The treatment of periodontal diseases by resective surgery : Osteoplasty and apically repositioned flaps

The title of this article makes clear that surgical treatment corrects the effects of a disease but does not cure it. It is a symptomatic treatment that cannot be considered to be beneficial unless it is integrated into an overall strategy of secondary prevention.

The evolution of surgical concepts

The history of periodontal surgery can be divided into two major periods, both of which focus on the symptom of the disease, the periodontal pocket. The first is dominated by the elimination (or reduction) of the pocket by the use of resective techniques. During the second period, the main attempt has been to achieve the regeneration of lost periodontal support. It is difficult to establish precisely the turning point between the two. Also, there are two intervals to consider, the period of initial research, then the application to normal clinical practice that inevitably follows the former. The medical literature teaches us that, since 1934, Beube and Silvers (cited by Cross, 1957) considered the possibility of regenerating bone defects by grafting with powdered bovine bone. ^{Nyman et al. (1982)} were the first to demonstrate that this approach could succeed with complete regeneration of the attachment apparatus.

It can be considered that grafting found its clinical application in the early 1980s and guided tissue regeneration with the use of barriers, in the middle of that decade. The favourable results that were recorded have led to disaffection with resective techniques ; the latter being considered to mutilate even further a periodontium already affected by the disease. It was towards the end of the 1990s that the first longitudinal studies have challenged regenerative treatments and have led once again to the return of the previous resective techniques.

Historical review of resective bone surgery

Up to about the 1980s, the identification of pocketing in periodontal disease was considered to be the essence of the disease and not an effect. Under this paradigm, resective bone surgery was indicated as a treatment that would, by eliminating the pocket, ipso facto, eliminate the disease itself. This logic was applied first to gingivectomy (^{Fauchard, 1961}), then to the apical repositioning of the soft tissues as close as possible to the level of the recontoured bone. It is this ability to raise a full thickness flap that permits a direct approach to the bone.

It seems likely that this treatment was in use from the th 19 century because, in 1883, Marshall published a report on a case involving bone surgery that must have been preceded by what later was called a flap operation. In 1965, Robicsek reported (^{Stern et al., 1965}) that his father was aware that resection of interdental craters was necessary to improve the bony architecture and practised it around 1884.

The first known complete description is that of ^{Neumann in 1920}, claiming that he had been practising the technique since 1911. According to him the procedure is indicated when bone loss exceeds one third of the bony support. He describes a muco-periosteal flap bounded by two relieving incisions in the region of the papillae and extending to the alveolar mucosa. The horizontal incision is intra-sulcular and is directed towards the alveolar bone margin. The raising of two flaps (buccal and lingual) is very important because the bone must be exposed to the level of the apices of the teeth. Neumann insists on the necessity of total elimination of root calculus as well as of the granulation tissue. The amount of bone removal is limited to the removal of the superficial infected bone (our emphasis). The flap is then replaced (idem) at its original level, except where there are deep pockets in which case a gingivectomy of about 2 mm is undertaken. In areas of shallow pocketing, Neumann is content to undertake curettage. He considers that this protocol can be applied to sectors involving up to six teeth and insists on the importance of the highest standards of oral hygiene. Although he devised the treatment, he only practised limited resection, localised mini-gingivectomy for pocketing and superficial bone removal for bone defects. The foresight of the author is surprising because, although he did not provide scientific proof, he nevertheless seized the essentials and opened the way for others. The first bibliographical reference to the technique is usually attributed to Widman. If we place them in the reverse chronological order, ^{Widman (1918)} after ^{Neumann (1920)}, it is because Neumann claimed previous use of the method, something that Widman implicitly recognised by referring to it as the « modified Neumann ». In 1918, Widman described a new method. We have not had access to his original text, only to later transcriptions. It is a question of « a muco-periosteal flap that is raised after making a scalloped incision to separate the pocket epithelium and [ emphasised by us ] the inflamed connective tissue from the non-flaming gingiva. The flap is bounded by two vertical incisions extending to the alveolar mucous membrane. The flap is then raised so as to expose 2 to 3 mm of the alveolar bone. The collar of soft tissue containing the pocket epithelium and [idem] the connective tissue is removed. The exposed root surfaces are scaled and the bone recontoured [ idem ] in order to re-establish a “physiological” shape to the alveolus [quotation marks are those of Widman ] . The edges of the flap are replaced at the level of the bone crest in order to reduce the pocket depth as far as possible ».

One point that is not very clear concerns the line of the horizontal incision. Amongst the advantages of the technique, Widman concludes that « compared with the gingivectomy, postoperative discomfort is reduced because healing is by first intention ». This seems to suggest the total elimination of the soft tissue wall at a variable level, according to the depths of the pockets, rather than a horizontal bevelled incision, or as we would say now, external. In this respect, Widman differs from Neumann in that the horizontal incision is intracrevicular. Nevertheless, the precise position of the incisions in these « flap operations » is not the essential feature. In neither case do these protocols consider the flap to be the end but only a means of access for easier debridement. It should be noted that ^{Widman (1920)}, although he did not consider the surface of the alveolar bone to be damaged (as being necrotic), he would remodel it.

The next stage involves more major osseous resection. ^{Neumann in 1920}, and again in ¹⁹²¹, recommended the complete elimination of the bony walls. ^{Black, in 1924}, also advised the elimination of interdental craters and the chamfering of the external buccal and lingual bone surfaces. It should be recalled that the dominant concept of the time was that of « extension for prevention », that Black had popularised in operative dentistry and which may have been gaining ground in periodontal surgery. In the same way that the extension of a cavity into healthy fissures was supposed to prevent caries, the removal of a certain quantity of bone could also prevent the extension of periodontal disease. At that time, in the period after the First World War, could the frequency of limb amputations have influenced dental surgery ? Even if that question cannot be answered today, in ^{1931 Crane and Kaplan} considered that the bone around periodontal defects was necrotic and should therefore be removed.

The event that marks the foundation of resective bone surgery was undoubtedly Carranza's thesis, published in Spanish in 1935. Carranza described the indications for resective bone surgery, which he called plastic remodelling of the bone. He considered that this remodelling stimulated the formation of « periodontal » bone. In a later publication (1942), he developed his theory that the immediate sacrifice of bone serves to enhance secondary bone formation. Today, one speaks of the rebound phenomenon.

It was ^{Schluger, in 1949}, who gave a complete description of resective bone surgery in his classic paper « Osseous resection - a basic principle in periodontal surgery ». It was no longer a question of delayed bony gain but of physiological contour. The first surgical stage relies on a gingivectomy (with the blade held horizontally at the level of ^{Black's [1924]} bleeding points) and the tissue that is replaced on the bony ridge consists solely of alveolar mucosa, tissue that is considered to be unsuited to resist the friction of food. That is to say, attention was directed to the end (RBS) and not to the means, the more appropriate flap. Even in 1954, when Nabers described a protocol for repositioning of the attached gingiva, he removed at least 2 mm of the marginal gingiva. Then the muco-periosteal flap was displaced apically, having made a single mesial relieving incision, and then being held in position by loose sutures.

It is interesting to note the strengths of old concepts in people's minds and the difficulty of changing opinions. It was only in ^{1957, that Ariaudo et al.} (Nabers being amongst them) completely abandoned the gingivectomy technique while preserving the gingival margin, thanks to what they call the repositioning incision. This involved an internal bevel incision directed toward the alveolar crest, with the objective of preserving the keratinised gingiva while refining its margin. There again, the evolution was progressive because the apical repositioning was made, as mentioned above, with only one (mesial) relieving incision. Shortly after, the same group proposed a way of facilitating access to the bone by making two relieving incisions so as to avoid the tension created in the tissues with small flaps (^{Ariaudo and Tyrell, 1957}).

At the same time ^{Friedman (1955)} developed Schluger's ideas on resective bone surgery and introduced two new words : « osteoplasty », that is a reshaping of the alveolar bone without removal of any supporting bone, and « ostectomy », where a part of the supporting bone is removed. This is a compromise designed to improve the architecture of bone that has been damaged by the disease, at the expense of a controlled removal of cortical bone.

It is necessary to remember that, in parallel with this current resective-type surgery, ideas on bone pathology were developing. Since ^{1935, Kronfeld} had stated that the bone surrounding periodontal pockets was not necrotic nor infected but rather destroyed by an inflammatory process. He proposed that an osteitis was present that was reversible if the aetiological factors were brought under control. This work, supported by histological studies on cadavers, was confirmed by ^{Orban in 1939}. This first scientific approach removed all biologic justification for the elimination of the marginal bone. One can say that, from this time, there were two different therapeutic approaches, gingivectomy, that was still used because it was practical and convenient in terms of public health at community level, and the resective bone surgery by flap surgery, the function of which was limited to achieving a new physiological architecture at the expense of the loss of part of the bone part that one knew being healthy.

The 1960s saw numerus, non-controversial studies published on this topic, all authors accepting Schluger's concept, later to be described again in his reference work (^{Schluger et al., 1978}). ^{Prichard (1961)}, although being one of the true promoters of regenerative surgery in the treatment of intrabony defects (1957), fully adhered to ^{Schluger (1949)} concept of RBS from the moment it was indicated. According to ^{Prichard (1961} and ¹⁹⁷²), the most important factor is the physiological gingival contour that can only be obtained with meticulous reshaping of the bone. Prichard considered that the gingival margins must be knife-edged, that the interdental papillae must be conical and concave interdentally, and that this result can only be obtained if the underlying bone has the same morphology. More or less simultaneously, another respected bone surgeon, ^{Ochsenbein (1960)}, put forward rules for bone reshaping. His recommendations hardly differ from those previously described, that is to say, the transformation of a negative to a positive architecture, except in areas of cratering. In this respect, he introduced a new concept by not remodelling both summits of the crater but only correcting one of them. This approach, even though it appears to be a little dogmatic, is none the less very astute. Ochsenbein considered that in the maxilla, where access for interdental cleaning is easier on the palatal side, the sacrifice of tissue to improve the palatal part of the crater is consistent with the concept of food shedding, with a summit shift only on the buccal side. In the mandible, the opposite is true, and Ochsenbein advises, also for reasons of access for interdental oral hygiene aids, to improve buccal access by eliminating the buccal part of the crater. Again, a positive architecture is recreated, this time with a lingual shift of the papilla.

It should be recognised that if all these authors followed the usual rules for bone removal, they were not dogmatic about it. ^{Carranza and Carranza (1956)} and ^{Ochsenbein (1960} and ¹⁹⁷⁷) defined the contra-indications and limitations of the technique. According to the first of these, bone reshaping encouraged a normal rearrangement of the fibres that is a prerequisite to « a complete and functional healing of the periodontium ». It was no less « limited by the foreseeable compromises imposed by the periodontal support of the adjacent teeth ». ^{Ochsenbein (1977)} specified also that the method was not indicated for cases of advanced periodontal disease or for deep and isolated craters. In addition, there was the question of the furcations. The reshaping of the craters in the posterior regions could open them up, especially to the mandible. ^{Ochsenbein (1977)}, by recommending the prior study of dry skulls, emphasised the necessity of acquiring knowledge of the morphological aspects of bone pathology. The compromise is sometimes of little importance, as the interdental bone anatomy is relatively flat in the posterior regions. This is something that has been known since the work of ^{O'Connor and Biggs (1964)}, and requires little updating.

To conclude this historical review, one can say that today, the protocol is very little different from that which Schluger described in 1949. He was not dogmatic but adapted his treatment of each case according to the specific pathological changes. That means that the indication for resective bone surgery is considered before intervention but the degree of resection and the type of reshaping done can only be decided direct vision.

Surgical technique

Resective surgery is a technique aiming to eliminate intrabony defects, the hemisepta, craters and their associated periodontal pockets, by removing the bony walls of the defect.

Following periodontal disease, the buccal and lingual (or palatal) became more apical than the interproximal bone (named reverse architecture), it is necessary to use osteoplasty and sometimes ostectomy (removal of supporting bone) to restore its normal scalloped shape. The gum is then repositioned apically on the recontoured bone. The soft and hard tissues are dealt with according to precise rules.

Management of the soft tissues

The goal of the resective surgery is not only to eliminate bone defects but also to obtain a gingival architecture that permits good maintenance, and pocket depths of less than 3 mm. That is why the correct management of the soft tissues is essential, from the first incision, the relieving incisions to the suturing. The initial flap is full thickness in order to gain access to the bone, then partial thickness to permit its repositioning on the bony crest or more apically. The incisions are made either within the sulcus or at a distance from the gingival margin (buccal, lingual or palatal), according to the depth of the lesions and the quantity of available attached gingiva.

Buccal incisions

The level of the buccal incision is determined according to the depth of the pockets and the width of the keratinised gingiva. The final aim is to replace the gingival margin at the level of recontoured bone.

The incision can be made at a distance from the gingival marginal. This option is rarely chosen where appearance matters and it requires a large band of attached gingiva.

In the posterior segments, where there are deep bony lesions and a sufficient quantity of attached gingiva, after detailed probing under anaesthesia, the incision can be shift buccaly from the gingival margin. The elimination of the pocket is carried out by this « gingivectomy of the flap » and not only by apical repositioning. Some authors even suggest not covering the interproximal bone completely in order to have, after healing by second intention, a very dense gingival tissue (^{fig. 1d} and ^1e) (^{Carnevale and Kaldahl, 2000}). On the other hand, particular attention is needed in areas of molar furcations so they are not denuded.

In the majority of the cases, in order to preserve the entirety of the gingival tissues, an intra-sulcular incision is made. By repositioning the flap apically, its coronal edge can be made to coincide with the new level of the bone crest.

In the case of small flaps, which have insufficient elasticity to permit apical repositioning, the relieving incisions will facilitate access to the underlying bone. These incisions must pass beyond the muco-gingival junction in order to facilitate the partial thickness dissection of the soft tissue.

In the anterior region, to give greater flexibility to the flap, the incision may be extended by one or two extra teeth. This may obviate the need for relieving incisions.

Lingual incisions

The same imperatives apply to these as to buccal incisions. The height of the attached gingiva is accurately assessed and it must not be sacrificed by making the incision too far from the gingival margin. Where there are of deep intrabony defects, apical repositioning of the flap to remove excess gingival tissue is achieved by making relieving incisions. However, in the distal regions, great care must be taken to avoid damage to the lingual nerve.

Palatal incisions

Because it is not possible to displace palatal flaps, the palatal incision will usually be made at some distance from the gingival margin. This scalloped incision is designed to remove the excess gingival tissue and to align the gingival and bone margins whilst thinning the palatal flap (^{fig. 1a}, ^1b, ^1c, ^1d and ^1e).

The level of the incision will be established by careful probing under anaesthesia around all the teeth to be treated and it corresponds to the depth of the pockets. It should coincide with, or be slightly more apical to the bone margin. Intrasulcular incisions made on the internal surface of the flap will facilitate removal of the inflammatory tissue. These three incisions were recommended by Widman in order to thin the palatal flap. A relieving incision sometimes permits better visibility of the underlying structures but is not undertaken distal to the premolars because of the risk of damage to blood vessels.

Sutures

Once the recontouring of the bone has been completed, the sutures will reposition the flaps at the level of the bony crest or slightly more apically (^{fig. 1d} and ^4d). These are generally interrupted sutures, anchored to the periosteum and at the tip of the papilla. The sutures, along with the relieving incisions, assist the apical repositioning.

Management the hard tissues : osteoplasty and ostectomy

If flap curettage is undertaken without correction of the underlying bone, the disharmony between the gingival architecture and that of the underlying bone can entail the formation of new pockets and their reinfection (^{Ochsenbein and Bohannan, 1963} ; ^{Ochsenbein, 1986}). Indeed, during the course of periodontal disease, we witness some important changes of the anatomy of the alveolar bone. The pathological process involves intrabony defects, hemisepta, a reversed bony architecture (where the level of the interproximal bone is situated more apically than that of the buccal and lingual bone) as well as the formation of buttresses and exososes.

The bony anatomy will, therefore, be modified in order to achieve a positive or flat architecture according to the depth of the defect and the anatomy of the tooth. The level of the interproximal bone will be located either at the same level as the buccal and lingual bone, or more coronally.

The quantity of bone to be removed must be carefully judged. It must not compromise aesthetics or increase tooth mobility.

^{O'Connor and Biggs (1964)} studied the anatomy of interproximal bone. In the molar regions it is flat and becomes more convex in the anterior sectors.

The best guide to the bony architecture in the normal state is to follow the line of the cement-enamel junction of the adjacent teeth. Clinical wisdom must be one's guide while carrying out these procedures and osteoplasty is preferred to ostectomy, while also avoiding damage to the supporting bone on the buccal surfaces.

For the treatment of the molars, in order to not to expose the furcations (especially interproximally in the case of upper molars), it is important to assess radiographically the height of the cervical trunk (^{fig. 2a}, ^2b and ^2c that is to say, the distance between the cement-enamel junction and the entrance to the furcation.

Intrabony and hemiseptal lesions

All walls of intrabony and hemiseptal defects are reduced. The base of the defect is then located level at, or more coronal than, the buccal and lingual bone (^{fig. 3a} and ^3b).

Depending on the bucco-lingual angulation of the tooth in the alveolus and the location of the bony defect, osteoplasty can be undertaken on the buccal or lingual and palatal surfaces. For example, in the case of lingually tilted mandibular molars with two-walled defects that are deeper on the lingual aspects, bone should be removed from the lingual surface with the effect of orientating the interproximal crest of the buccal surface toward the lingual surface (^{Carnevale and Kaldahl, 2000}).

Reversed bony architecture

When reversed bony architecture (^{fig. 4a}, ^4b, ^4c, ^4d, ^4e and ^4f , is encountered, as is frequently the case, bone should be removed from the buccal and lingual or palatal surfaces. If the amount to be removed is too great and risks exposing a furcation, the profile of the recontoured interproximal bone will be flat and not convex (as it is often normal in posterior zones).

^{Ochsenbein (1986)} has proposed a rational approach to this bony surgery, based on the depth of the defects and the height of the root trunk. The longer the trunk, the more the furcation will be protected and one can hope to achieve a bony profile that more nearly approaches the ideal (^{fig. 5}).

The slope of the palatal bone must make an angle of 10° with a horizontal line passing through the base of the crater (^{fig. 6}).

Once the reshaping on the palatal side has been achieved, the buccal bone will be reduced to the same level as the interdental bone or slightly more apically (^{Ochsenbein, 1986}). If the buccal bone is very thick, the reduction of the supporting bone will be minimal and there will be little post-operative recession. On the other hand, in some areas where it is very thin, such as the mesio-buccal root of the first molar, no bone should be sacrificed.

If the defect is deep and the trunk of the root is short, bone augmentation by graft or membrane will be indicated. However, very many maxillary and mandibular molars with root trunks of moderate height and bony craters of moderate depth can be successfully dealt with by resective bone surgery (^{Manson, 1976} ; ^{Ochsenbein, 1986}).

Bone hyperplasias and exostoses

These bony enlargements (^{fig. 7a}, ^7b and ^7c , often seen on buccal or lingual surfaces near interproximal lesions, interfere with good healing of the gingivae. They should be removed by the osteoplasty.

The thinning of this thickened bone and the creation of sluiceways will facilitate the adaptation of the flap.

Partial furcation defects (class II)

These defects (^{fig. 2a}, ^2b and ^2c, ^{fig. 8a}, ^8b and ^8c can be treated by resective surgery. The particular anatomy of the molars (including the size of the entrance to the furcation and the presence of root concavities) makes the blind cleaning of these zones difficult without surgical treatment (^{Matia et al., 1986} ; ^{Parashis et al., 1993}).

Surgical access permits perfect cleaning of the root surfaces, while resective bone surgery removes the bony defect and recreates a more « deflective » bone architecture that encourages gingival healing. A sluiceway can be created even from furcations either buccal or lingual ones. The excess gingival tissue will also be avoided by a small incision to enable the flap to be positioned at this level, so avoiding excess granulation tissue and recurrence of disease (^{fig. 8b}).

Instrumentation for resective bone surgery

Hand or rotary instruments can be used ; bone chisels, files, rongeurs, as well as various types of steel or diamond-tipped burs used at high or medium speed. Several studies have evaluated healing according to the instrumentation used. ^{Spatz (1965)} showed that high speed burs provoked an initial inflammatory reaction that was less severe than when low speed burs were used. However, there must be abundant irrigation and the pressures exercised by the operator must be light. The choice of burs (steel or diamond-tipped) can be at the discretion of the operator, knowing that the steel burs will eliminate the bone chips better while diamond-tipped burs act by progressively abrading and polishing the surface of the bone.

Particular care must be taken in narrow interproximal areas to avoid damaging the root surfaces.

Biological results of resective bone surgery

Hard tissues

Bone

Only the ostectomy part of resective bone surgery removes supporting bone. This is only a minor sacrifice if one believes Prichard when, in 1986, he recommended a « protocol of conservative ostectomy ». Resective bone surgery provokes an inflammatory response with an osteoclastic reaction that is capable, according to some (^{Grant, 1967}), of causing sequestration. In fact, the transient osteoclastic phase is followed by a rebound osteoblastic phase so that the bone loss is not clinically significant. ^{Wilderman et al. (1970)} studied the response to resective bone surgery in humans and found clear signs of bone remodelling one year after the intervention. They reported a mean loss of crestal bone of 0.8 mm, although with large individual variations.

Cementum

We have not found, in the medical literature, any comparative study measuring the damage to cementum according to whether there had been surgical treatment or not. It seems that the damage to the cementum is infinitely less during a surgical approach, where the raising of a flap allows calculus deposits to be removed under direct vision. Regardless of any physical damage to the cementum, there is a reaction to the raising of a flap. ^{Stallard and Hiatt (1968)}, in dogs, observed cementoid and osteoid formation 2 weeks rd th after surgery. This increased over the 3 and 4 weeks. In humans, ^{Listgarten (1972)} observed new cementum after 1 month. The resorption-apposition cycle doesn't seem to be an absolute model. New formation of cementum is not conditional upon a previous resorption (^{Linghorne and O'Connel, 1951}). Cementum left in place after surgery favours the formation of new attachment (^{Stallard and Hiatt, 1968}), whether it is over the root surface or in scattered fragments. In this latter case, it functions as a matrix facilitating the formation of new bone or cementum. The growth of new cementum is greatest in the apical parts of the roots. These electron microscope observations are interesting but may not have a clinical application when one knows that new cementum can also form directly on dentine (^{Listgarten, 1972}).

Soft tissues

Epithelial attachment

In monkeys, epithelial attachment has been observed 9 days after flap surgery and resective bone surgery (^{Caffesse et al., 1968}), whereas ^{Listgarten (1972)} observed it only after 1 month. The general consensus is from 3 weeks to 1 month, whether or not the surface of the cementum has been treated. Although the rate of growth of the epithelial cells is greater than that of connective tissue cells, it seems necessary to allow 3 weeks to 1 month for total restoration of the attachment in human subjects also (^{Dedolph and Clark, 1958}).

Connective tissue attachment

Having been observed and studied for decades, this phenomenon is now well understood (^{Schluger et al., 1978}). It begins with a fibrin clot separating the flap from the alveolar process. After 3 or 4 days, the clot begins to resorbe and triggers the concomitant development of the new connective tissue from the medullary and vascular spaces of bone, the coronal part of the ligament (the main source) and of the periosteum lateral and apical to the flap. Total resorption of the clot is achieved in 6 to 7 days. At this stage, the network of connective tissue fibres is very weak. One notes the presence of numerous inflammatory cells with a high concentration of polymorphonuclear leucocytes. Their progressive replacement by lymphocytes and macrophages signals the end of the inflammatory process. After 15 days, collagen fibres appear in the crestal area, with an orientation parallel to the root surface. Around the second month, the fibres begin to be replaced by surface osteoid tissue but it is a further 5 to 6 months before they become attached to the root surface.

Many workers (^{Björn et al., 1965} ; ^{Caton and Zander, 1976} ; and especially ^{Karring et al., 1984}) have warned of the risk of ankylosis and resorption that can occur after direct apposition of soft connective tissue on to a root surface, as is the case when the flap is replaced at the end of surgery. In fact, ankylosis and resorption are observed only very rarely. According to ^{Gottlow (1994)}, the rarity of these complications is due to the rapid migration of the epithelial cells in an apical direction, overtaking the connective tissue cells and acting as a barrier. Although this opinion is shared by many, it is permissible to wonder if it is correct insofar as direct connective tissue grafts in contact with root surfaces without any nearby epithelium, do not provoke resorption. It is more difficult to imagine a long epithelial attachment with the apically repositioned flap than with a replaced flap, nevertheless the repair takes place with a longer coronal epithelial part and a smaller apical connective tissue part.

Gingiva

Even if all authors agree that bone loss interdentally on the buccal and lingual (or palatal) aspects alters the gingival profile, there is less consensus with regard to the papillary zone. It seems that in this area, the gingival contour is influenced mainly by the shape and the volume of the interdental space as well as by the line of the cement-enamel junction (^{Takei, 1980}). It is possible that there is also a « tissue memory » that would explain why, after surgery aimed at the elimination of the pockets, residual pockets remain.

Clinical results of resective bone surgery

Studies comparing different therapeutic protocols are often disappointing and rarely long term. There are many assessment criteria but they vary considerably. In ^{1979, Knowles et al.} compare three different protocols : subgingival curettage without raising a flap, replaced flap without resective bone surgery and apically repositioned flap with resective bone surgery. The chosen criteria were pocket depth and the level of attachment. The survey, considered at the time to be long-term (8 years), suffered the loss of a considerable number of subjects (nearly half by 8 years), so that the results are difficult to interpret. The three protocols give similar overall results. It is when the results of the subgroups (determined according to pocket depths) are analysed in detail, that one observes slight variations. It seems that the deeper the pocket, more favourable results are achieved with resective bone surgery combined with apically repositioned flap.

Most of the later comparative studies are of shorter terms than that of ^{Knowles (1979)}. Those of ^{Isidor and Karring (1986)} and of ^{Ramfjord et al. (1987)} are of 5 years, that of ^{Becker et al. (1988)} lasted for 1 year and the one of ^{Kaldahl et al. (1988)} 2 years. The list of co-authors is a reliable indicator of the seriousness of these studies, the findings of which are similar. Whatever the operative protocol used - non-surgical, surgical, surgical with replaced flap, apically replaced flap with or without resective bone surgery - the results are similar. All stop the progression of the disease when measured in terms of pocket depths.

Despite everything, transposition in clinical practice seems difficult. In most of these studies, the frequency of aftercare and maintenance visits was optimal. It involved « ideal » patients, selected according to their reliability (low oral hygiene indices, regular maintenance visits). Considering the length of the studies, one is reduced to extrapolation : it would seem, if one believes ^{Ramfjord et al. (1987)}, that when one considers deep pockets (more than 7 mm), the reduction in depth is the greater with apically repositioned flaps with resective bone surgery. One point appears essential. It can be correlated with the last parameter considered by ^{Ramfjord et al. (1987)}, the « need for re-treatment ».

The authors speak of « a yearly subjective evaluation by a periodontist » and emphasise that this « need for re-treatment » is a highly subjective notion. Nevertheless, it can be an indication of differences in the long term success of the chosen initial treatments. It is noted that the need for re-treatment is decided according to the amount of bleeding and suppuration on probing, with the following results (^{Ramfjord et al., 1987}) :

- root planing, 44 (teeth, not of patients) ;

- sub-gingival curettage, 20 ;

- modified Widman flap, 21 ;

- apically repositioned flap with resective bone surgery, 16.

What adds to the credibility of this study is the fact that the protocol described by Ramfjord himself in 1974, as the « modified Widman flap », does not come out on top. At this stage, it is convenient to discuss the problem of root amputations. There is no study with a control group. Also, the criteria for inclusion are hazy. In spite of these limitations, the results for resective bone surgery are less good when combined with root amputation. The first long-term study (10 years), by ^{Bergenholtz (1972)} on 30 mandibular and 15 maxillary amputations, showed that the prognosis was good with a stable bone level. However, when recurrence of inflammation and pocketing does take place, it is in relation to the cut surface.

It seems that the results are closely correlated to the precise indication of the technique, to the quality of its execution and to the efficacy of the subsequent plaque control. Many authors have presented results that appear to be more an appreciation of their current practice than of the technique. All agree (^{Langer et al., 1981} ; ^{Bühler, 1988}, to mention two of the main papers only) that a larger number of fractures occur in the mandible than in the maxilla and there are more complications of an endodontic nature than periodontal (caries).

When one recommends a technique, it is necessary assess all factors in advance. Indeed, if we compare the results of Bühler and Langer (in these two studies about a third of the resected teeth were lost within 10 years) with those of ^{Hirschfeldt and Wasserman, (1978)} (where about a third of the teeth, without resection, were lost within 22 years), we are led to be cautious.

The controversy about vital and non-vital teeth is no longer necessary. A survey by a group of endodontists, ^{Filipowicz et al. (1984)}, has shown that after attempting conservation of the pulp, only 13 % of the teeth responded positively to vitality tests after 5 years.

Discussion

The problem posed is that of the relative importance of the factors concerned. Is it worth sacrificing supporting bone to eliminate the pocket ? There are only a few studies on this topic and they emanate mostly from the Ramfjord group in Michigan and the Lindhe group in Gothenburg. Although they had slightly differing methods of teaching of oral hygiene, they obtained comparable results whatever the protocol used. However these studies were designated, they are all short-term and to maintain a pathological situation over 5 years cannot be considered to be a success.

It seems that all authors agree that osteoplasty permits a better repositioning of the flap. The controversy concerns ostectomy. In workshop discussions, one notes more arguments against than for. If the general theme of discussion is public health, the earlier concepts recur. Why spend time reshaping bone surgically when it tends to remodel naturally (^{Patur and Glickman, 1962}) ? Why remodel bone when one knows that with a strict plaque-control program, the irregularities disappear and bony support increases (^{Rosling et al., 1976}) ?

Since, for obvious reasons, histological data from re-entry procedures are rare in man, one is often reduced to clinical impressions. However, they are not without value. An experienced practitioner who, over several decades, has treated the same group of patients has accumulated valuable data, the essence of which will have become clarified over time. Much of the bias is neutralized by two essential factors : the very long term and the very large numbers, elements that are not easily available to academic research groups. Hirschfeld and Wasserman reported, in 1978, a follow-up study of 600 patients that had lasted for some of them 40 years and, for one subject, 54 years (a record !). With private clinical practice, one moves away from a practice of dentistry founded on proof to something closer an art. It is obvious that all types of treatment must be adjusted to take into account numerous considerations. Every case, being unique, often requires a unique treatment.

Conclusion

Regardless of the acknowledged indications and contra-indications that have been widely discussed (for review, see ^{Watchtel, 1993}), it seems that in the posterior sectors, where there has been extensive but non-terminal bone loss, resectional bone surgery with apically repositioned flap is one of the more acceptable long-term solutions. Sometimes initially, the new morphology of the arches is a nuisance to patients because of food packing (from lateral origin), but plaque control may be facilitated. One could even say that it compels it. The most suitable instrument is the interdental brush, whose superiority over dental floss has been demonstrated especially in large spaces and concavities (^{Bergenholtz and Olsson, 1984}). It is also quicker and simpler to use (^{Gjermö and Flötra, 1970}). With regard to the root amputation, there is no long-term study that demonstrates an advantage of this method compared with treatment without amputatio

(1) Authors' note : Whilst this article contains a review of the literature from the 19th century (even 18th century with Fauchard), it is almost impossible to consult the older works. We confess not to have seen articles published before 1920. We have only had access to transcriptions of the earlier ones. Without doubting the integrity of authors, we do not ignore the possibility of inaccuracies that may have become incorporated into texts that have been serially reported several times. This is the reason why we are committed to the checking very carefully those details that could be taken out of context. Even if this historical review contains a few errors, we endeavour to describe the major trends that have successively dominated resective periodontal surgery.

(2) For details of terminology, the reader is directed to the glossary at www.parodontologie.com.

Demande de tirés à part

Catherine MATTOUT : 224, avenue du Prado - 13008 MARSEILLE - FRANCE.

BIBLIOGRAPHIE

• Ariaudo A, Tyrell H. Repositioning and increasing the zone of attached gingiva. J Periodontol 1957;28:106-115.
• Ariaudo A, Nabers C, Fraleight C. When is gingival repositionning an indicated procedure. J West Soc Periodont Abstracts 1957;26:106.
• Becker W, Becker BE, Ochsenbein C, Kerry G, Caffesse R, Morrison EC et al. A longitudinal study comparing scaling, osseous surgery and modified Widman procedures. Results after one year. J Periodontol 1988;59:351-365.
• Bergenholtz A. Radectomy of multirooted teeth. J Am Dent Assoc 1972;85:870-882.
• Bergenholtz A, Olsson A. Efficacy of plaque removal using interdental brushes and waxed dental floss. Scand J Dent Res 1984;92:198-203.
• Björn H, Hollender L, Lindhe J. Tissue regeneration in patients with periodontal disease. Odontologisk Revy 1965;16:317-326.
• Black AD. A work on special dental pathology. Chicago : Medico-Dental Publishing Co., 1924 : 196.
• Bühler H. Evaluation of root resected teeth. Results after 10 years. J Periodontol 1988;59:805-810.
• Caffesse RG, Ramfjord SP, Nasjleti CE. Reverse bevel periodontal flaps in monkeys. J Periodontol 1968;39:219-235.
• Carnevale JF, Kaldahl WB. Osseous resectrice surgery. Periodontol 2000 2000;22:59-87.
• Carranza FA. Tratamiento quirúrgico de la parodentosis. Buenos Aires : Imprenta Amorrotu, 1935.
• Carranza F, Carranza F Jr. The management of the alveolar bone in the treatment of the periodontal pocket. J Periodontol 1956;27:29-37.
• Caton J, Zander HA. Osseous repair of an infrabony pocket without new attachment of connective tissue. J Clin Periodontol 1976;3:54-58.
• Crane AB, Kaplan H. The Crane-Kaplan operation for prompt elimination of Pyorrhea alveolaris. Dental Cosmos 1931;73:643-652.
• Cross WG. Bone implant in periodontal diseases. A further study. J Periodontol 1957;28:184-191.
• Dedolph TH Jr, Clark HB Jr. A histological study of mucoperiostal flap healing. J Oral Surg Oral Med Oral Pathol 1958;16:367-376.
• Fauchard P. Le chirurgien-dentiste ou Traité des dents. Facsimilé de l'édition de 1747. Tome I. Paris : Julien Prélat, 1961 ; 494 p.
• Filipowicz F, Umstott P, England M. Vital root resection in maxillary molar teeth. A longitudinal study. J Endodont 1984;10:264.
• Friedman N. Periodontal osseous surgery : osteoplasty and osteoectomy. J Periodontol 1955;26:257-263.
• Gjermö P, Flötra L. The effect of different methods of interdental cleaning. J Periodontal Res 1970;5:230-236.
• Gottlow J. Periodontal regeneration. In : Proceedings of the 1st European Workshop on Periodontology. Ittingen (Switzerland) : Quintessence, 1994 ; 478.
• Grant D. Experimental periodontal surgery. Sequestration of alveolar bone. J Periodontol 1967;38:409-.
• Hirschfeld L, Wasserman B. A long term survey of tooth loss in 600 treated periodontal patients. J Periodontol 1978;49:225-237.
• Isidor F, Karring T. Long term effect of surgical and non-surgical periodontal teatment. A 5-year clinical study. J Periodont Res 1986;21:462-472.
• Kaldahl WB, Kalkwarf KL, Patil KD, Dyer JK, Bates RE. Evaluation of four modalities of periodontal therapy. Mean probing depth, probing attachment level and recession changes. J Periodontol 1988;59:783-793.
• Karring T, Nyman S, Lindhe J, Sirirat M. Potentials for root resorption during periodontal wound healing. J Clin Periodontol 1984;11:41-52.
• Knowles JW, Burgett FG, Nissle RR, Shick RA, Morrisson EC, Ramfjord SP. Results of periodontal treatment related to pocket depth and attachment level. Eight years. J Periodontol 1979;50:225-233.
• Kronfeld R. The condition of the alveolar bone underlying periodontal pockets. J Periodontol 1935;6:22.
• Langer B, Stein S, Wagenberg B. An evaluation of root resections - a ten year study. J Periodontol 1981;52:719.
• Linghorne WJ, O'Connell DC. Studies in the regeneration of supporting structures of the teeth. II. Regeneration of alveolar process. J Dental Res 1951;30:604-614.
• Listgarten MA. Electron microscopic study of the junction between surgically denuded root surfaces and regenerated periodontal tissus. J Periodont Res 1972;7:68-90.
• Manson JD. Bone morphology and bone loss in periodontal disease. J Clin Periodontol 1976;3:14-28.
• Marshall JS. A remarkable case of Phyorrhea alveolaris with reproduction of bone occuring in the practice of Dr Allport, Chicago. J Am Med Ass 1883;1:641-643.
• Matia J, Bissada N, Maybury J, Richetti P. Efficiency of scaling the molar furcation area with and without surgical access. Int J Periodont Rest Dent 1986;3:6-25.
• Nabers C. Repositioning the attached gingiva. J Periodontol 1954;25:38 [abstracts].
• Neumann R. Die Alveolar Pyorreha und ihre Behandlung. Hermann Meusser, 1920.
• Neumann R. Die radikal chirurgische Behandlung der Alveolarpyorrhea. Deutsche Vierteljahresschrift-für Zahnheikunde 1921;37:113-127.
• Nyman S, Lindhe J, Karring T, Rylander H. New attachment following surgical treatment of human periodontal disease. J Clin Periodontol 1982;9:290-296.
• Ochsenbein C. Rationale for periodontal osseous surgery. Dent Clin North Am 1960;4:27-39.
• Ochsenbein C. Current status of osseous surgery. J Periodontol 1977;48:577-590.
• Ochsenbein C. A primer for osseous surgery. Int J Periodont Rest Dent 1986;6:9-47.
• Ochsenbein C, Bohannan HM. Palatal approach to osseous surgery. I. Rationale. J Periodontol 1963;34:60.
• O'Connor TN, Biggs NL. Interproximal bony contours. J Periodontol 1964;35:326-330.
• Orban B. Gingivectomy or flap operation ? J Am Dent Assoc 1939;26:1276-1283.
• Parashis AO, Anagnoon V, Aretzides A, Demetriou N. Calculus removal from multirooted teeth with and without surgical access. II. Comparison between external and furcation entrance width. J Clin Periodontol 1993;20:63-68.
• Patur B, Glickman J. Clinical and roentgenographic evaluation of the posttreatment healing of infrabony pockets. J Periodontol 1962;33:164-171.
• Prichard J. The intrabony technique as a predictable procedure. Periodontol 1957;28:207.
• Prichard J. Gingivoplasty, gingivectomy and osseous surgery. J Periodontol 1961;32:275-290.
• Prichard J. Advanced periodontal disease. Philadelphie : Saunders, 1972.
• Prichard J. Reflexions on osseous therapy. Int J Periodont Rest Dent 1986;6:5-13.
• Ramfjord SP, Nissle RR. The modified Widman flap. J Periodontol 1974;45:601-615.
• Ramfjord SP, Caffesse RG, Morrison EC, Hill RW, Kerry GJ, Appleberry EA et al. 4 modalities of periodontal treatment compared over 5 years. J Clin Periodont 1987;14:445-452.
• Rosling B, Lindhe J, Nyman S. The effect of systematic plaque control on bone regeneration in infrabony pockets. J Clin Periodontol 1976;3:38-53.
• Schluger S. Osseous resection - a basic principle in periodontal surgery. Oral Surg 1949;2:316-327.
• Schluger S, Yuodelis RA, Page RC. Periodontal disease, basic phenomena, clinical management and occlusal and restorative interrelationships. Philadelphie : Lea and Febiger, 1978 : 737 p.
• Spatz S. Early reaction in bone following the use of burs rotating at conventional and ultra speeds. A comparison study. Oral Surg 1965;19:808-816.
• Stallard RE, Hiatt WH. The indications of new bone and cementum formations. I. Retention of mineralized fragments within the flap. J Periodontol 1968;39:273-277.
• Stern IB, Everett F, Robicsek K, Robicsek S. A pioneer in the surgical treatment of periodontal disease. J Periodontol 1965;36:265-278.
• Takei H. The interdental space. Dent Clin North Am 1980;24:2-37.
• Wachtel HC. Surgical periodontal therapy. In : Proceedings of the 1st European Workshop on Periodontology. Ittinger (Switzerland) : Quintessence, 1993 : 478.
• Widman L. The operative treatment of Pyorreha alveolaris. A new surgical method. [Rapport d'un travail de 1918]. Br Dent J 1920;1:293.
• Wilderman M, Pennel B, King K, Barron J. Histogenesis of repair following osseous surgery. J Periodontol 1970;41:551-563.