidence that there is no difference between ball and telescopic attachments (Cepa 2017). It is uncertain whether there is a difference in short‐term prosthodontic maintenance when ball attachments are compared with telescopic attachments.
Related to patient preference between attachment systems in cross‐over trials, as well as patient satisfaction and quality of life using a validated measure, there is limited evidence from the single cross‐over trial (Cune 2010), with no possibility of performing any data analysis on different sets of data. This could have been possible if there had been more than one cross‐over trial.
Patient satisfaction measured in three of the six trials was with different validated measures (Cune 2010; Naert 1999; Walton 2002) and cost analysis (treatment time and/or material costs), measured in two trials (Cristache 2014; Walton 2002) was done in different manners. Although improved patient satisfaction is undoubtedly predictable regardless of the attachment system used for mandibular overdentures, a preferred attachment system related to prosthodontic success, prosthodontic maintenance, patient preference and costs could not be identified.
Related to cost (treatment time and/or material costs), there is some evidence that there are higher costs when using magnetic attachments, as opposed to ball attachments (Cristache 2014).
Overall completeness and applicability of evidence
The trials identified are not sufficient to address the review question. Although the participants and interventions investigated were of direct relevance to our review question, the pooled included trials did not provide a sufficient number of participants and did not report on the attachment systems with maxillary overdentures. In addition, the studies did not report on long‐term outcomes, and failed to include results for important key outcomes in usable numerical form to allow for meta‐analysis. Therefore, there can be no consensus on the attachment system to be used with maxillary or mandibular implant overdentures based on this review findings.
There are distinct difficulties in conducting randomised clinical trials in prosthodontics and especially determining true economic aspects of hardware maintenance longitudinally. This is of utmost relevance in older studies given the attachment systems included in those studies are no longer commercially available, except in some instances by custom fabrication by the commercial companies at a high cost to the clinician. In addition, although some of the older studies have a relevance, they could not be included as they were conducted not in a randomised clinical trial format, but rather in a prospective study format which was the usual study design at that time.
Quality of the evidence
The body of evidence identified do not allow for a robust conclusion regarding the objectives of the review. The evidence was of very low quality for all pre‐specified outcomes (summary of findings Table for the main comparison; summary of findings Table 2; summary of findings Table 3). In all the included trials there were relatively few participants and few events, and most of the reported outcomes had wide confidence interval (CI) around the estimate of the effect that overlaps no effect and includes both appreciable benefit and appreciable harm. In addition, there were serious limitations in the trial design regarding incomplete outcome data and selective reporting of the outcomes. We evaluated the quality of evidence by using the GRADE approach. The evidence was of very low quality, we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect (Schünemann 2013).
Potential biases in the review process
The review authors followed the guidelines for conducting this systematic review under the strictest of conditions (Higgins 2011). We performed comprehensive searches to identify eligible trials, all abstracts were independently dual screened, and all references were assessed and had the risk of bias assessment carried out by at least two independent authors. All trials were subsequently reviewed and agreed by three of the review authors. Some corresponding authors did not reply to our requests of random sequence generation, allocation concealment, blinding of outcome assessment (detection bias), or additional primary outcome data on their trials and were therefore excluded from the analysis.
As the clinicians provided the evaluations related to the outcomes in the studies, there was a possibility of reporting bias and under‐reporting.
Agreements and disagreements with other studies or reviews
The present review included all the RCTs available to date. There have been five other related systematic reviews (Anas El‐Wegoud 2018; Assaf 2017; Cehreli 2010a; Kim 2012; Leão 2018).
Two of these were just on prosthodontic maintenance and complications of implant overdentures (Andreiotelli 2010; Cehreli 2010a) and both of these also included prospective studies as well as RCTs with a variation of 18 to 49 selected studies between them. Therefore direct comparisons with our findings could be misleading and difficult to interpret in these reviews. In addition, Kim 2012 also included prospective studies as well as RCTs up to August 2010, with duplication of the same studies at different time points in the 24 included studies. Implant survival was inaccurately attempted to be assessed simultaneously and lead to ambiguity. There were conflicting findings on prosthodontic maintenance between ball (stud) and bar attachments, and patient satisfaction was stated as "appearing" to be independent of attachment system. No meta‐analysis was performed.
The systematic review by Leão 2018 attempted to determine the influence of splinted and unsplinted overdenture attachment systems on prosthodontic maintenance and marginal bone loss and implant survival rate. Nine studies were included in the qualitative and quantitative analyses. A total of 984 implants were placed in 380 patients (mean age: 62.8 years). Splinted and unsplinted overdenture attachment systems achieved similar results with regard to prosthodontic maintenance. The meta‐analysis demonstrated no statistically significant differences between splinted and unsplinted attachment systems with regard to marginal bone loss (P = 0.39; mean difference (MD) ‐0.11, 95% confidence interval (CI) ‐0.37 to 0.14), complications (P = 0.31; risk ratio (RR) 1.26, 95% CI 0.80 to 1.99), and implant survival rate (P = 0.14; RR 0.37, 95% CI 0.10 to 1.36).
Similarly, Assaf 2017 from a total of 130 articles, found 33 studies that met the specified inclusion criteria for the review (14 RCTs, eight prospective clinical trials, three retrospective studies, and four systematic reviews). It was deduced that these articles provided evidence that a mean complication rate of prosthodontic maintenance was impossible to determine because of the multiplicity of contributing factors. No clear identification of the causes of prosthodontic maintenance were found, nor was there any clear evidence of superiority of any implant system or attachment design over another or both. It was concluded that prosthodontic maintenance with implant overdentures is inevitable. Further clinical studies were encouraged to achieve a constructive meta‐analysis that accounts for different parameters such as opposite arch, attachment functional variety, connection method, and prosthesis quality.
A more robust review was published recently by Anas El‐Wegoud 2018 who concluded that there was insufficient evidence to support bar or ball attachments being used with implant overdentures in edentulous patients to improve patient satisfaction and prosthesis retention. They included 10 trials (465 participants). After 5 years, one trial reported higher patient satisfaction when bar attachment was used (MD 1.30, 95% CI 0.20 to 2.40), and reported no difference between both systems in overdenture retention (MD ‐0.90, 95% CI ‐1.90 to 0.10). Two trials in Anas El‐Wegoud 2018 reported no implant failures after 1 and 5 years in both attachments and one of these (Naert 1999) was amongst the six included trials for this Cochrane Review. Downgrading of evidence was based on the unclear risk of bias of included studies and the wide confidence interval crossing the line of no effect.
Ball stud attachment systems.
Copyright© 2018 Payne A, Zarb G. Implant overdentures. In: Zarb G, Hobkirk J, Eckert S, Jacob R, editor(s). Prosthodontic Treatment for Edentulous Patients: Complete Dentures and Implant‐Supported Prostheses. 13th edition. St Louis, Missouri, USA: Mosby, 2013:330‐9 (Payne 2013): reproduced with permission.
Bar attachment systems.
Copyright© 2018 Payne A, Zarb G. Implant overdentures. In: Zarb G, Hobkirk J, Eckert S, Jacob R, editor(s). Prosthodontic Treatment for Edentulous Patients: Complete Dentures and Implant‐Supported Prostheses. 13th edition. St Louis, Missouri, USA: Mosby, 2013:330‐9 (Payne 2013): reproduced with permission.
Magnet telescopic attachment systems.
Copyright© 2018 Professor Ralf Kohal, Germany; Professor Yoshinobu Maeda, Japan: reproduced with permission.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 Ball versus bar attachment systems, Outcome 1 Success (short term; 1 to 3 years).
Comparison 1 Ball versus bar attachment systems, Outcome 2 Re‐treatment (repair) (short term; 1 to 3 years).
Comparison 1 Ball versus bar attachment systems, Outcome 3 Re‐treatment (replace) (short term; 1 to 3 years).
Comparison 2 Ball versus telescopic attachment systems, Outcome 1 Patrix replaced (short term; 1 to 3 years).
Comparison 2 Ball versus telescopic attachment systems, Outcome 2 Matrix activated (short term; 1 to 3 years).
Comparison 2 Ball versus telescopic attachment systems, Outcome 3 Matrix replaced (short term; 1 to 3 years).
Comparison 2 Ball versus telescopic attachment systems, Outcome 4 Reline of implant overdenture (short term; 1 to 3 years).
Comparison 3 Ball versus magnetic attachment systems, Outcome 1 Success (medium term; 3 to 5 years).
Comparison 3 Ball versus magnetic attachment systems, Outcome 2 Re‐treatment (repair) (medium term; 3 to 5 years).
Comparison 3 Ball versus magnetic attachment systems, Outcome 3 Costs (medium term; 3 to 5 years).
| Interventions for replacing missing teeth: attachment systems for implant overdentures in edentulous jaws | ||||||
| Patient or population: edentulous adults receiving implant overdentures in one or both jaws to overcome problems with conventional complete dentures | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | Number of participants | Certainty of the evidence | Comments | |
| Risk with bar attachment system | Risk with ball attachment system | |||||
| Success (short term) | Study population | ‐ | 130 | ⊕⊝⊝⊝ | Considerable heterogeneity (I2 = 97%). Pooling of data was not done | |
| See comment | See comment | |||||
| Re‐treatment (repair) (short term) | Study population | RR 3.11 | 130 | ⊕⊝⊝⊝ | ||
| 141 per 1000 | 437 per 1000 | |||||
| Re‐treatment (replace) (short term) | Study population | RR 1.18 | 130 | ⊕⊝⊝⊝ | ||
| 78 per 1000 | 92 per 1000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Risk of bias (serious), inconsistency (very serious), imprecision (very serious); downgraded by 3 levels. | ||||||
| Interventions for replacing missing teeth: attachment systems for implant overdentures in edentulous jaws | ||||||
| Patient or population: edentulous adults receiving implant overdentures in one or both jaws to overcome problems with conventional complete dentures | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | Number of participants | Certainty of the evidence | Comments | |
| Risk with telescopic attachment system | Risk with ball attachment system | |||||
| Patrix replaced (short term) | Study population | RR 6.00 | 22 | ⊕⊝⊝⊝ | ||
| 91 per 1000 | 545 per 1000 | |||||
| Matrix activated (short term) | Study population | RR 11.00 | 22 | ⊕⊝⊝⊝ | ||
| 0 per 1000 | 0 per 1000 | |||||
| Matrix replaced (short term) | Study population | RR 1.75 | 22 | ⊕⊝⊝⊝ | ||
| 364 per 1000 | 636 per 1000 | |||||
| Reline of implant overdenture (short term) | Study population | RR 2.33 | 22 | ⊕⊝⊝⊝ | ||
| 273 per 1000 | 635 per 1000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Risk of bias (serious), imprecision (very serious); downgraded by 3 levels. | ||||||
| Interventions for replacing missing teeth: attachment systems for implant overdentures in edentulous jaws | ||||||
| Patient or population: edentulous adults receiving implant overdentures in one or both jaws to overcome problems with conventional complete dentures | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | Number of participants | Certainty of the evidence | Comments | |
| Risk with magnetic attachment system | Risk with ball attachment system | |||||
| Success (medium term) | Study population | RR 0.84 | 69 | ⊕⊝⊝⊝ | ||
| 826 per 1000 | 694 per 1000 | |||||
| Re‐treatment (repair) (medium term) | Study population | RR 1.75 | 69 | ⊕⊝⊝⊝ | ||
| 174 per 1000 | 304 per 1000 | |||||
| Costs (medium term) | The mean costs (medium term) was EUR 2286.34 | MD 247.37 EUR lower | ‐ | 69 | ⊕⊝⊝⊝ | |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Risk of bias (serious), imprecision (very serious); downgraded by 3 levels. | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Success (short term; 1 to 3 years) Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 2 Re‐treatment (repair) (short term; 1 to 3 years) Show forest plot | 2 | 130 | Risk Ratio (M‐H, Fixed, 95% CI) | 3.11 [1.68, 5.75] |
| 3 Re‐treatment (replace) (short term; 1 to 3 years) Show forest plot | 2 | 130 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.18 [0.38, 3.71] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Patrix replaced (short term; 1 to 3 years) Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 2 Matrix activated (short term; 1 to 3 years) Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 3 Matrix replaced (short term; 1 to 3 years) Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 4 Reline of implant overdenture (short term; 1 to 3 years) Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Success (medium term; 3 to 5 years) Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 2 Re‐treatment (repair) (medium term; 3 to 5 years) Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 3 Costs (medium term; 3 to 5 years) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
