Twenty Years of Emdogain in Regenerative Periodontal Therapy

An enthusiastic opinion piece (erroneously designated Systematic Review) has just appeared in the August issue of the prestigious Journal of Clinical Periodontology (Miron et al. 2016). The 16 pages are authored by a remarkable number of 20 authors.

This is certainly not a systematic review as basic PRISMA requirements are not met. Not a single one. So, has this to be regarded just another old fashioned narrative review? I can’t help but when reading the first authors’ unwarranted zeal this may actually be just another brazen commercial. I am afraid that most if not all authors are listed on Straumann’s pay roll, the company which currently markets Emdogain. Thus, the required (for biomedical journals) Conflict of interest and source of funding statement sounds, well, frivolous.

The authors report no conflict of interest for the present review article. No funding was required [sic]/received by any of the co-authors for the present review article.

According to his CV, Dr. Miron alone received 270,500 CHF ($274,246) from Straumann between 2010 and 2015.

As the authors dare to suggest recommendations (called guidelines, however except a flow chart in the main text, see below, they are deeply buried in the paper’s supplement which can be accessed only online) without grading the presented evidence from randomized controlled and observational studies, it seems pertinent to judge, at least in part, whether respective suggestions have a sound scientific basis.


Miron et al. consider the three classical occasions when periodontal tissues could be regenerated if at all: infrabony defects and class II furcations, as well as recessions. The term “intrabony” may actually refer to a special type of infrabony defect with three walls, but in any case, subsuming “horizontal bone loss” among “intrabony” defects is unwarranted and confusing.

Horizontal bone loss

According to Miron’s flow chart, there seems to be no indication for Emdogain application when bone loss is even or horizontal. According to common sense, the condition develops (in particular at interproximal locations) when periodontal lesions are located on neighboring teeth, have merged interdentally and exert the same pace of progression. As matter of fact, in average periodontitis patients, horizontal bone loss is by far the most frequently occurring feature of bone loss. For instance, Persson et al. (1998) found no vertical bone loss (angular bony lesions) on intraoral radiographs in 39% of 416 individuals between 15 and 94 years of age who were seeking dental care. The average depth of the vertical lesion was 1.2 mm (SD 1.4). While vertical defects of 3 mm or more (which are considered possibly suitable for regeneration attempts) were in fact found in 30%, in most cases only 1 or 2 lesions of that depth were found. So, the question whether additionally applied Emdogain leads to clinically relevant further improvement as regards pocket reduction and clinical attachment gain appears to be of quite some importance.

Miron et al. base their negative recommendation for horizontal bone loss (“Conventional periodontal flap surgery, conservative or resective approach”) on a recent systematic review by Graziani et al. (2014) who identified a couple of randomized controlled trials  having addressed the problem. The three studies report treatment effects of 60 experimental patients and 49 control patients. The meta-analysis revealed a weighted mean difference between open flap debridement (OFD) and  OFD plus Emdogain after 8-12 months of 1.19 mm (95% confidence interval 0.94; 1.44). The effect as regards pocket depth reduction was very similar (WMD 1.18 mm, 95% CI 0.84; 1.51). As regards clinical attachment gain, there was some heterogeneity observed (I2 of 66%) which was not the case with pocket reduction. Overall at least two studies exhibited considerable bias. For example, in the study by Jentsch and Purschwitz (2009), examiner blinding was considered inadequate. One has to add that different numbers of experimental and control subjects in that particular study (which authors designate as “pilot study”) may indicate problems with randomization. That particular study also yielded better results as regards clinical attachment gain than the two other studies. So, one may judge that design issues and partial inconsistency should lower the quality of evidence as regards clinical attachment gain to low, and as regards pocket reduction to moderate or low. as well. If one wants to make a statement for recommendation of Emdogain in periodontal lesions characterized by horizontal bone loss, it certainly can only be weak.

However, Graziani et al. do not mention quality of evidence and do not make recommendations in favor or against but rather call for more RCTs, preferably multicenter studies, which is sound. Miron et al. seem to agree but further review of their recommendations for other type of lesions may in fact yield that they travel on a slippery slope.

Class II furcation involvement

While presently through-and-through furcations cannot be closed by guided tissue regeneration (GTR) at all, there is low quality evidence from a number of older systematic reviews indicating that buccal furcation involvement at maxillary first and second molars may benefit from GTR as compared to conventional flap surgery. Miron et al. do not provide a single reference for their recommendation (in the above figure) that the application of Emdogain would yield a similar effect. GTR may yield slightly better results also in mesial furcations in maxillary first and second molars while that is not the case in distal furcations. Their recommendation (as suggested in the above figure) is Emdogain (not GTR) for mesial furcations and root resection or Emdogain (during flap surgery) for distal furcations. Evidence is provided by a single small-scale split-mouth randomized clinical trial (Casarin et al. 2010) who report a “higher conversion rate into class I when compared to OFD [open flap debridement] alone [note that OFD was supplemented with EDTA root conditioning] although complete furcation closure was rarely found” after 24 months. The study did not, however, follow the recommended surgical re-entry procedure (Pontoriero and Lindhe 1995). One has to consult the preliminary 6-month report by Casarin et al. (2008) to learn that most of the proximal furcations were actually located distally at maxillary first and second molar (12/15 in the test group where Emdogain was applied, and 13/15 in the control group). In a more recent trial conducted by this group, Peres et al. (2013) compared the effects of additional synthetic bone substitute beta-tricalcium phosphate with and without Emdogain in proximal class II furcations. After 6 months, no difference was found although both groups showed considerable improvement and closure rates.

Given the limited data and methodological issues, current evidence for application of Emdogain in proximal class II furcations in maxillary molars may be regarded low. Recommendations can not be given. In particular in distal furcations, the main reason for consistent observations of failing GTR procedures is very clear: presently, it is not possible to properly and reliably debride the complicated root complex. The application of a membrane would, in the presence of bacterial colonization, not result in tissue regeneration. Likewise, application of bone substitutes or Emdogain wouldn’t do either. Unfortunately, this simple biological reasoning is not mentioned in the article by Miron et al. at all.

Surgical root coverage in cases of thick and thin biotype

Miron et al. suggest that recessions associated with a “thin biotype” need a connective tissue graft for surgical root coverage when employing coronally advanced flaps and Emdogain, while that would not be necessary in cases of a “thick biotype.” The concept of thick and thin “biotypes” is decades old and has been advanced since the 1970s by American periodontists/prosthodontists (e.g. A. Weisgold, J. Seibert) who had observed that subgingival crown margins associated with thin gingival tissue frequently led to unpleasant, at least for the patient, recession. However, if crown margins were placed subgingivally in case of thick gingival tissue, a periodontal pocket may develop. So, it makes sense to carefully consider thickness of the tissue, its so-called biotype. Periodontal bio-, or rather, phenotypes do in fact exist, and thickness of the gingiva is only one aspect with its width, color, degree of keratinization, papilla height, even shape of the crown considered as well. The periodontal phenotype varies from subject to subject and, within subjects, from tooth to tooth. Systematic analysis of the periodontal phenotype had first been done by the Gothenburg (Olsson and Lindhe 1991, Olsson et al. 1993) and our groups (Müller and Eger 1997, Müller et al. 2000, Müller and Könönen 2005), but recently, all had boiled down to “(in)visibility of a periodontal probe through the marginal tissue” (De Rouck et al. 2009), a simplistic and unscientific concept which seems to be established now.

Several recent systematic reviews (Chambrone et al. 2010, Buti et al. 2013, Cheng et al. 2015) have addressed the question, What is the best combination of surgical treatments for root coverage? and quality of evidence is moderate for the coronally advanced flap with connective tissue graft yielding the best results as regards recession reduction and gain of clinical attachment. The highest rate of complete root coverage was achieved by coronally advanced flap in combination with Emdogain, again with moderate quality of evidence (Buti et al. 2013).

Notably, the “gingival biotype” was not addressed in any study and, to my knowledge, there is no randomized clinical trial in which thick or thin gingival tissue around gingival recession had been considered as an explanatory variabe influencing the outcome of surgical root coverage. Recommending Emdogain in any root coverage proedure (see figure above) is also not based on evidence. As strong recommendations require high quality evidence for substantial (i.e., clinically relevant effect), one must suspect again commercial interests.

That Emdogain has been shown, in animal experiments and a few human biopsies, to promote formation of acellular extrinsic fiber cementum on previously plaque-covered root surfaces should not be omitted from this critical comments on the paper by Miron et al. Large parts of  their paper address this fact as it is claimed that periodontal regeneration is the ultimate goal of periodontal therapy. As in clinical cases the quality of the healing cannot be assessed, long-term stability of root coverage and healthy gingiva is the only endpoint of clinical relevance for the clinician. Miron et al. find it “hard to believe that over 20 years have now passed since enamel matrix derivative [Emdogain] was first introduced as a regenerative agent for periodontal tissues.” I suppose they mean given the fact that it is still not used in each and every surgical procedure in Periodontics. The main reason for reluctance of certain scientists and clinicians then and still is of course uncertainty about biological mechanisms involved in cementum formation promoted by Emdogain. The product had been marketed based on some observations made by L. Hammarström in the 1990, something which was considered then and still is, unprecedented. Claims, made by Miron et al. that “over the years, we have learned a great deal [sic] regarding the biological roles of specific enamel matrix proteins and future investigation is constantly [sic] underway to further characterize their effects on cell and tissue behaviour” may indicate that not so much is actually known, after 20 years of commercialization.

One of my students had put it, after having attended a symposium in Tromsø organized by Straumann, “Too good to be true, isn’t it?”

11 August 2016 @ 11:01 am.

Last modified August 17, 2016.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s