J Periodontol • January 2005

107

* Department of Prosthodontics and Periodontics, Division of Periodontics, Schoolof Dentistry at Piracicaba, University of Campinas (UNICAMP), Piracicaba, São Paulo,Brazil.

Osseointegrated titanium implantshave become an important alter-native to conventional prosthesis

in edentulous situations.1,2 Implant inte-gration into bone involves not only im-plant-related factors such as material,topography, shape, and surface chemistry,but also mechanical loading, surgicaltechnique, and patient variables, such asbone quantity and quality.3 Since osteo-penia and osteoporosis are estrogen defi-ciency-related bone diseases,4 whichresult in low bone mass, some studieshave investigated the impact of thesediseases on dental implants.5-10 Althoughthere is no clinical evidence suggestingthat osteoporosis affects implant successrates,5 animal studies have shown thatbone volume, bone area, and bone-to-implant contact are significantly decreasedaround titanium implants placed in estro-gen-deficient animals.6-10

Estrogen deficiency after ovariectomyor menopause plays a major role in theearly changes in the turnover of cancel-lous bone, leading to a reduction in bonemass and disruption of the trabecular net-work.11 Estrogen replacement therapy iscommonly used as a prophylactic andtherapeutic measure in order to preventbone loss in postmenopausal women andestrogen-deficient individuals.12 However,the possibility of estrogen side effects,such as breast swelling and tenderness,bloating, bleeding, and spotting, often leadpostmenopausal women to discontinuethis therapy.13 In addition, clinical contra-indications to estrogen therapy may also

Alendronate Therapy May Be Effective in thePrevention of Bone Loss Around TitaniumImplants Inserted in Estrogen-Deficient RatsPoliana Mendes Duarte,* Bruno César de Vasconcelos Gurgel,* Antonio Wilson Sallum,* Getúlio Rocha Nogueira Filho,* Enilson Antonio Sallum,* and Francisco Humberto Nociti Jr.*

Background: This study evaluated whether alendronate (ALD)influences bone healing around titanium implants inserted inovariectomized rats and whether it provides a residual effectafter its withdrawal.

Methods: Bilateral ovariectomies were performed in 87 Wistarrats and one screw-shaped titanium implant was placed in thetibiae. The animals were divided into the following groups: groupSHAM (N = 15): sham surgeries; group OVX (N = 15): ovariec-tomy; group AT (N = 15): OVX plus alendronate administrationfor 80 days; group AW (N = 14): OVX plus alendronate admin-istration for 40 days; group ET (N = 14): OVX plus 17β estradioladministration for 80 days; or group EW (N = 14): OVX plus 17βestradiol administration for 40 days. Bone-to-implant contact(BIC), bone area (BA) within the limits of implant threads, andbone density in a 500 µm-wide zone lateral to the implant (BD)were obtained and measured for the cortical (zone A) and cancel-lous (zone B) regions.

Results: In zone A, data analysis showed no significant differ-ences among the groups regarding BIC and BD (P >0.05), anda slight beneficial effect of estradiol on BA when compared withthe OVX, EW, and AW groups (P <0.05). In zone B, OVX nega-tively impacted bone healing around the implants, resulting inreduced BA and BD (P <0.05). ALD (continuous/interrupted) andestradiol (only continuous) positively affected BIC, BA, and BD,resulting in values at the same level as the control group (SHAM).

Conclusions: ALD may prevent the negative influence of estro-gen deficiency on bone healing around titanium implants insertedin OVX rats. This positive effect, in contrast to estradiol, is sus-tained following its withdrawal. J Periodontol 2005;76:107-114.

KEY WORDSAlendronate/therapeutic use; animal studies; bone and bones;dental implants; estrogen deficiency; estradiol/therapeutic use;titanium; wound healing.

40053.qxd 2/18/05 9:31 AM Page 107

108

Alendronate Therapy and Dental Implants Volume 76 • Number 1

occur, and therefore, development of therapeutic alter-natives is required.

Bisphosphonates are synthetic compounds exten-sively used for the treatment of systemic bone lossdue to estrogen depletion.14 Although the mechanismof action of the bisphosphonates is not yet fully under-stood, it has been demonstrated that these compoundsinhibit osteoclast-mediated bone resorption.15 It hasbeen shown that alendronate is a potent inhibitor ofbone resorption without significantly affecting boneformation.16 Because ALD accumulates in the bone,a very long half-life17 is observed even when therapyis discontinued.18

Therefore, the present study was designed to eval-uate, by histometric analysis, whether ALD therapymay prevent the negative influence of endogenousestrogen deficiency around titanium implants placed inovariectomized rats. Furthermore, little is known aboutthe effects of discontinuation of both ALD and estra-diol (E) therapies on bone healing and density aroundtitanium implants, and therefore, the present study alsoaimed to investigate the impact of the withdrawal ofboth therapies.

MATERIALS AND METHODSAnimalsThe experimental animals were 87 female Wistar rats(90 days old; average weight 210 g). During the experi-ment, the animals were kept in plastic cages withaccess to food and drinking water ad libitum, exceptthe ovariectomized (OVX) rats not on estradiol treat-ment (pair feeding).19 This protocol was approved bythe University of Campinas Institutional Animal Careand Use Committee.

OvariectomyThe animals were anesthetized with intramuscularadministration of ketamine (0.5 ml/kg). Ovariectomyor sham surgeries were performed at the beginning ofthe study. Bilateral ovariectomies were performed in 72rats from a dorsal approach. The remaining 15 animalswere subjected to sham surgeries in which the ovarieswere lifted up and returned intact to the original posi-tion. Postoperatively, the animals received antibiotic†

(1 ml/kg) given as a single intramuscular injection.

Experimental DesignAfter ovariectomies, the animals were randomly as-signed to one of six groups: Group SHAM (N = 15):sham surgery; group OVX (N = 15): ovariectomy; groupAT (N = 15): OVX plus 4 days/week subcutaneousinjections of ALD‡ at a dose of 5 mg/kg body weightfor 80 days; group AW (N = 14): OVX plus 4 days/weeksubcutaneous injections of ALD at a dose of 5 mg/kgbody weight for 40 days; group ET (N = 14): OVX anddaily subcutaneous injections of 17β estradiol,§ dis-

solved in 100% ethanol and diluted in mineral oil at adose of 20 µg/kg body weight for 80 days; or groupEW (N = 14): OVX and daily subcutaneous injectionsof 17β estradiol, dissolved in 100% ethanol and dilutedin mineral oil at a dose of 20 µg/kg body weight for40 days (Fig. 1).

Implant SurgeryTwenty-one days after ovariectomies, general anes-thesia was performed (ketamine, 0.5 ml/kg) and theskin was cleansed with iodine surgical soap. An inci-sion approximately 1 cm long was made and the bonesurface of the tibiae surgically exposed by blunt dis-section. Under profuse saline solution irrigation, bicor-tical implant beds were drilled at a rotary speed notexceeding 1,500 rpm. A screw-shaped, commerciallyavailable pure titanium implant 4.0 mm long × 2.2 mmdiameter was placed until the screw thread had beencompletely introduced into the bone cortex. Finally,soft tissues were replaced and sutured. Postoperatively,the animals received an antibiotic† given as a singleintramuscular injection (1 ml/kg).

Clinical AnalysesIn order to confirm the success of ovariectomy and estro-gen administrations, the estrous cycle was monitored2 weeks after the ovariectomy surgeries and 2 weeksafter the drug withdrawal. Changes in the vaginal smearduring 4 to 5 days of the estrus cycles were observedin each group. At autopsy, success of the ovariectomywas also confirmed by absence of ovaries and atrophyof uterine horns in ovariectomized rats. In addition, thepresence of normal uterine horns was also used todemonstrate the success of estrogen administration.

Alkaline Phosphatase AnalysisBlood samples were collected to measure plasmaconcentration of alkaline phosphatase at the time of

Figure 1.Illustration of the experimental design.

† Pentabiótico, Wyeth-Whitehall Ltda, São Paulo, SP, Brazil.‡ Teva Pharmaceutical Ltda., Petach Tikva, Israel.§ Sigma Chemical, St. Louis, MO.

40053.qxd 2/18/05 9:31 AM Page 108

109

J Periodontol • January 2005 Duarte, Gurgel, Sallum, Filho, Sallum, Nociti

sacrifice. Alkaline phosphatase activity was obtainedcolorimetrically using automated laboratory techniques.¶

Histometric ProcedureSixty days after implant placement (Fig. 1), the ani-mals were sacrificed and the tibiae removed and fixedin 4% neutral formalin for 48 hours. Undecalcified sec-tions were prepared, as previously described.8-10

Briefly, the blocks were dehydrated using an ascendingseries of ethanol (60% to 100%) and embedded in gly-colmethacrylate.# Subsequently, the sections (20 to30 µm) were obtained and stained using 1% toluidineblue staining. A masked examiner separately recordedthe percentage of bone-to-implant contact (BIC) andbone area (BA) within the limits of the threads of theimplants and bone density (BD); i.e., the proportion ofmineralized matrix in a 500 µm-wide zone lateral to theimplant surface for both sides of the implant in thecortical (zone A) and cancellous bone (zone B)areas.** One section representative of the mid-portionof the implant was used for the histometric analysis.

Morphological AnalysisIn order to parallel the histometric data with the mor-phological aspect of the contralateral tibiae, specimenswere observed at the scanning electron microscopy(SEM) level. Samples were fixed in 2.5% glutaralde-hyde in 0.05 mol/l cacodylate buffer, pH 7.4. Subse-quently, specimens were washed in five changes of tapwater for 15 minutes to remove the smear layer, fixed,post-fixed, dehydrated in ascending acetone concen-trations up to 100%, critical point dried,†† sputter-coatedwith gold,‡‡ and observed by SEM.§§ Sections wereobtained in a transversal direction. Representative areasof cortical and cancellous bone were photographed at35× magnification.

Statistical AnalysisData from zones A and B were separately averaged.Intergroup analysis was used to test the hypothesisthat the treatments had no influence on BIC, BA, andBD (Kruskal-Wallis test; alpha = 0.05). If a statisticaldifference was detected, Dunn’s method was used toisolate the groups that differed from the others. Further-more, one-way analysis of variance (ANOVA) (alpha =0.05) was used to test the hypothesis that the treat-ments did not influence the alkaline phosphatase serumlevels. If statistical difference was detected, a pair-wisemultiple comparison procedure was used (Bonferronitest). Finally, an intergroup analysis was used to testthe hypothesis that there was no difference among thegroups with respect to the animal’s body weight at theend of the experimental period (one-way ANOVA;alpha = 0.05). A pair-wise multiple comparison pro-cedure was used (Bonferroni test) if statistical differ-ence was detected.

RESULTSClinical ObservationsAll animals gained weight during the study. At the endof the experimental period, the animals from SHAM,OVX, AW, and EW groups weighed significantly morethan the animals from the AT and ET groups (P <0.05).The final mean body weight was 256.26 ± 16.08 g(SHAM); 261.66 g ± 17.16 g (OVX); 244.18 ± 20.4 g(AT); 266.56 ± 23.89 g (AW); 228.57 ± 15.76 g (ET);and 269.44 ± 35.85 g (EW).

Clinical appearance of uterine horns, absence ofovaries, and assessment of the estrous cycle confirmedthe success of ovariectomy and estrogen replacement.Groups OVX, AT, and AW presented diestrous smearand their reproductive organs atrophied, confirmingthe reduction of serum estrogen levels. In contrast, theanimals submitted to SHAM surgery presented the fourstages of estrous cycle (estrus, diestrus, proestrus, andmetaestrus) and pink and fluid-filled uteri. The ani-mals administered for 80 days with estradiol (ET)remained in the estrus stage and presented normaluteri, assuring that the serum estrogen levels were keptnormal in these animals. Finally, animals in which theestradiol therapy was interrupted by day 40 presenteddiestrous smear after the withdrawal, and their repro-ductive organs were atrophied at sacrifice.

Morphological AnalysisSimilar to the histometric observation, morphologicalanalysis confirmed the striking differences in the cor-tical and cancellous bone regions among the groups.SHAM, AT, AW, and ET presented a much more densetrabecullar bone than OVX and EW, while no visibledifferences could be observed among SHAM, AT, AW,and ET (Fig. 2).

Alkaline Phosphatase AnalysisAlkaline phosphatase serum concentrations (UI/L) andstandard deviation, performed at the time of sacrifice,were 27.6 ± 11.30 (SHAM); 80.46 ± 18.72 (OVX);40.93 ± 11.05 (AT); 44.28 ± 13.28 (AW); 33.2 ± 14.90(ET); and 91.23 ± 23.28 (EW). Alkaline phosphataselevels were statistically higher for the OVX and EWgroups (P <0.001) and, therefore, confirmed the highbone turnover in the animals in an estrogen-deficientstate.

Histometric ResultsBone-to-implant contact (BIC). Intergroup analysisdid not reveal a significant difference regarding BICin zone A (P >0.05). On the other hand, in zone B,

¶ Gold Analisa Diagnóstica, Belo Horizonte, MG, Brazil.# Technovit 7200; Heraeus Kulzer GmbH, Wehrheim, Germany.** Image-Pro; Media Cybernetics, Silver Spring, MD.†† CPD 030-BAL-TEC, Furstentum, FL, Liechtenstein.‡‡ MED 010-BAL-TEC, Furstentum, FL.§§ LEO 435 VP, LEO Electron Microscopy Ltd., Cambridge, U.K.

40053.qxd 2/18/05 9:31 AM Page 109

110

Alendronate Therapy and Dental Implants Volume 76 • Number 1

Figure 2.Transversal sections of the proximal tibial diaphysis from SHAM (A), AT(B), and OVX (C) observed at the SEM level. Note the reduced amountof trabecular bone (TB) in the OVX rats. Note also that the AT providedcomplete protection against bone loss.The trend for increased trabecularbone in SHAM and AT is very similar (original magnification, 35×).

data analysis revealed a statistically significant pos-itive effect of ALD and E therapies on the percent-

age of BIC, even after their withdrawal (P <0.05)(Table 1).

Bone area within the limits of the implant threads(BA). In zone A, although there was a slight difference,estrogen deficiency did not significantly influence BAaround the implants (P >0.05). Furthermore, statisticalanalysis demonstrated that continuous estradiol admin-istration to OVX animals promoted a significantlyhigher BA than the OVX, EW, and AW groups (P <0.05).In contrast, in zone B, data analysis showed that estro-gen deficiency may result in a lower percentage of BAthan the animals not submitted to ovariectomy (P <0.05).However, this negative effect was restored in the OVXrats, which were submitted to AT, ET, and AW, but notto EW (Table 1).

Bone density in a 500 µµm-wide zone lateral to theimplant surface (BD). Data analysis showed that in zoneA there was no significant difference regarding the per-centage of mineralized tissue in lateral region to theimplant surface (P >0.05). Means and standard devia-tions were 95.39% ± 1.34% (SHAM); 93.27% ± 3.8%(OVX); 94.44% ± 1.59% (AT); 94.34% ± 1.55% (AW);95.84% ± 1.75% (ET); and 92.81% ± 6.89% (EW). Inzone B, the results revealed that estrogen deficiencymay result in a lower BD than the SHAM group(P <0.05). Furthermore, AT, AW, and ET, but not EW,may protect the bone around the implants from the neg-ative effects of estrogen deficiency. Means and standarddeviations were 48.39% ± 9.37%; 8.2% ± 4.7%; 56.91% ±8.7%; 57.12% ± 5.2%; 50.97% ± 8.2%; and 32.81% ±10.7% for SHAM, OVX, AT, AW, ET, and EW groups,respectively. Figure 3 illustrates the histological results.

DISCUSSIONPostmenopausal osteoporosis and osteopenia are char-acterized by a progressive bone loss, which beginsafter natural or surgical cessation of the ovarian func-tion.4 These conditions strongly support the role ofestrogen in bone metabolism.20-24 Since dental implantoutcome is dependent on bone quality, estrogen defi-ciency has been investigated with respect to its impacton the osseointegration process.5-10 Although clinicaldata are not conclusive,5 animal studies have sug-gested that estrogen deficiency may negatively impactbone around titanium implants.6-10 In the present study,as previously reported,8-10 data analysis supports theobservation that bone is affected by an estrogen-defi-cient state. At this stage, however, caution should beused since further studies are required to determine ifimplant-supported prostheses may also be involved inthis phenomenon.

Pharmacological interventions for osteoporosisinclude mainly estrogen replacement therapy, calci-tonin and parathyroid hormones, and bisphosphonateadministrations.25 In the present study, the impact thatalendronate therapy (AT) and its withdrawal (AW) exert

40053.qxd 2/18/05 9:31 AM Page 110

111

J Periodontol • January 2005 Duarte, Gurgel, Sallum, Filho, Sallum, Nociti

on bone healing around titanium implants placed inovariectomized rats was investigated. In addition, con-tinuous (ET) and interrupted (EW) estrogen adminis-tration were also investigated with respect to theireffects on preventing the influence of endogenousestrogen deficiency on bone around the implants.

Estrogen replacement is commonly used as a firstline therapy for preventing and treating osteoporosis.The protective effect of estrogen therapy has beendemonstrated in clinical and animal studies withrespect to both bone mass loss and fracture inci-dence.12,26,27 The results of the current study, as inour previous studies,8-10 showed that ET immediatelyafter ovariectomy may neutralize the negative effectsof estrogen deficiency, mainly in cancellous bone, inall parameters analyzed; i.e., BIC, BA, and BD.Although the mechanisms involved remain to be inves-tigated, currently it has been demonstrated that boneresorption is characterized by two key molecules,receptor activator of nuclear factor-kappa ligand(RANKL) and osteoprotegerin (OPG).24 Bone resorp-tion is significantly reduced by RANKL function inhi-bition via its decoy receptor OPG.24,28 Since estrogenhas been reported to control bone resorption actingon OPG, it may be suggested that estrogen deficiencyinduces an imbalance in the RANKL/OPG systemfavoring bone resorption, and its replacement mayrevert this mechanism. However, the precise mecha-nisms of this pathway remain to be clarified.

In addition, for the first time we have demonstratedthat the interruption in the treatment with estrogenresults in BA and BD around titanium implants, in thecancellous bone, similar to the OXV group. Althoughlittle is known about the effects of discontinuation of

estrogen therapy on bone around titaniumimplants, these observations are in agree-ment with studies that have demonstratedthat EW causes a marked increase in boneturnover and physiological changes; i.e., lossof bone density and increased risk of frac-ture.29-30 Therefore, the pattern of bone lossobserved after estrogen withdrawal seems tobe comparable to that which occurs in ratsafter ovariectomy without treatment.

ALD is a bisphosphonate compound thathas been reported to inhibit bone loss andhas been extensively applied in the treatmentof osteoporosis.14 Several mechanisms havebeen investigated, including osteoclast devel-opment and activity inhibition, osteoclastapoptosis induction, and stimulation of anosteoclast inhibitory factor production.15,31

In the current study, alendronate, immedi-ately after ovariectomy, prevented bone lossand depressed bone turnover around tita-nium implants. The OVX rats which received

alendronate (AT) immediately after ovariectomy pre-sented with levels of BIC, BA, and BD similar to thesham operated animals (control) in the cortical andcancellous bone. These findings are consistent withthe skeletal effects of bisphosphonates in early post-menopausal and ovariectomized women.32 Our obser-vations are also in agreement with those reported byNarai and Nagahata,33 who demonstrated that removaltorque was lower for osteoporotic rats than ALD-treatedrats and that there was no significant differencebetween the ALD-treated and sham groups.

ALD side effects, including sensitivity to phosphatesand gastrointestinal upset, often lead postmenopausalwomen to discontinue therapy.18 Thus, in the presentstudy, we also investigated the effects of alendronatewithdrawal (AW) on bone healing and density aroundtitanium implants in OVX rats. Data analysis revealedthat all the histometric parameters (BIC, BA, and BD)in cancellous and cortical bone for the animals thatreceived an interrupted ALD therapy were similar tocontrol group (SHAM). Thus, in contrast to estrogenwithdrawal, OVX rats maintained normal bone massand low levels of bone turnover after alendronate treat-ment withdrawal. These findings are in agreement withstudies showing that accelerated bone loss is seen afterwithdrawal of estrogen therapy, but not after withdrawalof alendronate or combination therapy.34 The prolongedresidual skeletal effects of bisphosphonates are prob-ably a consequence of a strong affinity to the hydroxy-apatite crystals.35 Bisphosphonates bound to bonemineral are released during bone resorption by osteo-clasts. This could lead to a localized accumulation ofthe drug, which could directly perturb osteoclasts asexplained above.31,36

Table 1.

Mean and Standard Deviation (%) of Bone-to-Implant Contact (BIC) and Bone Area (BA) at ZonesA and B

BIC BA

Group Zone A Zone B Zone A Zone B

SHAM 52.40 ± 11.74 a* 51.08 ± 12.54 a 84.79 ± 3.70 ab 49.64 ± 5.60 ab

OVX 42.08 ± 8.20 a 36.75 ± 8.80 b 80.21 ± 3.80 a 36.59 ± 7.90 c

AT 55.66 ± 17.66 a 52.56 ± 9.38 a 83.85 ± 3.40 ab 56.84 ± 13.80 a

AW 57.97 ± 12.65 a 55.05 ± 15.04 a 82.99 ± 5.70 a 59.59 ± 7.80 a

ET 50.45 ± 12.76 a 52.99 ± 10.27 a 88.36 ± 3.39 b 58.54 ± 12.19 c

EW 51.08 ± 18.17 a 52.14 ± 13.66 a 81.04 ± 6.80 a 40.73 ± 7.05 bc

* Different letters indicate significant statistical differences (α = 0.05), within each column.BIC and BA determined by Kruskal-Wallis and Dunn’s tests.

40053.qxd 2/18/05 9:31 AM Page 111

112

Alendronate Therapy and Dental Implants Volume 76 • Number 1

Alkaline phosphatase has been described as an im-portant biochemical marker of high bone turnover usedto reflect the changes in bone metabolism during thediagnosis of osteoporosis.37,38 Serum levels of alkalinephosphatase were significantly higher in the OVX andEW groups, demonstrating a high bone turnover as aconsequence of an estrogen-deficient state. On theother hand, ET, AT, and AW demonstrated an alkalinephosphatase level similar to the control group (SHAM),indicating that these treatments were able to control

the high bone turnover induced by low level estrogen,thus providing support to the histometric and mor-phological findings.

Therefore, within the limits of the present investiga-tion, it may be suggested that ET and AT immediatelyafter ovariectomy may provide complete protectionagainst the negative influence of endogenous estrogendeficiency on bone healing and density around titaniumimplants. Furthermore, our findings suggest that theeffects of EW and AW are widely dissimilar. EW may

Figure 3.Photomicrographs illustrating the histological aspects observed within the limits of the threads and in a 500 µm-wide zone lateral to the implant surface inSHAM (A), OVX (ovariectomized rats) (B),AT (alendronate therapy) (C),AW (alendronate withdrawal) (D), ET (estrogen therapy) (E), and EW (estrogenwithdrawal) groups (F) (toluidine blue; original magnification, 6.25×).

40053.qxd 2/18/05 9:31 AM Page 112

113

J Periodontol • January 2005 Duarte, Gurgel, Sallum, Filho, Sallum, Nociti

be a high risk for subsequent bone loss around tita-nium implant, while AW maintained relatively normalcancellous bone. Nevertheless, further studies shouldbe considered in order to clinically evaluate the rele-vance of these findings.

ACKNOWLEDGMENTSDr. Duarte was supported by FAPESP (Fundação deAmparo à Pesquisa do Estado de São Paulo), process# 02/08555-7. The authors greatly appreciated the assis-tance of AS Technology, São Paulo, Brazil, for supplyingthe implants and Eliene Aparecida Orsini NarvaesRomani, Department of Morphology, School of Dentistryat Piracicaba, for helping with the SEM analysis.

REFERENCES1. Brånemark PI. Osseointegration and its experimental

background. J Prosthet Dent 1983;50:390-410.2. Zarb GA, Schmitt A. The longitudinal clinical effective-

ness of osseointegrated dental implants: The Torontostudy. Part II: The prosthetic results. J Prosthet Dent 1990;64:53-61.

3. Puleo DA, Nanci A. Understanding and controlling thebone-implants interface. Biomaterials 1999;20:2311-2321.

4. Harrison JE. Osteoporosis. In: Tam CS, Heersche JNM,Murray TM, eds. Metabolic Bone Diseases: Cellular andTissue Mechanisms. Boca Raton, FL: CRC Press; 1989:240-267.

5. Dao TTT, Anderson JD, Zarb GA. Is osteoporosis a riskfactor for osseointegration of dental implants? Int J OralMaxillofac Implants 1993;8:137-144.

6. Yamazaki M, Shirota T, Tokugawa Y, et al. Bone reac-tions to titanium screw implants in ovariectomized ani-mals. Oral Surg Oral Med Oral Pathol 1999;87:411-418.

7. Pan J, Shirota T, Ohno K, Mich K. Effect of ovariectomyon bone remodeling adjacent to hydroxyapatite-coatedimplants in the tibia of mature rats. J Oral MaxillofacSurg 2000;58:877-882.

8. Duarte PM, César-Neto JB, Gonçalves PF, Sallum EA,Nociti FH Jr. Estrogen deficiency affects bone healingaround titanium implants. A histometric study in rats.Implant Dent 2003;12:340-345.

9. Nociti FH Jr, Sallum AW, Sallum EA, Duarte PM. Effectof estrogen replacement and calcitonin therapies onbone around titanium implants placed in ovariectomizedrats: A histometric study. Int J Oral Maxillofac Implants2002;17:786-92.

10. Duarte PM, Cesar-Neto JB, Sallum AW, Sallum EA,Nociti FH Jr. Effect of estrogen and calcitonin therapieson bone density in a lateral area adjacent to implantsplaced in the tibiae of ovariectomized rats. J Periodon-tol 2003;74:1618-1624.

11. Dempster DW, Birchman R, Xu R, Lindsay R, Shen V.Temporal changes in cancellous bone structure of ratsimmediately after ovariectomy. Bone 1995;16:157-161.

12. Weinstein RS, Parfitt AM, Marcus R, Greenwald M,Crans G, Muchmore DB. Effects of raloxifene, hormonereplacement therapy, and placebo on bone turnoverin postmenopausal women. Osteoporos Int 2003;14:814-822.

13. Regan MM, Emond SK, Attardo MJ, Parker RA, GreenspanSL. Why do older women discontinue hormone replace-ment therapy? J Womens Health Gend Based Med 2001;

10:343-350.14. Rodan GA. Mechanisms of action of bisphosphonates.

Annu Rev Pharmacol Toxicol 1998;38:375-388.15. Hughes DE, Wright KR, Uy HL, et al. Bisphosphonates

promote apoptosis in murine osteoclasts in vitro andin vivo. J Bone Miner Res 1995;10:1478-1487.

16. Schenk R, Eggli P, Fleisch H, Rosini S. Quantitative mor-phometric evaluation of the inhibitory activity of newaminobisphosphonates on bone resorption in the rat.Calcif Tissue Int 1986;38:342-349.

17. Lin JH, Duggan DE, Chen IW, Ellsworth RL. Physiologicaldisposition of alendronate, a potent anti-osteolytic bis-phosphonate, in laboratory animals. Drug Metab Dispos1991;19:926-932.

18. Miller PD, Woodson G, Licata AA, et al. Rechallenge ofpatients who had discontinued alendronate therapybecause of upper gastrointestinal symptoms. Clin Ther2000;22:1433-1442.

19. Wronski TJ, Schenck PA, Cintron M, Walsh CC. Effectof body weight on osteopenia in ovariectomized rats.Calcif Tissue Int 1987;40:155-159.

20. Vaananen HK, Harkonen PL. Estrogen and bone metabo-lism. Maturitas 1996;23(Suppl.):S65-S69.

21. Turner RT, Riggs BL, Spelsberg TC. Skeletal effects ofestrogen. Endocr Rev 1994;15:275-300.

22. Pacifici R. Is there a causal role for IL-1 in postmeno-pausal bone loss? Calcif Tissue Int 1992;50:295-299.

23. Girasole G, Passeri G, Pedrazzoni M, Giuliani N, Passeri M.Interleukin-6: A pathogenetic role in the postmenopausalosteoporosis? Acta Biomed Ateneo Parmense 1995;66:125-138.

24. Lindberg MK, Erlandsson M, Alatalo SL, et al. Estrogenreceptor alpha, but not estrogen receptor beta, isinvolved in the regulation of the OPG/RANKL (osteo-protegerin/receptor activator of NF-kappa B ligand) ratioand serum interleukin-6 in male mice. J Endocrinol 2001;171:425-433.

25. Watts NB. Therapies to improve bone mineral densityand reduce the risk of fracture: Clinical trial results. JReprod Med 2002;47:82-92.

26. Wronski TJ, Dann LM, Qi H, Yen C-F. Skeletal effectsof withdrawal of estrogen and diphosphonate treatmentin ovariectomized rats. Calcif Tissue Int 1993;53:210-216.

27. Sims NA, Morris HA, Moore RJ, Durbridge TC. Estradioltreatment transiently increases trabecular bone volumein ovariectomized rats. Bone 1996;19:455-461.

28. Yasuda H, Shima N, Nakagawa N, et al. Identity of osteo-clastogenesis inhibitory factor (OCIF) and osteoprote-gerin (OPG): A mechanism by which OPG/OCIF inhibitsosteoclastogenesis in vitro. Endocrinology 1998;139:1329-1337.

29. Simon JA, Mack CJ. Counseling patients who elect todiscontinue hormone therapy. Int J Fertil Womens Med2003;48:111-116.

30. Tremollieres FA, Pouilles JM, Ribot C. Withdrawal of hor-mone replacement therapy is associated with significantvertebral bone loss in postmenopausal women. Osteo-poros Int 2001;12:385-390.

31. Tenenbaum HC, Shelemay A, Girard B, Zohar R, Frits PC.Bisphosphonates and periodontics: Potential applica-tions for regulation of bone mass in the periodontium andother therapeutic/diagnostic uses. J Periodontol 2002;73:813-822.

32. Uusi-Rasi K, Kannus P, Cheng S, et al. Effect of alendro-nate and exercise on bone and physical performance of

40053.qxd 2/18/05 9:31 AM Page 113

114

Alendronate Therapy and Dental Implants Volume 76 • Number 1

postmenopausal women: A randomized controlled trial.Bone 2003;33:132-143.

33. Narai S, Nagahata S. Effects of alendronate on theremoval torque of implants in rats with induced osteo-porosis. Int J Oral Maxillofac Implants 2003;18:218-223.

34. Greenspan SL, Emkey RD, Bone HG, et al. Significant dif-ferential effects of alendronate, estrogen, or combinationtherapy on the rate of bone loss after discontinuation oftreatment of postmenopausal osteoporosis. A randomized,double-blind, placebo-controlled trial. Ann Intern Med2002;137:875-883.

35. Fleisch MH. Bisphophonates-history and experimentalbasis. Bone 1987;8:S23-S28.

36. Rossini M, Gatti D, Zamberlan N, Braga V, Dorrizi R,Adami S. Long-term effect of treatment course with oralalendronate of postmenopausal osteoporosis. J BoneMineral Res 1994;11:1833-1837.

37. Sones AD, Wolinsky LE, Kratochvil FJ. Osteoporosisand mandibular bone resorption in the Sprague Dawleyrat. Calcif Tissue Int 1986;39:267-270.

38. Weiss RE, Gorn A, Dux S, Nimmi M. Influence of highprotein diets on cartilage and bone formation in rats.J Nutr 1981;111:804-816.

Correspondence: Dr. Francisco H. Nociti, Jr., Av. Limeira 901,Caixa Postal: 052, CEP: 13414-903, Piracicaba, S.P., Brazil.Fax: 55-19-3421-0144; e-mail: nociti@fop.unicamp.br.

Accepted for publication May 11, 2004.

40053.qxd 2/18/05 9:31 AM Page 114