Implant‐supported removable partial dentures compared to conventional dentures: A systematic review and meta‐analysis of quality of life, patient satisfaction, and biomechanical complications

Abstract Objectives The purpose of this systematic review and meta‐analysis was to compare implant‐supported removable partial dentures (ISRPDs) with distal extension removable partial dentures (DERPDs) in terms of patient‐reported outcome measures (PROMs: patients' quality of life and satisfaction) and to determine mechanical and biological complications associated with ISRPDs. Material and Methods An electronic search was performed on four databases to identify studies treating Kennedy class I or II edentulous patients and which compared ISRPDs with DERPDs in terms of PROMS and studies, which evaluated mechanical and biological complications associated ISRPDs. Two authors independently extracted data on quality of life, patient satisfaction, and biomechanical complications from these studies. The risk of bias was assessed for each study, and for PROMs, the authors performed a meta‐analysis by using a random‐effects model. Results Thirteen articles were included based on the selection criteria. The difference in mean scores for quality of life (30.5 ± 1.8; 95% confidence interval [CI], 24.9–36.1) and patient satisfaction (−20.8 ± 0.2; 95% CI, −23.7 to −17.8) between treatments with conventional and implant‐supported removable dentures was statistically significant (p < .05). Implant‐supported removable dentures improved patients' overall quality of life and satisfaction. Some mechanical and biological complications, such as clasp adjustment, abutment or implant loosening, marginal bone resorption, and peri‐implant mucositis, were noted in ISRPDs during patient follow‐up. Studies assessing PROMs were very heterogeneous (I 2 = 65%, p = .85; I 2 = 75%, p = .88). Conclusions ISRPDs significantly improved quality of life and patient satisfaction. Some mechanical and biological complications have been associated with ISRPDs treatment, requiring regular monitoring of patients to avoid the occurrence of these complications.


| INTRODUCTION
Distal extension removable partial dentures (DERPDs) are a suitable treatment option that improves stomatognathic functions, which are still widely used in the rehabilitation of Kennedy-Applegate class I or II partially edentulous patients (Gonçalves et al., 2014a;Vanzeveren et al., 2003). However, this type of prosthesis is associated with increased alveolar bone resorption, caries lesions on teeth, and psychologically less acceptable treatment (Knezović Zlatarić et al., 2002). Moreover, DERPDs present many biomechanical problems (unsatisfactory retention and stability), which may compromise masticatory efficiency (Cunha et al., 2008). Additionally, its limited functional and aesthetic properties (Shala et al., 2016) and its relatively high complication or failure rate (Knezović Zlatarić et al., 2002;Vermeulen et al., 1996;Wagner & Kern, 2000) explain why DERPDs can be a source of discomfort and dissatisfaction for patients (Armellini et al., 2008;Bilhan et al., 2012).
For these reasons, some patients rehabilitated with DERPD do not wear their prostheses regularly (Vanzeveren et al., 2003), hence the need for clinicians to consider other treatment alternatives as dental implants. According to the literature, dental implants are a highly successful treatment option for the replacement of missing teeth (Albrektsson et al., 1986;Howe et al., 2019;Pjetursson et al., 2012;Weber & Sukotjo, 2007). Their long-term survival rate was assessed in many systematic reviews (Hjalmarsson et al., 2016;Howe et al., 2019;Moraschini et al., 2015) which reported various results. The authors of these reviews concluded that this survival rate at 10-years follow-up was over 92.8% (95% confidence interval [CI]: 90-94.8) (Hjalmarsson et al., 2016;Howe et al., 2019;Moraschini et al., 2015). However, the presence of bone defects at the implantation site limits at times the availability of bone tissue for placing an adequate number of implants. Fortunately, there are other therapeutic solutions that make it possible to overcome this obstacle. It is notably about the use of dental implants with reduced dimensions (Threeburuth et al., 2018) or preliminary bone reconstruction of the edentulous jaw, a process that can uses the combination of bone substitutes with autologous mesenchymal stem cells or autologous bone grafting (Arinzeh et al., 2005;Finkemeier, 2002;Gjerde et al., 2017Gjerde et al., , 2018. Likewise, bone substitutes of animal, human or synthetic origin may be used alone to reconstruct small defects (Malard et al., 2007). However, these alternative treatments are often associated with increased cost, treatment time, and postoperative morbidity. Implant-supported removable partial dentures (ISRPDs) have been suggested as a minimally invasive approach for partially edentulous patients and are a suitable alternative to DERPDs without compromising implant success while improving the quality of life and satisfaction of patients when compared with DERPDs (De Carvalho et al., 2001;Ganz, 1991;Giffin & Dent, 1996;Kuzmanovic et al., 2004;Mijiritsky & Karas, 2004;Ohkubo et al., 2008;Uludag & Celik, 2006;Park et al., 2020). The IRSPDs provide cost-effective treatment. This treatment option not only increases the retention of the prosthesis and hence limits lateral and vertical displacement of the removable partial denture, but it also distributes masticatory forces more effectively along the prosthesis and the adjacent teeth (Cho, 2002). It also increases patient satisfaction and improves chewing ability, phonetics, and esthetics, since sometimes the unesthetic vestibular bracing arms can be removed (Ohkubo et al., 2007;Shahmiri & Atieh, 2010).
Previous studies have reported that ISRPDs are of both functional and aesthetic interest. It is a preferable treatment option for patients with complaints about their DERPDs (Mijiritsky & Karas, 2004;Uludag & Celik, 2006;Wismeijer et al., 2013). The relevant literature demonstrates that the additional retention provided by implants increases stability (Ohkubo et al., 2008) and thus improves masticatory efficacy and patient satisfaction (Cho, 2002;De Freitas et al., 2012;Goiato et al., 2018;Grossmann et al., 2009;Suzuki et al., 2017;Wismeijer et al., 2013;Zancopé et al., 2015). In addition to functional comfort, there is also an aesthetic gain through the removal of unsightly clasps in the anterior areas (Grossmann et al., 2008). ISRPDs also improve the quality of life of patients wearing appliances without the need for invasive surgery (bone grafts, sinus lift, etc.) (Cho, 2002;Goiato et al., 2018;Wismeijer et al., 2013). However, there is a lack of systematic reviews and meta-analyses providing clear scientific evidence of the long-term therapeutic efficacy of this prosthesis compared to conventional dentures (DERPDs). For these reasons, the purpose of this systematic review and meta-analysis was to compare ISRPDs and DERPDs in terms of the patient-reported outcome measures (PROMs: quality of life and patient satisfaction) and to determine the mechanical and biological complications associated with ISRPDs. The null hypothesis was that no difference would be found in the quality of life and satisfaction of patients rehabilitated with ISRPDs compared to those fitted with DERPDs.

| Protocol and study questions
This systematic review and meta-analysis were conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) recommendations (Moher et al., 2009) BANDIAKY ET AL. | 295 and the Cochrane Guidelines (Cumpston et al., 2019). The review was not registered in PROSPERO before data collection.
This study deals only with data from clinical studies published in bibliographic databases or specialized journals, its aspect does not require the approval of the ethics committee nor the "protection of human subjects and animals in research" or informed consent.

| Inclusion and exclusion criteria
The review included human clinical controlled studies evaluating patient-reported outcome measures and biomechanical complications associated with ISRPDs, and in which patients were rehabilitated first with a DERPD and then with an ISRPD. Articles from studies with no available data, prosthetic rehabilitations other than ISRPDs and DERPDs, clinical report cases, and literature reviews were excluded from this analysis. Similarly, studies that did not compare ISRPD to DERPD in terms of patients' quality of life or satisfaction, that did not evaluate the clinical complications of ISRPDs, or with fewer than 10 participants were excluded from this analysis.

| Risk of individual bias of the studies
The risks of bias were evaluated for the totality of the studies included according to a modified MINORS scale (Methodological Index for Non-Randomized Studies) of Tsirogiannis et al. (2016). This scale consisted of 10 items, with 2 additional items proposed for in vivo studies. Each item is scored from 0 to 2; for most items, 0 indicates that the content of the item is not reported, 1 indicates that the content is reported but inadequately, and 2 indicates that it is sufficiently reported. The risk of bias could be weak, moderate, or high (Table 1).

| Synthesis of results
Data from the various studies were extracted, and the results were synthesized. For studies in which the authors reported results as medians and interquartile ranges, the values were converted to means and SDs using the formula (q1 + median + q3)/3, where q1 indicates the 25th percentile and q3 the 75th percentile, as proposed in the study by Nagarkar et al. (2018). An approximation of the SD was obtained by applying this formula: (q3 − q1)/1.35. When several data points were reported by the authors, the most negative ones were used for the quantitative synthesis. The same was true for patient follow-up, where the data from the longest follow-up were retained. Meta-analysis was performed by using R Commander™ software, and a random-effects model (Gonçalves et al., 2014a;Higgins & Thompson, 2002). The choice of this model was justified by the fact that most of the studies were small (number of patients <30) and that the effect of the intervention measured (quality of life and patient satisfaction) was different for each of these studies, given their heterogeneity. Thus, a random-effects model will give more weight to these small studies. When studies used the same type of intervention and comparison groups with the same outcome measure, the results were pooled with mean differences for continuous outcomes and risk ratios for dichotomous outcomes and calculated 95% CIs and p values for each outcome (DerSimonian and Laird method) (DerSimonian & Laird, 1986). Heterogeneity between the studies was assessed by using the Higgins I 2 statistic (Higgins & Thompson, 2002). We considered an I 2 value of 50% or more to indicate substantial heterogeneity. A sensitivity analysis based on the risk of bias of the included studies (low risk of bias vs. high or unclear risk of bias) was conducted.

|
reported but inadequate, 2: reported and adequate.  | 305 (p < .05) (Figure 3). The study by Jensen et al. (2017) was also excluded from quantitative synthesis due to the missing data on patients' satisfaction at baseline. The study in which the investigators assessed the patient satisfaction were heterogeneous (I 2 = 75%, τ 2 = 0.65, p = .88).

| Secondary outcomes of the studies
Ten studies included in the systematic review evaluated the mechanical and biological complications associated with ISRPDs and reported various results regarding implant survival rates, marginal bone loss around implants, abutment loosening or mobility, implant mobility, bleeding on probing or deep probing depth. They also reported the prosthetic complications such as fracture of the denture base, rest, clasps or resin. Some authors (Gonçalves et al., 2014b;Mijiritsky et al., 2013;Oh et al., 2021) reported an implant survival rate of 100% after the follow-up period ranging from 2 to 180 months, while for others (Bellia et al., 2020;Bortolini et al., 2011;Gates et al., 2014;Grossmann et al., 2008;Jensen et al., 2017;Ortiz-Puigpelat et al., 2014;Payne et al., 2017) this survival rate was 91.6%-97% (Table 3). The number of implants that failed was ranging from 1 to 6. The mean marginal bone loss around implants ranged between 0.64 and 2.11 mm and the mean deep pockets varied from 2 to 4 mm. Some authors (Bortolini et al., 2011;Gates et al., 2014;Grossmann et al., 2008) reported implant bleeding on probing, mobility or deep probing depth, abutment loosening, loose healing cap (Payne et al., 2017) or ball attachments replacement. Bortolini et al. (2011) reported that the peri-implant soft tissues and marginal gingiva of most patients were slightly inflamed. In the study of Gates III et al. (2014), prosthetic complications involved clasp, fracture of denture tooth, reline of the denture base, and reprocess of DERPDs were primarily minor and could be managed within a single clinical visit. Gonçalves et al. (2014b) found stable periodontal conditions around the implants, no intrusions or mobility of teeth, and no radiographic changes in bone level after 2 months follow-up. For their part, Jensen et al. (2017) reported that posterior implants demonstrated significantly more complications than anterior implants (peri-implant mucositis). In their study, Bellia et al. (2020) concluded that the use of short implants for retaining DERPDs may be considered a viable treatment option for patients with distal edentulism and contraindications for more complex implant rehabilitation.
As shown in Table 1, the level of risk of bias was moderate overall for all studies. The studies were of low methodological quality because half of the studies were retrospective case series.

| DISCUSSION
ISRPDs can be considered as an alternative to DERPDs and implantsupported fixed partial prostheses when placement of an adequate number of implants is limited by bone height and thickness or by financial reasons. In this situation, a small number of conventional or T A B L E 2 (Continued) Abbreviations: DERPD, distal extension removable partial denture; IARPD, implant-assisted removable partial dentures; ISRPD, implant-supported removable partial denture; KA, Kennedy-Applegate; M, molar; Mn, mandible; Mx, maxilla; No, number; NR, not reported; OHRQOL, oral health-related quality of life; PM, premolar; RCT; randomized controlled trial; RNT, remaining natural teeth; RPDs, removable partial dentures.
mini-implants can be placed to retain and stabilize the DERPDs, pro-  (Campos et al., 2015).
In this review, patients included in the studies that evaluated the PROMs were first rehabilitated with DERPDs which were converted to ISRPDs following placement of implants in the premolar or molar region. The loading of these implants as well as the insertion of the attachment systems were carried out at least 3 months later, which was sufficient time to achieve osteointegration. In their study, Ortiz-Puigpelat et al. (2014) reported that the treatment of partially edentulous patients with ISRPDs improves the PROMs without the need for extensive bone regeneration surgeries and prosthodontic rehabilitation. ISRPDs improve also prosthesis performance, overall patient satisfaction with respect to retention, comfort, and masticatory capacity (Gonçalves et al., 2014b). Chikunov et al. (2008) reported other advantages related to the ISRPDs: a smaller number of implants, lower cost, fewer time-consuming clinical and laboratory procedures, simplified hygiene when compared with fixed dental prostheses, better distribution of the masticatory loads to the abutment teeth and implants, preservation of residual bone around the implants and remaining teeth, better comfort because of minimal rotational movement, treatment compliance, and possible later conversion into a complete overdenture. Most of the implants placed in the patients F I G U R E 2 Mean score OHIP questionnaires between 2 treatment modalities (DERPD vs. ISRPD). Wilcoxon signed-rank test. Significant at p < .05. CI, confidence intervals; DERPD, distal removable partial denture; ISRPD, implant-supported removable partial dentures; OHIP, oral health impact profile; OHRQOL, oral health-related quality of life; SD, standard deviation F I G U R E 3 Mean VAS score assessing participant's satisfaction. Significant at p < .05. DERPD, distal removable partial denture; ISRPD, implant-supported removable partial dentures; SD, standard deviation; VAS, visual analog scale included in these studies were conventional types. However, Bellia et al. (2020) reported that the use of short implants for retaining DERPDs may be considered a viable treatment option for patients with distal edentulism and contraindications for more complex implant rehabilitation. The conversion of the already well-accepted and patientintegrated DERPDs into an ISRPDs brings more comfort during wear by limiting the prosthesis' dislocation from its supporting surfaces, particularly during mastication. Indeed, these DERPDs are known to be more vulnerable to lifting forces (Wismeijer et al., 2013). This is probably one of the main reasons why patients resort to implants to obtain a more stable and retentive prosthesis limiting food accumulation underneath the distal extension bases of the removable partial Abbreviation: ISRPD, implant-supported removable partial dentures. denture and decreasing the pressure on the resilient mucosa. In addition, less relining of the intaglio surface is required with implant support but hygiene maintenance of the natural teeth and implant attachment systems will be required. Therefore, our results should be interpreted with caution, our review included both prospective and retrospective studies for the evaluation of PROMs. This mix of design studies constitutes a bias in the interpretation of the results. In addition, the instruments (OHRQoL, OHIP-49, SF-36, VAS, and questionnaires) used to evaluate these PROMS differ from one study to another. This shows that these studies are highly heterogeneous even if the participants are their own control. In addition, the characteristics of the participants were different, some of whom were already unable to wear their DERPDs, which constitutes a selection bias.
For implant survival rate, our results were consistent with those of previous systematic reviews (De Freitas et al., 2012;Lemos et al., 2021;Park et al., 2020;Zancopé et al., 2015) which reported a low proportion of implant failure rates over a follow-up period ranging from 6 to 180 months. Our results can be explained by the fact that most of the implants used in the selected studies were of conventional length and diameter. Indeed, it has been described in the literature that these types of implants had a better survival rate than mini-implants (Lemos et al., 2016;Papaspyridakos et al., 2018).
However, Threeburuth et al. (2018) found no difference in terms of implant survival rate between conventional-size and mini dental implants 12 months after surgery. Some authors concluded that the mini dental implants can be applied for retaining mandibular Kennedy years follow-up period. In the study by Jensen et al. (2017), the mean peri-implant bone loss was 1.06 ± 0.59 and 1.10 ± 0.53 in the premolar and molar regions respectively. Posterior implants demonstrated significantly more complications than anterior implants (periimplant mucositis). These results on bone loss were similar to those of Bellia et al. (2020). The average pocket depth varies from one study to another. Other biological (abutment loosening, bleeding on probing), and mechanical complications described in Table 3 have been reported by the authors of the different studies. All of these results demonstrated that the ISRPDs did not compromise the longevity of dental implants (Lemos et al., 2021), but careful planning is crucial to ensure success and prevent or minimize future problems, such as periodontal and peri-implant bone changes. The studies included for the evaluation of these parameters were very heterogeneous due to differences in patient characteristics (age, gender, number of residual teeth, occlusal pattern, duration of follow-up, the position of, and size of implants.
Our work has limitations, and its results should be interpreted with caution because of the low methodological quality of the included studies, the small number of participants, and the short follow-up period for some studies. These are mainly retrospective studies with a low level of scientific evidence. There is a lack of randomized controlled studies dividing patients into parallel groups and evaluating their quality of life and level of satisfaction and the biomechanical complications associated with each type of prosthetic rehabilitation. However, some positive points emerge from this study, and the patients served as their own controls, which limits the interindividual variability of the results.

| CONCLUSIONS
A qualitative and quantitative synthesis of the data reported in the included studies indicates that: 1. ISRPDs significantly improved patients' quality of life and satisfaction compared to DERPDs.
2. Some mechanical and biological complications were observed following the completion of the ISRPDs.

ACKNOWLEDGMENTS
The authors thank Amany Laham for her help in translating the paper into English.

CONFLICT OF INTEREST
The author declares that there is no conflict of interest.

AUTHOR CONTRIBUTIONS
Conducting the literature review, selecting articles, supervision, | 309

DATA AVAILABILITY STATEMENT
The datasets collected and/or analyzed during the current systematic review and meta-analysis are available from the corresponding author on reasonable request.