Dental implants have proven to be a reliable treatment modality for replacing missing teeth. The original surgical protocols required the implants to be placed into healed alveolar ridges either as a one-stage (Schroeder et al. 1976) or two-stage surgical procedure (Brånemark et al. 1977). Relatively soon after the introduction of these protocols, a clinical study was published documenting the placement of root-form implants into fresh extraction sockets (Schulte et al. 1978). This led to a phase of development from the mid-1970s to the early 1990s of immediate implant placement after tooth extraction. During the same period, techniques based on the principles of guided bone regeneration or GBR (Dahlin et al. 1988, Dahlin et al. 1989) were applied to promote bone regeneration in the periimplant extraction defects of immediate implants (Lazzara 1989, Becker & Becker 1990, Lang et al. 1994).

The original treatment concept was to load the implants only after extended undisturbed healing periods to ensure that osseointegration had taken place, with healing times of 3 to 6 months typically recommended (Schroeder et al. 1976, Adell et al. 1981, Buser et al. 2000). This was based on the assumption that premature loading of an implant could jeopardize osseointegration (Adell et al. 1981). However, clinical studies soon emerged to challenge this paradigm by demonstrating that immediate loading of implants in fully edentulous sites could be achieved when 4 implants were placed with cross-arch splinting (Ledermann 1979, Babbush et al. 1986). The concept of immediate loading continued to evolve (Schnitman et al. 1990) and by the late 1990s, interest turned to a challenging clinical scenario in which an immediate provisional prosthesis was connected to an implant placed in an fresh extraction socket in the anterior maxilla (Wohrle 1998). 

The evidence gathered until then suggested that immediate implants had survival rates that were comparable to implants placed in healed sites (Schwartz-Arad & Chaushu 1997a, Mayfield 1999). During the same period, preclinical and clinical studies were published to demonstrate histologically that implants could be placed into fresh extraction sockets and successfully achieve osseointegration (Barzilay et al. 1988, Barzilay et al. 1991, Paolantonio et al. 2001).

At the 3rd ITI Consensus Conference 2003 in Gstaad, Switzerland, a systematic review on immediate and early implant placement confirmed that survival rates of immediate and early implants were similar to that in healed ridges (Chen et al. 2004). However the authors noted that there was a lack of data on long-term success and esthetic parameters. At this consensus conference, a classification of 4 categories for the timing of implant placement after extraction (type I – IV) was proposed (Hämmerle et al. 2004), and further modified by the addition of descriptive terminology in the third volume of the ITI Treatment Guide series (Chen & Buser 2008). This classification system has now been widely adopted (Schropp & Isidor 2008, Shi et al. 2015, Huynh-Ba et al. 2018).

In the mid- to late 2000s, the focus turned to esthetic outcomes with immediate implants, since the esthetic zone was the predominant location where this treatment protocol was used and reported on. In a systematic review prepared for the 4th ITI Consensus Conference 2008 in Stuttgart, Germany, the authors concluded that bone augmentation procedures were effective in achieving predictable bone fill and defect resolution in the peri-implant defects at post extraction sites (Chen & Buser 2009). However, esthetic outcomes of immediate (type 1) implant placement based on the positional change of the facial periimplant mucosa were highly variable. 20% to 30% of immediate implants were at risk of mid-facial mucosal recession of 1 mm or more (Fig. 1). The authors concluded that the variability in outcomes could be due to the majority of included studies not having well-defined inclusion and exclusion criteria that took into account the condition of the socket bone walls at the time of extraction.

By the mid- to late 2000s, experimental and clinical data had emerged to show that the facial bone of extraction sockets underwent significant resorption in the first weeks and months after tooth extraction. The “bundle bone theory” proposed that the thin facial wall of an extraction socket is comprised almost entirely of bundle bone. Since this bundle bone is developmental in origin, it naturally undergoes resorption once a tooth is removed. This resorption subsequently leads to dimensional alteration of the ridge and explains the typical resorptive pattern observed on the buccal aspect of the alveolar ridge once a tooth has been extracted (Araujo & Lindhe 2009). It was also observed that the placement of an implant into a fresh extraction socket did not prevent facial bone resorption and the subsequent dimensional alteration of the alveolar ridge (Araujo et al. 2005). Furthermore, studies have confirmed that thin facial bone (<1 mm) of the socket is likely to resorb three times more in an apicocoronal plane than thick facial bone (≥1 mm) when immediate (type 1) implants are placed (Ferrus et al. 2010). Other risk factors include a damaged facial bone wall (Kan et al. 2007), thin periodontal phenotype (Evans & Chen 2008, Cordaro et al. 2009) and inadvertent facial malposition of the implant within the socket at the time of placement (Evans & Chen 2008, Chen et al. 2007).

At the 5th ITI Consensus Conference 2013 in Bern, Switzerland, a systematic review reported on esthetic outcomes of immediate (type 1) and early (type 2 and 3) placement, and the influence that simultaneous bone augmentation procedures may have on the results (Chen & Buser 2014). This review highlighted that, compared to earlier reports, studies published after 2008 began to include stricter selection criteria that included intact facial bone of the socket and a medium-to-thick soft tissue phenotype. The authors also commented that interpretation of data relating to immediate (type 1) implants was difficult due to the heterogeneity of the surgical techniques used, which included a number of adjunctive factors and procedures. These adjunctive factors included type of surgery (flapless versus open flap procedures), bone augmentation (graft versus no graft), type of graft (autogenous, allograft or xenograft), use of connective tissue (CT) for soft tissue augmentation, and the simultaneous connection of a provisional prosthesis (immediate loading). There appeared to be a trend in which studies of immediate (type 1) implant placement that used adjunctive techniques to minimize and compensate for resorption of the facial bone (flapless, bone graft, CT graft and provisional restoration) had less variability in the positional change of the facial mucosa between studies.

In 2018 in Amsterdam, the Netherlands, the 6th ITI Consensus Conference included a systematic review by Gallucci and co-workers that further clarified the outcomes of immediate (type 1) implants (Gallucci et al. 2018). In this review, a new classification system was proposed that combined the implant placement time and implant loading protocols (Table 1).

Thus with immediate implant placement, 3 subcategories were listed:

• Type 1A: immediate placement AND immediate restoration/loading
• Type 1B: immediate placement AND early loading
• Type 1C: immediate placement AND conventional loading

The implant placement and loading protocols were in accordance with the classification established by the ITI. Immediate implant placement (type 1) was defined as the placement of an implant in a fresh extraction socket on the same day as tooth extraction and part of the same procedure (Chen et al. 2004, Hämmerle et al. 2004, Chen & Buser 2009). Immediate loading was defined as the connection of a prosthesis to the implant within one week of implant placement. Early loading was defined as the attachment of a prosthesis between one week and two months after implant placement. Conventional loading was defined as the connection of a prosthesis to the implants more than two months after implant placement (Weber et al. 2009, Gallucci et al. 2014).

The analysis of the survival rates of each of these treatment protocols follows in the next section.