Autologous platelet concentrates for treating periodontal infrabony defects

Massimo Del Fabbro; Lorena Karanxha; Saurav Panda; Cristina Bucchi; Jayakumar Nadathur Doraiswamy; Malaiappan Sankari; Surendar Ramamoorthi; Sheeja Varghese; Silvio Taschieri

doi:10.1002/14651858.CD011423.pub2

. 2018 Nov 26;2018(11):CD011423. doi: 10.1002/14651858.CD011423.pub2

Autologous platelet concentrates for treating periodontal infrabony defects

Massimo Del Fabbro ^1,^2,^✉, Lorena Karanxha ¹, Saurav Panda ^1,³, Cristina Bucchi ⁴, Jayakumar Nadathur Doraiswamy ⁵, Malaiappan Sankari ⁵, Surendar Ramamoorthi ⁶, Sheeja Varghese ⁵, Silvio Taschieri ^1,²

Editor: Cochrane Oral Health Group

PMCID: PMC6517213 PMID: 30484284

Abstract

Background

Periodontal disease is a condition affecting tooth‐supporting tissues (gingiva, alveolar bone, periodontal ligament, and cementum), with the potential of introducing severe adverse effects on oral health. It has a complex pathogenesis which involves the combination of specific micro‐organisms and a predisposing host response. Infrabony defects are one of the morphological types of alveolar bone defects that can be observed during periodontitis. Recent approaches for the treatment of infrabony defects, combine advanced surgical techniques with platelet‐derived growth factors. These are naturally synthesized polypeptides, acting as mediators for various cellular activities during wound healing. It is believed that the adjunctive use of autologous platelet concentrates to periodontal surgical procedures produces a better and more predictable outcome for the treatment of infrabony defects.

Objectives

To assess the effects of autologous platelet concentrates (APC) used as an adjunct to periodontal surgical therapies (open flap debridement (OFD), OFD combined with bone grafting (BG), guided tissue regeneration (GTR), OFD combined with enamel matrix derivative (EMD)) for the treatment of infrabony defects.

Search methods

Cochrane Oral Health's Information Specialist searched the following databases: Cochrane Oral Health's Trials Register (to 27 February 2018); the Cochrane Central Register of Controlled Trials (CENTRAL; 2018, Issue 1) in the Cochrane Library (searched 27 February 2018); MEDLINE Ovid (1946 to 27 February 2018); Embase Ovid (1980 to 27 February 2018); and LILACS BIREME Virtual Health Library (from 1982 to 27 February 2018). The US National Institutes of Health Ongoing Trials Register (ClinicalTrials.gov) and the World Health Organization International Clinical Trials Registry Platform were searched for ongoing trials on 27 February 2018. No restrictions were placed on the language or date of publication when searching the electronic databases.

Selection criteria

We included randomised controlled trials (RCTs) of both parallel and split‐mouth design, involving patients with infrabony defects requiring surgical treatment. Studies had to compare treatment outcomes of a specific surgical technique combined with APC, with the same technique when used alone.

Data collection and analysis

Two review authors independently conducted data extraction and risk of bias assessment, and analysed data following Cochrane methods. The primary outcomes assessed were: change in probing pocket depth (PD), change in clinical attachment level (CAL), and change in radiographic bone defect filling (RBF). We organised all data in four groups, each comparing a specific surgical technique when applied with the adjunct of APC or alone: 1. APC + OFD versus OFD, 2. APC + OFD + BG versus OFD + BG, 3. APC + GTR versus GTR, and 4. APC + EMD versus EMD.

Main results

We included 38 RCTs. Twenty‐two had a split‐mouth design, and 16 had a parallel design. The overall evaluated data included 1402 defects. Two studies were at unclear overall risk of bias, while the remaining 36 studies had a high overall risk of bias.

1. APC + OFD versus OFD alone

Twelve studies were included in this comparison, with a total of 510 infrabony defects. There is evidence of an advantage in using APC globally from split‐mouth and parallel studies for all three primary outcomes: PD (mean difference (MD) 1.29 mm, 95% confidence interval (CI) 1.00 to 1.58 mm; P < 0.001; 12 studies; 510 defects; very low‐quality evidence); CAL (MD 1.47 mm, 95% CI 1.11 to 1.82 mm; P < 0.001; 12 studies; 510 defects; very low‐quality evidence); and RBF (MD 34.26%, 95% CI 30.07% to 38.46%; P < 0.001; 9 studies; 401 defects; very low‐quality evidence).

2. APC + OFD + BG versus OFD + BG

Seventeen studies were included in this comparison, with a total of 569 infrabony defects. Considering all follow‐ups, as well as 3 to 6 months and 9 to 12 months, there is evidence of an advantage in using APC from both split‐mouth and parallel studies for all three primary outcomes: PD (MD 0.54 mm, 95% CI 0.33 to 0.75 mm; P < 0.001; 17 studies; 569 defects; very low‐quality evidence); CAL (MD 0.72 mm, 95% CI 0.43 to 1.00 mm; P < 0.001; 17 studies; 569 defects; very low‐quality evidence); and RBF (MD 8.10%, 95% CI 5.26% to 10.94%; P < 0.001; 11 studies; 420 defects; very low‐quality evidence).

3. APC + GTR versus GTR alone

Seven studies were included in this comparison, with a total of 248 infrabony defects. Considering all follow‐ups, there is probably a benefit for APC for both PD (MD 0.92 mm, 95% CI ‐0.02 to 1.86 mm; P = 0.05; very low‐quality evidence) and CAL (MD 0.42 mm, 95% CI ‐0.02 to 0.86 mm; P = 0.06; very low‐quality evidence). However, given the wide confidence intervals, there might be a possibility of a slight benefit for the control. When considering a 3 to 6 months and a 9 to 12 months follow‐up there were no benefits evidenced, except for CAL at 3 to 6 months (MD 0.54 mm, 95% CI 0.18 to 0.89 mm; P = 0.003; 3 studies; 134 defects). No RBF data were available.

4. APC + EMD versus EMD

Two studies were included in this comparison, with a total of 75 infrabony defects. There is insufficient evidence of an overall advantage of using APC for all three primary outcomes: PD (MD 0.13 mm, 95% CI ‐0.05 to 0.30 mm; P = 0.16; 2 studies; 75 defects; very low‐quality evidence), CAL (MD 0.10 mm, 95% CI ‐0.13 to 0.32 mm; P = 0.40; 2 studies; 75 defects; very low‐quality evidence), and RBF (MD ‐0.60%, 95% CI ‐6.21% to 5.01%; P = 0.83; 1 study; 49 defects; very low‐quality evidence).

All studies in all groups reported a survival rate of 100% for the treated teeth. No complete pocket closure was reported. No quantitative analysis regarding patients' quality of life was possible.

Authors' conclusions

There is very low‐quality evidence that the adjunct of APC to OFD or OFD + BG when treating infrabony defects may improve probing pocket depth, clinical attachment level, and radiographic bone defect filling. For GTR or EMD, insufficient evidence of an advantage in using APC was observed.

Plain language summary

Autologous platelet concentrates for treating periodontal infrabony defects

Review question

Does the addition of autologous platelet concentrates (APC) improve surgical treatment outcomes of bone defects in gum disease?

Background

Teeth are maintained in their position by soft and hard tissues (gums and surrounding bone). Gum disease or periodontitis, is an inflammatory condition of all these tissues caused by the bacteria present in the dental plaque. If left untreated, gum disease can cause teeth to loosen and eventually lead to tooth loss. The destruction of jaw bone around teeth (called the alveolar bone) during gum disease, can be horizontal (where the whole level of bone around the root is reduced) or vertical, forming a bone defect within the bone (infrabony defect). There are several available surgical treatments for infrabony defects, including: 1. open flap debridement in which the gum is lifted back surgically in order to clean the deep tartar; 2. bone graft in which a portion of natural or synthetic bone is placed in the area of bone loss; 3. guided tissue regeneration in which a small piece of membrane‐like material is placed between the bone and gum tissue in order to keep the gum tissue from growing into the area where the bone should be; and 4. the use of enamel matrix derivative, a gel‐like material which is placed in the area where bone loss has occurred and promotes its regeneration. In order to accelerate the healing process, autologous platelet concentrates have been recently used. They are concentrates of the platelets of patient's own blood containing growth factors that are thought to promote tissue regeneration. The aim of this review was to assess if the addition of APC brings any benefits in the treatment of infrabony defects when combined with different surgical treatments.

Study characteristics

Authors from Cochrane Oral Health carried out this review and the evidence is up to date to 27 February 2018. We included 38 studies and a total of 1042 infrabony defects. We considered four different types of surgical treatments and compared each technique with the same one when APC was added. Overall we considered these comparisons: open flap debridement with APC versus without APC; open flap debridement and bone graft with APC versus without APC; guided tissue regeneration with APC versus without APC; and enamel matrix derivative with APC versus without APC.

Key results

There is very low‐quality evidence that the addition of APC to two types of treatment: open flap debridement and open flap debridement with bone graft, may bring some advantages in the treatment of infrabony defects. However, for the other two types of treatment, guided tissue regeneration and enamel matrix derivative, there is insufficient evidence of a benefit.

Quality of evidence

We judged the quality of the evidence to be very low due to problems with the design of the studies.

Summary of findings

Background

Description of the condition

Periodontitis is a disease of the periodontium characterized by the irreversible loss of connective tissue attachment and supporting alveolar bone (Pihlstrom 2005). For its onset, the presence of specific micro‐organisms together with an altered response of the host, are necessary. Despite its many variations, a typical course of periodontitis starts with pocket formation induced by bacterial plaque and a subsequent alveolar bone destruction typical of chronic periodontitis. Bone destruction during periodontitis can be of different morphological patterns including suprabony (horizontal) defects and infrabony (vertical) defects (Kinane 2001). An infrabony defect represents the anatomic sequelae resulting from the apical advancement of the dental plaque during the progression of the disease (Waerhaug 1979). Such defects, if left untreated, easily promote periodontitis progression and further loss of attachment (Papapanou 1991). Because infrabony defects are common in periodontitis (Vrotsos 1999), there is a considerable interest in approaches that will convert such defects, at risk for disease progression, to easily maintainable shallow probing sites (Crea 2014).

Description of the intervention

The ultimate goal of periodontal therapy is to preserve the natural dentition for as long as possible and enhance patient's comfort and aesthetic features by maintaining and improving the health and function of all tooth‐supporting tissues (gingiva, periodontal ligament, cementum, alveolar bone). Conventional treatment of periodontal disease may arrest bone destruction but usually does not restore the already lost alveolar bone or periodontal connective tissue. Various surgical techniques have been developed as an attempt to provide an efficient treatment to periodontitis. Open flap debridement (OFD) is among the earliest and most promising procedures to be used (Caffesse 1986; Cortellini 1996). Its main objective is to reduce the presence of micro‐organisms which develop and maintain the inflammatory process. By doing so, it consequently promotes the regeneration properties of the host, despite not being a regenerative procedure. Later, the combination of conventional OFD with various biomaterials such as bone grafts, enamel matrix derivative or membranes (guided tissue regeneration), resulted in the development of regenerative treatment protocols which introduced significant clinical benefits (Cochran 2003; Cortellini 1996; Esposito 2009; Hoidal 2008; Needleman 2006).

Despite advances in surgical procedures and materials, a complete and predictable regeneration, defined as the development of new bone, periodontal ligament and cementum on a root surface previously exposed to periodontal disease, remains a challenge (AAP 1992). Consequently, the concept of tissue engineering (Rose 2002) which requires the presence of cells, scaffold and signalling molecules, gained particular attention in terms of periodontal regeneration. Bone grafts and membranes used in guided tissue regeneration (GTR) can serve as scaffolds but there always exists a need of signalling molecules.

Recently, polypeptide growth factors have been investigated as possible signalling factors for enhancing periodontal regeneration. As preliminary evidence for their potential applications in periodontal wound healing, several polypeptide growth factors have been identified in the human periodontal tissues by immuno‐histochemistry and in‐situ hybridisation (Giannobile 1996). An abundant source of such growth factors are platelets, easily utilisable in the form of autologous platelet concentrates (APC). Therefore, the adjunctive use of APC in combination with periodontal surgery has emerged as a possible tool to enhance the predictability of infrabony defects treatment.

APCs based on their preparation protocol, can be of various types, including platelet‐rich plasma (PRP) (Marx 1998), platelet‐rich fibrin (PRF) (Choukroun 2001), and plasma‐rich growth factors (PRGF) (Anitua 2001). Several commercial techniques for obtaining platelet concentrates are available. However, their indication of use has been confusing because each method leads to a different product with different biological properties and possible applications. PRP represents the first generation of platelet concentrate, and shows a release of an array of growth factors for 7 days, with a peak release on its first day of application (Dohan Ehrenfest 2009). PRF represents the second generation APC, and its technique of preparation is simplified when compared to PRP. Moreover, PRF showed a sustained growth factors release for a period of 21 days with a peak release at 7 days (Carroll 2005). PRGF is also a second generation platelet concentrate, whose main difference when compared to PRP is the absence of leucocytes and the small blood volume required for its preparation (Anitua 2001). Following an upgrade in their classification (Dohan Ehrenfest 2009), platelet concentrates can be divided into four categories, based on the presence of leucocyte and fibrin: P‐PRP (pure PRP, without leucocytes, which includes PRGF), L‐PRP (leucocyte and platelet‐rich plasma), P‐PRF (pure PRF), and L‐PRF (leucocyte PRF).

How the intervention might work

The contribution of blood‐derived platelets to the bone healing process is thought to be based on the growth factors stored in their granules and released upon activation. The main growth factors released from platelet aggregates are the following: platelet derived growth factor (PDGF), transforming growth factor‐beta (TGF‐β), vascular endothelial growth factor (VEGF), epithelial growth factor (EGF), insulin‐like growth factor‐1 (IGF‐1), and basic fibroblast growth factor (bFGF), as well as three blood proteins known to act as cell adhesion molecules for osteo‐conduction (fibrin, fibronectin and vitronectin). The set of these factors serve as biological mediators with the ability to regulate cell proliferation, chemotaxis, and differentiation.

Why it is important to do this review

The considerably increased interest in combining APC with surgical techniques for better outcomes in the treatment of infrabony defects, has made it necessary a thorough investigation of the actual benefits that can be obtained. The first systematic review that evaluated the effect of PRP on clinical applications in dentistry reported beneficial effects of PRP in the treatment of periodontal defects (Plachokova 2008). Another systematic review that evaluated the effect of a PRP adjunct in treatment of intraosseous defects, underlined the limits and the heterogeneity of available data and cautiously concluded that the specific selection of the graft type and the surgical procedures combined with PRP may be important (Kotsovilis 2010). A subsequent systematic review also evaluated the effect of platelet rich plasma in various regenerative procedures of periodontal defects, and concluded that PRP may be advantageously used as an adjunct to grafting procedures treatment for infrabony defects (Del Fabbro 2011). Such review also suggested that the use of PRP is ineffective when GTR procedure is used for treating infrabony defects.

Despite the numerous reports on the adjunctive use of autologous platelet concentrate to periodontal surgical procedures, its efficacy remains controversial. This is partly due to a large heterogeneity among different studies (Del Fabbro 2011; Del Fabbro 2013), concerning methods, study design, protocols for platelet concentrate preparation, participants selection criteria, outcome variables assessed, etc. Therefore, a review of the current state of the evidence is crucial in order to clarify if the adjunct of APCs eventually produces better outcomes in the treatment of infrabony defects, and if their effect is particularly enhanced when combined with a specific surgical technique. By doing so, clear and relevant guidelines can be addressed to clinicians.

Objectives

To assess the effects of autologous platelet concentrates used as an adjunct to periodontal surgical therapies (open flap debridement (OFD), OFD combined with bone grafting, guided tissue regeneration, OFD combined with enamel matrix derivative) for the treatment of infrabony defects.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials, of both parallel and split‐mouth design.

Types of participants

Patients affected by infrabony defects requiring surgical treatment, regardless of their age or gender.

Types of interventions

Experimental intervention: autologous platelet concentrates (APCs) (irrespective of the type: platelet‐rich plasma (PRP), plasma‐rich growth factors (PRGF), or platelet‐rich fibrin (PRF)) used in conjunction with a specific surgical technique (open flap debridement (OFD), OFD + bone grafts (BG), guided tissue regeneration (GTR), enamel matrix derivative (EMD)).

Comparison (control) intervention: the same surgical techniques when used alone (without the adjunct of APCs).

Types of outcome measures

Primary outcomes

Change in probing depth (PD), change in clinical attachment level (CAL), and change in radiographic bone defect filling (RBF).

Secondary outcomes

Tooth survival, pocket closure, and oral health‐related quality of life.

Search methods for identification of studies

Electronic searches

Cochrane Oral Health's Information Specialist conducted systematic searches in the following databases for randomised controlled trials and controlled clinical trials. There were no language, publication year or publication status restrictions:

Cochrane Oral Health's Trials Register (searched 27 February 2018) (Appendix 1);
Cochrane Central Register of Controlled Trials (CENTRAL; 2018, Issue 1) in the Cochrane Library (searched 27 February 2018) (Appendix 2);
MEDLINE Ovid (1946 to 27 February 2018) (Appendix 3);
Embase Ovid (1980 to 27 February 2018) (Appendix 4);
LILACS BIREME Virtual Health Library (Latin American and Caribbean Health Science Information database; 1982 to 27 February 2018) (Appendix 5).

Subject strategies were modelled on the search strategy designed for MEDLINE Ovid. Where appropriate, they were combined with subject strategy adaptations of the highly sensitive search strategy designed by Cochrane for identifying randomised controlled trials and controlled clinical trials as described in the Cochrane Handbook for Systematic Reviews of Interventions Chapter 6 (Lefebvre 2011).

Searching other resources

The following trial registries were searched for ongoing studies:

US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (clinicaltrials.gov; searched 27 February 2018) (Appendix 6);
World Health Organization International Clinical Trials Registry Platform (apps.who.int/trialsearch; searched 27 February 2018) (Appendix 7).

An adjunctive search was performed on the reference lists of the included articles and reviews retrieved.

Moreover, a handsearch was performed on the issues since January 2010 (including the 'early view' or equivalent section) of the following journals: International Journal of Periodontics and Restorative Dentistry, Journal of Clinical Periodontology, Journal of Periodontal Research, Journal of Periodontology, Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontology (last search was performed on 2 March 2018). Two review authors independently performed the searches (Saurav Panda (SP), Cristina Bucchi (CB)).

We also searched for grey literature, such as conference abstracts, proceedings and theses on the following databases: www.greylit.org; www.opengrey.eu (last search was performed on 2 March 2018, see Appendix 8).

Data collection and analysis

Selection of studies

Following the electronic search, two review authors (Jayakumar Nadathur Doraiswamy (JND), Malaiappan Sankari (MS)) independently screened the titles and abstracts (if available) to exclude all articles clearly not meeting the inclusion criteria. The search was designed to be sensitive and include controlled clinical trials, these were filtered out early in the selection process if they were not randomised. Of all the remaining articles, full texts were obtained and assessed independently by two review authors (JND, MS) and only articles fully meeting the inclusion criteria were considered. In cases of disagreement between the two review authors, a third review author (Massimo Del Fabbro (MDF)) was consulted. Detailed reasons were stated for all excluded studies. This process is summarised in Figure 1.

Data extraction and management

Three review authors (SP, Lorena Karanxha (LK), CB) independently extracted and recorded data on ad hoc forms. Any disagreement was solved through discussion, or a third review author was consulted (MDF). In case of missing or unclear information, we contacted the authors of the included reports by email to provide clarification or missing information. In case of missing or incomplete data and absence of further clarification by study authors we excluded the report from the analysis.

We recorded the following data for each included report:

demographic characteristics of the population;
defect characteristics (PD, CAL, RBF);
type of platelet concentrate used (PRP, PRF, PRGF);
outcome characteristics (outcome variables assessed such as CAL and PD, follow‐up duration);
when possible, we also recorded the expertise of the clinician (years of experience with using platelet concentrates); and
source of funding.

Assessment of risk of bias in included studies

Three review authors (LK, SP, CB) independently assessed the risk of bias in the included studies. In case of disagreement a fourth review author (MDF) was consulted. Since some of the authors of one of the randomised controlled trials included (Panda 2016) are also authors of this review (SP, MDF, Silvio Taschieri (ST)), the risk of bias assessment for that study was carried out by other review authors not involved in the study (LK, CB).

The assessment was conducted following the instructions and the approach described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). For each study, the following domains were considered: selection bias (random sequence generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data addressed), and reporting bias (selective reporting).

For each domain the risk was judged either low, unclear or high. If one study had low risk for all domains, the study was judged at low risk of bias. If it had an unclear risk for at least one domain, the study was judged at unclear risk of bias. If it had a high risk for at least one domain, the study was judged at high risk of bias. It was considered that blinding of patient and clinician might be difficult/impossible, as for many studies involving surgical procedures where interventions are quite different from each other.

We categorised the overall risk of bias of individual studies. Studies were categorised as being at low, high, or unclear risk of bias according to the following criteria:

low risk of bias (plausible bias unlikely to seriously alter the results) if all domains were at low risk of bias;
high risk of bias (plausible bias that seriously weakens confidence in the results) if one or more domains were at high risk of bias; or
unclear risk of bias (plausible bias that raises some doubt about the results) if one or more domains were at unclear risk of bias.

These assessments are reported in the Characteristics of included studies table and also graphically.

Measures of treatment effect

For continuous outcomes (e.g. PD, CAL, RBF), mean differences (change score) along with 95% confidence intervals (CIs) were used to summarise data for each treatment group. We expressed the data in mm for PD and CAL and in percentage for RBF, as they were reported in the studies.

Unit of analysis issues

The statistical unit of analysis in parallel studies was the patient, unless the study provided data only for defects. We considered one infrabony defect per patient in studies with parallel design. In the case of split‐mouth studies, the unit of analysis was the defect; a single defect per patient per group was considered.

Dealing with missing data

In case of missing data, we contacted the corresponding author of the article through e‐mail to obtain complete data. In case of no response, the same e‐mail was sent to co‐authors for a maximum of three times. If no answer was obtained, the study was excluded from the analysis. When feasible, missing standard deviations were estimated using the methods described in Section 7.7.3 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Assessment of heterogeneity

Heterogeneity among studies was assessed with Cochran's test for heterogeneity, with a significance threshold of P < 0.1. The quantification of the heterogeneity was calculated with I² statistic. For the interpretation of I² the ranges suggested in Section 9.5.2 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) were considered.

Assessment of reporting biases

We assessed publication bias by testing for funnel plot asymmetry, as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). If asymmetry was evident, we investigated this and described possible causes.

Data synthesis

The meta‐analysis was performed only with studies with similar comparisons reporting the same outcome measures. We combined mean differences for continuous data, using random‐effects models if at least four studies were included in the meta‐analysis, while if there were less than four studies a fixed‐effect model was chosen. The software RevMan 5 (Review Manager 2014) was used for meta‐analysis computations. Data from split‐mouth and parallel‐group studies were combined (Elbourne 2002). The appropriate standard errors were estimated where they were not present in the trial reports (Follmann 1992). For the split‐mouth studies the standard error was calculated assuming an intraclass correlation coefficient of 0. The generic inverse variance procedure in RevMan 5 was used to combine these two subgroups in the analyses.

Subgroup analysis and investigation of heterogeneity

In addition to the different surgical protocols for different types of infrabony defects, duration of the follow‐up was investigated as a factor possibly affecting the outcome. The subgroups included data up to 6 months (3 to 6 months) and longer than 6 months (9 to 12 months).

Sensitivity analysis

Sensitivity analysis was performed in order to evaluate the effect of risk of bias and source of funding on the overall effects (e.g. omitting studies at unclear or high risk of bias or those sponsored by the manufacturer of the product under investigation). The effect of excluding specific studies that eventually appeared to be outliers was also investigated.

Summary of findings

We produced a 'Summary of findings' table for each comparison in which there was more than one study. We included the change in PD, CAL and RBF of the all follow‐up periods of each comparison group. We used GRADE methods, and GRADEpro software (GRADEpro GDT 2015) for developing 'Summary of findings' tables. We assessed the quality of the body of evidence for each comparison and outcome by considering the overall risk of bias of the included studies, the directness of the evidence, the inconsistency of the results, the precision of the estimates, and the risk of publication bias. We categorised the quality of each body of evidence as high, moderate, low, or very low.

Results

Description of studies

Results of the search

The electronic search retrieved 855 records, four trials were identified by handsearching and none by searching the grey literature. After discarding the duplicates, two review authors (Jayakumar Nadathur Doraiswamy (JND), Malaiappan Sankari (MS)) screened 480 titles and abstracts and rejected 402. The full text was obtained for 78 potentially eligible articles and of these, 40 were excluded with reasons (see Characteristics of excluded studies table). Finally, after agreement among the review authors 38 studies were included in this review (Figure 1).

Included studies

Design

Of the 38 included studies, 22 had a split‐mouth design, reporting for a total of 371 participants and 701 teeth (Agarwal 2014; Agarwal 2015; Agarwal 2016; Arabaci 2017; Aydemir 2016; Camargo 2009; Christgau 2006; Elgendy 2015; Gupta 2014; Hanna 2004; Hassan 2012; Kaushick 2011; Khosropanah 2015; Naqvi 2017; Ozdemir 2012; Panda 2016; Patel 2017; Ravi 2017; Rosamma Joseph 2012; Sezgin 2017; Shukla 2016; Thorat 2017); 16 studies had a parallel design with a total of 645 patients and 721 teeth (Chandradas 2016; Demir 2007; Döri 2007a; Döri 2007b; Döri 2008a; Döri 2008b; Döri 2009; Garg 2017; Kanoriya 2016; Martande 2016; Okuda 2005; Piemontese 2008; Pradeep 2015; Pradeep 2016; Sharma 2011; Thorat 2011). Of the 38 included studies only one was a multicentric study (Elgendy 2015). Finally, two studies declared that they were supported in part by companies whose products were used in the trials (Döri 2008a; Döri 2008b).

Sample size calculation was reported only by 15 studies (Döri 2007a; Döri 2007b; Döri 2008a; Döri 2008b; Döri 2009; Kanoriya 2016; Panda 2016; Patel 2017; Pradeep 2015; Pradeep 2016; Ravi 2017; Rosamma Joseph 2012; Sezgin 2017; Sharma 2011; Thorat 2011), meaning that in almost 60% of cases there was no rationale regarding the choice of the sample size.

Participants

The age range of the participants of included studies was between 17 and 74 years. However, four studies did not report the age of the participants (Agarwal 2016; Gupta 2014; Naqvi 2017; Shukla 2016) and 10 studies (Agarwal 2015; Aydemir 2016; Chandradas 2016; Demir 2007; Elgendy 2015; Hassan 2012; Khosropanah 2015; Okuda 2005; Ozdemir 2012; Sezgin 2017) reported only mean ages, ranging from 36.03 and 55.5 years.

35 studies included both men and women, but with different proportions, and three studies did not report this information (Gupta 2014; Elgendy 2015; Kaushick 2011). Finally, most of the studies did not include smokers (Agarwal 2014; Agarwal 2015; Agarwal 2016; Arabaci 2017; Aydemir 2016; Chandradas 2016; Döri 2007a; Döri 2007b; Döri 2008a; Döri 2008b; Döri 2009; Garg 2017; Gupta 2014; Hassan 2012; Kanoriya 2016; Kaushick 2011; Khosropanah 2015; Martande 2016; Naqvi 2017; Okuda 2005; Ozdemir 2012; Panda 2016; Patel 2017; Piemontese 2008; Pradeep 2015; Pradeep 2016; Ravi 2017; Rosamma Joseph 2012; Sezgin 2017; Sharma 2011; Shukla 2016; Thorat 2011; Thorat 2017).

Interventions

The general comparison was between a group that received autologous platelet concentrates (APC) as an adjunct to surgical treatment (experimental group), and a group that received surgical treatment alone (control group). Four different types of comparisons were assessed, based on the treatment type:

APC + open flap debridement (OFD) versus OFD alone (12 trials): Agarwal 2016; Arabaci 2017; Chandradas 2016; Kanoriya 2016; Martande 2016; Patel 2017; Pradeep 2015; Pradeep 2016; Rosamma Joseph 2012; Sharma 2011; Thorat 2011; Thorat 2017
APC + OFD + bone graft (BG) versus OFD + BG (17 trials): Agarwal 2014; Agarwal 2015; Demir 2007; Döri 2009; Elgendy 2015; Garg 2017; Gupta 2014; Hanna 2004; Hassan 2012; Kaushick 2011; Khosropanah 2015; Naqvi 2017; Okuda 2005; Ozdemir 2012; Piemontese 2008; Sezgin 2017; Shukla 2016
APC + guided tissue regeneration (GTR) versus GTR (7 trials): Camargo 2009; Christgau 2006; Döri 2007a; Döri 2007b; Döri 2008a; Panda 2016; Ravi 2017
APC + enamel matrix derivative (EMD) versus EMD (2 trials): Aydemir 2016; Döri 2008b.

Outcomes

Primary outcomes

Change in probing depth (PD), reported by all 38 included studies.
Change in clinical attachment level (CAL), defined relative attachment level (RAL) in some studies, reported by all 38 included studies.
Change in radiographic bone defect filling (RBF), reported by 31 studies.

Secondary outcomes

All articles in all groups reported a survival rate of 100% for the treated teeth. No complete pocket closure was reported. No quantitative analysis regarding patients' quality of life was possible.

Excluded studies

We excluded 40 studies from the review, for the following reasons (see Characteristics of excluded studies table):

no randomisation (Aleksić 2008; Jovicić 2013; Saini 2011)
no control group (Camargo 2002; Camargo 2005; Lekovic 2012)
gingival recession, not infrabony defects (Aroca 2009; Dogan 2015; Huang 2005; Jankovic 2010; Padma 2013; Shepherd 2009; Shivakumar 2016; Thamaraiselvan 2015)
same patients reported in a previous study (Cetinkaya 2014; Döri 2013; Moder 2012; Yajamanya 2017)
non‐independence of analysing unit (Gupta 2014b; Pradeep 2012a)
incomplete data (Cieplik 2018; Harnack 2009; Keceli 2008; Keles 2006; Menezes 2012; Shah 2015; Yassibag‐Berkman 2007; Yen 2007)
no APCs (fibrin glue) (Cortellini 1995; Trombelli 1995; Trombelli 1996)
APC not the only difference between groups (Cheung 2004; Eren 2014; Jankovic 2012)
studies with mixed (parallel/split‐mouth) design (Agarwal 2017; Bajaj 2017; Chatterjee 2017; Ouyang 2006; Pradeep 2017; Qiao 2016).

Risk of bias in included studies

The risk of bias in included studies is summarized in Figure 2 and Figure 3. Two studies were at unclear overall risk of bias (Ravi 2017; Rosamma Joseph 2012). The remaining 36 studies had a high overall risk of bias.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Random sequence generation

The randomisation was performed correctly in most of the studies. The methods used were the tossing of a coin (Agarwal 2014; Agarwal 2015; Camargo 2009; Demir 2007; Gupta 2014; Hanna 2004; Khosropanah 2015; Okuda 2005; Ozdemir 2012; Panda 2016; Patel 2017; Piemontese 2008; Ravi 2017; Rosamma Joseph 2012; Sezgin 2017; Sharma 2011; Thorat 2011), the block approach (Döri 2007a; Döri 2007b; Döri 2008a; Döri 2008b; Döri 2009), the use of a freeware link (Chandradas 2016), computerized generated scheme (Aydemir 2016; Kanoriya 2016; Martande 2016; Pradeep 2016; Shukla 2016; Thorat 2011), biased coin randomisation (Hassan 2012), lottery method (Naqvi 2017), and a table of random numbers (Christgau 2006; Pradeep 2015). The randomisation method was not described in five articles, which were considered to be an unclear risk of bias (Agarwal 2016; Elgendy 2015; Garg 2017; Gupta 2014; Kaushick 2011).

Allocation concealment

The concealment of the allocation was correctly done in 19 studies (Arabaci 2017; Aydemir 2016; Camargo 2009; Chandradas 2016; Christgau 2006; Demir 2007; Döri 2007a; Döri 2007b; Döri 2008a; Döri 2008b; Döri 2009; Khosropanah 2015; Okuda 2005; Ozdemir 2012; Panda 2016; Patel 2017; Piemontese 2008; Ravi 2017; Rosamma Joseph 2012). In the remaining 19 studies, insufficient information was provided regarding the exact method used for allocation concealment (Agarwal 2014; Agarwal 2015; Agarwal 2016; Elgendy 2015; Garg 2017; Gupta 2014; Hanna 2004; Hassan 2012; Kanoriya 2016; Kaushick 2011; Martande 2016; Naqvi 2017; Pradeep 2015; Pradeep 2016; Sezgin 2017; Sharma 2011; Shukla 2016; Thorat 2011; Thorat 2017).

Blinding

Blinding of participants and personnel (performance bias)

Being the intervention surgical in nature, blinding of participants and treating clinicians is almost unfeasible either in a parallel or split‐mouth design: 36 out of 38 studies had a high risk of performance bias. For two studies an unclear risk of performance bias was assigned given that it was stated in the paper that blinding of the operator was performed but without specifying how (Ravi 2017; Rosamma Joseph 2012). The blinding of the personnel was also evaluated, which was reported in most of the studies except for eight studies (Agarwal 2016; Christgau 2006; Elgendy 2015; Garg 2017; Gupta 2014; Kaushick 2011; Okuda 2005; Ozdemir 2012). However, again for the fact that the intervention has a surgical nature, it is unlikely that blinding or not of the personnel could influence the outcome. Therefore such parameter did not influence the assignment of the risk of performance bias.

Blinding of outcome assessment (detection bias)

The blinding of the outcome assessor was done in most of the studies. However, it was not reported in seven studies, which were considered to be at unclear risk of detection bias (Agarwal 2016; Elgendy 2015; Garg 2017; Gupta 2014; Kaushick 2011; Martande 2016; Ozdemir 2012).

Incomplete outcome data

The completeness of outcome data was adequate in all but three studies in which the number of subjects that finished the study was not clear (Elgendy 2015; Garg 2017; Gupta 2014).

Selective reporting

All studies properly reported data for all patients.

Effects of interventions

See: Table 1; Table 2; Table 3; Table 4

Summary of findings for the main comparison. APC + OFD compared to OFD (9‐12 months follow‐up) for treating periodontal infrabony defects.

APC + OFD compared to OFD (9‐12 months follow‐up) for treating periodontal infrabony defects
Patient or population: patients affected by infrabony defects requiring surgical treatment  Settings: tertiary care  Intervention: APC + OFD  Comparison: OFD
Outcomes	*Illustrative comparative risks^ (95% CI)**		Relative effect  (95% CI)	Number of participants/defects  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	OFD	APC + OFD
Change in probing depth (PD) (mm) (9‐12 months follow‐up)	Mean PD change (gain) across control groups ranged from 2.40 to 3.68 (2.36) mm Mean PD baseline value was 7.92 mm (95% CI 6.25 to 9.54)	The mean PD change (gain) in the intervention groups was 1.29 mm higher (1.00 to 1.58 higher)	Mean difference 1.29 (1.00 to 1.58) mm	510  (12 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is evidence of an advantage in using APC
Change in clinical attachment level (CAL) (mm) (9‐12 months follow‐up)	Mean CAL change (gain) across control groups ranged from 1.27 to 4.14 (2.03) mm Mean CAL baseline value was 6.78 mm (95% CI 5.56 to 7.54)	The mean CAL change (gain) in the intervention groups was 1.47 mm higher (1.11 to 1.82 higher)	Mean difference 1.47 (1.11 to 1.82) mm	510  (12 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is evidence of an advantage in using APC
Change in radiographic bone defect filling (RBF) (%) (9‐12 months follow‐up)	Mean RBF change (gain) across control groups ranged from ‐3.60% to 54.20% (16.90%)	The mean RBF change (gain) in the intervention groups was 34.26% higher (30.07 to 38.46 higher)	Mean difference 34.26% (30.07 to 38.46)	401  (9 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is evidence of an advantage in using APC
^The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  APC: autologous platelet concentrates; CAL: clinical attachment level; CI: confidence interval; OFD: open flap debridement; PD: probing depth; RBF: radiographic bone defect filling.
GRADE Working Group grades of evidence  High quality: further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: we are very uncertain about the estimate.

APC + OFD + BG compared to OFD + BG (all follow‐ups) for treating periodontal infrabony defects
Patient or population: patients affected by infrabony defects requiring surgical treatment  Settings: tertiary care  Intervention: APC + OFD + BG  Comparison: OFD + BG
Outcomes	*Illustrative comparative risks^ (95% CI)**		Relative effect  (95% CI)	Number of participants/defects  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	OFD + BG	APC + OFD + BG
Change in probing depth (PD) (mm) (All follow‐ups)	Mean PD change (gain) across control groups ranged from 1.90 to 5.30 (3.54) mm Mean PD baseline value was 7.32 mm (95% CI 5.94 to 8.65)	The mean PD change (gain) in the intervention groups was 0.54 mm higher (0.33 to 0.75 higher)	Mean difference 0.54 (0.33 to 0.75) mm	569  (17 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is evidence of an advantage in using APC
Change in clinical attachment level (CAL) (mm) (All follow‐ups)	Mean CAL change (gain) across control groups ranged from 1.30 to 4.70 (3.20) mm Mean CAL baseline value was 7.34 mm (95% CI 5.21 to 9.82)	The mean CAL change (gain) in the intervention groups was 0.72 mm higher (0.43 to 1.00 higher)	Mean difference 0.72 (0.43 to 1.00) mm	569  (17 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is evidence of an advantage in using APC
Change in radiographic bone defect filling (RBF) (%) (All follow‐ups)	Mean RBF change (gain) across control groups ranged from 9.20% to 57.20% (40.54%)	The mean RBF change (gain) in the intervention groups was 8.10% higher (5.26 to 10.94 higher)	Mean difference 8.10% (5.26 to 10.94)	420  (11 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is evidence of an advantage in using APC
^The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  APC: autologous platelet concentrates; BG: bone graft; CAL: clinical attachment level; CI: confidence interval; OFD: open flap debridement; PD: probing depth; RBF: radiographic bone defect filling.
GRADE Working Group grades of evidence  High quality: further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: we are very uncertain about the estimate.

APC + GTR compared to GTR (all follow‐ups) for treating periodontal infrabony defects
Patient or population: patients affected by infrabony defects requiring surgical treatment  Settings: tertiary care  Intervention: APC + GTR  Comparison: GTR
Outcomes	*Illustrative comparative risks^ (95% CI)**		Relative effect  (95% CI)	Number of participants/defects  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	GTR	APC + GTR
Change in probing depth (PD) (mm) (All follow‐ups)	Mean PD change (gain) across control groups ranged from 3.19 to 6.00 mm (4.40 mm) Mean PD baseline value was 8.67 mm (95% CI 6.29 to 10.31)	The mean PD change (gain) in the intervention groups was 0.92 mm higher (‐0.02 lower to 1.86 higher)	Mean difference 0.92 mm (‐0.02 to 1.86)	248  (7 studies)	⊕⊝⊝⊝  very low^{1, 2, 3}	There is insufficient evidence of an advantage in using APC
Change in clinical attachment level (CAL) (mm) (All follow‐ups)	Mean CAL change (gain) across control groups ranged from 3.38 to 5.20 mm (4.38 mm) Mean CAL baseline value was 9.40 mm (95% CI 5.97 to 11.40)	The mean CAL change (gain) in the intervention groups was 0.42 mm higher (‐0.02 lower to 0.86 higher)	Mean difference 0.42 mm (‐0.02 to 0.86)	248  (7 studies)	⊕⊝⊝⊝  very low^{1, 2, 3}	There is insufficient evidence of an advantage in using APC
^The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  APC: autologous platelet concentrates; CAL: clinical attachment level; CI: confidence interval; GTR: guided tissue regeneration; OFD: open flap debridement; PD: probing depth; RBF: radiographic bone defect filling.
GRADE Working Group grades of evidence  High quality: further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: we are very uncertain about the estimate.

APC + EMD compared to EMD (all follow‐ups) for treating periodontal infrabony defects
Patient or population: patients affected by infrabony defects requiring surgical treatment  Settings: tertiary care  Intervention: APC + EMD  Comparison: EMD
Outcomes	*Illustrative comparative risks^ (95% CI)**		Relative effect  (95% CI)	Number of participants/defects  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	EMD	APC + EMD
Change in probing depth (PD) (mm) (All follow‐ups)	Mean PD change (gain) across control groups ranged from 3.87 to 5.90 mm (4.89 mm)	The mean PD change (gain) in the intervention groups was 0.13 mm higher (‐0.05 lower to 0.30 higher)	Mean difference 0.13 mm (‐0.05 to 0.30)	75  (2 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is insufficient evidence of an advantage in using APC
Change in clinical attachment level (CAL) (mm) (All follow‐ups)	Mean CAL change (gain) across control groups ranged from 3.30 to 5.00 mm (4.15 mm)	The mean CAL change (gain) in the intervention groups was 0.10 mm higher (‐0.13 lower to 0.32 higher)	Mean difference 0.10 mm (‐0.13 to 0.32)	75  (2 studies)	⊕⊝⊝⊝  very low^{1, 2}	There is insufficient evidence of an advantage in using APC
Change in radiographic bone defect filling (RBF) (%) (All follow‐ups)	Only 1 study reported RBF outcome with a mean change in control groups of 18.30%	The mean RBF change (gain) in the intervention group was 0.60% lower (‐6.21 lower to 5.01 higher)	Mean difference ‐0.60 (‐6.21 to 5.01)	49  (1 study)	⊕⊝⊝⊝  very low^{1, 2}	There is insufficient evidence of an advantage in using APC
^The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  APC: autologous platelet concentrates; CAL: clinical attachment level; CI: confidence interval; EMD: enamel matrix derivative; OFD: open flap debridement; PD: probing depth; RBF: radiographic bone defect filling.
GRADE Working Group grades of evidence  High quality: further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: we are very uncertain about the estimate.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	12	510	Mean Difference (Random, 95% CI)	1.29 [1.00, 1.58]
1.1 Split‐mouth studies	5	158	Mean Difference (Random, 95% CI)	1.86 [1.07, 2.66]
1.2 Parallel studies	7	352	Mean Difference (Random, 95% CI)	0.99 [0.90, 1.07]
2 Clinical attachment level (mm)	12	510	Mean Difference (Random, 95% CI)	1.47 [1.11, 1.82]
2.1 Split‐mouth studies	5	158	Mean Difference (Random, 95% CI)	2.36 [1.19, 3.54]
2.2 Parallel studies	7	352	Mean Difference (Random, 95% CI)	0.99 [0.84, 1.14]
3 Radiographic bone defect filling (%)	9	401	Mean Difference (Random, 95% CI)	34.26 [30.07, 38.46]
3.1 Split‐mouth studies	2	49	Mean Difference (Random, 95% CI)	27.32 [20.92, 33.72]
3.2 Parallel studies	7	352	Mean Difference (Random, 95% CI)	35.77 [31.20, 40.35]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	17	569	Mean Difference (Random, 95% CI)	0.54 [0.33, 0.75]
1.1 Split‐mouth studies	12	360	Mean Difference (Random, 95% CI)	0.47 [0.24, 0.71]
1.2 Parallel studies	5	209	Mean Difference (Random, 95% CI)	0.81 [0.58, 1.03]
2 Clinical attachment level (mm)	17	569	Mean Difference (Random, 95% CI)	0.72 [0.43, 1.00]
2.1 Split‐mouth studies	12	360	Mean Difference (Random, 95% CI)	0.67 [0.35, 0.99]
2.2 Parallel studies	5	209	Mean Difference (Random, 95% CI)	0.89 [0.49, 1.29]
3 Radiographic bone defect filling (%)	11	420	Mean Difference (Random, 95% CI)	8.10 [5.26, 10.94]
3.1 Split‐mouth studies	8	270	Mean Difference (Random, 95% CI)	7.73 [4.50, 10.97]
3.2 Parallel studies	3	150	Mean Difference (Random, 95% CI)	9.66 [5.39, 13.94]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	11	272	Mean Difference (Random, 95% CI)	0.62 [0.30, 0.94]
1.1 Split‐mouth studies	10	252	Mean Difference (Random, 95% CI)	0.58 [0.25, 0.92]
1.2 Parallel studies	1	20	Mean Difference (Random, 95% CI)	0.84 [0.60, 1.07]
2 Clinical attachment level (mm)	11	272	Mean Difference (Random, 95% CI)	0.47 [0.11, 0.84]
2.1 Split‐mouth studies	10	252	Mean Difference (Random, 95% CI)	0.40 [0.02, 0.77]
2.2 Parallel studies	1	20	Mean Difference (Random, 95% CI)	1.0 [0.93, 1.07]
3 Radiographic bone defect filling (%)	6	162	Mean Difference (Random, 95% CI)	4.76 [1.27, 8.25]
3.1 Split‐mouth studies	5	142	Mean Difference (Random, 95% CI)	3.59 [0.13, 7.05]
3.2 Parallel studies	1	20	Mean Difference (Random, 95% CI)	10.0 [4.90, 15.10]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	10	381	Mean Difference (Random, 95% CI)	0.50 [0.31, 0.69]
1.1 Split‐mouth studies	6	192	Mean Difference (Random, 95% CI)	0.49 [0.26, 0.72]
1.2 Parallel studies	4	189	Mean Difference (Random, 95% CI)	0.58 [0.09, 1.06]
2 Clinical attachment level (mm)	6	192	Mean Difference (Random, 95% CI)	0.84 [0.62, 1.06]
2.1 Split‐mouth studies	6	192	Mean Difference (Random, 95% CI)	0.84 [0.62, 1.06]
3 Radiographic bone defect filling (%)	6	282	Mean Difference (Random, 95% CI)	9.99 [6.44, 13.55]
3.1 Split‐mouth studies	4	152	Mean Difference (Random, 95% CI)	10.16 [6.18, 14.14]
3.2 Parallel studies	2	130	Mean Difference (Random, 95% CI)	8.87 [1.03, 16.71]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	7	248	Mean Difference (Random, 95% CI)	0.92 [‐0.02, 1.86]
1.1 Split‐mouth studies	4	166	Mean Difference (Random, 95% CI)	1.52 [0.54, 2.51]
1.2 Parallel studies	3	82	Mean Difference (Random, 95% CI)	0.25 [‐0.15, 0.64]
2 Clinical attachment level (mm)	7	248	Mean Difference (Random, 95% CI)	0.42 [‐0.02, 0.86]
2.1 Split‐mouth studies	4	166	Mean Difference (Random, 95% CI)	0.67 [0.20, 1.14]
2.2 Parallel studies	3	82	Mean Difference (Random, 95% CI)	0.09 [‐0.32, 0.50]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	3	134	Mean Difference (Random, 95% CI)	1.07 [‐0.71, 2.86]
1.1 Split‐mouth studies	3	134	Mean Difference (Random, 95% CI)	1.07 [‐0.71, 2.86]
2 Clinical attachment level (mm)	3	134	Mean Difference (Random, 95% CI)	0.54 [0.18, 0.89]
2.1 Split‐mouth studies	3	134	Mean Difference (Random, 95% CI)	0.54 [0.18, 0.89]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	5	164	Mean Difference (Random, 95% CI)	0.68 [‐0.66, 2.02]
1.1 Split‐mouth studies	2	82	Mean Difference (Random, 95% CI)	1.53 [‐0.85, 3.91]
1.2 Parallel studies	3	82	Mean Difference (Random, 95% CI)	0.25 [‐0.15, 0.64]
2 Clinical attachment level (mm)	5	164	Mean Difference (Random, 95% CI)	0.27 [‐0.39, 0.93]
2.1 Split‐mouth studies	2	82	Mean Difference (Random, 95% CI)	0.51 [‐0.72, 1.73]
2.2 Parallel studies	3	82	Mean Difference (Random, 95% CI)	0.09 [‐0.32, 0.50]

Methods	Trial design: randomised, split‐mouth trial Location: Department of Operative Dentistry and Periodontology, University of Regensburg, Germany Number of centres: 1 Recruitment period: not stated Source of funding: reported, Robert Mathys Foundation, Bettlach, Schweiz Ethical approval: ethical committee of the medical facility of University of Regensburg Number of surgeons: 1
Participants	Inclusion criteria: patient having 1 pair of contralateral deep, intrabony, inter‐proximal periodontal defects with a PPD of at least 6 mm, radiographic evidence of angular bone loss of at least 4 mm at baseline, none of the defects to show furcation involvement  Exclusion criteria: not meeting the inclusion criteria  Age at baseline: 26 to 62 years (median 42 years)  Gender: F 15/M 10  Smokers: 5 patients (smoking 8 cigarettes per day)  Teeth treated: not specified  Number randomised (participants/teeth): 25/50  Number evaluated (participants/teeth): 25/50
Interventions	Comparison: β‐TCP/GTR + APC versus β‐TCP/GTR Test group: β‐TCP/GTR + APC (n = 25) Control group : β‐TCP/GTR (n = 25) Surgical technique: intrabony defects were treated with β‐TCP/GTR bioresorbable barrier membrane at control site and APC was additionally applied on test group Follow‐up duration: 12 months
Outcomes	Clinical: papillary bleeding index, approximal plaque index, CAL, gingival recession, PPD, depth of osseous defect Radiographic: digital subtraction radiography ‐ bone density Other: vertical relative attachment gain
Notes	Sample size calculation: not reported Radiographs were taken at baseline and at end of follow‐up to analyse by digital subtraction radiography Comparability at baseline: yes, assessed Complications reported: yes Dropouts: reported, no dropouts
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "For randomized treatment allocation, a randomizing table was created by our mathematician (K‐AH) using the SPSS software (Version 13.0, SPSS Inc., Chicago, IL, USA)" Comment: likely to have been done properly
Allocation concealment (selection bias)	Low risk	Quote: "The randomization table comprised the patient numbers (1–25) and the corresponding defect numbers (1 and 2) per patient. The therapy methods (test or control) were randomly allocated to the defect numbers. By entering the study, the patient numbers were consecutively allocated to the patients and the defect numbers were allocated to the 2 teeth to be treated. Treatment allocation was concealed to the surgeon until the beginning of the surgery" Comment: likely to have been done properly
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Impossible to blind the operator given the surgical nature of the treatment
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "Clinical examination was performed by 2 masked examiners .." Comment: blinding of outcome assessment done properly
Incomplete outcome data (attrition bias)   All outcomes	Low risk	All randomised patients completed the study
Selective reporting (reporting bias)	Low risk	All data were properly reported

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Probing depth (mm)	2	75	Mean Difference (Random, 95% CI)	0.13 [‐0.05, 0.30]
1.1 Split‐mouth studies	1	49	Mean Difference (Random, 95% CI)	0.13 [‐0.05, 0.31]
1.2 Parallel studies	1	26	Mean Difference (Random, 95% CI)	‐0.10 [‐1.32, 1.12]
2 Clinical attachment level (mm)	2	75	Mean Difference (Random, 95% CI)	0.10 [‐0.13, 0.32]
2.1 Split‐mouth studies	1	49	Mean Difference (Random, 95% CI)	0.12 [‐0.12, 0.36]
2.2 Parallel studies	1	26	Mean Difference (Random, 95% CI)	‐0.2 [‐1.06, 0.66]
3 Radiographic bone defect filling (%)	1	49	Mean Difference (Random, 95% CI)	‐0.6 [‐6.21, 5.01]
3.1 Split‐mouth studies	1	49	Mean Difference (Random, 95% CI)	‐0.6 [‐6.21, 5.01]

Methods	Trial design: randomised, parallel trial Location: Department of Periodontics, Sree Mookambika Institute of Dental Sciences, India Number of centres: 1 Recruitment period: not specified Source of funding: nil Ethical approval: institutional ethics committee Number of surgeons: 1
Participants	Inclusion criteria: systemically healthy patients diagnosed with chronic periodontitis based on the international workshop for the classification of periodontal disease, having ≥ 20 teeth and ≥ 30% of sites with > 4 mm clinical attachment loss, PD ≥ 5 mm, and presence of intrabony defect ≥ 3 mm (measured from alveolar crest to the base of the defect on intraoral periapical radiograph) Exclusion criteria: patients with use of tobacco or tobacco‐related products; systemic or local application of antibiotics within the previous 6 months; patients with poor oral hygiene (PI ≥ 3) after the revaluation of cause‐related therapy  Age at baseline: 44.4 years  Gender: F 18/M 18  Smokers: excluded  Teeth treated: maxilla and mandible  Number randomised (participants/teeth): 36/36  Number evaluated (participants/teeth): 36/36
Interventions	Comparison: group A, PRF + DBM; group B, PRF alone; and group C, control (OFD) Test groups: PRF + DBM (n = 12), PRF alone (n = 12) Control group: OFD (n = 12) Surgical technique: OFD Follow‐up duration: 9 months
Outcomes	Clinical: GI, GR, PD, relative attachment level was measured from apical border of the stent to the base of the pocket Radiographic: linear bone growth and percentage in bone fill
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Allotment of participants within the groups was performed randomly by creating a randomization list by means of a freeware link (http://www.graphad.com/quickcalcs/randomize1.cfm)" Comment: likely to have been done properly
Allocation concealment (selection bias)	Low risk	Quote: "The treatment allocation of the patients was prepared and sealed in the numbered opaque envelopes and were opened during surgery immediately after completing the defect debridement. Allocation protocol was unavailable to the periodontal examiner (RS) throughout the study" Comment: correct method for allocation concealment
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Impossible to blind the operator given the surgical nature of the treatment. The patients were blinded to their treatment group allocation. However blinding of the patients is unlikely to influence treatment outcome again because of the surgical nature of the treatment
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "The pre‐ and postoperative assessments were performed by another examiner (RS) without knowledge of the nature of intervention" Comment: blinding done correctly
Incomplete outcome data (attrition bias)   All outcomes	Low risk	All randomised patients completed the study
Selective reporting (reporting bias)	Low risk	Data for outcomes of this review were reported appropriately

Methods	Trial design: randomised, split‐mouth, double‐blinded trial Location: The University of Texas Health Science Center at Houston, Texas, USA Number of centres: 1 Recruitment period: not reported Source of funding: not stated Ethical approval: yes, Committee for the Protection of Human Subjects at the University of Texas Health Science Center at Houston, Texas, USA Number of surgeons: 1
Participants	Inclusion criteria: patients between 35 to 75 years of age; exhibited plaque score of 20% or less prior to the surgical phase; teeth with mobility less than  Miller's Class III or mobile teeth requiring splinting; and teeth responding normally to vitality testing or with stable endodontic therapy  Exclusion criteria: known systemic diseases and/or drug therapy known to interfere with wound healing; known drug allergies to any of the medications  used in the study; using systemic antibiotics or having received antibiotic therapy in the last 3 months; abnormal platelet counts disclosed by a complete  blood count (CBC) test performed within 1 month prior to surgery; and participation in other dental clinical trials  Age at baseline: 37 to 74 years  Gender: F 8/M 5  Smokers: yes, 1 heavy smoker (> 20 cigarettes/day)  Number randomised (participants/teeth): 13/26  Number evaluated (participants/teeth): 13/26
Interventions	Comparison: BDX + PRP versus BDX alone Test group: BDX + PRP (n = 13 defects) Control group: BDX alone (n = 13 defects) Surgical technique: OFD + intrabony defects treated with BDX alone in control group and additionally PRP was applied in test group Follow‐up duration: 6 months
Outcomes	Clinical: GI, PI, PD, CAL, recession as the position of the gingival margin from the CEJ, and BOP Radiographic: none reported Other: none
Notes	Sample size calculation: not reported Comparability at baseline: yes, assessed Complications reported: yes Dropouts: reported, no dropouts
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization was performed immediately following defect debridement by the flip of a coin" Comment: correct method for random sequence generation
Allocation concealment (selection bias)	Unclear risk	Quote: "Randomization was performed immediately following defect debridement by the flip of a coin" Comment: not sufficient information provided regarding the method of allocation concealment
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Impossible to blind the operator due to the surgical nature of the treatment
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "13 patients were enrolled in a randomized, split‐mouth, double‐masked clinical trial" Comment: blinding of outcomes likely to have been done properly
Incomplete outcome data (attrition bias)   All outcomes	Low risk	All randomised patients completed the study
Selective reporting (reporting bias)	Low risk	All outcomes properly reported

Methods	Trial design: randomised, longitudinal, triple‐masked, parallel trial Location: Department of Periodontics, Government Dental College and Research Institute, Bangalore, India Number of centres: 1 Recruitment period: November 2013 to July 2014 Source of funding: not stated Ethical approval: Institutional Ethical Committee and Review Board of the Government Dental College and Research Institute, Bangalore, India Number of surgeons: 1
Participants	Inclusion criteria: presence of intrabony defect ≥ 3 mm deep (distance between alveolar crest and base of the defect on an intraoral periapical radiograph (IOPA)) along with an interproximal PD ≥ 5 mm after phase I therapy (scaling and root planing) in asymptomatic maxillary/mandibular molar teeth  Exclusion criteria: aggressive periodontitis patients; patients with systemic conditions known to affect the periodontal status; medications known to affect the outcomes of periodontal therapy; haematological disorders and insufficient platelet count (< 200,000/mm³); pregnancy/lactation; smoking and tobacco use in any form; and immunocompromised individuals. Those having unacceptable oral hygiene (PI > 1.5) after re‐evaluation of phase I therapy were also excluded. In addition, teeth with furcation defects, non‐vital teeth, carious teeth warranting restorations and mobility of at least grade II were also excluded  Age at baseline: mean = 41 years  Gender: F 68/M 68 (for all 4 groups)  Smokers: excluded  Teeth treated: maxillary and mandibular molar  Number randomised (participants/teeth): 64/64 (126/126 for all 4 groups; 136 eligible but 10 excluded at time of surgery)  Number evaluated (participants/teeth): 60/60 (120/120 for all 4 groups)
Interventions	Comparison: OFD alone versus OFD + PRF Group 1: OFD alone (n = 30) Group 2: OFD + PRF (n = 30) Surgical technique: in group 1, only OFD was done, without addition of any regenerative material into the bone defect; in group 2, PRF of the required size was filled into the intrabony defect after OFD Follow‐up duration: 9 months
Outcomes	Clinical: site specific PI, modified sulcus bleeding index, relative attachment level, gingival marginal level Radiographic: radiographic intrabony defect depth Other: none
Notes	Sample size calculation: reported Radiographs were taken with a bite block for ensuring reproducibility Comparability at baseline: yes, assessed Complications reported: yes Dropouts: reported, 4 dropouts (6 for all 4 groups) 3rd and 4th group data (OFD + 1% metformin and OFD + PRF + 1% metformin) not included in this review
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "These patients were divided randomly (computer generated tables) into 4 groups" Comment: random sequence generation properly done
Allocation concealment (selection bias)	Unclear risk	Insufficient information provided for allocation concealment method
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Impossible to blind the operator given the surgical nature of the treatment. For the same reason, blinding of the patient, even though it was done, does not influence the outcome of the treatment
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quotes: "This was a randomized, single‐centre, longitudinal, triple‐masked (investigators, individuals and statistician), parallel arm design study" and "One operator (KN) performed all the surgeries, whereas another operator (ARP) performed all the clinical and radiographic measurements without knowledge of the groups" Comment: blinding of outcome assessment properly done
Incomplete outcome data (attrition bias)   All outcomes	Low risk	4 dropouts (6 for all 4 groups)
Selective reporting (reporting bias)	Low risk	All outcomes properly reported

Study	Reason for exclusion
Agarwal 2017	Mixed design ‐ randomised controlled trial
Aleksić 2008	No randomisation
Aroca 2009	Gingival recession (not infrabony defects)
Bajaj 2017	Mixed design ‐ randomised controlled trial
Camargo 2002	No control group
Camargo 2005	No control group
Cetinkaya 2014	Same participants of Keles 2006
Chatterjee 2017	Mixed design ‐ randomised controlled trial
Cheung 2004	Autologous platelet concentrates not the only difference between groups
Cieplik 2018	Incomplete data
Cortellini 1995	No platelet concentrate (fibrin glue)
Dogan 2015	Gingival recession (not infrabony defects)
Döri 2013	Same participants of Döri 2008b
Eren 2014	Autologous platelet concentrates not the only difference between groups
Gupta 2014b	Non‐independence of analysis unit
Harnack 2009	Incomplete data
Huang 2005	Gingival recession (not infrabony defects)
Jankovic 2010	Gingival recession (not infrabony defects)
Jankovic 2012	Autologous platelet concentrates not the only difference between groups
Jovicić 2013	No randomisation
Keceli 2008	Incomplete data
Keles 2006	Incomplete data
Lekovic 2012	No control group
Menezes 2012	Incomplete data
Moder 2012	Same participants of Christgau 2006
Ouyang 2006	Mixed design ‐ randomised controlled trial
Padma 2013	Gingival recession (not infrabony defects)
Pradeep 2012a	Non‐independence of analysis unit
Pradeep 2017	Mixed design ‐ randomised controlled trial
Qiao 2016	Mixed design ‐ randomised controlled trial
Saini 2011	No randomisation
Shah 2015	Incomplete data
Shepherd 2009	Gingival recession (not infrabony defects)
Shivakumar 2016	Gingival recession (not infrabony defects)
Thamaraiselvan 2015	Gingival recession (not infrabony defects)
Trombelli 1995	No platelet concentrate (fibrin glue)
Trombelli 1996	No platelet concentrate (fibrin glue)
Yajamanya 2017	Same participants of Chatterjee 2017
Yassibag‐Berkman 2007	Incomplete data
Yen 2007	Incomplete data

Draft the protocol	Massimo Del Fabbro, Saurav Panda
Develop a search strategy	Saurav Panda, Surendar Ramamoorthi
Search for trials	Saurav Panda, Cristina Bucchi
Obtain copies of trials	Massimo Del Fabbro
Select which trials to include (two + one arbiter)	Jayakumar Nadathur Doraiswamy, Malaiappan Sankari, Massimo Del Fabbro
Extract data from trials (three people)	Lorena Karanxha, Saurav Panda, Cristina Bucchi
Enter data into Review Manager	Massimo Del Fabbro, Lorena Karanxha, Cristina Bucchi
Carry out the analysis	Massimo Del Fabbro, Lorena Karanxha, Saurav Panda, Sheeja Varghese, Cristina Bucchi
Interpret the analysis	Sheeja Varghese, Jayakumar Nadathur Doraiswamy, Silvio Taschieri
Draft the final review	Massimo Del Fabbro, Lorena Karanxha, Malaiappan Sankari, Cristina Bucchi
Update the review	Massimo Del Fabbro, Lorena Karanxha, Saurav Panda

PERMALINK

Autologous platelet concentrates for treating periodontal infrabony defects

Massimo Del Fabbro

Lorena Karanxha

Saurav Panda

Cristina Bucchi

Jayakumar Nadathur Doraiswamy

Malaiappan Sankari

Surendar Ramamoorthi

Sheeja Varghese

Silvio Taschieri

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

1.

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Summary of findings

Results

Description of studies

Results of the search

Included studies

Design

Participants

Interventions

Outcomes

Primary outcomes

Secondary outcomes

Excluded studies

Risk of bias in included studies

2.

3.

Allocation

Random sequence generation

Allocation concealment

Blinding

Blinding of participants and personnel (performance bias)

Blinding of outcome assessment (detection bias)

Incomplete outcome data

Selective reporting

Effects of interventions

Summary of findings for the main comparison. APC + OFD compared to OFD (9‐12 months follow‐up) for treating periodontal infrabony defects.

Summary of findings 2. APC + OFD + BG compared to OFD + BG (all follow‐ups) for treating periodontal infrabony defects.

Summary of findings 3. APC + GTR compared to GTR (all follow‐ups) for treating periodontal infrabony defects.

Summary of findings 4. APC + EMD compared to EMD (all follow‐ups) for treating periodontal infrabony defects.

1. Autologous platelet concentrates (APC) + open flap debridement (OFD) versus OFD