Effect of root surface alterations on periodontal healing. I. Surgical denudation

Abstract The present study was undertaken to evaluate the effect of root surface denudation on periodontal healing. Twelve teeth, distributed in four squirrel monkeys, were extracted and reimplanted after surgically denuding the coronal root surface of connective tissue fibers and cementum by root planing. The reimplantation schedule provided three teeth for histologic analysis at 1, 3, 7 and 21 days after reimplantation. One day after reimplantation a zone of fibrin enmeshing erythrocytes and inflammatory cells was interposed between the root surface and the remaining periodontal fibers attached to the alveolar bone. Epithelium migrated rapidly along the denuded root, had reached the alveolar crest at 3 days, and was within the ligament space at 7 days. At 21 days, the epithelium was at the apical limit of root instrumentation, which corresponded to the level of attached connective tissue fibers on the root surface. No evidence of new connective tissue attachment was observed on the denuded root surface. It was concluded that the absence of fibers on the root surface resulted in apical migration of the epithelium, and precluded formation of new connective tissue attachment.     

Fibrin linkage: a precursor for new attachment

Previous studies indicated that connective tissue attachment to the root surface appeared to be dependent upon a chronologic healing sequence related to fibrin and collagen interactions. It was the purpose of the present study to try and substantiate this hypothesis by using histologic techniques designed to differentiate between fibrin and collagen during healing at the root surface interface. In four squirrel monkeys, 24 normal teeth were extracted and reimplanted after either (i) surgically denuding the coronal root surface of connective tissue fibers and cementum by root planing or, (ii) surgical denudation followed by topical application of citric acid (pH = 1; 3 minutes). Three specimens were available for histological analysis 1, 3, 7 and 21 days after reimplantation. Mallory's phosphotungstic acid hematoxylin staining technique was used to differentiate between fibrin and collagen. Epithelium migrated rapidly along the denuded, non-acid-treated, root surfaces, had reached the alveolar crest at 3 days, and was within the ligament space to the level of root denudation at 21 days. Epithelium did not migrate apically along denuded root surfaces treated with citric acid. At 1 and 3 days, inflammatory cells were enmeshed in a fibrin network which appeared to be attached to the root surface by arcade-like structures. At 7 and 21 days, the region had repopulated with connective tissue cells, and collagen fibers had replaced the fibrin. It was concluded that collagen fiber attachment to the root surface was preceded by fibrin linkage, and that the linkage process occurred as an initial event in the wound healing response.     

Healing following reimplantation of teeth subjected to root planing and citric acid treatment

The aim of the present study was to examine the effect of citric acid treatment on periodontal healing around teeth which were extracted, root planed and then reimplanted. Maxillary incisors and mandibular incisors, premolars and molars of 5 monkeys were used. The teeth were divided into 3 experimental groups. In 1 group, the teeth were extracted and immediately reimplanted into their own sockets. In a 2nd group, the teeth were extracted, root planed to a level corresponding to 50-75% of the root length and then reimplanted. In the 3rd group, periodontal breakdown extending to 50-75% of the root length was first induced by placing orthodontic elastic ligatures around the teeth. They were then extracted and root planed and transplanted into the sockets of the contralateral, periodontally healthy teeth which had just been extracted. Half the number of the teeth of groups 2 and 3 were treated with citric acid before reimplantation or transplantation. The animals were sacrificed after 6 months of healing. The jaws were removed and histological specimens prepared for microscopic examination. With the exception of a limited coronal regrowth of new cementum in the apical part of the planed portion of a few roots, connective tissue attachment failed to reform on most root surfaces deprived of their periodontal ligament tissue. Healing was most frequently characterized by root resorption and ankylosis. These were the most predominant features of healing both on root surfaces which had been deprived of the ligament tissue by mechanical means or during a course of experimentally-induced periodontal disease, and occurred in citric acid as well as non-citric acid treated roots. The findings imply that the result of healing following tooth reimplantation or transplantation is determined by the type of cells that repopulate the wound area adjacent to the denuded root surface.     

Connective tissue regeneration to periodontally diseased teeth

A study was made to investigate the biological principle of new attachment of connective tissue to periodontally diseased root surfaces utilizing the previous finding that topical application of citric acid to the affected area may stimulate periodontal regeneration. Reconstructive surgery was performed on ten surfaces of nine teeth involved in advanced chronic periodontal disease. A muco-periosteal flap was raised and the apical extent of existing subgingival calculus was demarcated by a notch made through the calculus and into the root. Following thorough instrumentation, a saturated solution of citric acid was applied to the root surface for five minutes and the flap repositioned and sutured. Four months later the teeth with attached periodontal tissues were removed and processed for histological analysis. Connective tissue regeneration characterized by deposition of new cementum, and more coronally, by tightly apposed soft connective tissue had occurred in all specimens. The junctional epithelium ended 1.2–2.6 mm coronal to the apical border of the notch in the various specimens. The results demonstrate that regeneration of periodontal tissues to a root surface that has become denuded as a result of chronic, destructive periodontal disease and that has been covered by calculus is, in fact, a biological possibility. This study does not establish whether or not acid conditioning of the root surface is a prerequisite for new attachment.     

Periodontal healing with citric acid and fibronectin treatment in cats

The glycoprotein fibronectin is thought to act as a biological adhesive and may be useful in attaining fibrous reattachment following periodontal surgery. In the present study the effects of fibronectin, with and without citric acid preconditioning of the root surfaces, were studied following periodontal surgery in cats. The experimental procedures included treatment of exposed root surfaces with citric acid, citric acid and fibronectin together or fibronectin alone, while the control tooth was root planed only. The effects were assessed in the laboratory 24 to 28 days after surgery. When citric acid and fibronectin were applied together, no epithelial proliferation along the root surface occurred and there was evidence suggesting a new fibrous tissue reattachment. Citric acid alone resulted in some new attachment at the base of the defect but epithelium had proliferated along the root surface. When fibronectin was used alone, the epithelium proliferated but there was evidence of windows of connective tissue reattachment to the root surface. Root planing in control teeth was followed by epithelial proliferation to the extent of the root planing. These results suggest that fibronectin when used in combination with citric acid demineralization may lead to enhanced connective tissue reattachment.     

Citric acid conditioning of roots affects guided tissue regeneration in experimental periodontal wounds

The effect of citric acid conditioning of roots on the formation of a new connective tissue attachment was evaluated in the presence of a selective cell population. Fenestration wounds of standard sizes were made on the buccal aspects of mandibular premolars in 6 beagle dogs. The exposed root surfaces were curetted thoroughly and conditioned with citric acid (experimental) or distilled water (controls) for 3 minutes. The wounds were covered with Millipore filters to facilitate population of curetted root surfaces by cells from the adjacent periodontal ligament. Histologic analysis was made after 3 months of healing.
The extent of new connective tissue attachment varied in both the experimental and control specimens. The percentage of surgically denuded root surface showing new cementum with inserting fibers was significantly lower in the experimental group compared with the controls. Root resorption was seen in both experimental and control specimens but involved a significantly larger percentage of denuded root surface in the experimental specimens. Ankylosis occurred more frequently in the experimental group compared with the controls, while no difference was seen in the degree of bone regeneration.
The results indicate that new connective tissue attachment can form on denuded root surfaces by a selective cell population. Citric acid conditioning of roots appears to either delay or complicate healing in the presence of a selective cell population.     

Cell and fiber attachment to demoralized dentin A comparison between normal and periodontitis-affected root surfaces

The purpose of the present study was to compare and contrast cellular, connective tissue, and epithelial responses to dentin specimens derived from the roots of either normal or periodontitis-affected human teeth after surface demineralization. Rectangular dentin specimens, with opposite faces of root and pulpal dentin, were derived from beneath root surfaces covered by periodontal ligament (normal) or calculus-covered areas of periodontitis-affected teeth. In each of the groups, the specimens were treated with citric acid (pH 1 for 3 min), whereupon they were implanted transcutaneously into incisional wounds on the dorsal surface of rats with one end of the implant protruding through the skin. 4 specimens were available in each group at 10 days after implantation. Histologic and histometric analyses of the root surfaces of the implants included counts of adhering cells, evaluation of connective tissue fiber relationships, and assessment of epithelial migration. New connective tissue attachment with inhibition of epithelial migration occurred in both groups. Cementum formation was not present. Comparisons between the groups showed no significant differences regarding length of implant surface adjacent to connective tissue, number of attached cells, or density and diameter of attached fibers. The fiber attachment system which had developed on these demineralized surfaces seemed intrinsic to the connective tissue location, and differed morphologically from corresponding fibers attaching the root surface in a normal periodontium. It was concluded that there were no observable differences between the new connective tissue attachment systems which developed on demineralized dentin from either normal or periodontitis-affected root surfaces.     

The regenerative potential of the periodontal ligament

Abstract The aim of the present study was to examine if new cementum and new attachment may form during healing of a wound prepared in such a way that preference is given to periodontal ligament ceils to repopulate the wound area adjacent to a root which has been surgically deprived of its periodontal ligament and cementum layer. The maxillary lateral incisors and mandibular canines in three monkeys were used for experimentation. Following elevation of a mucoperiosteal flap, the buccal and approximal alveolar bone was removed within an area extending from the mid-root level to a level 2 mm apical to the marginal bone crest. Following bone removal, the root surfaces were curetted in order to remove the cementum layer. Notches were prepared in the roots to demarcate the denuded root portion. Prior to repositioning of the tissue flap a millipore filter was placed over the treated area in order to prevent the gingival connective tissue from coming into contact with the root surface during healing. The animals were sacrificed 6 months after surgery. The jaws were removed and histological sections of the experimental teeth and surrounding periodontal tissues were produced. New cementum with inserting collagen fibers was observed on the curetted root surfaces. However, this result of healing did not consistently occur along the entire length of the curetted root portion. In the coronal part of the wound, healing was frequently characterized by connective tissue adhesion to the root surface without signs of cementum formation and fibrous attachment. The results of the experiment suggest that the periodontal ligament cells possess the ability to reestablish connective tissue attachment.     

Fine structure of new connective tissue attachment following acid treatment of experimental furcation pockets in dogs

Periodontal furcation pockets were created in mandibular premolar teeth of three beagles and subsequently treated by reconstructive procedures including flap surgery, root planing and application of citric acid. Block biopsies removed after 1, 2, 3 and 6 weeks showed new attachment of connective tissue to the periodontally involved root surface. This was established by progressive interdigitation of new collagen fibrils and the acid-exposed fibrils of the dentin matrix along the root planed surface as well as within the orifice of dentinal tubules. The results supplement those of a previous study (Ririe, Crigger & Selvig 1980) and show that attachment of soft connective tissue to a root planed and acid-conditioned dentin surface can be achieved by the same mechanism irrespective of whether the root surface has been surgically denuded or has been exposed to the environment of an experimental periodontal pocket.     

The Production of a Root Surface Smear Layer by Instrumentation and its Removal by Citric Acid

The purpose of this study was to characterize, using the scanning electron microscope, the nature of root surfaces denuded by (1) root planing alone or (2) root planing plus citric acid treatment. Six teeth were extracted from three squirrel monkeys, and the coronal half of the root surface was planed to remove attached periodontal ligament fibers and cementum. Citric acid, pH 1, was topically applied to the denuded surfaces of 3 teeth. The surface of only root-planed specimens had an irregular surface which corresponded to a smear layer. Root surfaces that had received acid treatment after root planing exhibited funnel-shaped orifices of dentinal tubules, and intertubular zones with a fibrillar, mat-like morphology. The implications of the different root-surface characteristics are discussed.     

Cell and fiber attachment to demineralized dentin from normal root surfaces

The study assessed connective tissue and epithelial responses to dentin specimens (obtained from normal roots of human teeth) after surface demineralization. Rectangular dental specimens with opposite faces of root and pulpal dentin were prepared from beneath root surfaces covered by periodontal ligament. One-half of the specimens were treated with citric acid, pH 1, for 3 minutes, while the remainder served as untreated control specimens. Specimens were implanted vertically into incisional wounds on the dorsal surface of rats with one end of the implant protruding through the skin. Four specimens in each group were available 1, 3, 5 and 10 days after implantation. Histologic and histometric analyses included counts of adhering cells, evaluation of connective tissue fiber relationships and assessment of epithelial migration. Analyses within each group comparing root and pulpal surfaces showed no differences between any of the parameters. Comparisons between experimental and control groups showed that demineralized surfaces had a greater number of cells attached, fiber attachment occurred and epithelial downgrowth was inhibited. The fiber attachment to experimental specimens differed morphologically from fiber attachment to normal root surfaces: the number of fibers attached per unit length and the diameter of attached fibers were significantly less on experimental specimens. Demineralized specimens at 10 days had a distinct eosinophilic surface zone. Surface demineralization of dentin predisposed toward a cell and fiber attachment system which inhibited migration of epithelium.     

Healing following surgical treatment and root demineralization in monkeys with periodontal disease

Abstract The aim of the present investigation was to study the potential for new attachment to root surfaces which 1) had become devoid of attachment either by mechanical removal or in conjunction with experimental periodontal tissue breakdown and 2) subsequently had been conditioned with citric acid. Three adult monkeys (Macaca cynomolgus) were used. Periodontal pockets were produced during a 2–6-month period around the maxillary central incisors and first and second premolars and around the mandibular central incisors by the placement of orthodontic elastics. Surgical treatment of the pockets was carried out with the use of a flap procedure. During surgery, the roots of the premolars on the right side of the jaws were first scaled and planed and subsequently treated with citric acid. Citric acid conditioning was not performed on the premolars of the left side. In the front tooth regions the surgical treatment included citric acid conditioning of the mandibular central incisors in two animals and of the maxillary central incisors in one animal. Surgical procedures were also carried out around the mandibular second premolars and first molars which had not been exposed to periodontal tissue breakdown. Folio wing flap elevation, the buccal bone plate was mechanically removed within an area corresponding to that of bone loss produced by the experimental periodontitis model in the maxillary premolars. The cementum layer of the surgically denuded portion of the roots was removed by root planing. In all treated teeth, a notch was prepared in the buccal root surfaces at the level of the alveolar bone crest. Six months following treatment the animals were sacrificed and histologic sections prepared. Linear measurements were made along the root surface from the apical border of the prepared notch to the apical termination of the junctional epithelium and to the crest of the alveolar bone. The results demonstrated that cementum formation and new connective tissue attachment did not occur on root surfaces previously exposed to periodontal pockets and subsequently subjected to scaling and root planing or on root surfaces surgically deprived of their supporting bone and previously “non-exposed” cementum layer. Cementum formation and new connective tissue attachment also failed to occur following citric acid conditioning of root dentin surfaces. In all sections representing all three treatment groups, the attachment between the gingiva and the root was established by epithelium.     

New connective tissue attachment in beagles with advanced natural periodontitis

A pilot study using advanced, natural periodontitis in beagle dogs was performed to determine the possibility of achieving periodontal regeneration to an extent that would bring the attachment back to the original level at the cemento-enamel junction. Nine mandibular premolars in two beagles were treated with a mucoperiostal flap procedure including citric acid conditioning of the root surfaces and coronal placement and suturing of the flaps.
Histologic examination of buccal and lingual root surfaces demonstrated new connective tissue attachment including the formation of new cementum back to or near the cemento-enamel junction in most specimens. The mean epithelial migration apical to the cemento-enamel junction amounted to only 0.3 mm. The formation of new crestal alveolar bone, however, was limited. The histological analyses of the furcation areas showed new connective tissue attachment in six of the nine specimens.
The result indicates that it is biologically possible to obtain new connective tissue attachment up to the original level of attachment at the cemento-enamel junction following treatment of advanced natural periodontitis in beagle dogs. It seems to be possible to accomplish this connective tissue attachment not just in isolated areas, but at several circumferential locations on the teeth as well.     

Ultrastructural study of initial attachment by human fibroblast-like cells on tooth roots in vitro

The purpose of this study was to determine the longitudinal effect of demineralized treatment of root surfaces on initial attachment, growth and differentiation of human periodontal ligament fibroblast-like cells in vitro. Cementum and dentin fragments were prepared from intact extracted human teeth for orthodontic reason. The root fragments of one group were not demineralized. Those of the other groups were demineralized by either citric acid (pH = 1.0, 3 min) or EDTA (pH = 7.4, 30 min). Plastic sheets served as controls. Human periodontal ligament fibroblast-like cells were incubated on root fragments and plastic sheets. After incubation, the root fragments and plastic sheets were examined by electron microscopy. The collagen fibers were exposed to the root surface by demineralized root surface. The exposed collagen fibers showed an effect on the cell attachment and growth, and the cells produced collagen fibers in the extra-cellular space of the root surface. Demineralization of dentin fragments were more strongly affected in cell attachment, growth and differentiation than demineralization of cementum fragments. Citric acid demineralization of dentin fragments had a greater effect on cell attachment, growth and differentiation than EDTA demineralization of dentin fragments. The results suggest that citric acid demineralization of dentin fragments may provide the most effective dental surface for the establishment of connective tissue attachment after periodontal treatment.     

Cell and fiber attachment to demineralized cementum from normal root surfaces

The purpose of this study was to assess connective tissue and epithelial responses to cementum (from normal human root surfaces covered by periodontal ligament) after surface demineralization with citric acid. Each rectangular specimen had a face of cementum and an opposite surface composed of pulpal dentin. One half of the specimens were treated with citric acid (experimental group), while the remainder served as untreated control specimens. Specimens were implanted vertically into incisional wounds on the dorsal surface of rats with one end of the implant protruding through the skin. Four specimens in each group were available for examination 1, 3, 5, and 10 days after implantation. Histologic and histometric analyses of the implants included counts of adhering cells, evaluation of attached connective tissue fiber density and diameter, and assessment of epithelial migration. At day 1, a distinct lighter staining zone was present on the surface of both cementum and dentin in the experimental group which corresponded to a zone of surface demineralization produced by the acid treatment. Histometric comparisons between experimental and control groups at 10 days showed a greater number of cells attached to demineralized cementum surfaces. Also, a connective tissue fiber attachment system had developed on these experimental surfaces, but which differed morphologically from periodontal ligament fiber attachment to normal cementum. It was concluded that citric acid treatment can surface demineralize cementum from normal roots, and that the surface demineralization of this cementum facilitated a cell and fiber attachment to the cementum surface.     

Connective tissue attachment formation following exclusion of gingival connective tissue and epithelium during healing

The present investigation was designed to evaluate the potential for reformation of connective tissue attachment on exposed and planed root surfaces by preventing the dentogingival epithelium and the gingival connective tissue from interfering with healing following periodontal surgery. Following the elevation of soft tissue flaps, the buccal and proximal alveolar bone of 24 teeth (48 roots) was removed to mid-root level in 6 monkeys and the exposed root surfaces were carefully planed in order to remove the root cementum. Before the flaps were repositioned and sutured, a membrane (Millipore® filter) was placed over the denuded part of the root surfaces of 16 teeth (test teeth) in order to prevent the epithelium and the gingival connective tissue from interfering with healing. The membrane was adjusted to cover the tooth surfaces from midcrown level to approximately l mm apical to the bone crest. No membranes were placed around the remaining 8 teeth (control teeth) before flap repositioning. The animals were sacrificed 6 months after surgery. The jaws were removed and histological sections of test and control teeth including their buccal periodontal tissues were produced. Nine of the test teeth had to be excluded from examination due to technical failures in the surgical procedure or tissue preparation. New cementum with inserting collagen fibers was observed on all remaining 14 test roots. The length of this newly formed fibrous attachment corresponded to approximately 50% of the distance from the apical extension of root planing to the cemento-enamel junction. In the majority of the control teeth no new attachment had formed but a “long” junctional epithelium was lining the root surfaces to the apical extension of root planing. In 3 control roots a small amount of new cementum with inserting collagen fibers was found in the most apical area of root planing. The results showed that the reformation of a connective tissue attachment was considerably favored by the placement of membranes which prevented the dentogingival epithelium and the gingival connective tissue from interfering with healing.     

New attachment formation as the result of controlled tissue regeneration

The present study was designed to examine whether new attachment forms on root surfaces previously exposed to plaque by preventing the oral epithelium and the gingival connective tissue from participating in the process of healing following treatment. 4 roots in each of 3 monkeys were used as test units while the roots of contralateral teeth served as controls. A surgical procedure was first used to expose the coronal half of the buccal root surfaces. Plaque was allowed to accumulate on the exposed surfaces for a period of 6 months. Subsequently, soft tissue flaps were raised and the root surfaces were carefully scaled and planed. The crowns of the test and control teeth were resected and the mucosal flaps were repositioned and sutured in such a way that the roots were properly covered. Immediately prior to suturing, membranes (Millipore filter or Gore-tex membrane) were placed over the denuded root surfaces of the test teeth in order to prevent granulation tissue from the soft tissue flaps from reaching the roots during healing. The monkeys were sacrificed 3 months later. The jaws were removed and histological sections of test and control roots including their periodontal tissues were produced. New cementum with inserting collagen fibers was observed on the previously exposed surfaces of both test and control roots. However, the test surfaces exhibited considerably more new attachment than the control surfaces, indicating that the placement of the membrane favoured repopulation of the wound area adjacent to the roots by cells originating from the periodontal ligament.     

Evaluation of human recession defects treated with coronally advanced flaps and either enamel matrix derivative or connective tissue. Part 2: Histological evaluation

BACKGROUND: A number of surgical procedures are effective in covering denuded root surfaces. The first paper in this series evaluated the subepithelial connective tissue graft and the coronally advanced flap with enamel matrix derivative (EMD). That paper revealed no significant difference in the percent of root coverage between the two treatments (P = 0.82). There is limited human histological evidence of the type of attachment achieved with these types of procedures. This paper presents a human case report detailing the histological nature of the attachment of these two treatments to the root surfaces previously exposed by recession. METHODS: One patient presented with two hopeless teeth that were randomized to receive either a subepithelial connective tissue graft or a coronally advanced flap plus EMD. The surgery was accomplished in accordance to the protocol previously described. The teeth and a small collar of tissue were removed at 6 months and underwent histological analysis. RESULTS: Histological evaluation of the subepithelial connective tissue graft revealed a connective tissue attachment between the tooth and graft, and no histological evidence of cementum, bone, or periodontal ligament (PDL) and, therefore, regeneration. In addition, there appeared to be some resorption of the dentin adjacent to the graft. Histological evaluation of the coronally advanced flap with EMD revealed new cementum, organizing PDL fibers and islands of condensing bone at a constant distance from the root surface. CONCLUSIONS: The subepithelial connective tissue graft in this study was found to have adhered to the root surface primarily by a connective tissue attachment with some evidence of root resorption. The coronally advanced flap with EMD was found histologically to have all the tissues necessary for regeneration: new cementum, organizing PDL fibers, and islands of condensing bone. These histologic sections strongly suggest that enamel matrix derivative works in a biomimetic fashion by mimicking the natural process of tooth development.     

Healing after root reimplantation in the monkey

The aim of the present investigation was to evaluate the regenerative potential of the periodontal tissues following tooth reimplantation using a model which excluded the dentogingival epithelium from the process of healing. Maxillary and mandibular incisors, premolars and molars of 5 monkeys were used. Following root filling of all experimental teeth, the teeth were divided into 3 experimental groups. In 1 group, the teeth were extracted following the elevation of full thickness flaps. The crowns were separated from the roots at the level of the buccal cemento-enamel junction and the roots immediately reimplanted into their sockets. The flaps were replaced and sutured to accomplish complete coverage of the roots. In a 2nd group, the teeth were subjected to the same experimental procedure, but in addition, the buccal alveolar bone was removed to about half its original height prior to root reimplantation. The teeth of the 3rd group were subjected to identical experimental procedures as for group II with the addition that the buccal root surfaces were planed to the level of the surgically created bone crest. The animals were sacrificed after 6 months of healing. The jaws were removed and histological specimens prepared for microscopic examination. The results showed that a complete fibrous re-attachment formed onto roots on which the original periodontal ligament tissue was preserved. This occurred irrespective of whether the roots were reimplanted into sockets with normal (group I) or reduced (group II) bone height. When the original periodontal ligament tissue was removed by root planing before reimplantation (group III), healing resulted in a significant amount of new connective tissue attachment. However, coronal to the newly formed fibrous attachment, the root surface frequently showed signs of resorption and particularly so in those roots which remained covered by the soft tissue during the entire course of healing. In the majority of the roots which perforated the covering soft tissue during the early phase of healing, the dentogingival epithelium had migrated apically into contact with the coronally generated fibrous attachment. In these cases, root resorption was never discernible. New bone formation occurred to a variable extent in the roots of groups II-III. No relationship was found, however, between the amount of connective tissue reattachment or new attachment and newly formed alveolar bone, which in turn indicates that bone tissue regrowth and periodontal ligament regeneration are unrelated phenomena.     

Citric acid treatment of periodontitis-affected cementum

Previous studies have described an inconsistent histological occurrence of a zone of surface demineralization on periodontitis-affected cementum following treatment with citric acid, and a lack of connective tissue attachment to the latter surfaces. In view of these findings, the purpose of the present study was to use scanning electron microscopy to examine the surface morphology of cementum from normal and periodontitis-affected root surfaces following citric acid treatment for differences in the effects of the demineralizing solution on these surfaces. Cementum surfaces were derived from the roots of extracted human teeth from areas beneath attached periodontal ligament fibers (normal) and calculus deposits (periodontitis-affected). 5 specimens were evaluated in both groups. Periodontal ligament fibers were removed from normal root surfaces with a curette, and calculus deposits were removed from periodontitis-affected root surfaces using an ultrasonic scaler. The resultant 5 specimens in each group were then sectioned in half, one-half serving as the untreated control and the other as the experimental, citric acid treated specimen. Experimental specimens were immersed in a saturated solution of citric acid, pH 1 for 3 min and then rinsed in tap water. Both control and experimental specimens were dehydrated in ethanol, critical-point dried, sputter-coated with gold and examined in the scanning electron microscope for morphological characteristics. Citric acid treatment of cementum from normal root surfaces produced an undulating, markedly fibrillar surface morphology which is consistent with the exposure of a fibrillar, collagen substrate. Periodontitis-affected cementum, however, was not appreciably altered in appearance by the citric acid treatment, having only a faint mat-like surface texture. (ABSTRACT TRUNCATED AT 250 WORDS)