Success Rate of the EVL Evolution Implants (SERF): A Five-Year Longitudinal Multicenter Study

The EVL evolution implant was designed to answer the 1-stage surgical technique, while also allowing a submerged approach. The transmucosal part of this implant includes a retrievable ring, which is mirror-polished and premounted at fabrication (Figure 1). At the time of prosthetic loading, this part can be either removed or preserved. This concept allows a 1-stage soft tissue healing, and an esthetic prosthetic connection at loading time as well (Figure 2). The purpose of this study was to evaluate the success rate of this implant over a long time period1 for patient indications commonly encountered in private practice.

This is a multicentric longitudinal study conducted on a population of patients treated with 1 or several EVL evolution implants (SERF, Décines, France).2^,3 These implants are grade 2 CP titanium, according to the ISO 5832–2 standards. They include a mirror-polished surface in the transmucosal section. Their endosseous part is sandblasted with highly pure aluminum particles. The EVL evolution system offers 3 different diameters (3.3, 4, and 5 mm), and 6 lengths (7, 8, 10, 12, 14, and 16 mm).

The studied population included patients treated between January 1, 1996 and December 31, 2000 by 3 implantologists familiar with the EVL evolution system. The 3 private practices are located in 3 geographically distinct cities (Grenoble, Nice, and Paris). In the Parisian clinic, only the patients who received a full treatment (implant + prosthesis) were included in the study.

The sample included 413 patients: 222 women and 191 men. They were treated with a total number of 1198 EVL evolution implants; 492, 144, and 562 implants were respectively placed in Grenoble, Nice, and Paris in respectively, 175, 51, and 187 patients. The mean number of implants per patient was 2.9.

The EVL evolution implants are indicated for a 1-stage surgical approach. Implant selection was performed according to the preimplant clinical and radiographic examinations. The surgical technique respected the conventional imperatives, and used the ancillary material provided by the manufacturer, which includes a specific disposable final drill. Implants were inserted under clean but not sterile conditions as defined by Sharf and Tarnow.4 

For each implant, patient sex, implant placement and loading times, implant dimensions (length and diameter), and implant location were recorded. An implant was considered as anterior when placed between teeth 14 and 24 (or 34 and 44), and inversely as posterior. This terminology does not take into account the prosthetic functional criteria but is based on anatomical criteria; the regions are delimited by a vertical line drawn at the mental foramina. For each implant site, bone quality and quantity were assessed according to the Lekholm and Zarb (1985) classification, which is directly related to the NFS91–154 standard. Bone quantity was graded from A to E, and bone quality from 1 to 4; these data were recorded for each site at the time of implant placement.

A total of 497 implants were placed in the maxilla and 701 in the mandible. Their distribution on the arches is illustrated in Figure 3. Their dimensions are indicated in relation to their location in Table 1. The number of 4-mm diameter implants was significantly higher in the anterior region (P < .001). At the time of placement, the available bone quantity and quality were variable (Tables 2 and 3). The type B ridges were most frequently encountered in both sexes. The percentage of type A ridges was higher in women than in men, while the type C ridges were more frequent in men (P < .001; Table 2). Also, independent of the location, type B ridges were more frequent. However, they were proportionally more frequent in the upper maxilla. Types A and C had about the same frequency in the maxilla, while in the mandible, type A ridges were more frequent than type C (P < .001). In the anterior region, type C was more frequent than type A, unlike the posterior region (P < .001; Table 3).

The criterion for treatment evaluation or success was a qualitative variable related to 4 possible treatment outcomes, in reference to the 4-entry table described by Albrektsson and Zarb:

• Success: no clinical implant mobility. No radiographic peri-implant translucency. After 1 year of functional loading, the vertical bone loss was less than 0.2 mm per year. No neuropathic sign, pain, or infection symptoms were observed.

• Failure: the implant was not functional or was removed.

• Ailing: the implant was still in the patient's mouth, but could not be classified as a success or as a failure.

• Lost (dropout patients): the situation of the implant was unknown because the patient did not come for follow-up visits.

A chi-square test was performed to analyze implant site bone quantity and quality at the time of implant placement, according to sex and location, and at the end of the clinical follow-up, according to success. This test also allowed evaluation of outcome in relation to the time period following placement, to sex and to location.⁵

A survival curve was established over 5 years according to the Kaplan Meyer method. It took into account the ailing cases, given that these implants were still in the patient's mouth, and therefore potentially at risk (Figure 4).

The statistical analyses were performed using the SPSS version 11.5 (SPSS Inc, Chicago, Ill).

The clinical follow-up was 3.1 ± 1.2 years, ie, 1 to 6 years. At the end of this follow-up period, 1163 implants were classified as successful, 19 as failure, 12 as ailing, and 4 as lost (dropout patients). If the 4 lost implants are not taken into account in the chi-square tests, the success rate was independent from the time period following implant placement (P = .220), from the region (P = .336), from the treated arch (P = .999), and from sex (P = .058). However, for the sex, the failure/success ratio tended to be higher in the men (Table 4), and the success rate was independent from the recorded bone quality and quantity data (Table 5).

Finally, the success rate was evaluated in relation to the diameter and length of the implants (Table 6). No failure was recorded for the smallest diameter (3.3 mm). Inversely, the 5-mm diameter implant showed a slightly lower success rate (94.20%) compared with the mean success rate of the study. Concerning the length, no failure was recorded for the 7-mm and 16-mm implants. The success rate was similar for lengths 10, 12, and 14 mm. Only the 8-mm length showed a lower success rate (86.75%).

This study allowed demonstration of the efficiency of the EVL evolution system used in clinical conditions encountered in everyday practice, while respecting the conventional indications and contraindications of implant therapy, according to the NIH Consensus Statement.⁶ The implant sites were selected according to the clinical decision of 3 clinicians. The prosthetic indications ranged from single unit restorations to bridges of various spans (Figures 5 and 6). The results obtained by each of the clinicians are nevertheless very homogenous,7^,8 and the mean success rate is very high.

Froum et al2 have demonstrated in a histologic study the quality of the bone-to-implant interface offered by the surface properties of the EVL evolution implant. Numerous studies also have shown the advantage of rough surfaces vs machined surfaces.9^,10 

The success rate obtained is 97.08%, which is higher than the 85% recommended by Albrektsson in 1986,11 and the 95% over 5 years reported by Smith and Zarb in 1989.12 It is also higher than the 93.6% reported for fixed partial prostheses, and similar to the 97.5% reported for single crowns by Lindh et al.13 Khang et al,14 when comparing machined implants to implants having undergone a surface processing, obtained a success rate of 96.8% after 3 years, which was higher than the one observed for the machined implants (84.8%). These results confirm those of Testori15 (ie, 97.5% after 3 years) and those of Stach and Kohles16 (ie, 98.4% after 4 years).

Our results may sometimes appear slightly lower because in the statistical analyses, cases considered as lost (dropout patients) are counted. By excluding the dropout patients from the present study, the success data would certainly be higher, but these patients represent a situation that can occur in any dental office, and must therefore be included in the calculation process of the truly observed success rate.¹⁷

According to the studies of Zarb and Zarb,18 no significant difference was observed in the success/failure ratio for the women and men at the end of this study.

The success rate was not influenced by the location in either the upper or lower maxilla. Therefore, our results are in contradiction with those of Versteegh et al19 who reported a success rate of at least 95% for the ITI implants in the mandible, and 85% in the maxilla in a clinical follow-up of 5 years. In the same way, according to Albrektsson and Sennerby,20 the Branemark implant showed a 91% success rate in the mandible, and 84% in the maxilla. Lastly, Salvi et al21 reported a 100% success rate for 67 ITI SLA implants at 1 year in the posterior mandible.

Nedir et al22 have emphasized the advantage of using short ITI implants (≤11 mm) in the posterior regions to avoid complex imagery examinations. These authors recommended, besides an accurate clinical examination, a simple panoramic or periapical radiograph; in these conditions, they obtained 99.4% success rate over 7 years. Our results tend to confirm this opinion, given that we obtain the maximum success rate for the 10- and 12-mm implants; only the 8-mm implants show a higher failure rate, while for the 7-mm implants, no failure was observed.

However, in a long-term study, implant length must not be considered as the only criterion; the mechanical lever yielded by the implant-supported prosthesis should also be accounted for. It is likely that the prosthetic criteria significantly influence the variations observed in the results reported among the different authors.

Finally, respect of the indications and strict use of a 1-stage surgical protocol allow to obtain results which are similar to those already reported in the literature.^23,24

This study conducted in private practice clinical conditions, and not in a hospital or a university, allowed evaluation of the SERF EVL evolution system, which offers a 97.08% success rate, thus similar to other implant systems. Moreover, this system specifically provides the same results in both maxillas, and in both the anterior and posterior regions as well. It is likely that patient selection and the employed prosthetic strategies influence significantly the results of this study. Further studies describing and analyzing the associated prosthetic designs are required for a better understanding of these excellent results.