Partial Extraction Therapy: A Review of Human Clinical Studies

Following tooth loss or extraction, significant alveolar ridge alteration, buccal plate resorption, and reduction of soft-tissue volume can occur due to loss of the periodontal ligament (PDL), resulting in insufficient blood supply to the bundle bone.¹  Together, these dimensional changes represent significant challenges for the restorative and implant rehabilitation of edentulous areas. In complex cases in which esthetically demanding patients present with edentulous areas in the anterior esthetic zone, high smile lines, and significant hard- and soft-tissue deficiencies, site development and subsequent dental implant placement can be especially challenging.² 

Over the years, various strategies have been developed to limit postextraction ridge resorption, including alveolar ridge preservation, guided bone regeneration, and immediate implant placement. Although immediate implant placement and ridge preservation are predictable procedures, they do not completely prevent resorption of the buccal bone.³  Several techniques have been proposed to minimize dimensional changes to the facial contour such as incorporation of bone graft materials in the implant-socket gap, subepithelial connective tissue grafts, or bone grafting on the facial aspect of the extraction socket.⁴  These techniques have limited predictability and often require multiple surgical interventions, which can lead to increased cumulative treatment cost and patient morbidity.

Partial extraction therapy (PET) techniques have gained significant popularity. PET comprises a set of surgical techniques including the socket shield technique (SST), root membrane technique (RMT), proximal shield technique (PrST), pontic shield technique (PtST), and root submergence technique (RST), which aim to preserve the periodontium and peri-implant tissues during implant and restorative therapy with the maintenance of the patient's existing root structure and PDL-derived blood supply.^5,6 

First described by Hürzeler et al in 2010,⁷  SST is the most popular and widely studied PET technique. During tooth extraction and subsequent immediate dental implant placement, the clinician leaves behind a root section against the buccal plate to preserve the PDL and associated blood supply, enabling the maintenance of the bundle bone, thus preventing significant hard- and soft-tissue resorption.⁷ 

As SST evolved, Gluckman et al, who coined the term PET, suggested that, if present, a gap between the immediate implant fixture and the shield should always be grafted with particulate bone graft material.^6,8  In contrast, Siormpas et al described the RMT in support of preservation of the PDL and associated vascular supply to the bundle bone. Compared with the SST method, Siormpas et al endorsed the concept that it may not be necessary to graft the space between the root fragment and the immediate implant.⁹  Both the SST and RMT have changed considerably since their initial conception and have become more similar to one another. However, the main difference between these 2 techniques lies in the sequence of shield preparation and implant placement. In the SST, the shield is prepared first and the implant inserted afterward.⁶  For the RMT, the osteotomy for the implant is prepared through the root before preparing the shield.⁹  Ultimately, these techniques achieve the preservation of buccal hard- and soft-tissue architecture through the PDL-mediated vascular supply and bundle bone maintenance.

Other variants of PET include PrST and PtST. The PtST preserves the alveolar ridge at sites intended for pontic development for tooth- or implant-supported prostheses.¹⁰  Alternatively, the PrST in conjunction with immediate implant placement and provisionalization is beneficial in the maintenance of the interimplant papilla when replacing a nonrestorable tooth adjacent to an implant restoration.¹¹ 

For decades, clinicians have attempted to preserve alveolar ridge volume and prevent bone loss by intentionally leaving root remnants.¹²  The concept of a submerged root to maintain PDL and stabilize the alveolar ridge associated with pontic regions of fixed dental prostheses and complete dentures has been repeatedly described since the 1950s, but its application was a controversial topic. For instance, Helsham reported in a study with 2000 patients that 16.2% of retained roots resulted in pathological signs, especially when exposed to the oral environment.¹³  Since the 1970s, numerous studies have explored the submergence of both vital root and endodontically treated roots.^14–18  However, the RST remained controversial until Salama et al revisited and modernized the concept for the creation of an esthetic result in adjacent multiple tooth replacement cases.¹⁹  RST may be used to preserve the alveolar morphology to maintain soft-tissue esthetics for future pontic sites or to preserve the stability and support offered by a denture-bearing residual ridge.

The aim of this review is to provide a comprehensive evaluation of human clinical studies on PET techniques and discuss their clinical benefits and limitations. More specifically, we aim to examine PET with regard to implant osseointegration, dimensional changes in hard and soft tissue, esthetic outcomes, and complications.

A systematic search was conducted in concordance with the PRISMA checklist in the following electronic bibliographic databases: PubMed, EMBASE, and Dentistry & Oral Sciences Source. Gray literature was searched to identify additional candidates for potential inclusion. A manual search of the most relevant journals and reference list of relevant previous publications was performed. No time point restriction was applied. Articles were screened by a group of 4 reviewers (coauthors) using the Covidence software.

The search strategy with detailed key words and Boolean operators are listed as follows: (Socket shield OR tooth socket OR tooth socket OR dental alveolus OR alveolus dentali OR root membrane OR Pontic Shield OR Proximal shield OR partial extraction OR root submergence OR tooth root OR tooth roots OR tooth extraction OR tooth extractions OR Tooth Socket) AND (Dental implant OR dental implants OR surgical dental protheses OR surgical dental prothesis OR dental implantation OR end osseous dental implantation OR osseointegrated dental implantation OR end osseous implantation OR osseointegration OR peri-implant endosseous healing OR peri implant endosseous healing OR peri-implant endosseous healings))

Only human clinical studies were eligible for inclusion. Case reports, prospective or retrospective case series, cohort, case-control studies, and randomized controlled trials with at least 6 months follow-up were included.

In vitro and in vivo preclinical animal studies, studies providing insufficient information on the surgical protocol and follow-up, and articles written in languages other than English, French, Spanish, Portuguese, Chinese, Korean, Farsi, and Vietnamese were excluded.

The title and abstract of all articles were assessed by 2 independent reviewers (D.T.W., T.T.N.) to identify relevant articles for the full-text screening phase. Full-text screening and data extraction were performed by 2 independent groups of researchers to reduce the risk of bias. Any disagreements were resolved through discussions between the researchers involved.

A predetermined data extraction sheet was used by the 2 groups of researchers to independently extract data from each included study to limit bias. Information on the data extraction sheet include author, year of publication, PET technique used, type of study performed, number of subjects, number of sites treated, number of implants placed, tissue dimension change, esthetic outcomes, and complications. A descriptive data synthesis and narrative review using a systematic literature search was performed.

A systematic search of the literature yielded 5714 results. After removal of duplicates and abstract and title screening, 64 articles were selected for full-text, eligibility assessment. Forty-two studies met the inclusion criteria and were included for data extraction. An additional 12 studies were screened and added to the synthesis after a manual search of the reference lists. A total of 54 studies were examined in this review (Figure 1). The studies included are summarized in Tables 1–3.

Because PET consists of a group of surgical techniques, we provide a detailed overview of the current literature and summarize the available human clinical evidence for each of the techniques. The SST was originally reported by Hurzeler et al in 2010.⁷  Although initially limited to use by a select few experienced clinicians, the technique has now begun to achieve widespread popularity in everyday clinical practice (Figure 2). Consequently, the number of publications evaluating SST and other PET variations have steadily increased, with the highest number of studies published in 2018 and 2019 (Table 1). In total, this review compiled 33 studies examining the SST and its variations including the PrST first described by Kan and Rungcharassaeng in 2013¹¹ (Figure 3) and the PtST first introduced by Gluckman et al in 2016¹⁰ (Figure 4).

In addition, this review examines the RMT originally introduced by Siormpas and Mitsias et al in 2014.⁹  Since then, 5 studies have been published, including case series and a retrospective clinical study with follow-up varying from 2 to 5 years (Table 2).^49–52  Finally, root submergence has been reported in the literature for more than 5 decades (Figure 5). A total of 16 case reports and case series were included in this review reporting evidence on both vital root submergence and nonvital root submergence (Table 3).

Articles included in this review were 25 case reports, 19 case series, 5 retrospective studies, 3 prospective case-control studies, and 2 randomized clinical trials. Although most of the publications to date are case studies, PET techniques have gained widespread attention in the clinical arena, increasing the number of available studies to contribute higher-level clinical data in the hierarchy of evidence.

The sample size of clinical studies varied between case reports of single clinical cases and retrospective clinical studies with up to 182 patients and 250 sites examined (Tables 1–3).⁵¹  The follow-up time reported in the studies varied significantly, ranging from 6 months up to 10 years, with most publications stating follow-up times of about 1 year (Tables 1–3).

All 50 studies examining SST, PrST, and RMT reported on implant osseointegration (Tables 1 and 2). Implant success ranged from 87.9% to 100% across the human clinical studies examined. As the follow-up time and cohort size were increased, we observed a trend for a decrease in implant osseointegration and implant success due to complications. For instance, a retrospective case series by Gluckman et al evaluating 128 socket shield cases with up to 4-year follow-up reported 96.10% implant osseointegration.²⁹ 

Similarly, a retrospective clinical study evaluating root membrane cases by Siormpas et al followed 182 patients and 250 implant sites for up to 10 years with a mean follow-up time of 49.94 months. This study reported a 96.5% 10-year cumulative patient-level implant survival rate and an 87.9% 10-year cumulative implant success rate considering mechanical and biological complications.⁵¹  Although complications were reported, a combination of nonsurgical and surgical management was sufficient to prevent the implant loss, explaining the notable discrepancy between the implant success rate (87.9%) and implant survival rate (96.5%).

Most studies reported dimensional changes in hard and soft tissue. More specifically, these reports provided information regarding the preservation of the alveolar ridge architecture and/or soft-tissue buccal to the dental implant or prosthesis. A detailed description of tissue dimension changes documented in each study is provided in the tables (Tables 1–3). A high degree of heterogeneity was observed in the methods used to evaluate dimensional stability. Studies employed a combination of clinical and radiographic assessment tools, including periapical radiographs, cone-beam computerized tomography scans, and digital 3-dimensional scans of dental casts compared with digital superimposition.

Overall, most of the groups reported favorable outcomes for use of the SST and RMT to achieve dimensional stability in the anterior esthetic zone. A randomized clinical trial by Bramanti et al reported an average buccal marginal bone resorption of 0.605 ± 0.06 mm in the SST group at 3 years compared with 1.115 ± 0.131 mm in the control group.²⁸  Similarly, in a subsequent randomized case-control study, Sun et al reported that the SST group exhibited higher buccal bone width and height compared with the control.⁴³  Finally, in a recent RCT, Abd-Elrahman et al reported that the SST group exhibited less horizontal and vertical bone loss compared with the control group, although the amount reported may not be clinically significant.⁴⁵ 

Several studies reported on the soft-tissue esthetic outcomes of PET techniques. Assessments were performed with a variety of methods, including clinical evaluation of soft-tissue stability, gingival recession, and interdental papilla fill, in addition to the patient-perceived outcomes regarding esthetic appearance. The Pink Esthetic Score (PES) was the most commonly used objective evaluator for esthetic outcomes.

Among the articles included, 8 studies reported the esthetic outcomes of PET with PES. Most groups reported significantly higher scores with the intervention compared with controls. For instance, a randomized controlled trial with 40 patients by Bramanti et al compared the PES of SST with immediate implant to conventional implant and found that after a 3-year follow-up period, PES was 12.15 ± 0.87 for the SST group compared with 10.3 ± 1.59 for the control group.²⁸  Similarly, additional studies have reported superior esthetic outcomes using PET in conjunction with immediate implant placement. Zhu et al reported in a case series with 9 patients an average PES of 13.5 after 12–24 months of follow-up.³²  Also, in their case-control study of 12 patients with 1-year follow-up, Xu et al reported that the SST group had a PES of 13.25 ± 0.75 compared with 11.83 ± 0.94 for the control group.⁴⁰ 

Abd-Elrahman et al. reported in an RCT including 40 patients that the PES of the socket shield group increased from 11 to 12, whereas the PES of the control group decreased from 13 to 9.⁴⁵  In contrast, other studies have reported comparable esthetic outcomes comparing PET to conventional methods. In an RCT with 24 months of follow-up, Sun et al. reported that there was no difference in PES.³² 

As for root submergence, most of the studies were either performed before the concept of PES or did not use PES as an evaluator of soft-tissue esthetics. Recent studies have reported good esthetic outcomes such as preservation of harmonious soft-tissue contours.^19,64,65 

Although numerous clinical studies have demonstrated the benefits of PET, case reports have demonstrated that complications can occur but are typically manageable. For SST, the most common complication appears to be shield exposures and mobility/migration. Gluckman et al reported exposure of the shield that required a surgical procedure, as it did not have complete soft-tissue coverage and required buccal flap advancement for closure.¹⁰  Bäumer et al reported the occurrence of soft-tissue recession.⁶⁶  Gluckman et al. reported 19.5% combined complication (25/128 sites). Five of 128 implants failed to osseointegrate and were removed. The remaining 20 complications were managed with implant survival, including 16 shield exposures, 3 infections, and 1 shield migration.²⁹  Stuani et al. reported a radiolucent area around the apical region of the implant and a fenestration. The complication was managed surgically, and the implant survived.³⁹  Zuhr et al. reported a case of shield mobility after 4 years with deep buccal probing depth. The complication was managed, and the implant survived.⁴⁴  In a retrospective case series with 15 sites, Durrani et al. reported a complication of pus coming from shield, which was managed.⁴⁷ 

For RMT, complications reported in the literature include root fragment and implant-associated complications. In a retrospective case series examining 250 sites with up to 10-year follow-up, Siormpas et al. reported 8 complications including root fragment infection with fistula, peri-implant mucositis, peri-implantitis, and implant mobility in the absence of infection.⁵¹  Additional case series have reported localized bleeding on probing, indicating peri-implant mucositis, and apical root resorption of retained root fragment that did not interfere with implant osseointegration.^9,52 

For RST, the most commonly reported complications include dehiscence and incomplete soft-tissue coverage.^{53–55,57,62}  Von Wowvern and Winther reported a high incidence of failures due to the exposure of the coronal root surface through mucosal perforations.⁵⁷  Gongloff reported in a case series examining 122 sites, 26% failures (32/123 sites), divided into 4 categories. Immediate failures consisted of flap breakdowns (20/32; 62.5%), intermediate failures consisted of denture irritations with ulceration and dehiscence (2/32; 6.25%), and late failures consisted of eruption of the root through the oral mucosa and gingiva (10/32; 31.25%).¹⁶  Other complications include the obliteration of labial and lingual sulci in exchange for tension free flap for primary closure with no root lost.¹⁶ 

For vital root submergence, one commonly reported complication is the development of pulpal pathoses requiring endodontic treatment or extraction. Garver et al reported in a case series evaluating 45 sites of vital root submergence, 7 pulpal pathoses requiring root canal therapy, and 1 pulpal pathoses requiring extraction.⁵⁴ 

PET techniques have been used by clinicians for both implant therapy as well as fixed and removable prosthodontics. SST, RMT, and PtST have been used to maintain buccal tissue architecture for immediate implant therapy and associated fixed prosthesis, PrST to maintain interproximal bone and papilla, and RST to maintain alveolar ridge architecture for pontic site development or removable dentures.

SST has evolved significantly since its inception. The original technique published by Hurzeler et al involved creating the implant osteotomy directly through the tooth root, with extraction of the palatal portion occurring after the drilling sequence was completed.⁷  Similarly, the RMT, originally described by Siormpas et al, entailed preparation of the implant osteotomy through the root following decoronation.⁹  The rationale was to preserve the PDL and buccal bundle bone to minimize dimensional changes to the facial contour after extraction. However, these 2 approach have drawbacks. Drilling directly through the root may generate cracks, which can compromise the shield. In addition, implant placement directly against the shield surface leads to compression against the root fragment and may place excessive pressure on the buccal plate. Thus, clinicians began to extract the palatal root fragment prior to immediate implant placement and place the implant toward the palatal aspect of the socket away from the shield while using SST.⁶ 

For a decade, clinicians have adopted many modifications to the original SST. Kan and Rungcharassaeng adapted SST into PrST to preserve the interproximal bone and papilla.¹¹  Gluckman et al. adapted SST into PtST to preserve buccal contour and esthetics for pontic sites.¹⁰  Roe et al. proposed a minimally invasive approach to performing this using a semilunar flap and buccal window to access the apical portion of the root following decoronation.⁴ 

With advancements in digital dentistry, clinicians have incorporated computer-aided design (CAD)/computer-aided manufacturing (CAM) and dynamic digital navigation systems to PET. Saeidi Pour et al. reported the use of CAD/CAM for the fabrication of a surgical guide and immediate fabrication of definitive restoration in a case of PET. This method significantly reduced the number of appointments needed for (1) preoperative planning, (2) surgical treatment, and (3) prosthetic rehabilitation.⁶⁷  In addition, Chen reported the use of a dynamic navigation system with SST, which further enhances the precision of immediate implant placement.⁶⁸ 

In parallel, Salama et al revisited the root submergence concept and applied RST to maintain the natural attachment apparatus of the tooth in pontic sites for complete preservation of the alveolar bone and to assist in the creation of esthetic results in adjacent multiple tooth replacement cases.¹⁹ 

Several factors are important for success. First, the root fragment must have no mobility, no fracture, and no exposure, as these may lead to potential infection and poor esthetics. Other factors to consider include the shape and topography of the alveolar socket, implant position and design, and behavior of the patient.

Because immediate implant placement occurs in a fresh extraction socket as opposed to an intact, edentulous site, there is often a gap between the implant fixture and the residual socket walls. A key decision for clinicians using PET is whether or not to graft the gap between the implant fixture and the retained root fragment and, if so, which graft materials to use. Gluckman et al suggested that, if present, the gap between the implant fixture and the shield should always be grafted with particulate graft material.⁶  In contrast, Siormpas and Mitsias, the pioneers of RMT, support the concept that it is not necessary to graft this space. Their rationale is that this gap will be filled with blood clot, which will remodel into bone following the body's innate ability for wound healing.⁹  Currently, there is limited evidence demonstrating the benefit of grafting the gap between the shield and dental implant. One recent study by Pohl et al. demonstrated the benefit of using autogenous dentin particles or cortical tuberosity bone to graft the shield gap, which resulted in a favorable outcome leading to minimal soft-tissue ingrowth between the socket shield and dental implant.⁶⁹ 

Various graft materials have been assessed in human clinical studies, including allograft, xenograft, and alloplast (Tables 1 and 2). The 2 most common graft materials used are deproteinized bovine bone mineral (BioOss, Geistlich & GenOs, OsteoBiol) and freeze-dried bone allograft (MinerOss, BioHorizons & Puros, Zimmer). Additional materials used include calcium phosphosilicate alloplast bone graft (NovaBone, Halma), enamel matrix derivative (Emdogain, Straumann), and platelet-rich fibrin.^35,42,67  Currently, there is only one study in the literature comparing the clinical outcomes of PET performed with different graft material used to graft the gap.⁶⁹  Thus, this represents a largely unexplored area of research for a future prospective split-mouth clinical study or randomized clinical trial.

Another important technical consideration in PET is the apical coronal position of the residual root fragment relative to the crest. Human clinical studies have reported shield position, above, at, or below the alveolar crest. Creators of the RMT purport that the supracrestal position of the shield preserves the dentogingival fibers and thus maintains soft-tissue esthetics; however, there are no studies comparing the outcomes of shield position. As for the shape and mesiodistal position of the shield, the literature reports a C-shaped shield extending from the mesial to distal line angle.⁵ 

Because most studies in the literature were case reports and case series with small sample sizes and shorter follow-up periods ranging from 6 months to 3 years, the reported osseointegration rates were 100%. As the follow-up time and cohort size were increased, there was a trend for a decrease in implant osseointegration and implant success due to complications. As for changes in tissue dimension, most of the studies reported favorable outcomes for the use of PET to achieve dimensional stability in the anterior esthetic zone. However, many studies did not specify the evaluation method used and subjectively stated that the hard and soft tissue was stabilized. Finally, because of the heterogeneity between study design, sample size, and follow-up time, we cannot compare the esthetic outcome of PET with conventional methods. The review of available human clinical studies suggests that PET may provide patients with a superior esthetic outcome, although randomized clinical trials with a larger patient sample and longer follow-up time are needed to provide more robust evidence.

In terms of complications, there is a trend that the incidence of complications increases with longer follow-up and larger sample size. Fewer complications were reported at short-term follow-up. Root fragment or submerged root exposure was common to all PET techniques. Most of the complications reported in the literature were successfully managed surgically, and the implants' osseointegration was maintained. Management ranged from no treatment/observation in the case of shield exposure with no signs of inflammation to shield reduction in the case of inflammation and extraction of the shield if mobile.

The incidence of implant failure using the SST was reported only by Gluckman et al in a case series in which 128 sites were examined; 5 implants failed to osseointegrate and were removed.²⁹  As for RMT, 5 implant failures were reported by Siormpas et al in a case series examining 250 sites. As for RST, the most common complication reported was dehiscence in the soft tissue.⁵¹  In cases in which vital roots were submerged, the most common complication was the development of pulpal pathoses, which resolved with endodontic treatment.⁵⁴  With PET techniques gaining popularity in the clinical arena, it is paramount for practitioners to be educated about the possible complications and strategies to manage them. This will be corroborated with additional clinical trials. Ultimately, the key to successfully reducing the occurrence of complications may be careful case selection and meticulous execution by experienced clinicians.

Given the various clinical applications of PET techniques in implant dentistry, fixed prosthodontics, and removable prosthodontics, this study contributes a comprehensive review of human clinical studies in the literature. Our limitation is the lack of quantitative analysis of clinical outcomes due the variations in clinical protocol, instrumentation, follow-up duration, and clinical and radiographic documentation. A comparison between PET techniques and their clinical outcomes was not feasible at this time. Moving forward, future randomized controlled clinical studies using standard protocols, rigorous follow-up timelines, and high-quality clinical documentation will enable clinicians and researchers to evaluate the predictability of PET in preserving periodontal and peri-implant tissues compared with other treatment modalities.

PET is an evolving field. The main challenge with PET is the limited amount of evidence with long-term follow-up. In fact, only 5 case series have presented 4- to 5-year follow-up results.^{9,29,37,47,66}  In addition, currently available clinical studies have technical variations and have made modifications to the original reported techniques, making a comparison between studies difficult, which may alter the clinical results and incidence of complications. Thus, for future studies, standardizing the surgical protocol is important for assessing the clinical performance of PET. In addition, the skill level of the clinician may be a subject of study, as PET techniques are highly dependent on the clinician's skill level and experience. Finally, patient-reported outcomes are key to assessing whether PET is truly beneficial. These outcomes may include patients' perceptions of postoperative pain and morbidity as well as esthetic results. Currently, no studies have reported patient-reported outcomes. Future clinical trials assessing these parameters by providing patients with a quality-of-life questionnaire while comparing PET to conventional approaches will be useful for the clinical community.

PET techniques offer clinicians several advantages: (1) preservation of buccal bone postextraction and limitation of alveolar ridge resorption, (2) reduced need for invasive ridge augmentation procedures, and (3) achievement of soft-tissue dimensional stability in the esthetic zone and overall esthetic outcomes. Although numerous clinical studies have demonstrated the benefits of PET, case reports have demonstrated that complications can occur but are largely manageable. Thus, further randomized clinical studies with a larger number of subjects are needed for a better understanding of the clinical effectiveness, complications, and long-term predictability of PET.

The authors would like to thank Dr. Michelle Bass, PhD, MSI, at Harvard Countway Library for her assistance in the literature search.