Survival of Single Immediate Implants and Reasons for Loss: A Systematic Review

: Background: Immediate implant placement (IIP) or Type I implants have become more attractive than conventional implant placements as it reduces the number of surgical procedures and allows faster delivery of the ﬁnal restoration compared to conventional implant placements. However, the survival of Type I implants varies depending on multiple factors. Purpose: To evaluate the survival rate of Type I implants, and to describe the factors inﬂuencing their failure. Materials and methods: A developed search strategy was applied to identify randomised controlled trials on single-unit immediate implants including at least six human participants with a minimum follow-up time of 12 months and published between 1 January 1999 and 1 January 2020 in several databases. The data were extracted independently using validated data extraction forms. Information on survival rates, number of implants placed, loading protocols, setting of the study, location of implants in the jaw, antibiotic protocol, grafting methods, and implant geometry were obtained and assessed. Results: Twenty-six randomised controlled trials with an average follow-up time of 24 months (range = 12–120 months) were included and analysed to give a survival rate ranging between 83.7 and 100%. Fifteen studies reported implant failures, of which twelve reported early losses (loss before deﬁnitive restoration). Nine early losses were due to lack of osseointegration, two did not report the reason for implant failure, and one was reported as iatrogenic. Of the eleven studies with 100% survival rates, the common trend observed was the use of titanium implants and an antibiotic regimen using amoxicillin. Conclusions: The survival rate for immediate single implant placement ranged from 83.7 to 100%. Implant failure was not consistently reported and when reported, failure due to lack of osseointegration prior to placement of the deﬁnitive restoration was the most common descriptor. Other attributed reasons included infection abscess, mobility after immediate loading, and iatrogenic complications.


Summary Box
What is known: There are several systematic reviews on the survival of immediate implants which did not differentiate between single-unit and multiple-unit implants, or between different loading protocols. These reviews also have limitations, such as inclusion of non-randomised controlled trials and no risk of bias assessment.
What this study adds: The current review was designed to overcome the limitations of these previous systematic reviews and to update the current knowledge on the survival rate of single-unit immediate implants. This study suggests that immediate implants can be a predictable procedure with high survival rates based on the most current randomised controlled trials.

Introduction
Implants are an attractive treatment option for single tooth replacement, especially when traditional restorative options may be too destructive or inconvenient for the patient, such as a conventional 3-unit, cantilevered, or resin-bonded fixed partial denture, or a removable partial denture. Despite the increasing popularity of implants, they require complex, multidisciplinary treatment planning and strict inclusion criteria [1].
Conventional implant placement typically requires longer periods of healing before the final restoration can be placed, which may increase the psychological impact of tooth loss [2]. Type I or immediate implant placement (IIP), therefore, has become the more appealing option for both patients and dentists due to the reduced number of surgical procedures and, hence, shorter treatment time [1]. However, IIP should be based on case selection as successful placement is not always guaranteed [1]. Primary stability, which is paramount in the success of dental implant treatment is often difficult in IIP due to the lack of hard tissue immediately post-extraction. In order to achieve primary stability, a 3-5 mm apical, palatal, and intraradicular bone is needed [2]. Furthermore, bony defects and unfavourable bony morphology post-extraction present a challenge to osseointegration [3]. When an implant fails to osteointegrate, its removal can cause trauma. As implants do not display bundle bone, the remaining defect after removal of failed implants do not behave like postextraction sockets [4]. Therefore, a re-attempt at implant placement may not be possible and the patient may be left with less or even no option to replace their missing dentition [5,6].
The survival of an implant is defined as the presence of the implant upon recall examination, despite its conditions. Conversely, implant failure is the absence of the implant on recall examination. These definitions are derived from the Third International Team of Implantology (ITI) consensus meeting [7]. Implant failure can be further grouped into four main reasons: biological, mechanical, iatrogenic, or inadequate patient adaptation requiring removal of implants. Biological issues are related to osseointegration and can be classified into early and late loss depending on whether it was lost before or after implant loading, respectively [8].
There are several systematic reviews on the survival of immediate implants with different independent variables applied across a range of implant systems from numerous manufacturers [9][10][11][12][13][14][15]. Some of these studies did not differentiate between single-unit and multiple-unit implants, or between different loading protocols. Some of these reviews also included non-randomised controlled trials and did not investigate the risk of bias (refer to Table 1). The current review is designed to overcome the limitations of these previous systematic reviews. The objective of this systematic review is to evaluate the survival rate of single-unit immediate implants using recent randomised controlled trials, and to establish a link between the reasons for failure and factors that may influence its survival. Included multiple-unit Implants Did not mention whether the implants were immediate or delayed placements Did not define survival Included non-randomised controlled trials Did not investigate loading protocols Did not assess bias risk

Materials and Methods
This systematic review was registered as a protocol in the International Prospective Register of Systematic Reviews (PROSPERO) platform (CRD42020173150), and the reporting was carried out following the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [16].

Search Strategy
A detailed search strategy was used for the PubMed database to identify all articles published between 1 January 1999 and 1 January 2020 in relation to the stated aims of this review. In addition, a manual search of Clinical Oral Implants Research and the European Journal of Oral Implantology was attempted to identify any relevant studies. The reference list from the included studies was also screened for further inclusions into this study.
Focused question: what is the survival rate of single-unit immediate implant and what is the reason for implant failure?
The following PICO strategy was designed to select the studies to be included in this review [17]: Participants: Subjects requiring a single implant in the maxillary and mandibular areas. Intervention: Implant placement using the immediate placement protocol (Type 1). Comparison: Delayed implant treatment protocols (Type 4) used for the replacement of a single tooth in the maxillary and mandibular region.
Outcomes: Implant survival, implant failure, and the reasons for failure. Eligibility criteria: For a study to be included it must meet the following inclusion criteria: Study included a minimum of six human subjects or more, including split mouth studies; • Used single-unit immediate implants; • Minimum follow-up time of one year; • Full-text study published in English

Study Selection
After the initial electronic search of titles by two authors (N.K. and B.K.), the titles and abstracts of all the studies identified via electronic searches were independently scanned by two reviewers (N.K. and L.A.M.). The next step was to review all selected abstracts to determine selection of full-text articles after applying the inclusion criteria. The full texts of all studies of possible relevance were then obtained for independent review and assessment by the two reviewers. Disagreements among reviewers were resolved by discussion. All studies meeting the inclusion criteria then underwent data extraction. Studies rejected at this or subsequent stages were removed and the reasons for exclusion were recorded (Figure 1).

Results
The first electronic search yielded 6042 citations that could be reviewed. After the abstracts were screened, 5674 of these were rejected. Full text assessment was conducted on 368 studies. A total of 193 articles were excluded: 69 studies were excluded on the grounds of mixed results with multi-unit restorations, 55 were rejected as they did not follow-up implant placement to the minimum 1-year mark, 30 were not longitudinal studies, 13 had a sample size less than six, 14 were published prior to 1 January 1999, 10 studies did not do IIP, one study was not published in English, and one study was not on humans (see Table 4).
Furthermore, 17 additional studies were included after manually searching the reference lists of non-excluded studies. An overall total of 192 longitudinal studies were accepted. A further 164 articles were rejected as they were not randomised controlled trials, and 28 articles were accepted; however, two studies were published twice at two different time points during their experiment so their results were combined [21][22][23][24]. A total of 26 randomised controlled trials were included in this study, with an average follow-up time of 24 months (range 12 to 120 months). A diagram detailing the search strategy is shown in the Figure 1.
Subsequently, the bias was assessed using the Cochrane Risk of Bias Tool for randomised controlled trials. A summary is provided in Table 3.

Data Extraction
The data were independently extracted by a group of seven review authors (N.K., W.T., P.S., P.G.S., J.G., C.T., D.C.) using validated data extraction forms. Any discrepancies between the reviewers were resolved by discussion and consensus after consultation with the other author (L.A.M.).
The implants in the included studies were grouped into four categories based on implant placement and loading protocol: immediate placement and immediate loading (IPL); immediate placement and immediate restoration with a non-occluding provisional crown (IPR); immediate placement and delayed restoration, which includes both early and conventional loading (IPDL); and delayed placement, regardless of loading technique (DP). This review classifies implant placement and implant loading protocols according to the Third International Team of Implantology (ITI) Consensus conference in 2003 [18,19].
The information on survival rates, number of implants placed, loading protocols, setting, location of implants in the jaw, antibiotic protocols, grafting methods, and implant systems were obtained (see Table 2). These parameters were assessed to determine if they influenced the survival rates reported by the studies.

Risk of Bias Assessment
The quality of the included studies was assessed by six independent reviewers (W.T., P.S., P.G.S., J.G., C.T., D.C.) following the Cochrane Risk of Bias tool for randomised controlled trials [20]. This tool encompasses seven criteria: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias. All studies were judged against these criteria as having low, unclear, or high risk of bias. The overall risk of bias was low if all criteria were considered to be at low risk of bias, unclear if there was at least one criterion with unclear risk of bias, and high if there was at least one criterion with a high risk of bias.
[Diagram is in a PDF file as requested by CIDRR author guidelines.]

Results
The first electronic search yielded 6042 citations that could be reviewed. After the abstracts were screened, 5674 of these were rejected. Full text assessment was conducted on 368 studies. A total of 193 articles were excluded: 69 studies were excluded on the grounds of mixed results with multi-unit restorations, 55 were rejected as they did not follow-up implant placement to the minimum 1-year mark, 30 were not longitudinal studies, 13 had a sample size less than six, 14 were published prior to 1 January 1999, 10 studies did not do IIP, one study was not published in English, and one study was not on humans (see Table 4).
Furthermore, 17 additional studies were included after manually searching the reference lists of non-excluded studies. An overall total of 192 longitudinal studies were accepted. A further 164 articles were rejected as they were not randomised controlled trials, and 28 articles were accepted; however, two studies were published twice at two different time points during their experiment so their results were combined [21][22][23][24]. A total of 26 randomised controlled trials were included in this study, with an average follow-up time of 24 months (range 12 to 120 months). A diagram detailing the search strategy is shown in the Figure 1.
Subsequently, the bias was assessed using the Cochrane Risk of Bias Tool for randomised controlled trials. A summary is provided in Table 3.  , and an iatrogenic mistake where the clinician did not provide enough relief between an implant and the provisional denture [43].

Patient Selection Criteria
A list of the patient selection criteria in each study is summarised in Table 5. Most studies required their patients to be systemically healthy (n = 24), with no acute infection, either periapical or periodontal, in the area of implant placement (n = 22) and having an intact tooth socket or sufficient buccal bone after extraction (n = 22). These studies are listed in Table 6.  239]. A further 162 articles were excluded as they were not randomised controlled trials; these articles are not tabulated above but are included in the references [36, 69,77,83,101,107,112,113,118,189,198,210,226,.

Follow-Up
Multiple-Units Sample Size Year of Publication Not a Clinical Study Non-Immediate Implants Not in English Animal Study       List of articles that specified the above patient inclusion criteria.

Systemically Healthy No Acute Periodontal/Peri-Apical Infection in Area Intact Tooth Socket or Sufficient Buccal Bone
Ten studies did not specify inclusion or exclusion of smokers and ten studies included smokers of ≤10 cigarettes a day. Three studies that did not specify smoking and six studies that included smoking ≤10 cigarettes a day achieved 100% survival rates. Three studies excluded smoking completely and three studies included smokers of ≤20 cigarettes a day, and all achieved survival rates >95%. Of the eleven studies achieving 100% survival rates across all categories, most included patients who smoked ≤10 cigarettes a day. Smoking inclusion is tabulated in Table 7.

Loading Protocol
Three studies investigated the effect of IPL implants. Four out of 51 IPL implants failed from the three studies. Only one out of these three studies reported a survival rate of <90%. One of these studies compared IPL and IPDL with survival rates of 96.6% (1 out of 26 implants failed) and 93.3% (2 out of 29 implants failed), respectively [27]. Even though the survival rate for the IPL implants was slightly higher, they did not find that the difference was significant.
One study by Cucchi et al. [44] loaded the implants early where definitive crowns were placed one week after implant placement. Two implants were lost out of the 49 implants that were placed in fresh extraction sockets (95.5% survival rate).
The remaining studies placed definitive restorations 3-6 months after placements (IPR or IPDL implants). All 11 studies that achieved a 100% survival rate had at least one experimental group using IPR or IPDL implants. Twelve studies reported survival rates between 91.3 and 98.9%, and three studies reported survival rates between 80 and 84.6% for IPR or IPDL implants. These are tabulated in Table 8.

Antibiotic Therapy
Most studies used amoxicillin (n = 21). Among them, six studies used amoxicillin with clavulanic acid, six studies allowed substitution with clindamycin, and two studies allowed substitution of amoxicillin with azithromycin [46] or clarithromycin [34] if the patients were allergic to penicillin. All but one of these studies yielded a survival rate greater than 90% [25], with eleven of these studies yielding 100% survival across all categories of placement and loading. One study did not specify which antibiotic they used, but the study yielded 100% survival [40]. Two studies used phenoxymethylpenicillin [48] and cephalosporin [28], respectively, and both studies yielded survival rates below 90%. Two other studies did not use antibiotics, but yielded survival rates of 96.6% and 98.9%. The studies are listed in Table 9.
Twelve studies used antibiotics both pre-operatively and post-operatively. Of these studies, four had 100% survival rates and the rest had survival rates greater than 90%. Four studies only used antibiotics post-operatively, three of which had survival rates of 100% and one with a survival rate less than 90%. Five studies only used antibiotics pre-operatively, four of which had 100% survival rates and one had a survival rate lower than 90%. Three studies only used post-operative antibiotics if a graft was used, and survival rates ranged from 40 to 95.9%. The studies are listed in Table 10.

Setting
Twelve studies were conducted in a private practice (PP), of which nine were multicentre (Mc) studies, eleven studies were conducted in a university setting (U), and three studies did not specify the location of the study (NS).
Of the studies that achieved 100% survival rates for immediate implants, five were conducted in private practices, four were in universities, and one with an unspecified setting. Of the studies with a survival rate of less than 90% for immediate implants, one was conducted in a private practice [25] and had survival rates of 40% for IPL and 80% for IPR implants. The other two did not specify the setting [28,48]. The studies are listed in Table 11.

Location of the Implant
Eighteen RCTs placed implants in the maxilla only, where eleven reported survival rates of 100%, six reported survival rates between 92 and 98.9%, and one RCT in the IPR group reported a survival rate of 84.6% [28]. Two out of the eighteen RCTs placed immediate implants in the maxillary anterior region only. Both studies reported 100% survival rates for IPR and DP implants. Five studies reported survival rates of IPR implants in the aesthetic zone (14-24), where three of the studies reported 100% survival rates and the other two reported 96.7% survival rates. Ten studies also included the second premolars. Six out of the ten RCTs reported 100% survival rates for the IPR group. The lowest survival rate for IPR implants involving maxillary anterior and premolar teeth was 84.6% reported by Block et al. [28]. The study also reported a 96.6% survival rate for the IPDL group. One other RCT only involved the maxillary posteriors and reported a survival rate for IPDL implants of 98.9%.
Eight RCTs involved placement of implants in both arches. Two RCT reported 100% survival rate for both IPDL and DP implants replacing posterior teeth. Cucchi et al. also noted similar results for DP posterior implants; however, the IPDL implants survival rate was only 95.5% [44]. Three other RCTs reported survival rates between 91.3 and 96.6% for IPDL posterior implants. One other study [25] investigated IPL, IPR, and DP implants involving both arches and reported survival rates of 40%, 80%, and 97%, respectively. The lowest survival rate for IPDL posterior implants was 83.7% [48]. The studies are listed on Table 14.

Discussion
This systematic review analysed 26 randomised controlled trials to evaluate the survival rates of immediately placed single implants and describe the reasons for failure.
After analysis of the included articles, the survival rate of immediate single implants, regardless of loading, ranged from 40 to 100% over an average follow-up period of 24 months (range 12-120 months). One study that produced a drastically low survival rate of 40% for IPL and 80% for IPR was excluded as an outlier due to limitations, such as low sample size of IPL implants (n = 5), the use of zirconia implants, and that only one out of the four clinicians involved in the study had experience with zirconia implants [25]. After exclusion of this outlier, the survival rate range is 83.7-100%. Eleven RCTs presented with 100% survival rates in immediate implants. Similar to other reviews [10,11,13], most of the studies that were included in this review reported survival rates above 90%, except for three studies that reported survival rates of immediate implants ranging from 40 to 84.6% [25,28,48].
A previous systematic review on single immediate implants found a survival rate ranging between 94 and 100%, which is higher than our present study [14]. However, they included studies with a follow-up of less than a year. There was also a recent meta-analysis on single immediate implants, which resulted in a survival rate of 94.9% over a follow-up period of 12-96 months [13]. Two other meta-analyses on immediate implants also found similar survival rates, however they did not differentiate the results for single and multipleunits. Lang et al. [10] found a 2-year survival rate of 98.4% and a 4-year survival rate of 97.5%. Mello et al. [9] found a survival rate of 95.2% over a follow-up period of 6 months; however, they did not define 'survival', whereas the other reviews defined survival as the presence of the implant in the mouth, in accordance with the present review.
Most studies reported implant failure before placement of the definitive prosthesis and the main reason was failure of osseointegration, which is consistent with a previous systematic review [13]. Failure of osseointegration is commonly assessed by: clinical mobility of the implant, radiolucency between implant and bone, and the sound when a metal instrument taps the implant [389]. Two studies did not name a reason for implant failure, including Block et al. that reported a survival rate below 90% for IPR implants [28,46]. According to the study by Levin, a failure of osseointegration is typically due to overheating, trauma and contamination during surgery, lack of primary stability, micromotion, and overloading [390].
Of the 11 studies with 100% survival rates for immediate implants, the only consistent trend is the use of titanium implants and an amoxicillin antibiotic regime either preand/or post-operatively [21-24,26,29,31,34,35,37,39,40,47]. The preference for placement of implants in the maxilla and the use of grafting was also observed among these studies. Furthermore, the 100% survival rate trend may also be reflective of additional factors, such as operator experience and reporting bias.
In the study by Urban et al., 15 out of 92 implants failed [48]. A total of 11 out of the 15 failed implants were placed after tooth extraction due to apical periodontitis; however, the study did not utilise post-operative antibiotics nor did they debride the socket prior to the insertion of the implant. Most studies reporting high survival rates in apically infected sites debrided the extraction socket prior to insertion and implemented a post-operative antibiotic therapy to prevent bacterial transmission to the implant site [7,391]. Moreover, Urban et al. did not exclude smokers and used bone grafts in all implants [48], and previous studies have found significantly higher rates of implant failure in patients who smoke cigarettes, especially in those patients with bone grafts [392].
The effect of smoking on implant failure has been well-documented in the literature. Nicotine in cigarette smoke is detrimental to healing and osseointegration of implants [392,393]. On the other hand, smoking did not seem to play a visible role in survival in our current review, as similar survival rates were found in studies that included smokers in the inclusion criteria as compared to those that excluded them. It was noted that most studies either included smokers of less than or equal to ten cigarettes per day (n = 10) or did not exclude smoking (n = 10) (See Table 4). Not excluding smokers makes it difficult to discern the true number of smokers in the study and may hide the true impact of smoking on implant survival rates. Other than smoking, most studies tended to have stringent inclusion criteria and included only optimal situations where the patients are systemically healthy, and the site of implant placement has sufficient bone and lack of infection. The present study did not observe other patterns related to the remaining inclusion criteria.
The setting of a study may indicate the presence of factors that influence the results, such as operator experience and potential reporting bias. Of the included studies, there were an almost equal number of studies from private practices and universities, with three done in unknown settings. If the study by Cannizarro et al. [25] was excluded as an outlier, private practice survival rates ranged from 91.3 to 100%, whereas it ranged from 92 to 100% in universities. As both settings yielded similar results, bias arising from operator experience and selective reporting may be minimal.
Out of the 26 RCTs included in this study, only 3 investigated IPL implants [25,113,375]. Even though IPL implants significantly shorten treatment time, definitive prosthetic rehabilitation of an implant earlier than 3-4 months has not been recommended in the literature as it may jeopardise its stability [25,272,394,395]. Loading the implant with occlusal forces before complete healing can create micromotion which can prevent osseointegration and production of a fibrous scar tissue between the implant and bone [394]. This is supported by the high failure rate of IPL implants reported by Cannizarro et al. [25]. On the contrary, if the study by Cannizarro et al. [25] is excluded as an outlier, this review found that IPL implants have excellent results with survival rates of 96.6-100%. Even though this was concluded from a small sample size of two RCTs [27,113], this is supported by the findings from a previous systematic review that found a survival rate of 95.6% [396].
DP implants that were immediately loaded or loaded one week after placement also showed excellent survival rates, suggesting this may be a suitable treatment alternative. Again, as this was founded only on two RCTs [25,44], there is not enough evidence to corroborate the findings. Nevertheless, other studies have reported similar results [272,397].
Providing immediate non-occluding provisional restoration on the same day as implant placement is more desirable, as the patient does not have to be left without a tooth during the healing period while avoiding overloading the implants. Even though the provisional restorations can introduce micromovements that may interrupt osseointegration [398], immediate provisionalisation of post-extractive implants have been described as a reliable technique [36]. The current review found that IPDL implants have similar survival rates compared to IPR implants, where the survival rates for each group ranged from 83 to 100% and 80 to 100%, respectively.
According to the literature, the highest rate of implant failure was reported with post-extraction implants placed in the posterior maxilla due to poorer bone quality and the location of the base of maxillary sinus, which prevents the implant from achieving primary stability [224,399]. In agreement with this, the included studies that only placed IPDL implants in the posterior regions of both arches reported that most of the lost implants were placed in the posterior maxilla [44,48].
The majority of the RCTs that placed implants in the premaxillary zone achieved 100% survival rates, indicating that implants placed in this area can have predictable outcomes. According to the current review, IPR implants placed in the anterior maxilla only achieved 100% success rates. However, there were only two RCTs that exclusively placed implants in the anterior maxilla [31,37].
Interestingly, of the four studies that did not restrict the location of implant placement, none reported a 100% survival rate [25,27,43,45]. From these four studies, two failed implants replaced the mandibular second premolar [25,27]. The posterior mandible often receives heavy masticatory forces during function which may contribute to failure of these implants [399]. Furthermore, the height between the mandibular canal and the alveolar crest often limits the length of the implant which may be insufficient to achieve primary stability [399]. Hence, careful case selection and planning is required when placing implants in this area. Most of the included studies did not report the location of the lost implants, hence further investigation is required.
There is contradicting evidence on the benefits of antibiotics in implant therapy [10,13,[400][401][402][403][404][405]. Furthermore, problems including antibiotic resistance and allergies can arise. Among the studies in the current review, only two studies [42,43] did not use an antibiotic regimen but both yielded high survival rates (96.6-98.9%). Whereas the studies yielding low survival rates used an antibiotic regimen. Hence, this review could not conclude that higher survival rates are solely due to the use of antibiotics.
There has been evidence that pre-operative antibiotics reduce early implant failure, by reducing the amount of bacteria in the surgical site [403,[406][407][408][409]. On the other hand, the evidence on the benefit of post-operative antibiotics on implant therapy has been unclear [406,408]. In immediate implants, however, two systematic reviews found that the prescription of pre-and post-operative or only post-operative antibiotics significantly reduced early implant failure, especially when compared to use of pre-operative antibiotics only [10,13]. The present review found that half of the studies prescribing antibiotics used it both pre-and post-operatively. It was noted that none of these studies had survival rates below 90%, suggesting that there may be some benefit to this regimen. However, this suggestion should be treated with caution considering that only four of the eleven studies with 100% survival rates used this regimen.
In terms of the type of antibiotic, amoxicillin was the most commonly used (n = 21) and was used by all studies achieving 100% survival rates. Two of the three studies achieving survival rates less than 90% did not use amoxicillin as an antibiotic [28,48]. Hence, when antibiotics are used, amoxicillin may achieve higher survival rates. Another study also found amoxicillin as the most frequently prescribed antibiotic after implant placement [410]. Its efficacy is likely due to its moderate spectrum that encompasses odontogenic bacteria, along with an acceptable dosing schedule for good patient compliance [411]. There is limited data in the present review about the efficacy of alternative antibiotics for IIP in the case of a penicillin allergy. Clindamycin was used as an alternative to amoxicillin in only six studies [25,32,33,35,41,45], and clarithromycin and azithromycin were only used in one study each [34,46].
The use of bone grafts did not appear to influence survival rates in the studies which used bone grafts compared to those that did not. However, there is a consensus in the literature that bone grafts are advantageous in the inhibition of peri-implant bone loss in immediate extraction sockets [412]. When comparing the survival rates of the present included studies using autografts and allografts to other graft materials, xenografts and alloplasts had more consistently high survival rates. The accelerated resorption rates of autografts have made other grafting materials more desirable as they do not completely resorb and so the stability is retained long-term [413]. Another retrospective study using both immediate and delayed implants found the clinical survival rate for autografts to be 94.4-97.9% within a two-year follow-up, compared to 100% for bovine xenografts [414].
In the present study, only 1 in the 26 studies used zirconia implants, which is likely due to the lack of long-term data compared to the well-researched titanium implant [415]. Zirconia is an attractive alternative as it is said to attract less bacterial plaque, produce a low inflammatory infiltrate, and provide good tissue integration, which makes it desirable in limiting peri-implant biological complications [415]. However, a systematic review and meta-analysis with immediate and delayed placement of zirconia implants reported a 92% survival rate after one year whereas titanium implants boasted 97% after five years [416]. It is possible that the use of zirconia implants in the study by Cannizarro et al. [25] contributed to its poor survival rate.
Tapered implants were a popular design choice amongst the studies. The larger diameter threads at the coronal portion compared to the tapping portion led to better bone compaction at the crest at which the implants were placed [417]. This improved primary stability and may prevent early failure as seen in the nine studies using tapered implants, all reaching a survival rate of 100%. Platform-switching was the second most common design and is likely attributed to the numerous studies that have supported the significant reduction in peri-implant marginal bone loss [418,419]. The literature has reported that the survival rates between platform-switching implants and platform-matching implants were comparable and the type of platform was not considered to be the determinant of implant survival [419]. For the studies that used titanium implants only, the survival rates of platform-switched implants ranged from 91 to 100% whilst the survival rates for platform-matched implants ranged from 84 to 100% which suggests that platformswitching may influence survival. Wide diameter implants, defined as ≥4.5 mm (Renouard and Nisand 2006), presented mostly 100% survival rates (seven out of eight studies) [418]. Its ability to close the implant socket gap and engage more bone makes it easier to achieve primary stability, which reduces the need for bone grafts [23,419]. The wide diameter also prevents the implants from being overloaded which can diminish osseointegration [419].
A meta-analysis has reported that wide implants had a strong survival rate of 92% after five years [419].
The limitations of the studies are as follows. First, most of the included studies were of an unclear risk of bias and eight had a high risk of bias. Secondly, other confounding factors such as follow-up period and the patient history could present heterogeneity amongst the included studies. Finally, the small number of samples per placement or loading category makes it difficult to draw definitive conclusions.

Conclusions
This systematic review investigated 26 randomised control trials and found a survival rate of 83.7-100% for single immediately placed implants. Implant failure was not consistently reported and when reported, failure due to lack of osseointegration prior to placement of the definitive restoration was the most common descriptor. Others attributed reasons included infection abscess, mobility after immediate loading, and iatrogenic complication. Several factors may influence the survival of immediate implants, such as loading protocols, location of implants in the jaw, antibiotic protocol, grafting methods, and implant geometry; however, the current literature lacks a large volume of homogenous studies reporting on immediately placed implants and so further investigation is required.