Much has evolved about the placement protocols and usage of endosseous dental implants. Early on, the osseointegration of dental implants was imagined as a way to restore patients with total or near-total edentulism, through the stabilization of a removable prosthesis. Initial protocols called for several months of healing after tooth removal. It was recognized that such healing was accompanied by both vertical and horizontal loss of bone, and loss of the gingival scallop, but that mattered little for such applications as an implant-anchored denture. After a second surgery for placement of the implant in the healed ridge, as much as six more months was required for osseointegration. Only then, at a third surgery, would the implants be uncovered and a prosthesis be attached to the dental implants. As more was learned about the behavior and predictability of these cylindrical implants, pioneers began to experiment with implant design, surface treatments, and accepted protocols for integration to obtain even better and faster results. They also began to employ implants for the replacement of single lost teeth. Today, single tooth implants make up a substantial proportion of all implants placed. Among the advances in the evolution of implant therapy was the immediate placement of a dental implant into an extraction socket. In relatively healthy sockets this has proven predictable and successful. Another advance involved the placement of a healing abutment at the time of surgery (the so called One Stage approach) rather than placing a cover screw and suturing the tissue closed for the osseointegration period. Today’s implant therapy looks different than that of thirty years ago.
In the past, patients who faced the prospect of losing a tooth (especially an anterior tooth) had several options: a) a removable appliance, which is largely considered undesirable or b) a fixed bridge, which provided an immediate tooth replacement but required the reduction of adjacent teeth, or c) an implant, which took months of healing, during which a removable appliance was liable to be used. Such a prosthesis, if very carefully designed, could create essentially an ovate pontic receptor site-like depression on top of the implant and thus support the healing soft tissue, but more often could compromise the soft tissue architecture and lead to a less than optimal result. The concept of immediate loading of implants was introduced, but it was primarily used on fully edentulous cases where rigid splinting of the several implants by a passively fitting fixed bridge was employed. This limited any micromotion of the newly placed implants (which could result in a fibrous union, rather than true osseointegration, and probable failure) and ensured the likelihood of implant survival. The idea of further pushing the envelope, by immediately placing a fixed provisional on an unsplinted single implant placed in a fresh extraction socket, seemed riskier still, but the concept has been proven. Wohrle 1 achieved success with 14 such cases. Kumar and Jaffin 2 studied 56 implants in the esthetic zone treated in this fashion, none of which were lost. It resulted in esthetics that were good as, if not better than, delayed loading. Certain guidelines were developed to obtain consistently good results. Chief among those is good primary stability at the time of implant placement. Wohrle recommended 45 Ncm initial torque value, which he felt eliminated micromotion of the implant.3 Next would be the elimination of contact on the provisional in centric occlusion and all excursions. Finally, the patient must be instructed to maintain a soft diet until osseointegration is complete.
No longer, though, is simple survival of an implant adequate; the expectations of patients have been raised, especially for restorations in the anterior “esthetic zone”. “When placing a single-tooth implant-supported restoration in the anterior region, the esthetic criteria represent the fundamental aspect of success, since the prosthesis will be compared to the adjacent dentition and their supporting tissues and/or conventional fixed prostheses that now achieve esthetic results on a routine basis”.4 It may then have come as a surprise to some when Belser et al 5 found, after a review of esthetic outcomes reported in the implant literature, that the single tooth replacement might be the only clinical situation where predictable esthetic outcomes could be consistently achieved. Esthetic determinants of dental therapy have emerged as highly significant among both clinicians and patients. For anterior dental implants, esthetic outcomes are strongly related to peri-implant mucosal architecture.6 Various methods, such as the Pink Esthetic Score have been introduced as a way to objectively measure the soft tissue parameters.7 There is now significant evidence that not only does the immediate fixed provisionalization of an immediately placed single implant not jeopardize the outcome, but that such a protocol yields results that are predictably as good as delayed restoration. Gelb 8 retrospectively reported on 2056 implants that were placed immediately and provisionalized immediately, and compared them to 5284 implants that were placed immediately but had delayed restoration. He concluded that immediate restoration of an implant does not contribute to a less favorable outcome. Cooper’s five-year prospective study of 94 implants showed no differences in the hard and soft tissue levels after immediate provisionalization of dental implants in extraction sockets or healed ridges. He concluded that this “higher risk” approach of immediate implant placement coupled with immediate provisionalization yielded excellent long term stable results.9 And, as mentioned earlier, unless steps like this are taken to preserve surrounding crestal bone and the soft tissue architecture, it is possible that the papillae may be lost. The recreation of the papillae may prove difficult, if not impossible.10
The use of fixed provisionals allows the “grooming” of soft tissue in both the anterior and posterior regions of the mouth, as will be demonstrated in the case reports to follow. It is not enough, however, to simply create or maintain soft tissue architecture that resembles that around a healthy tooth. Once the above procedures have been applied to yield beautiful, anatomically shaped soft tissue, the transfer of information to the lab about the soft tissue from which the restoration must emerge can be vastly improved by the customization of impression copings. The cross-sectional size and shape of a groomed socket can be substantially different than that of a stock impression part. Adding on to the stock part to “fill the gap” allows one to precisely record the shape of the socket that has been carefully groomed. This can yield predictably superior results and simplify the completion of the case.
A similarity exists between today’s implant restorative esthetics and the era when porcelain laminate veneers were first introduced. Having taken an introductory course in the concepts of veneers, many a practitioner, armed with their depth-cutting burs, would conservatively prepare teeth and take impressions. They and their patients would then pray that the ceramists’ vision of what the restored teeth should look like matched their own. Far too many remakes were required. Only when we began to rely on diagnostic wax-ups as a blueprint for the fabrication of our restorations did the treatment become more predictable.
Employing standard methods for restoration of implant cases, using stock parts, may yield a mixed bag of results, sometimes frustrating dentists and patients alike. Nowadays, to strictly use stock cylindrical healing abutments, whose cross-sectional shape and size does not remotely resemble that of the tooth to be replaced, and to transfer information to the lab with standard impression copings, means that the laboratory technician will once again be entrusted to “guess” at the appropriate size and shape of the restoration needed to properly support the surrounding soft tissue. With so little control over the outcome, it is no surprise that practitioners too often struggle to seat abutments or crowns that are appropriately robust in size. One could liken it to trying to squeeze a quarter into a hole the size of a dime. Creating soft tissue surroundings for our restorations that mimic the appearance of a healthy dentition, with intact papillae, and gingival zeniths at the appropriate height and location, requires more control of the variables.
Many an implant representative has volunteered their observations that methods such as fixed provisionals and customized impression copings are too seldom employed. Implants may be immediately placed as part of a one stage approach, but often a removable appliance is placed over a stock cylindrical healing abutment instead of fabricating a provisional crown, which seems too labor intensive. Rather than groom the soft tissue, the appliance may deform it. Standard impression copings faithfully record the position of the implant, but provide little information about the emergence profile, which the laboratory needs to predictably make a restoration that will mimic the healthy tooth. One could reasonably argue that spending the time on fixed provisionals and customization of impression copings is time well spent. Not only will most patients prefer the function and esthetics of a fixed provisional but it is a procedure for which we will be compensated. The use of custom impression copings usually allows the seating of our restorations to be accomplished without a lengthy struggle (the stress and wasted time for which goes uncompensated). This article will discuss methods to make the management of such cases more predictable and successful.
Role of the Restorative Dentist
While the number of general dentists performing their own implant surgeries is on the rise, the removal of a tooth and the surgical placement of the implant is still most frequently the province of the specialist, be it oral surgeon or periodontist. The fabrication of the final prosthesis is, of course, accomplished by the lab technician. What, then, is the role of the restorative dentist? His or her job should be to quarterback the implant team, coordinating the timing and involvement of surgeon and technician, and to manage the soft-tissue interface where implant and restoration meet. By some measures, an implant’s “success” is defined simply by whether or not it is still in the patient’s mouth (Figs. 1, 2). Our goal however, as previously stated, should be to mimic the natural dentition and surrounding soft tissue, and our efforts will be judged by comparison with the adjacent teeth or restorations. The following case histories will illustrate some of the many methods of immediate provisionalization as well as the customization of impression copings.
Statistically a success; esthetically a failure.
Another esthetic failure.
Case History 1
This 70-year-old male had undergone significant dental work on teeth #s 15-24 in the past, including all ceramic crowns on teeth #15, 14, 23 and 24, a veneer on #13, and older PFM crowns on #s 12, 11, 21 and 22 (Fig. 3). Nine years prior to this round of treatment, a post & core and crown had come out of tooth 11. A heroic recementation (C&B Metabond, Parkell) had been performed for esthetic reasons but, due to the lack of ferrule, the patient had been advised that either forced eruption of the root for the purposes of restoration, or removal of the root and replacement with an implant was recommended. When the patient presented this time, he had lost the crown on tooth #23 with the entire clinical crown inside. Once again, there was virtually no tooth structure remaining supragingivally, and this time the patient expressed interest in an implant solution (Fig. 4). The patient also acknowledged that the restoration on tooth #11 had recently become mobile, and so he was interested in an implant at that location as well. It was hoped that the patient would be a candidate for immediate provisionalization, but consideration was given to the possibility that he might not. After simply placing the lost crown #23 atop the root, alginate impressions (Jeltrate, Dentsply) were taken to fabricate an Essix retainer as a backup interim prosthesis (Fig. 5).
The patient was seen by the periodontist, who atraumatically removed the roots of teeth #11 and 23 and both were replaced with 4.1 mm wide Straumann bone-level implants. He determined that, due to the bone quality, there was insufficient primary stabilization. Stock cylindrical healing abutments were placed and the patient returned to my office where the Essix retainer was inserted. Three months later, after torque-testing the implants to verify osseointegration, the periodontist called, anticipating that I would be taking impressions for fabrication of final restorations. Recognizing, however, the inadequacy of the development of proper soft tissue architecture (Fig. 6, 7), a decision was made to instead place fixed provisional restorations to groom the tissue. Straumann PEEK (phenylethyl ethyl ketone) plastic temporary abutments, whose cross sectional size more closely resembled the adjacent teeth, were placed, which caused temporary blanching of the soft tissue but it returned to its natural appearance in just minutes (Fig. 8). Note the difference in their cross sectional size relative to the size of the stock healing abutments (Fig. 9). To illustrate that difference, after only several minutes the temporary abutment was removed and the stock-healing abutment was momentarily reinserted. The difference is significant even with only moments of grooming (Fig. 10). The temporary abutments were then prepared (mostly extraorally) to resemble crown preparations and their screw holes were occluded with PTFE tape (Fig. 11). Using a pre-treatment Polyvinylsiloxane impression (Cinch Vinyl, Parkell) and bisacryl temporary resin (ExactaTemp, Exacta Dental), cementable provisional crowns were fabricated, inserted, and checked to ensure that occlusal contact was limited. A discrepancy was noted between gingival heights of teeth #11 and 12. This was later corrected by the periodontist by a minor gingivectomy on #11. After roughly eight weeks of healing/grooming the tissue appeared ready for impressions (Fig. 12).
Upon removal of the temporary abutments, it was obvious that stock impression copings could not adequately record the newly groomed “sockets” (Fig. 13, 14). The anticipated rapid collapse of the soft tissue, even during the few minutes that would be needed to take an impression, necessitated the customization of the impression copings. To accomplish that, each temporary abutment was screwed down to an analog, and then the provisional crown was seated upon it (Fig. 15). Each assembly was then seated into a rigidly-setting polyvinylsiloxane bite registration material (Chocolate Bite, Discus Dental), analog down, until the cervical one-third of the crown was covered, thus recording the shape of the portion of the temporary restoration that had groomed the soft tissue (Fig. 16). After the material set, the crown and then temporary abutment were removed, leaving the analog buried in bite registration material and thus recording the emergence shape of the socket (Fig. 17). Placement of the stock impression coping showed just how inadequate it would be for recording such an enlarged socket (Fig. 18). The powder/liquid addition of acrylic (GC Pattern Resin, GC America) to the roughened surface of the impression coping compensated for the discrepancy (Fig. 19). The two customized impression copings were then taken to the mouth and seated (Fig. 20). A full-arch open stock-tray (Impress Tray, CBite DDS) impression was then taken using polyvinylsiloxane (Affinity, Clinician’s Choice) (Fig. 21). The laboratory used this to fabricate two custom titanium abutments (Atlantis), which would properly support the tissue shape as recorded and allow appropriate placement of finish lines (Fig. 22). Next, they fabricated two PFM crowns (Fig. 23). At the insertion visit, the temporary crowns and abutments were removed, the custom abutments were inserted with no difficulty and torqued down to 35 Ncm, and the PFM crowns were cemented with temporary cement (TempBond NE, Kerr). The soft tissue architecture has been preserved and has proven to be stable, as evidenced by this 26-month follow-up (Fig. 24, 25).
Pre-treatment condition, including heroic recementation of #11. Note slight gingival discrepancy between #11 and 12.
Fracture and loss of crown and buildup tooth #23.
Essix retainer fabricated as backup provisional.
Healing abutment #11 with inadequately developed soft tissue.
Healing abutment #23 with inadequately developed soft tissue.
PEEK temporary abutment.
Improved soft tissue support from larger diameter temporary abutment.
Improved emergence obtained immediately.
Prepared temporary abutments.
Soft tissue after eight weeks with fixed provisional.
Figure 13 Figure 14
Groomed soft tissue tooth #11. Groomed soft tissue tooth #23.
Abutment/crown on analog.
Recording shape of provisional’s emergence profile.
Analog with record of provisional’s shape.
Inadequacy of stock impression coping for recording enlarged “socket”.
Customized impression coping.
Customized impression copings seated at #11 and 23.
Impression of implants for restoration.
Soft-tissue model with Atlantis custom abutments.
PFM crowns #11 and 23.
26 month follow-up of abutment-supported crown #11.
26 month follow-up of abutment-supported crown #23.
Case Report 2
A 69-year-old female called one Saturday night, reporting that one of her front teeth had fractured off at the gumline while eating. At a Sunday morning emergency visit, I learned that the patient’s brother, a lab technician, had fabricated crowns on teeth #12 through 22 forty years earlier. Now, however, tooth #12 was fractured, leaving inadequate ferrule for restoration (Fig. 26). After informing the patient of the options, she elected to have the tooth removed and an implant placed. The tooth was recemented (C&B Metabond, Parkell) (Fig. 27) to allow an impression to be taken (Cinch Vinyl, Parkell) which would facilitate the fabrication of a provisional crown following implant placement. The patient was then referred to the periodontist. Three weeks later, the root of tooth #12 was atraumatically removed, a Straumann 4.1 mm bone level implant was placed with adequate primary stabilization, some bone grafting material was added (Bio Oss, Geistlich) and a stock cylindrical healing abutment was inserted (Fig. 28). The patient came directly to my office with a Straumann VITA CAD (polymethyl methacrylate) temporary abutment, provided by the surgeon (Fig. 29). My initial expectation was that I would prepare this temporary abutment as if for a crown, and then use my pre-treatment impression to fabricate a cementable provisional crown. Upon inserting the abutment, however, it appeared that prepping it would leave little more than the hole to support a provisional crown (Fig. 30, 31). I instead modified the temporary abutment to resemble the missing tooth. The location of the screw hole, on the labial-incisal, indicated that the final restoration would require a custom abutment and cementable crown but for the sake of temporization, this screw retained provisional would be very stable and more than adequate esthetically (Figs. 32-34). After three months, the surgeon removed the composite from the screw hole to verify osseointegration. The patient then came directly to my office for final impressions. The maintenance of the papillae was obvious (Fig. 35). Removal of the screw-retained provisional revealed a beautifully groomed “socket” (Fig. 36, 37) but it was apparent that this shape could not be adequately recorded with a stock impression coping (Fig. 38).
The process of customizing the impression coping began by attaching an analog to the screw-retained provisional and then burying the analog in bite registration material (Fig. 39). Upon setting, the provisional was unscrewed and removed (Fig. 40). The stock impression coping was roughened, attached to the analog, and this time flowable composite resin was employed to fill the gap (Fig. 41). This customized impression coping was taken to the mouth and inserted prior to the taking of a closed tray polyvinylsiloxane impression (Affinity, Clinicians Choice) (Fig. 42). The laboratory then fabricated a custom zirconium oxide abutment and lithium disilicate crown (e.max, Ivoclar Vivadent) (Fig 43, 44). The results remain stable after 20 months follow-up.
Fracture resulting in loss of crown #12.Tooth #12 after heroic recementation of crown.
Tooth #12 after heroic recementation of crown.
Immediately after removal of tooth #12, implant replacement, and bone grafting.
Straumann Vita-CAD temporary abutment.
Figure 30 Figure 31
Temporary abutment #12. Temporary abutment #12, incisal view.
Modified temporary abutment #12, incisal view.
Modified temporary abutment #12, facial view.
Modified temporary abutment #12, screw hole filled with composite resin.
Soft tissue maturation three months after implant placement.
Figure 36 Figure 37
Soft tissue emergence #12, facial view. Soft tissue emergence #12, incisal view.
Inadequacy of stock impression coping for recording enhanced “socket”.
Recording emergence profile of temporary abutment.
Analog with record of emergence profile of temporary abutment.
Customized impression coping.
Customized impression coping seated at #12.
Custom zirconium oxide abutment #12.
e.max crown #12. Note maintenance of papilae.
Case Report 3
A 60-year-old male had been seen, by a different periodontal surgeon, for the removal of a failing tooth #47 and augmentation of the ridge by bone grafting. Months later, a single stage Nobel Biocare Speedy Replace 5 mm implant had been placed and a stock cylindrical healing abutment was attached. Three months after that, once osseointegration was complete, the patient was referred to me for restoration. Upon examination, one could see that the shape and size of the healing abutment, and thus the soft tissue emergence, in no way resembled the cross-sectional shape of the adjacent tooth at the level of the gingiva (Fig. 45, 46). In order to better manage the soft tissue emergence shape for the impending restoration, and to minimize the embrasure space that would result between teeth #46 and 47, it was decided that further grooming of the soft tissue would be beneficial. A unique after-market healing abutment called a Contour Healer, which is available in numerous sizes and styles compatible with many of the most commonly placed implants, was chosen. When the use of this healing abutment is anticipated, there are instructions to guide the final positioning of the implant fixture such that the abutment can be placed without modification, ideally at the time of surgery. Since that had not been the case, my trial insertion in healed tissue revealed that the orientation of the abutment was not aligned with the arch form (Fig. 47). Modification of the Contour Healer (by reducing two corners and adding to the other two corners with flowable composite resin), corrected the alignment (Fig. 48, 49).
After three weeks, the tissue’s emergence shape was significantly altered and was ready for impressioning (Fig. 50). The Contour Healer was attached to an analog, and the assembly was buried, up to the gingival level of the abutment, in bite registration material (Fig. 51, 52). Removal of the Contour Healer and attachment of the stock impression coping to the buried analog revealed, once again, how inadequate the stock part would have been for recording the soft tissue’s new shape (Fig. 53). Again, flowable composite resin was used to fill the gap (Fig. 54). The customized impression coping was taken to the mouth and secured to the implant fixture (Fig. 55), whereupon a full arch stock tray polyvinylsiloxane impression was made (Fig. 56). The laboratory fabricated a screw-retained lithium disilicate crown (e.max, Ivoclar Vivadent) with a titanium base (Fig. 57) which was inserted with no resistance from the soft tissue and torqued down to 35 Ncm (Fig. 58). The occlusal opening was occluded with PTFE tape and composite resin. The lifelike result remains stable after 11 months of follow-up (Fig. 59).
Stock cylindrical healing abutment #47.
Non-atomical emergence created by stock healing abutment.
Figure 47 Figure 48
Misaligned Contour Healer Modification of Contour Healer.
Alignment of Contour healer corrected by modification.
Enhanced emergence profile created by Contour Healer.
Attaching analog to Contour Healer for customization.
Recording shape of Contour Healer.
Inadequacy of stock impression coping for recording enhanced soft tissue emergence.
Addition of flowable composite resin to customize impression coping.
Customized impression coping seated at #47.
Impression taken with customized Impression coping.
e.max crown with titanium base.
Simplified seating of e.max screw-retained crown on implant #47.
Eleven-month follow-up of implant-supported crown #47. Note how soft tissue adaptation resembles that on the adjacent natural tooth.
The use of stock cylindrical healing abutments (with or without removable interim prostheses) for the period of osseointegration, while simple, wastes an opportunity to more ideally groom the soft tissue surrounding an implant. Fixed provisional restorations or anatomically shaped healing abutments allow for the creation of much more anatomically accurate soft tissue emergence profiles. Use of stock impression copings fails to capitalize on the faithfully recreated soft tissue architecture, whereas customization of the stock part enables this ideal form to be accurately recorded. Failure to employ such measures means that the laboratory has two choices: either 1) fabricate a restoration which inadequately supports the soft tissue, potentially creating a mismatch with the tissue surrounding the adjacent teeth, or 2) fabricate a restoration with more ideal shape and size, which will create difficulties for the restorative dentist at the insertion appointment. Many a practitioner has struggled mightily to seat an appropriately sized abutment or screw-retained crown in an undersized socket. This situation is time consuming and uncomfortable for patient and doctor alike. Minimally, it may require the administration of local anesthetic to accomplish the insertion. Worse yet, it might require the use of a releasing incision in the soft tissue to seat the restoration, which not only warrants an uncomfortable healing period for the patient, but also may result in some remodeling of the soft tissue margin, potentially to a less desirable final position.
While the patient may not be aware that a better alternative scenario was possible, they are liable to be less than pleased to have undergone such a distressing denouement to their treatment, and thus be less likely to become a referral source of other patients. The restorative dentist will have similarly endured a difficult insertion visit, accompanied by much consternation and perspiration. The minimal investment of time needed to groom the soft tissue with a fixed provisional or anatomic healing abutment and the accurate recording of that soft tissue with a customized impression coping will yield large dividends at the insertion visit. The uncomplicated placement of the final restoration, which is a more ideal reproduction of the tooth to be replaced, will make for a stress-free visit for the doctor, and an outcome that is sure to please the patient. OH
Dr. Rosenfeld is a graduate of Cornell University and Northwestern University Dental School. He served a one year General Practice Residency, and a second year as Chief Resident, at Long Island Jewish Medical Center. He is a “wet-fingered” general dentist who has practiced in Westwood, New Jersey since 1993. He has advanced training in Esthetic dentistry and has served as a mentor at the Nash Institute in Charlotte, North Carolina. Dr. Rosenfeld has had numerous articles published in Dentistry Today, Inside Dentistry, Dental Economics, Dentaltown, and Dental Products Report. He serves as an assistant attending in the Department of Dentistry at Hackensack University Medical Center in New Jersey. He has a “key opinion leader” relationship with several dental manufacturers and has lectured on various subjects in restorative dentistry. He was inducted as a Fellow of the International Academy of Dental Facial Esthetics.
Oral Health welcomes this original article.
1. Wohrle, P.S. (1998). Single-tooth replacement in the aesthetic zone with immediate provisionalization: fourteen consecutive case reports. Practical Periodontics and Aesthetic Dentistry, Nov-Dec; 10(9): 1107-1114.
2. Kumar, A. & Jaffin, R. (2002). Immediate provisionalization of single tooth ITI implants placed into fresh extraction sockets. Academy of Osseointegration News, Volume 13, Number 3: 1, 8-11.
3. Wohrle, P.S. (1998). Single-tooth replacement in the aesthetic zone with immediate provisionalization: fourteen consecutive case reports. Practical Periodontics and Aesthetic Dentistry, Nov-Dec; 10(9): 1107-1114.
4. Touati, B, Guez, G., Saadoun, A. (1999). Aesthetic soft tissue integration and optimized emergence profile: provisionalization and customized impression coping. Practical Periodontics and Aesthetic Dentistry, April: 11 (3): 305-314.
5. Belser, U., Schmid, B., Higginbottom, F., Buser, D. (2004). Outcome analysis of implant restorations located in the anterior maxilla: a review of the recent literature. International Journal of Oral and Maxillofacial Implants. Volume 19, Supplement: 30-42.
6. Cooper, L.F. (2008). Objective criteria: Guiding and evaluating dental implant esthetics. Journal of Esthetic Restorative Dentistry. Vol 20(3): 195-205.
7. Furhauser, R., Florescu, D., et al (2005). Evaluation of soft tissue around single-tooth implant crowns: the pink esthetic score. Clinical Oral Implants Research. Volume 16(6): 639-644.
8. Gelb, D.A. (2016). Unique benefits of immediate implants with immediate provisional restoration: outcome analytics from 26 years of documentation. Inside Dentistry. April:35-41.
9. Cooper, L.F., Reside, G.J., Raes, F., et al (2014). Immediate provisionalization of dental implants placed in healed alveolar ridges and extraction sockets: a 5-year prospective evaluation. International Journal of Oral & Maxillofacial Implants. May-June; 29(3): 709-717.
10. Chu, S.J., Hochman, M.N., Tarnow, D.P. (2015). Loss of the interdental papilla. CDE World Continuing Dental Eduction, American Academy of Periodontology, April: `1-4.