Original study - ZZI 04/2011
Implant-retained prostheses: ball vs. conus attachments – A randomized controlled clinical trial
S. Cepa1, B. Koller1, M. Wolkewitz2, R. Kohal1
Purpose: The aim of the present study was to evaluate implant survival, peri-implant soft and hard tissue conditions, prosthodontic maintenance and patient satisfaction for implant-supported mandibular overdentures retained with ball attachments or prefabricated conical crown attachments during a 2-year period.
Materials and methods: 25 patients with edentulous mandibles received two oral implants in the mandibular interforaminal region. The denture attachment system was chosen randomly. Twelve patients received ball attachments and 13 patients received prefabricated conical crown attachments. Implant survival, peri-implant soft and hard tissue conditions such as probing depth (PD), modified plaque index (mPlI), bleeding on probing (BOP), modified gingival index (mGI), horizontal bone loss (HBL) and patient satisfaction were assessed for each patient. In addition, detailed prosthodontic maintenance interventions were monitored during the follow-up period.
Results: The implant survival rate was 100 %. There were no significant differences in PD, mPlI, BOP, mGI and HBL between the two groups. Within the two-year evaluation period prosthodontic intervention was required for 80 % of the patients in the ball attachment group and 75 % of the patients in the conical crown group. After two years, patient satisfaction in the ball group (82 %) was higher than in the conical crown group (60 %).
Conclusion: The results indicate that both attachments (ball attachment, conical crown attachment) used on two implants in the edentulous mandible show similar clinical results. However, the maintenance of both attachment groups is high. Due to the need for extensive maintenance, regular recall is fundamental for both attachment systems.
Keywords: implant-retained mandibular overdentures; SynCone; ball attachments
The management of patients with edentulous mandibles by means of implant-retained hybrid prostheses, that is, combined fixed and removable overdentures, is a simple, practicable and cost-effective solution . Use of implants in the edentulous mandible improves the stability of the prosthesis and leads to an improvement in patient satisfaction compared with conventional total overdentures [3, 5, 10, 12, 17, 25, 44].
For these reasons, according to the McGill conference (2002), management with a hybrid overdenture retained on two implants should be the standard treatment of the edentulous mandible [13, 39]. This type of treatment is financially much more economical for the patient because of the very simple dental laboratory work than, for example, a fixed implant-retained restoration .
Implant-retained mandibular overdentures can be attached by different retaining elements such as ball attachments, locators, magnets, telescopic crowns or bars [18, 31, 32, 40, 41]. As regards maintenance, differences between splinted (bars) and non-splinted retention elements (ball heads, locator attachments etc.) have been described [6, 14, 34, 40]. However, the choice of retention element has no influence on implant survival rates [7, 11, 31, 38, 41]. In the edentulous mandible, these are 86 % to 100 % after five and ten years respectively [14, 23, 24, 26, 31].
The aim of this randomized prospective clinical study was to investigate the influence of two different retention elements (ball head and prefabricated conical crowns [SynCone]) with two interforaminal implants on different clinical parameters and horizontal bone loss. In addition, the study investigated which attachment design requires greater prosthodontic maintenance and which leads to greater patient satisfaction.
The null hypothesis states that there is no difference between SynCone attachments and ball head attachments with two interforaminal implants with regard to the aforementioned parameters.
Materials and methods
After approval of the study by Freiburg Regional Council and approval by the Ethics Committee of Freiburg University Hospital (decision number 105/04) and after obtaining written informed consent from the patients, 25 patients were treated with implant-retained mandibular prostheses in the Department of Dental Prosthetics of Freiburg University Clinic in a randomized clinical study. Either prefabricated conical crowns (ANKYLOS SynCone, DENTSPLY Friadent, Mannheim, Germany) or ball head attachments (ANKYLOS ball heads, DENTSPLY Friadent, Mannheim, Germany) were used for prosthesis retention.
The patients for this study were selected from healthy adult individuals who wanted an implant-retained restoration for their edentulous mandible. The 25 patients recruited for the study were dissatisfied with their previous full mandibular overdenture and wanted a better-fitting prosthesis. The exclusion criteria included: infectious or metabolic diseases, cardiovascular disease, pregnancy or planned pregnancy within the following twelve months, lactation, local contraindications (e. g., tumors, ulcers) to oral surgical procedures, alcohol abuse or smoking, psychological disease, non-compliance and intolerance of pre- or postoperative medication in the presence of chronic disease. All forms of opposing dentition were accepted provided these were sufficient.
Four interforaminal implants were inserted according to the manufacturer’s recommendations by one clinician (RJK) in the mandibles of 25 edentulous patients. Under local anesthesia, a mucoperiosteal flap was dissected from the region of the second left premolar to the second right premolar and the mental foramina were exposed. Two conventional implants (ANKYLOS, DENTSPLY Friadent, Mannheim, Germany) were placed in the canine region of the mandible without using a drilling template. The implants were selected according to the available bone and had a diameter of 3.5 or 4.5 mm and a length of 14 mm. For test purposes, two further smaller implants (ANKYLOS, DENTSPLY Friadent, Mannheim, Germany) were inserted between these two implants in positions 31 and 41 in artificially created bone defects. These had a diameter of 3.5 mm and a length of 8 mm. The two middle experimental implants were removed at a later time using a trephine drill to investigate two bone substitutes in another study. After relieving slits in the periosteum, the mucoperiosteal flaps were approximated tension-free with interrupted and continuous sutures. The sutures were removed ten days after implant insertion. Following suture removal, the existing lower dentures were adjusted to the new situation with a soft relining material (Soft-Liner, GC Europe, Leuven, Belgium).
After a healing period of at least two months, the implants were exposed and provided with healing abutments. About four weeks after implant exposure, an open impression of the implants was taken with a custom tray, transfer posts (DENTSPLY Friadent, Mannheim, Germany) and a polyether impression material (Impregum Penta, 3M Espe, Neuss). Fabrication of working models and registration templates was followed by facebow transfer and centric bite registration. A wax-up of the teeth (Vita Physiodens, VITA Zahnfabrik, Bad Säckingen) was then tried in the mouth. After checking fit, function and appearance, the metal frameworks of the prostheses were constructed in the next step. The prostheses were then transferred to acrylic (Aesthetic Autopolymerisat, Candulor AG, Wangen, CH). After insertion of the ball head abutments (Ankylos Ball Attachment, DENTSPLY Friadent, Mannheim) (Fig. 1) or the prefabricated conical crowns (SynCone Abutments 4°, DENTSPLY Friadent, Mannheim) (Fig. 2) in the implants, the matrices were polymerized into the metal framework spacers with a self-curing resin (Pattern Resin LS, GC Europe, Leuven, Belgium) after prior Rocatec treatment in the patient directly.
The patients were randomized to the two groups. Randomization was carried out by a statistician not involved in the study using a sealed envelope. The envelope with information about the abutments to be employed was opened by a dental nurse on the day of final incorporation. Blinding of the investigators was not possible because the prosthetic reconstruction was visible.
The patients were given precise oral hygiene instructions after the prostheses were fitted.
The patients were investigated with regard to different peri-implant soft tissue parameters on the day of prosthesis incorporation and after twelve and 24 months. The probing depth (PD) in mm, the Mombelli modified plaque index mPlI (grade 0?III; grade 0 = no plaque, grade 1 = non-visible thin film of plaque, which can be detected by scraping the tooth surface with a probe, grade 2 = visible plaque, grade 3 = massive plaque deposits which fill the interdental space) , the bleeding index BOP (bleeding on probing, 0 = no bleeding, 1 = bleeding)  and the Mombelli modified gingiva index mGI (grade 0–III; grade 0 = no inflammation, grade 1 = slight change in color and surface, grade 2 = erythema, edema, gingival bleeding on probing, grade 3 = marked erythema and edema, spontaneous gingival bleeding, ulceration)  were recorded. The probing depth and bleeding index were measured in mesial, distal, vestibular and lingual positions. The means of these four measurements were obtained and used for statistical analysis.
To evaluate horizontal bone loss (HBL) of the implants, orthopantomographs were taken after implantation, when the prosthetic restoration was fitted (baseline) and after twelve and 24 months. The marginal bone level was measured mesial and distal to the implants. The upper edge of the implants was used as reference point for the marginal bone measurement. The marginal bone levels were recorded for the baseline and 1-year and 2-year follow-up. Using the real and digital implant length, the real distance of the implant shoulder from the first bone-implant contact was calculated using a rule of three. This was then compared with the measurements at the previous examinations (baseline) and the real difference in mm was calculated. The difference between the two values then gives the bone loss. The digital orthopantomographs were measured using computer software (measurement module, Dental Vision, Computer Forum GmbH, Elmshorn, Germany) on a 17-inch monitor (Neovo DR-17 model, Dürr Dental AG, Bietigheim-Bissingen, Germany). The monitor resolution was 1280 x 1024 pixel. Magnification was adjusted to provide optimal measurement conditions.
The patients were asked about their satisfaction with the new implant-retained mandibular restoration at the different annual follow-ups. A distinction was made simply between satisfied and dissatisfied. Satisfaction was not divided into different sub-variables. Any necessary prosthodontic interventions were also performed and documented.
A linear mixed model with a random effect was adapted (“subject = patient”). For each soft tissue parameter and horizontal bone loss parameter, the continuous “response” variable was modeled as a linear function of the abutment type and time
(0 years, 1 year and 2 years). In addition, the corresponding time-type interaction was modeled as an explanatory variable. Variance components were used as covariant structures. Paired differences of the means (least square means) were calculated. All calculations were performed with the statistical software SAS System Version 9.1 using the “PROC MIXED” procedure. The statistical significance level was set at p ? 0.05. Subgroup analysis was performed according to Tukey-Kramer with adjusted p values.
Of the 25 patients recruited for the study, 25 underwent surgical and prosthetic treatment. In the ball head abutment group, twelve patients were fitted with overdentures. These were aged 65.2 ± 7.02 years at the time the overdentures were fitted. These twelve patients were followed up after one year. Eleven participants were followed up after two years because of one death.
13 participants were randomized to the prefabricated conical crown group. These were aged 62.9 ± 8.98 years at the time the overdentures were fitted. Twelve of these were followed up after one year. One study participant had died. Eleven patients were followed up after two years. One participant left the study as the prefabricated conical crowns were exchanged for ball heads on account of poor denture fit (Fig. 3).
Implant survival rates
No implant was lost in either group within 24 months. The implant survival rate after 24 months was therefore 100 %.
No significant differences were found between the two abutment groups for the parameters probing depth, BOP, modified gingival index, mesial HBL and distal HBL at the different examination times (p > 0.05) (Table 1–6, Fig. 4–9). The changes over time in the two groups showed no significant differences. The mean modified plaque index (mPlI) demonstrated a significant difference only at baseline (t = 0) with the ball heads (0.4 ± 0.3) and with the prefabricated conical crowns (1.2 ± 0.3) (p = 0.0322).
Prosthodontic intervention was required 13 times in the ball head group up to the one-year follow-up. In eleven cases, retention was too weak and was therefore improved by activation of the matrices in seven patients and exchange of the matrices in three patients. Three overdentures had to be relined.
In the SynCone group, prosthodontic maintenance treatment was required on 14 occasions. The conical crown retention was too weak in four cases. In one of these cases, new matrices were polymerized into the prosthesis. The other three patients did not want any subsequent improvement. In seven cases, the retention was too strong. In one of these cases, the prefabricated conical crowns were exchanged for ball heads. This patient was therefore withdrawn from the study. In the other cases, the overdentures were sent to the dental laboratory for improvement. The retention was reduced there by carefully rubberizing the secondary crowns. In two cases, the overdentures had to be relined. One abutment (Fig. 10) fractured. The broken abutment had to be drilled out of the implant so that it could be replaced by a new conical crown.
Further treatment was required on eight occasions in each of the study groups up to the two-year follow-up.
Loss of retention was found three times in the ball head group. To improve this, the matrices were activated in one case and in two cases the matrices were exchanged for new ones. A need for relining was found in five cases.
In the SynCone group, the retention was too weak in five cases at the two-year follow-up. One of these patients did not want any prosthodontic measures to improve denture fit. In one patient, new SynCone matrices were polymerized and in three patients the abutments were exchanged for ball heads. These patients were withdrawn from the study from this time and treated as drop-outs. In one patient, prosthetic retention was very strong. However, this patient did not want any corrective measures either. In two patients, the overdenture had to be relined.
If prosthodontic maintenance is regarded as failure, this results in the Kaplan-Meier graph seen in Figure 11. After
24 months, only 20 % of the restorations attached with ball heads are in situ unchanged, and the figure is 25 % for overdentures attached with prefabricated conical heads.
All study participants were dissatisfied with their full denture prior to the start of the study. After one year, the absolute satisfaction in the ball head group was 83 % (10 of 12 patients). Satisfaction was 82 % in the prefabricated conical crown group (9 of 11 patients, 1 patient did not respond). After two years, 82 % of the patients (9 of 11 patients) in the ball head group were satisfied. In the prefabricated conical crown group, satisfaction fell to 60 % (6 of 10 patients, 1 patient did not respond) (Table 7, Fig. 12).
This randomized clinical study investigated various clinical parameters, prosthodontic maintenance and patient satisfaction over a two-year period with the attachment of mandibular overdentures retained on two implants with classical ball heads compared with prefabricated conical crowns.
The round heads employed in our study have now been described by numerous authors as sufficient retention elements for mandibular overdentures [6, 7, 10, 14, 20, 40].
Double crowns have hitherto been used mainly as dental retention elements for the restoration of partially dentate jaws using telescopic overdentures. This type of restoration has been well investigated [9, 27, 33, 35, 42]. The survival rate of tooth-borne telescopic overdentures is between 90 % and 95.1 % after four and 5.3 years respectively . Resilient telescopic crowns have been described in clinical studies as a promising treatment option for implant-retained overdentures [15, 20]. One study showed lower prosthodontic maintenance with resilient telescopic crowns compared with ball heads . This contrasts with the results of the present study in which the prosthodontic maintenance was roughly equal with both types of reconstruction. This may be attributable to greater sensitivity to technique with the non-resilient conical crowns used in this study compared with resilient telescopic crowns.
The SynCone conical crowns in the present study have previously been described as retention elements for the immediate restoration of edentulous jaws with four implants [8, 21]. The advantages compared with directly splinted retaining elements such as bar constructions are the improved oral hygiene capability and the less elaborate dental laboratory work as the matrices can be polymerized directly in the patient [8, 21, 43].
Promising long-term results were apparent in an in-vitro study of SynCone abutments. Over a simulated period of five years, the adhesive force of the SynCone abutments was stable . This in-vitro study contrasts with the present clinical study, where adjusting the retention of the prefabricated conical crowns proved problematic. Loss of denture retention occurred frequently so that new matrices had to be polymerized in. Excessive denture retention was also observed often. This also required further work, which in turn contributed to the prosthodontic maintenance.
When the prosthodontic maintenance for the two investigated abutments is considered, it is apparent that numerous corrections were necessary with both the ball heads and prefabricated conical crowns. With the ball heads, the loss of denture retention was the most frequently encountered problem. Retention had to be increased by the dentist by activation of the matrices. If this was not possible, the matrices had to be exchanged. The problem of loss of retention when ball heads are employed has also been described by other authors [4, 14, 18, 32]. So far, there are no clinical studies for SynCone abutments used with two implants in the mandible for retention of an overdenture.
The need for relining was greater in the ball head group in the second year of the study than in the first and was generally higher than in the prefabricated conical head group. The manner of construction of the ball heads can result in greater loading of the distal denture bearing area compared with the conical crowns, which might explain the necessity of relining.
Retention problems occurred with the prefabricated conical crowns. In the first year of the study, too weak and too strong retention were both found. Appropriate adjustment of retention was difficult as it was not possible because of the construction to create greater or weaker retention by activation or deactivation of the matrices. In the second follow-up period, the number of subjects with prefabricated conical crowns whose overdenture retention was too weak increased markedly. It was therefore decided to exchange the abutments for ball heads in some of the patients and withdraw the patients from the study.
In the first year of follow-up, one abutment fracture occurred in the prefabricated conical crown group. The possible reasons for this might be excessively strong retention and associated excessive force when removing the overdenture or highly tilting forces due to incongruence between the denture base and the denture bearing area. It should be noted that the risk of fracture is greater with rigid attachment elements like such conical crowns than with resilient retention elements or those that can swivel around a point or axis. In patients who generate high forces, e. g., due to bruxism, SynCone abutments should therefore not be used with mandibular overdentures attached to only two implants on account of the high bending load and the long level arm.
The null hypothesis could not be rejected for the peri-implant soft tissue parameters of probing depth (PD), BOP, modified plaque index (mPlI), gingival index (mGI), the horizontal bone loss (HBL) and prosthodontic measurement. No significant differences were found between ball heads and prefabricated conical crowns for the various parameters. These clinical results are in agreement with those of other studies, in which no differences in peri-implant parameters were found between the different attachment systems [6, 14, 26, 30, 34].
With the prefabricated conical crowns, the rates of bone loss after 24 months were 0.4 mm distally and 1.0 mm mesially. It is striking that the mesial horizontal bone loss is greater than distal horizontal bone loss with both investigated abutment types. The background to the greater mesial bone loss is that the small test implants between the study implants were removed at the exposure operation. For this, the soft tissue had to be mobilized more widely in the mesial region of the study implants than in the distal region. This might have caused the greater mesial bone loss.
Patient satisfaction increased clearly due to reconstruction with a mandibular overdenture retained on two implants. This result is in line with other studies [2, 3, 5, 10, 12, 17, 25, 44]. In addition, it was found that more patients were satisfied with reconstruction by ball heads than when prefabricated conical crowns were used. This might be attributable to difficulties in adjusting retention and the associated extra effort required with the prefabricated conical crowns.
However, a criticism of this is that patient satisfaction was surveyed by a single global question. No sub-variables were investigated. Batteries of questions consisting of several individual questions  in combination with rating (Likert) scales or visual analogue scales (VAS) would have been useful (VAS) . This would have allowed differentiation between different aspects of satisfaction .
Both of the types of abutment studied proved to be retention elements for economical restoration of the edentulous mandible with an implant-retained overdenture that require intensive maintenance. While other studies [8, 21] demonstrated good results in the restoration of edentulous mandibles with four SynCone abutments, restoration with two SynCone abutments for retention of a mandibular overdenture offers no advantages compared with ball heads and is therefore not recommended. Regarding the present study, it must be commented that the small size of the two study groups can be regarded critically with regard to the interpretation of the results.
Approvals: this study was approved by Freiburg Regional Council. This study was approved by the Ethics Committee of Freiburg University Hospital (decision number 105/04).
Conflict of interests: This study was supported financially by DENTSPLY Friadent (contract ZVS 20050426).
Dr. Sandy Cepa
Abteilung für Zahnärztliche Prothetik
Universitätsklinikum für Zahn-, Mund- und Kieferheilkunde
Klinikum der Albert-Ludwigs-Universität Freiburg
Hugstetter Straße 55
79106 Freiburg i. Br.
1 Department of Dental Prosthetics, University Oral and Dental Medicine Clinic, Albert Ludwig University Freiburg Clinic
2 Medical Biometrics and Medical Informatics Institute, Albert Ludwig University Freiburg