1. Introduction
The rehabilitation of edentulous patients with dental implants constitutes a stable and highly predictable acquisition of modern dentistry, corroborated by well-documented clinical results and long survival
[1][2]. The presence of an adequate bone volume at the edentulous site still represents a basic requirement for dental implants to be correctly placed. Dimensional reduction of the alveolar process, however, usually happens after tooth extraction, and adjunctive loss of bone may be caused by trauma, periodontal disease, or atrophy
[3]. In these cases, additional reconstructive surgeries, e.g. vertical bone augmentation procedures or sinus lift, may be applied to allow standard-length implant placement, although they are invariably associated with relevant limitations, such as longer treatment time, additional costs, increased postoperative morbidity, and a higher risk of complications
[4][5][6][7][8].
A widely documented alternative to augmentation surgical procedures is represented by the rehabilitation of atrophic edentulous sites with implants of reduced length, which have reached a considerable diffusion in the last few years with very promising clinical results
[9][10][11][12].
Several advantages are linked to the use of short implants, including ease of handling, reduced surgical invasiveness, and a low risk of injuring noble anatomical structures, thus sustaining the concept of a “stress minimizing surgery”
[9]. Short implants, however, are not free from risks and complications, due to the higher crown-implant ratio and the lower bone-to-implant contact area with respect to longer fixtures
[13].
The clinical performance of short implants has been widely investigated in the recent literature, with no univocal findings. Higher failure rates have been historically associated with short implants compared to longer ones
[14][15]. More recent studies showed, for short implants, clinical outcomes comparable to, or even better than, longer implants placed either in native or augmented bone, as confirmed by several systematic reviews
[8][12][16][17][18][19][20][21][22][23][24][25].
The reasons for such a great variability may be found in several elements of heterogeneity among studies. The definition of “short dental implant”, for instance, greatly differs from <10 mm to ≤8 mm and, more recently, ≤7 mm
[15][26][27][28]. In this sense, in the 2015 EAO Consensus Conference on the use of short implants for dental rehabilitation
[11], the researchers recommend, for future research, to identify the minimum length for a predictable survival of implants (redefining the concept itself of “short” implant) and to perform clinical trials with an appropriate design to reliably compare these therapeutic concepts in the long term. Furthermore, one should consider that the greatest part of the existing studies is designed to compare short implants placed in pristine atrophic bone with long implants inserted in augmented sites. Finally, short- and medium-term data (1 to 5 years from loading) are prevalent, and longer follow-ups are lacking.
2. Current Insights
The present meta-analysis exclusively included randomized controlled trials comparing short (≤6 mm) and longer (≥8.5 mm) implants placed in sites without the need for bone augmentation.
Universally accepted definitions of a “short”, “standard”, or “long” implant in the literature do not yet exist, and these concepts continuously evolved throughout the years
[27][29][30]. In accordance with the 2018 ITI Consensus Report
[12], the researchers considered truly “short” implants of length ≤ 6 mm, in comparison to control implants with a length ≥ 8.5 mm.
Short implants have been extensively proposed in atrophic sites as an alternative to longer implants associated with bone augmentation, finding in this treatment their natural term of comparison.
Dental implants placed in vertically augmented bone showed a high survival rate (98%, range 95–100%), comparably to short implants placed in atrophic sites (96%, range 86.7–100%) after periods of 1 to 5 years in function, although they were associated with a higher number of complications, surgical time, and treatment costs
[12].
The researchers believe, however, that comparisons of short and long implants in the same clinical scenario could be the only one able to correctly evaluate their performance avoiding the effect of confounding factors related to the different characteristics of the edentulous sites, especially those related to the augmentation procedure itself.
Only two other meta-analyses, among those available in the literature, addressed this issue, trying to analyse, by specific sub-analyses, the influence on the reported outcomes of the variable “bone augmentation” among the included trials
[8][25], thus representing the main term of comparison for the researchers' results. With respect to those meta-analyses, however, the researchers were able to include a higher number of more recent RCTs
[31][32][33][34][35][36][37][38] with longer follow-up, as well as further additional data directly obtained by the researchers themselves.
The first relevant finding of the present meta-analysis is the comparable survival rate found between test and control groups at all the available follow-up times, although the low number of available studies limits the strength of the evidence for this outcome. In fact, the available sample size is smaller than the optimal information size and this affects the rating of evidence by the GRADE methods, indicating a moderate quality, as well as the TSA, which revealed the weak power of the meta-analysis findings. Recent systematic reviews on short implants separately performed additional analyses focused on those RCTs comparing short and longer implants placed in non-augmented bone
[8][25]. Ravidà et al. (2019) showed homogeneous results in terms of survival rates between trials with or without bone augmentation procedures, thus suggesting no influence of this variable on this outcome. It must be underlined that just one study of only one follow-up period (one year) was included in the “non-augmentation” group by the researchers. Yu et al. (2021) performed a similar sub-group analysis on a larger number of studies and follow-up periods (1, 3, and 5 years). In this case, accordingly to the researchers' analysis, a trend for higher risk of failure for short implants at 1 and 3 years was found, which reached the significance, differently from the researchers' results, at 5 years of follow-up with an RR of 0.955 (95% CI 0.912–0.999,
p < 0.05). Several factors must be considered for a correct interpretation of such different findings, including the type of edentulism, the loading time, and the prosthetic design. Indeed, if the researchers look at the trials reporting a higher risk of failure for short implants at 5 years in the study by Yu et al. (2021), the researchers find that all the implants were early loaded and supported single crowns or 2/3-units fixed partial prostheses. In contrast, two over the three additional data sets (from two studies
[32][33]) included in the researchers' analysis regarded splinted implants, conventionally loaded, supporting full-arch rehabilitations of mandibular edentulism. Moreover, details on other important variables (such as bruxism, smoke habit, bone quality, and implant stability) in the patients experiencing implant loss were rarely provided by the primary studies, making it more difficult to correctly analyse possible causes and risk factors of implant failures.
Peri-implant marginal bone level maintenance over time is of pivotal importance for the long-term success of dental implants. The present meta-analysis showed that only 1- and 5-year follow-up data from implant placement showed a significantly higher bone loss for the long implant group. However, such results must be considered with caution due to the very low number of included studies and the high heterogeneity found, with a significant impact on the overall evaluation given by two articles
[33][39] derived from the same study. This latter result was the only one included in the “no augmentation” group by Ravidà et al.
[8] in their meta-analysis, resulting in a significantly higher bone loss for the long implant group at the 1-year follow-up. Also, Yu et al.
[25] performed a sub-group analysis for MBLc on “augmentation” vs. “no augmentation” studies one year after two baseline points (implant placement and prosthetic loading). They did not find a difference between long and short implants one year after prosthetic loading and, different from the research results, also one year after implant placement. Also in this case, the low number of included studies and their heterogeneity impose caution.
In addition, the incidence of the reported biological and technical complications did not show a statistically significant difference between short and long implants, in line with the analyses performed by previous similar reviews up to 5 years of follow-up
[8][25]. At longer follow-ups, a statistically significant higher number of technical complications was found after 10 years
[38], although the value of this finding was sustained by only one trial. It is hard, however, to compare the results with those from other similar studies, in light of the different biological complications considered (only the occurrence of peri-implant mucositis and peri-implantitis was analysed in the research), and, moreover, of the different diagnostic criteria adopted for such diseases among the studies.
Looking at the main limitations of the present meta-analysis, the low number of includible studies and their mainly short/medium follow-ups must be cited. Furthermore, the variety of experimental conditions among studies limits the power of the obtained results, and no further analyses could be performed based on variables including implant location (mandible vs. maxilla), type of edentulism (total vs. partial), implant type, smoke habit, implant loading, periodontal health, systemic conditions due to the reduced number of available studies, or the lack of information retrievable about these variables. No subgroup analysis could be performed for confounding factors, such as smoking and periodontal status, or for the level of analysis (implant or patient-level). Implant-level and patient-level data were pooled together in the MBLc meta-analysis, with a consequent possible underestimation of the confidence intervals for the pooled estimate.
Further, well-designed RCTs comparing clinical and radiological outcomes of short and long implants placed in similar conditions (location, type of edentulism, prosthetic rehabilitation, etc.) and with adequate analysis of confounding factors should be performed to obtain more solid evidence about the efficacy of short implants used in atrophic sites.
This approach would also help to extend their clinical indications, supporting the hypothesis of their routine use in non-atrophic edentulous sites instead of traditionally long implants, in the groove of a minimally invasive, low-stress, simplified implant therapy, with benefits for both patients and clinicians.
The ideal length of implants supporting prosthetic rehabilitations, indeed, is a relative concept that underwent a progressive reduction throughout the years from the origin of modern implantology to the present
[27][29][30], also thanks to the outstanding progress made in the last decades in terms of constitutive materials and macro-design of dental implants, as well as of micro/nano-topographic and chemical modifications of modern implant surfaces
[40][41].
The advantages of using short implants in non-atrophic sites compared to longer ones include simplified surgical management during placement, with a lower risk of involving noble anatomic structures, as well as a simplified and less invasive surgical procedure, if biological complications, over time, might require fixture removal, with the chance to leave an amount of residual bone enough for a new rehabilitation. On the other hand, due to the reduced fixture length, some conditions, such as low-density bone and operative protocols, e.g., post-extraction placement, and immediate/early loading, could reduce the reliability and predictability of this rehabilitative approach.
This is the reason, in the researchers’ opinion, because further RCTs should be performed in which all the possible relevant variables and risks are considered, analysed, and controlled in order to obtain more solid evidence on the performance of short implants in each specific clinical scenario, starting from the “low-risk” one (healed sites, good bone quality, conventional loading, splinted implants), to the others.
This entry is adapted from the peer-reviewed paper 10.3390/ma15093138