European Journal of Orthodontics 23 (2001) 671681 2001 European Orthodontic SocietyTissue reaction to orthodontic tooth movementa new paradigm*Birte MelsenDepartment of Orthodontics, Royal Dental College, Aarhus University, Denmark Direct or indirect resorption are both perceived as a reaction to an applied force.This is in contrast to orthopaedic surgeons who describe apposition as the reaction to loadingof bone. The article reviews the literature on intrusion of teeth with periodontal breakdown,and on the basis of clinical and experimental studies. The conclusion is reached thatintrusion can lead to an improved attachment level, and that forces have to be to low andcontinuous. The tissue reaction to a force system generating translation of premolars and molars inthe five Macaca fascicularis monkeys is described. Three force levels, 100, 200, and 300 cNwere applied for a period of 11 weeks. Undecalcified serial sections were cut parallel to theocclusal plane and a grid consisting of three concentric outlines of the root intersected bysix radii was placed on each section so that areas anticipated to be subject to differingstress/strain distributions were isolated. A posteriori tests were utilized in order toseparate areas that differed with regard to parameters reflecting bone turnover. Based on these results and a finite element model simulating the loading, a newhypothesis regarding tissue reaction to change in the stress strain distribution generated byorthodontic forces is suggested. The direct resorption could be perceived as a result of lower-ing of the normal strain from the functioning periodontal ligament (PDL) and as such as astart of remodelling, in the bone biological sense of the word. Indirect remodelling could beperceived as sterile inflammation attempting to remove ischaemic bone under the hyalinizedtissue. At a distance from the alveolus, dense woven bone was observed as a sign of aregional acceleratory phenomena (RAP). The results of the intrusion could, according to thenew hypothesis, be perceived as bending of the alveolar wall produced by the pull fromSharpeys fibres.SUMMARYIntroductionThe orthodontic literature is characterized bymany controversies in relation to tooth movement.This article will suggest possible solutions,supported by the results of studies carried outon intrusion of periodontally damaged teeth. Inaddition, the biological reaction of alveolar boneto orthodontic tooth movement will be analysedin the light of the strain distribution of theperiodontal tissues and, based on this, a newperception of the tissue reaction to orthodonticforces will be suggested.*Sheldon Friel Memorial Lecture. A possible explanation for the controversiesregarding tooth movement may be that thedesign of the clinical or experimental studiesvaries, resulting in conclusions that cannot becompared. It is important to differentiate betweenwhat is possible in an adult and in a growingindividual. It is likewise important to note thatthe local environment has a major influence;the healthy and inflammed periodontium reactsdifferently. The force magnitude is often the onlyparameter considered by investigators. Differenttypes of tooth movements do, however, generatedifferent force distributions. Since it is mostlikely that tissue reaction is a result of changes inthe stress/strain distribution in the periodontium,

672information on the moment to force ratio of theforce system, on the anatomy of the tooth, andon the constraints determined by the perio-dontium is required. Another explanation for the controversies inthe orthodontic literature could be the lack ofuniformity and definition. An example of the useof similar words for different events is the termintrusion, which has been used to describethe reaction following both an occlusal onlay,the displacement of a tooth after tipping, and theresult of vertical forces developed by differentorthodontic appliances. Compression of themarginal periodontium on the side toward which atooth is tipped as described by Polson et al. (1984)may be comparable to the compression of theperiodontal ligament (PDL) described by Reitan(1964) after occlusally directed force application. The evaluation of the amount of intrusion isalso controversial. The amount of true intrusiondepends on the type of appliance used (Sanderet al., 1996; Melsen et al., 1997; Nanda, 1997).Weiland et al. (1996) demonstrated that intrusionoccurred only when applying the segmented archapproach, whereas this is not the case when acontinuous arch is used. The predictive valueof treatment time, force level, and amount ofintrusion vary between studies, and differencesin appliance design renders it impossible tocompare the results. The different results can, onthe other hand, be explained by the differencein force level and in the load deflection rate usedin the two approaches. A third controversy regarding intrusion is itsrelationship to root resorption. Reitan (1964)demonstrated that intrusion caused both severeroot resorption and reduction in the height ofthe alveolar process. Dellinger (1967), however,when intruding primate premolars, emphasizedthe importance of the force level for rootresorption. Stenvik and Mjr (1970) analysedhuman premolars following intrusion andconfirmed this. The increased risk of apical rootresorption seen in relation to intrusion has beenreported in several studies (McFadden et al., 1989;Costpoulos and Nanda, 1996; Baumrind et al.,1996; Faltin et al., 1998; Parker and Harris, 1998). Intrusion seems to be a treatment with a highrisk of adverse effects in patients with a deepB. M E L S E Noverbite caused by an over-eruption of theincisors. If they are to be treated orthodonticallywith a normal periodontium it can be anticipatedthat the height of the alveolar process will bereduced as a result of this intrusion. However,this is not desirable in patients with a reducedperiodontium who, in addition to a deep over-bite, suffer from increased clinical crown length.If intrusion would result in root resorption anda reduced height of the alveolar process asdescribed above, such treatment would seem tobe unwise in these patients. The major concern is therefore to find ananswer to the following question: How doesintrusion affect the clinical crown length and themarginal bone level in patients with a reducedperiodontium?Subjects and materialsA prospective clinical study of 30 consecutivelytreated subjects was conducted, where patientswith a reduced periodontium and a deep over-bite with an identified treatment aim to intrudethe upper incisors were selected (Melsen et al.,1989). All patients underwent a prophylaxisprogramme including motivation and instruction.Once the oral hygiene was satisfactory, a modifiedWidman flap surgery was performed and thegingiva repositioned apically, so that no pocketsabove 3 mm were present. Following healing,impressions were made and study casts produced.A lateral head film and four periapical radio-graphs were taken, the latter by means of a filmholder allowing for reproducible exposure of theindividual teeth. The orthodontic treatment wasthen initiated with an appliance comprising threepassive segments of full wire size stainless steel:two lateral segments connected transpalatallyand an anterior segment consolidating thefour upper incisors. The intrusion of the fourupper incisors was undertaken using a 0.017 0.025-inch TMA base arch adjusted for 50 g ofintrusive force corresponding to 12.5 g per tooth.The point of force application and the line ofaction of the force were monitored accordingto the need of the patient, so that either anintrusion in combination with a proclination ora retroclination was obtained. The appliance

T I S S U E R E AC T I O N A N D O RT H O D O N T I C S673clinical crown height and increase the area of thealveolus. As a consequence of these results anotherquestion arose: How does intrusion of teeth witha reduced periodontium affect the attachmentlevel? To answer this question an experiment wascarried out on six adult Macaca fascicularismonkeys. Periodontal breakdown was producedby the placement of orthodontic elastics aroundthe premolars and the upper incisors, which werereplaced with new elastics until pockets of 4 mmwere obtained. Periapical radiographs verified abreakdown of the marginal bone correspondingto a minimum of 2 mm. At this point, reversebevel flap surgery was performed and areference notch corresponding to the most apicalpart of the epithelial junction was produced witha round burr (Figure 2). Following surgery theright side of the jaw was kept meticulously cleanwith brushing and chlorhexidine rinsing twice aweek, while the other side served as the control.One week following surgery an orthodonticappliance for the intrusion of upper incisorsand premolars was inserted in five monkeyswhereas one served as the control. The appliancecomprised passive anchorage units of cast stain-less steel caps for the molars, passive steel wireconsolidating the incisors, and active U-shapedintrusion wires (0.018-inch TMA) extendingfrom the molar splint and loading the occlusalsurfaces of the premolar with 10 g per tooth.Intrusion of the incisors was undertaken withthe same force level by means of a base arch(0.018-inch TMA) extending from the molarsplints. The observation time was 3 and 1216weeks. At the end of the observation period, theanimals were killed under Ketalar anaesthesiawith perfusion of buffered formalin. The jawswere excised and the tissue blocks decalcifiedin EDTA and double embedded in paraffincelloidin. Eight-m thick serial sections were cutparallel to the mesio-distal axis, and stainedalternatively with hematoxylin and eosin, andvan Giesen connective tissue stain. The following parameters were evaluatedpost-treatment: the relationship between theattachment level at the start of treatment, asindicated by the apical limit of the notch (AM),was adjusted monthly until sufficient clinicalmovement was obtained. The treatment timevaried between 8 and 18 months. At the end oftreatment a further set of records was obtained. The amount of intrusion was measured withrespect to a co-ordinate system, where the X-axisrepresented the palatal plane and the Y-axisa perpendicular through the pterygomaxillarypoint. The displacement of a best-fit templatetransferred from the first to the last cephalogramwas used in order to minimize the error of themethod. Intrusion of the apex, the centroid, andthe incisal edge (incision) was recorded. Theclinical crown height was measured on the studycasts with an electronic calliper to within 0.01 mm.For each of the four upper incisors the distancebetween the incisal edge, and the highest point ofthe gingival margin was measured at the begin-ning and end of treatment, and the differencerecorded. Based on the periapical radiographs,drawings were made, and the area of the alveolusfor the individual teeth was calculated andcorrected for shortening due to root resorption(Melsen et al., 1988).ResultsThe results revealed that the intrusion of theindividual teeth and the single reference pointsvaried according to the type of movementperformed (Table 1 and Figure 1A). The clinicalcrown height was reduced in 28 subjects and thearea of the alveolus was increased in 25 (Tables 2and 3; Figure 1B,C). Only two patients withoutreduction of the clinical crown height hadpockets above 3 mm at the end of treatment. Based on these results it was concluded thatclinical intrusion with low continuous forces ofperiodontally healthy teeth is recommended,and that this type of treatment may reduce theTable 1True intrusion-range in mm.MinApexCRIss2 02Max546

674B. M E L S E NFigure 1 (A) Examples of tooth movements performed in an individual patient. (B) Immediatelyfollowing periodontal surgery. Note that the clinical crown length is increased, but that there has beenprevious shortening of the incisors by grinding the incisal edge once the teeth had started to elongate.(C) Following orthodontic intrusion. Note the improved gingival level and aesthetic appearance.Reproduced with kind permission from Melsen et al. (1989).Table 2 Change in bone support area of bonyalveolus (expressed as a percentage).x676Min15Max22Table 3 Reduction in crown length (mm).x1.1Min2.1Max4Twenty-five out 30 had an increased area.Twenty-eight out 30 had a reduction.

T I S S U E R E AC T I O N A N D O RT H O D O N T I C S675Figure 2 (A) Photograph of the teeth of a monkey during surgery. An important breakdown of the periodontium has takenplace. (B) Reference points used for the evaluation of the changes occurring during treatment. (C) Clinical photograph ofan intruded tooth (c), where no attention has been given to the periodontal health. The tooth is intruded almost to the levelof the neighbouring tooth (d). (D) Microphotograph of the same situation. Note that the attachment level (a) is still at thenotch and that there is a clear osteoclast activity reducing the height of the alveolar ridge (b). The intruded tooth (c) hasbeen intruded almost to the level of the apex of the neighbouring tooth (d). (E) Composite of a periapical radiograph andthe histological image of the same area. Note that the intruded tooth (a) has kept the same bone level (b) as the non-intruded tooth and the intrusion has thus been into bone. (F) Micrograph illustrating the new attachment generated duringintrusion. Note that the notch (n) is significantly below the bone level, and the collagen fibres are inserting into the cellularcementum coronally to the previous attachment level (a, the apical level of the notch). The new epithelial junction (e) hasbeen displaced coronally. Reproduced with kind permission from Melsen et al. (1988).

676the bone level (BL), the gingival level (GM)and the level of the junctional epithelial (EJ)(Figure 2B). In addition the width of the PDL,the angulation between the collagen fibres ofthe PDL and the alveolar wall were measured.The cellular density close to the alveolar wall wasestimated and finally stereometric measurementswere applied for the determination of thenuclear volume of the cells in the vicinity of thealveolar wall (Melsen and Kragskov, 1992). The results clearly demonstrated that the tissuereaction to intrusion of periodontally-damagedteeth depended on the periodontal status ofthe teeth. Where inflammation occurred, abreakdown of the alveolar bone was observed(Figure 2C,D). In the case of a good oral hygienelow constant forces passing close to the long axisof the teeth resulted in a coronal displacement ofthe attachment level in all teeth varying from 0.7to 2.4 mm, with an average of 1.3 mm (Table 4,B. M E L S E NFigures 2E,F). The width of the PDL wasnarrowed, the collagen fibres were stretched inan apical direction and the cell density in thevicinity of the alveolar wall was increased(Table 5). Ischaemic areas were avoided and thecells were stimulated to an increased DNAsynthesis, as reflected by the shift of the cellpopulation to a larger component with a largevolume of the nucleolus. This study led to theconclusion that, in the case of a healthy perio-dontium, intrusion improved the periodontalstatus, since the attachment level was displacedin an apical direction in all intruded teeth. The discrepancy in the literature regardingintrusion, particularly of periodontally-involvedteeth, can be explained by difference in perio-dontal health (Ericsson and Thilander, 1978;Polson et al., 1984). The existence of pockets,even if kept clinically healthy, may still presenta risk factor. The tissue reaction generated byTable 4 Variations in apical limit of the notch (AM), the bone level (BL), the gingival level (GM), and thelevel of the junctional epithelium (EJ) with and without orthodontic intrusion of the teeth.With intrusion+ HygienexAMBLAMGMAMEJ1.33.71.5SD0.290.990.76 Hygienex0.32.80.9SD0.280.940.75Without intrusion+ Hygienex0.42.10.3SD 0.213.7 0.32 Hygienex 0.1 1.70.1SD0.50.480.67Table 5 Variation in histological parameters of the periodontal ligament of orthodontically intruded teeth,teeth with mechanical periodontal support and teeth without periodontal disease.Intruded teethTeeth with reducedsupportx247.4469.5544.02182.03SD24.093.264.4345.08Teeth with normalsupportx188.5357.9942.41177.36SD9.755.573.2650.02xThe width of the periodontal ligamentAngulation of the collagen fibres to the root surfaceThe relative extension of the root surface covered by cellsNuclear volume138.5341.2759.16235.30SD9.623.903.3244.61

T I S S U E R E AC T I O N A N D O RT H O D O N T I C S677occurrence of hyalinization of the PDL followinghigh force application. Bone biologists, on theother hand, relate loading to formation (Frost,1986). The question to be asked was thus: Whydo orthodontists relate loading to resorption,while bone biologists relate it to formation? One of the major theories that relatesmechanical load to biological reaction is themechanostat theory developed by Frost (1987),who described a relationship between variousstrain values and the balance between modellingand remodelling of bone (Figure 3). In the caseof low strain values, a net loss of bone occursas a consequence of increased remodelling. Withincreasing strain, modelling is initiated and apositive balance is achieved. Where the straincurve crosses the neutral line resorption andapposition are in balance, and the newly-formedbone consists of lamellar bone. In contrast,woven bone is formed as a result of even largerstrains. Still higher strains will result in a negativebalance, since repair cannot keep up with theoccurrence of micro-fractures (Burr et al., 1985).The borderline between a noxe, i.e. a traumaticstimulus, and mechanical stimulus resulting inan increased bone mass, has not yet beenestablished. It is likely that the present modelevokes strain values perceived as trauma, as wellas strain values perceived as a mechanical usageprovoking a structural adaptation to mechanicalan orthodontic force system resembles thatof inflammation and, if not kept sterile, willresult in an uncoupling of tissue resorption andformation, leading to a negative balance, i.e.more breakdowns will occur. Waerhaug(1978) demonstrated that, in the case of a pocketdeeper than 4 mm, it is practically impossible toobtain satisfactory cleansing by means of scalingwithout surgical exposure. Therefore, it canbe concluded that intrusion of teeth with areduced periodontium should only be carriedout in patients with a healthy periodontiumwithout pathologically increased pockets. Differences in the type of force delivery mayalso contribute to the different results reportedin the literature (Polson et al., 1984; Melsen,1986; Goerigk et al., 1992; Wilson et al., 1994;Costopoulos and Nanda, 1996; Baumrind et al.,1996; Parker and Harris, 1998). Intrusion shouldbe completed with light and constant forces(515 g per root) so that the visco-elasticity ofthe PDL is taken into consideration. High forceswill result in a trauma to the PDL followed byrepair, which starts a breakdown of the alveolus.Forces above a certain magnitude result inischaemia of the compressed PDL with initiallyno true intrusion occurring. This reaction iseasily distinguished from the stretching of thePDL fibres seen when the tooth is intruded withlight forces along its long axis. With light forcesno compression of the marginal part of thealveolus occurs; on the contrary a stretching ofthe fibres and increased cellular activity isobserved (Melsen and Kragskov, 1992). Theincreased cellular activity could also explainthe coronal displacement of the attachment level. Another controversy in relation to tissuereaction to orthodontic forces is found betweenthe orthodontic and the orthopaedic literature(Rubin, 1998). Orthodontic forces are knownto generate a compression and tension zone.Orthodontists have traditionally looked uponthe resorption zone as a result of compressionand apposition as a result of tension. Related tocompression, either direct or indirect resorptioncan be observed. Direct resorption of thealveolar wall from the PDL occurs in the case oflow forces, while indirect resorption startingfrom the marrow spaces is related to theFigure 3 Graphical illustration of bone biological reactionto variation in strain values. In the case of low strain value(1) remodelling is turned on and a negative balance will bethe result. With higher strain values (2), modelling is turnedon and a formation of lamellar bone is occurring. Furtherincrease in strain (3) results in the formation of wovenbone. Severe overloading (4) results in a negative balanceas a result of the repair process related to the micro-fractures occurring at this strain level.678usage [SATMU] (Frost, 1986). Frost (1992) listedthe threshold strain values for lamellar bonefibrous tissue. A typical minimum effective strain(MES) is reported to be 15003000 m forlamellar bone to start modelling, and if the strainis below 100300 remodelling is activated as aresult of inactivity. Based on the loading of bonetransferred from implants to trabecular bone,the values recorded for trabecular bone seemto be slightly higher (Melsen et al., 1996). Alarge body of evidence concerning structuraladaptation to changes in strain has beengenerated experimentally and confirmed Frosts(1992) hypothesis (Goodship et al., 1979; Lanyonet al., 1979; Rubin and Lanyon, 1984). It is logicalto ask whether the biological reactions toorthodontic tooth movement could be explainedby the mechanostat theory. In this perspective,resorption observed in the direction of the forcecould be either a result of under- or over-loading.The main question that arose from thesespeculations was thus: How is the stress straindistribution in the periodontium related to thebiological reaction of the supporting tissues? In order to answer this question an inves-tigation combining strain levels and biologicalreactions was needed. The tissue reaction wasstudied in six Macaca fascicularis monkeys inwhich the first and the second molars had beenextracted, and an appliance was inserted forapproximation of the second premolar and thethird molar by translation. (Melsen, 1999). After11 weeks the monkeys were killed under Ketalaranaesthesia by perfusion with neutral formalin.The alveolar process was subsequently excised,embedded in methylmethacrylate, and preparedfor the production of 1520-m parallelhorizontal sections between the marginal bonelevel and the apex of the teeth cut with adiamond saw. The sections were then stainedwith fast green and analysed by histomorpho-metric measurements. The fractional resorptionsurface [Sfract (f)) m2/m2: the extent ofresorption lacunae as a fraction of the totaltrabecular bone surface] and fractional formationsurfaces [Sfract (f)) m2/m2 the extent of osteoidcovered surfaces as a fraction of the trabecularsurface] were measured. The fractional restingsurface was calculated as 100 per cent minus theB. M E L S E Nfraction recorded as resorption or apposition.Bone density was additionally evaluated in theareas mesial and distal to the third molars andsecond premolars (Gundersen et al., 1988). The type of tooth movement registered variedfrom translation to controlled tipping arounda centre localized at varying distances abovethe apices of the teeth. The quantity of toothmovement varied between 0.21 and 0.9 mmper month, measured at the bone margin.No relationship between force magnitude, thetype of tooth movement, and the amount ofdisplacement could be verified. All parametersmeasured in areas surrounding the loaded teethdeviated markedly from the correspondingvariables from the unloaded control teeth. Therewas an increase in the relative extension ofresorption from 35 per cent observed in thespecimens from the control teeth to 713 percent of the total cancellous surfaces surroundingthe loaded teeth. There was likewise an increasein the extension of appositional surfaces from1520 per cent in the control, to 3549 per centaround the loaded teeth. At the time of observation the alveolar wall inthe direction of the tooth movement was com-pletely resorbed, while woven bone formationwas seen in the alveolar bone ahead of thedirection of the tooth movement. Relative to thecontrol teeth, the density in the direction oftooth movement was increased by a factor of23. The alveolar surface opposite to the toothmovement was clearly appositional (Figure 4). In order to identify the range of strain levels inthe alveolar bone during orthodontic toothmovement, a three-dimensional finite elementmodel was constructed representing a conically-shaped root and the surrounding supportingstructures. Youngs modulus of the root was20 GPa, while that of the alveolar bone variedbetween 0.5 and 2 GPa. The 0.2-mm thick PDLwas modelled by non-linear spring elements. Incompression, their stiffness corresponded to aYoungs modulus of 0.1 MPa, while in tensionthis stiffness steadily increased from 0.1 MPa at0 mm deformation to 1000 MPa at maximumdeformation. Different loading situations wereanalysed, where the point of force applicationshifted along the longitudinal axis of the root

T I S S U E R E AC T I O N A N D O RT H O D O N T I C S679Figure 4 Micrograph of a molar that has been submittedto 11 weeks of orthodontic force in the direction of thearrow. Note that the alveolar wall in the direction of theforce is resorbing and that the bone in the direction ofthe movement consists of dense woven bone (W).Reproduced with kind permission from Melsen (1999).Figure 5 Stress profile of the tissue surrounding anorthodontically loaded tooth moving to the right. Note thatthe stress of the alveolar wall in the direction of the force isvery low, whereas it reaches a high value at the marginalbone level corresponding to the insertion of the stretchedperiodontal fibres.from the level of the apical crest, to a little overhalf the length of the root with a horizontalmovement of the root of 25 m at the level of thepoint of force application, the maximal tensilestresses in the alveolar bone were found tooccur when the point of force application waslocated at one-quarter of the roots length. Themagnitude of these stresses was then 0.2 MPa,while on the compression side the compressivestresses reached a peak value of 0.04 MPa.The magnitude of these stresses appeared tobe almost independent of the stiffness of thebone. The corresponding strain levels in the bonewere 400 strain, on the tension and 80 strainon the compression side. When taken intoconsideration that bone is inhomogeneousand porous, the strain values corresponded to2.000 strain in tension and 400 strain in com-pression (Figure 5). The strain values found in the direction of thedisplacement were below the MES. In the lightof the mechanostat theory this would releaseunderload remodelling, thus explaining directresorption in the so-called compression side (Figure3, region 1). On the other hand, the stretching ofthe PDL fibres on the opposite side, generated astrain level corresponding to modelling, thusexplaining new bone formation on the so-calledtension side (Figure 3, region 2). The woven bone formation seen ahead ofthe alveolus in the direction of the displacementcould be interpreted as an expression of aso-called regional acceleratory phenomena (RAP)(Figure 3, region 3). According to Frost (1986)any regional noxious stimulus of sufficient magni-tude can evoke a RAP. The extension ofthe affected region and the intensity of theresponse varies directly with the magnitude andthe nature of the stimulus. The indirect resorptiontakes place in the PDL when ischaemia andhyalinization occurs. This is probably due to thedisappearance of the lining cells necessary for thecommunication of the osteocytes. The hyalinizedtissue of the PDL is removed by non-clast cellsand the underlying bone is simultaneouslyresorbed by osteoclasts, as a repair response to thedamaged tissues (Figure 3, region 4). In the studies reported, the nature of thestimulus was a controlled orthodontic force. Thestrain developed, however, varied in magnitudeboth due to the difference in force, and tobiological variations related to root size andstructure of the bone surrounding the teeth(Melsen et al., 1996). In that investigation,the most conspicuous result was the increase inboth activation level and density of the bonesubjected to compression in the direction oftooth movement.680ConclusionsThe studies regarding tissue reaction to ortho-dontic forces were initiated with the purposeof finding an explanation of the controversiesin the literature. The intrusion of periodontallydamaged teeth was chosen as this movement isparticularly controversial. It was found thatthe biological reaction was dependent on theforce level and the stress/strain distribution.Histological examination demonstrated that thePDL fibres were stretched and formation activitywas found along the major part of the alveolus.Only apical fibres were not observed to stretch.Interpreting these findings retrospectively inthe light of the mechanostat theory, it can beconcluded that stretched fibres will most likelylead to the delivery of strain values corres-ponding to modelling. In the apical region, theloading is minimal as the PDL fibres seem notto be stretched. Apical bone resorption can thusbe interpreted as a remodelling. The stretchingof the fibres may induce a slight bending ofthe alveolar wall. This bending in terms of anincrease in the curvature of the alveolar wall haspreviously been suggested by Epker and Frost(1965) as the cause of new bone formation, in anattempt to harmonize the orthodontic and theorthopaedic perception of the tissue reaction tomechanical load. Based on the above-mentionedexperiments it is suggested that the mechanostattheory can also be used as a model for under-standing orthodontic tissue reaction (Figure 3).Direct resorption will be perceived as aremodelling induced by under-loading of thealveolar wall and apposition as modelling inducedby the load exerted by stretched fibres. Thewoven bone formation observed in the directionof the tooth movement could finally be explainedas a RAP phenomenon developing as a reactionto overloading (Frost, 1994). The indirect resorptiontakes place when ischaemia and hyalinization ofthe PDL is generated by high stress values.Disappearance of the lining cells (Marotti, 1996)necessary for the communication of the osteocytesprobably leads to pathological reaction of theunderlying overloaded bone, which is probablyaffected by microfractures. Investigations on stress levels in the alveolarbone in the direction of the force and measurementB. M E L S E Nof the occurrence of microcracks in the case ofindirect resorption will contribute to the validationof this paradigm.Address for correspondenceProfessor Birte MelsenDepartment of OrthodonticsRoyal Dental CollegeAarhus UniversityVennelyst Boulevard 9DK-8000 Aarhus CDenmarkAcknowledgementM. Dalstra is thanked for finite element analysis.Figures 1 and 2 reproduced with kind permissionof the American Journal of Orthodontics andDentofacial Orthopedics, and Figure 4 with kindpermission of The Angle Orthodontist.ReferencesBaumrind S, Korn E L, Boyd R L 1996 Apical root resorption in orthodontically treated adults. American Journal of Orthodontics and Dentofacial Orthopedics 110: 311320Burr D B, Martin R B, Schaffler M B, Radin E L 1985 Bone remodeling in response to in vivo fatigue microdamage. Journal of Biomechanics 18: 189200Costopoulos G, Nanda R 1996 An evaluation of root resorption incident to orthodontic intrusion. American Journal of Orthodontics and Dentofacial Orthopedics 109: 543548Dellinger E L 1967 A histologic and cephalometric investigation of premolar intrusion in the Macaca speciosa monkey. American Journal of Orthodontics 53: 325355Epker B N, Frost H M 1965 Correlation of bone resorption and formation with the physical behavior of loaded bone. Journal of Dental Research 44: 3341Ericsson I, Thilander B 1978 Orthodontic forces and recurrence of periodontal disease. An experimental study in the dog. American Journal of Orthodontics 74: 4150Faltin R M, Arana-Chavez V E, Faltin K, Sander F G, Wichelhaus A 1998 Root resorptions in upper first premolars after application of continuous intrusive forces. Intra-individual study. Journal of Orofacial Orthopedics 59: 208219Frost H M 1986 Intermediary organization of the skeleton. CRC Press Inc., Boca RotonT I S S U E R E AC T I O N A N D O RT H O D O N T I C S681cellular reaction studied in trabecular bone by means of afinite element analysis. In: Davidovitch Z (ed.) Biologicalmechanisms of tooth movement and craniofacialadaption. Harvard Society for the Advancement ofOrthodontics, Boston, USA pp. 213221Melsen B, Konstantellos V, Lagoudakis M, Planert J 1997 Combined intrusion and retraction generated by cantilevers with helical coils. Journal of Orofacial Orthopedics 58: 232241Nanda R 1997 Correction of deep overbite in adults. Dental Clinics of North America 41: 6787Parker R J, Harris E F 1998 Directions of orthodontic tooth movements associated with external apical root resorption of the maxillary central incisor. American Journal of Orthodontics and Dentofacial Orthopedics 114: 677683Polson A, Caton J, Polson A P, Nyman S, Novak J, Reed B 1984 Periodontal response after tooth movement into infrabony defects. Journal of Periodontology 55: 197202Reitan K 1964 Effect of force magnitude and direction of tooth movment on different alveolar bone types. Angle Orthodontist 34: 244255Reitan K 1967 Clinical and histological observations on tooth movement during and after orthodontic treatment. American Journal of Orthodontics 53: 721745Rubin C T, Lanyon L E 1984 Regulation of bone formation by applied dynamic loads. Journal of Bone and Joint Surgery 66: 397402Rubin C T, Lanyon L E 1985 Regulation of bone mass by mechanical strain magnitude. Calcified Tissue International 37: 411417Rubin R M 1998 Moving moments in orthodontics. Angle Orthodontist 68: 483, Comments Angle Orthodontist 69: 6Sander F G, Wichelhaus A, Schiemann C 1996 Intrusion mechanics according to Burstone with the NiTi-SE-steel uprighting spring. Journal of Orofacial Orthopedics 57: 210223Stenvik A, Mjr I A 1970 Pulp and dentine reactions to experimental tooth intrusion. American Journal of Orthodontics 57: 370385Waerhaug J 1978 Healing of the dento-epithelia junction following subgingival plaque control. II: As observed on extracted teeth. Journal of Periodontology 49: 18Weiland F J, Bantleon H P, Droschl H 1996 Evaluation of continuous arch and segmented arch leveling techniques in adult patientsa clinical study. American Journal of Orthodontics and Dentofacial Orthopedics 110: 647652Wilson A N, Middleton J, Jones M L, McGuinness N J 1994 The finite element analysis of stress in the periodontal ligament when subject to vertical orthodontic forces. British Journal of Orthodontics 21: 161167Frost H M 1987 The mechanostat: a proposed pathogenetic mechanism of osteoporoses and the bone mass effects of mechanical and nonmechanical agents. Bone and Mineral 2: 7385Frost H M 1992 The role of changes in mechanical usage set points in the pathogenesis of osteoporosis. Journal of Bone Mineral Research 7: 253361Frost H M 1994 Wolffs Law and bones structural adaptions to mechanical usage: an overview for clinicians. Angle Orthodontist 64: 175188Goodship A E, Lanyon L E, McFie H 1979 Functional adaption of bone to increased stress. An experimental study. Journal of Bone and Joint Surgery 61: 539546Goerigk B, Diedrich P, Wehrbein H 1992 Intrusion of the anterior teeth with the segmented-arch technic of Burstonea clinical study. Fortschritte der Kieferortho- pdie 53: 1625Gundersen H J G et al. 1988 The new stereological tools: disector, fractionator and point sampled intercepts and their use in pathological research and diagnosis. Acta Pathologica, Microbiologica, et Immunologica Scandinavica 96: 857881Lanyon L E, Magee P T, Baggott D G 1979 The relationship of functional stress and strain to the processes of bone remodelling. An experimental study on the sheep radius. Journal of Biomechanics 12: 593600Marotti G 1996 The structure of bone tissues and the cellular control of their deposition. Italian Journal of Anatomical Embryology 101: 2579McFadden W H, Engstrom C, Engstrom, H, Anholm J M 1989 A study of the relationship between incisor intrusion and root shortening. American Journal of Orthodontics and Dentofacial Orthopedics 96: 390396Melsen B 1986 Tissue reaction following application of extrusive and intrusive forces to teeth in adult monkeys. American Journal of Orthodontics 89: 469475Melsen B 1999 Biological reaction of alveolar bone to orthodontic tooth movement. Angle Orthodontist 69: 151158Melsen B, Kragskov J 1992 Tissue reaction to intrusion of periodontally involved teeth. In: Davidovitch Z (ed.) The biological mechanisms of tooth movement and craniofacial adaption. The Ohio State University, College of Dentistry, Columbus, USA, pp. 423430Melsen B, Agerbaek N, Eriksen J, Terp S 1988 New attachment through periodontal treatment and ortho- dontic intrusion. American Journal of Orthodontics and Dentofacial Orthopedics 94: 104116Melsen B, Agerbaek N, Markenstam G 1989 Intrusion of incisors in adult patients with marginal bone loss. American Journal of Orthodontics and Dentofacial Orthopedics 96: 232241Melsen B, Olesen M, Oestergaard L, Agerbaek N, Lang N K 1996 The relationship between strain values and