The Internationa! Journal of Periodontics & Restorative Dentistr

521

Histologie Analysis of a FracturedImplant: A Case Report

Gérard Brunei

Serge A rmand"

Neat Miller***

Jacqueline Rue****

This study presents the histoiogic anaiysis of an impiant retrieved 74 months aiter

loading because of a fracture in the collar region. The implant ¡BioventJ was

removed with part of the penimplant tissues, and the block was prepared using

cutting and grinding equipment to obtain 3 sections approximately 30 firn thicfc.

The examination evidenced a high degree oi osseointegration, with a faone-to-

Implant contact of 74% ± T3%. Tho lamellar bone was dense and in c'ose relation

with the hydroxy apatite coating of the implant. The connection between the 0 4-

mm hydroxyapatite coating and the metal was always very tight. The thickness of

the meta/at the breaic point tvas assessed to be í,ó mm. Besides the relative

weakness ofthe metal at the neck of the implant, other possible causes of failure

are discussed, {Int J Periodontics Restorative Dent 2000;20:521-52ó,)

•Professor. Department of Oral Biology, Faculty of Odontology, PaulSabatier University, Toulouse, France.

••Maïtie de Conférence de l'Université, Department of Prosthadontics,Faculty of Odontology, Paul SabatJer University, Toulouse, France.

"Mai'tie de Conférence de l'Université, Department of Periodontology,

Faculty of Odontology, Nancy University, Nancy, Prance.'»Technician, Department of Oral Biology, Faculty of Odontology, Paul

Sabatier University, Toulouse, France.

Reprint requests. Dr Gérard Brunei, Faculté de Chirurgie Dentaire, 3Chemin des Maraîchers, 31062 Toulouse Cedei, Fiance, e-mail:brunel@,dct fr

In spite of a high percentage of long-

term successes, some implant fail-

ures occur for different reasons,

including impaired healing, micro-

btal contamination, or mechanical

problems. Among these possible

complications, fractures hold an

important place,'"" Subsequent

analyses of failed implants, whether

histologie,^ histochemical,* or elec-

tromicroscopic,*'^ provide invaluable

data that contribute to the evolution

of implant systems and the devel-

opment of measures to prevent fail-

ures. It was in thisframeof mind that

the histoiogic analysis of a Biovent

(CoreVent) implant was performed

following the fracture of its collar.

Clinical report

A 50-year-old partially edentulous

patient consulted for a complete

rehabilitation of his maxillary arch.

After clinical and radiographie eval-

uation, it was decided to construct

a dentoimplant-supported fixed

prosthesis on 8 vital residual teeth

(maxillary right second molar,

canine, and lateral incisor and left

Volume 20, Number 5, 2000

522

Fig la (left) Clinical view of the implantbroken in the cervical region.

Fig 1 b (right) fmpiant with its surround-ing tissues after removal and before histo-logie preparation.

lateral incisor, canine, first premolar,and first and second molars) with asubsidiary implant in the place ofthe right first premolar. The Bioventimplant, 3,5-mm diameter and 13mm long, was placed under normalconditions in compliance with themanufacturer's instructions, and aprovisional fixed partial denture(FPD) made of acrylic resin on a Co-Cr cast substructure was cementedin place and worn during the entirehealing period, avoiding all pres-sure on the implant site. After 8months, second-stage surgery wasperformed and the provisional FPDwas modified. The final ceramic

restoration spanning from the right

second molar to the left secondmolar was placed 3 months later.During the fol lowing year, thepatient was content with his maxil-lary prosthesis until he returned tothe office because of a persistentinflammation in the cervical periim-plant zone. In spite ofthe absenceof mobility, the radiographie controlsuggested a fracture ofthe implantcollar. Removing the ceramometal-lic FPD and the implant abutmentevidenced the fracture (Fig la), butbecause of the patient's profes-sional occupation the provisionalFPD was replaced after suturing the

mucosa over the implant site. It wasnot until 18 months later that thepossibility to remove the implantoccurred. This was accomplishedwith a 5.5-mm trephine, and a blockcomposed of the implant and itsperiimplant tissues (Fig 1 b) was pre-pared for histology. Healing wasevaluated by computed tomo-graph ic{CT) scan after 6 months. Anew implant (Frialit, Friatec) 3.8 mmin diameter and 13 mm long wasplaced (Fig 2), and another fixedprosthesis was cemented 10months later, with no other incidentfor more than 4 years.

The International Journal of Peiiodontics S Restorative Dentistry

523

Histologie preparation

The specimen was fixed in 10% neu-tral buffered formalin soiution, dehy-drated with ethanol containing 0.3%basic fuchsin, and embedded inmethyl methacryiate as describedby Stich.^ The block was sectionedusing a bandsaw and further groundto a thickness of approximately 30|jm. The 3 ground sections werecounterstained with light green andexamined under optic microscopy.The histologie observation assessedthe quality ofthe periimplant bone,the bone-to-implant interface, andthe implant design. Histomorpho-metric analysis was performed oneach section using an image-analy-sis system (Optilab, Graftek) coupledwith a video camera on the micro-scope. The bone-to-implant-contact(BIC) length was measured at a mag-nification of 1OOX as a percentage oftotal implant length."

Results

Figure 3 shows a global view of theimplant and its periimplant tissues. Inspite of the loss of these tissues inthe most cervical region during tre-phination, the degree of osseointe-gration appeared to be quite highon the light microscopic level. Theimplant was coated with hydroxyap-atite (HA), the thickness of which wasassessed to be about 0.4 mm (Figs4 and 5). The HA coating was in closerelation with the alveolar bone, andthe BIC mean was 74% ± 13%. Thebone structure was lamellar anddense, with many cells and few

medullar spaces (Fig 4), Some osteo-cytes were directly in contact withthe HA coating (Fig 5). A vascular-ized connective tissue layer pro-duced a very localized résorption ofthe HA coating (Fig 6). The bondingbetween the implant metal and theHA layer was very tight and contin-uous at all points (Fig 5). The designof the implant neck included a nar-rowing of the metal, which was 1.6mm thick in the fracture zone (Fig 7).

Discussion

The histologie examination demon-strated that the Biovent implant wascorrectly osseointegrated, as is thecase with most broken implants,around which authors have de-scribed lamellar bone, both matureand compact, providing 70% to100% contact.''•'•'° The choice of anHA-ccated implant was made be-cause the osteoconductive charac-teristics of this material enable a re-duction of the healing period,^'although some consider the prog-nosis to be uncertain,'^ Although asmall résorption spot was detected inthe HA coating (Fig 6), it is impon:antto insist on the quality of bone-HAand HA-metal interfaces in opposi-tion to other studies that describeHA particles in the cytoplasm cfmultinucleated cells," more or lessextensive dissolution of the coating,or an interruption of HA-metal con-nection.^ In the present case, thequality ofthe osseointegration char-acterized by a BIC of more than 70%can be explained by the fracture just1 year after loading, similar to the

Volume 20, Number 5, 2000

524

Fig 2 Radiograph of the Fnabt impiantrepiacing the Biovenr impiant in the site ofthe ma;<)7l3ry eight first premolar.

Fig 3 Genera! urew ofthe most mediansection. The implant appears well osseoin-tegrated for 'A of its height. The bone inthe cervicai third was iost during removalwith the trephine. The cifcied fragnienf is apiece of metai coated with hydroxyapatitethat was ripped off during the specimengrinding. (Original magnification >' 16;fuchsin and light green slam,J

Fig 4 Higher-power view of section afrom Fig '3. The periimpiant bone appearsnormaiiy dense and presents few meduilarspaces Between the metal ofthe implantand the bone, the thin gray outline is thehydroiyapatite coating. fOfiginai magnifi-cation - 40; fuchsin and iight green stain.)

Fig 5 Bone-hydroxyapatile-metal inter-face. The osseous lissue contains nunierousceils (osteocytes), some of wi^ich are directly¡n contact witfi the hydroxyapatite coating(red arrow], Ciose contacts eiist betweenthe bone and the hydroxyapafite, as well asbetween the hydroxyapatile and the blackmetal. The fhicl ness of ihe coating wasassessed to be 0.4 mm. (Original magnifica-tion / 250: fuchsin and light green stain.)

Fig 6 Higher-power uieiv of section bfrom Fig 3. The direct contacf of a medulla;space wiiii (he hydraxyapalile layer canlocally produce a résorption (arrows) thatreduces the thickness oí coating, (Originalmagnification X ¡60; fuchsin and lightgreen stain,)

Fig 7 Higher-power view of section cfrom Fig 3. On the part of the jmpJant col-isr that is still intact, the thiclfness o'ihemetal is only 1.6 mm, (Original magnifica-tion X 100, fuchsin and light green stain.)

The International Journal of Periodontics & Restorative Dentistry

525

report by Piattelli and Trisi,'^ How-

ever, unlike these authors,^ we did

not observe any osteoid border

along the surface of the coating,

which seems normal for an implant

that had not been loaded for 18

months.

Mechanical ruptures are still

responsible for too many failures in

implantology. In a retrospective sur-

vey of 8 years and 230 implants,

Piattelli et al" determined that of all

oftheirfailures, 40% were because of

fractures. As in this case, the frac-

tures occurred relatively soon after

loading—70% of the fractures were

early. This corroborates the work of

other authors, some of whom ob-

served fractures during the first

year,^'' with most during the first 5

years.'^ Ahead of all other causes,

mechanical tensions are responsible

for implant fractures,^•^''•^* and risks

seem greatest in the posterior re-

gions with short implants of stan-

dard diameters.'" The small-diame-

ter Biovent implant used to replace

the maxillary first premolar was in a

position strategic enough to impose

a canine-guided occlusion to avoid

excessive functional forces. However,

one must admit that there isa sort of

antagonism between the relative

mobility of natural teeth and the

ankylosisofan implant. A fixed pros-

thesis on natural teeth keeps a cer-

tain degree of mobility that produces

repetitive forces concentrated at the

connection between the suprastruc-

ture and the implant. In all implant

systems that have an internal antiro-

tational hexagon like Biovent, the

collar is the region subjected to the

most strain, which at length can

weaken the metal and finally pro-

duce a fracture.^''"^''" These obser-

vations question the indication of

mixed dentoimplant restorations.'•'"

Another risk factor is crestal résorp-

tion, which is inevitable with time'^

and may unfavorably increase the

crown-to-root ratio, thus accentuat-

ing mobility and weakening the cer-

vical region ofthe implants.

Of course, the design of certain

implant systems is also a source of

fragility,'^ like the holes in the hol-

low-cylinder implants.'" Indeed, the

thickness ofthe metal at the neck of

the Biovent implant did not seem

sufficient and could have presented

an aggravating risk in the presence

of unfavorable biomechanical con-

ditions.

This histoiogic analysis suggests,

though it does not present absolute

proof, that the strains caused by the

type of dentoimplant restoration

could have been the cause of the

fracture of the irnplant collar. The

thinness ofthe metal in this region

surely provided a "locus minoris

resistentiae" that increased the risk

of fracture.

Volume 20, Number 5, 2000

526

References

1, Lekholm U, van Steenberghe D, HerrmannI, Bolendef C, Folmer T, Gunne J, et alOsseointegrated implants in the treatmentof partially edentulous jaws-A prospective5-year multicenter study, Int J OralMaxilWac Implants 1994;9:627-635.

2, Friberg B, Jemt T, Leiiholm U, Early failuresin 4,641 consecutively placed Branemarkimplants: A study from stage 1 surgery tothe connection of completed prostheses,Int J Oral Maxillofac Implants 1991;6:142-146,

3, Balshi TJ, An analysis and management offractured implants: A clinical report. Int JOral Maxillofac Implants 1996;11 bbQ-ùbb.

4 Piattelli A, Scarano A, Piattelli M. Histologieobservations on 230 retrieved dentalimplants: 8 years' experience (1939-1996).J Periodontol 1998.69:178-184

5. Albrektsson T, Astrand P, Seeker W,Eriksson AR, Lekholm U, Malmquist J,Sennerby L. Histologie studies of faileddental implants: A letrieval analysis of fourdifferent oral implant designs Clin Mater1992:10:225-232

6. Piattelli A, Trisi P A light and laser scanningmicroscopystudyofbone/hydroxyapatite-coated titanium implants interface:Histochemical evidence of unmineralizedmaterial in humans. J Biomed Mater Res1994,28 529-536

7. Takeshita F, Matsushita Y, Ayukawa Y,Suetsugu T, Fractures of hydroxyapatite-coated blade implants connected with nat-ural teeth. A histobgical study using SEM,light microscopy, and an image processingsystem, J Penodontol 1996;67:86-92.

8. Stich H. Herstellung der Schliffpräparate.ln:SchroederA(ed) Orale Implantologie.Stuttgart: Thieme, 1988.109-115,

9. Gotfredsen K, Nimb L, Buser D, Hjarting-Hansen E. Evaluation of guided bone gen-eration around implants placed into freshextraction sockets: An experimental studyin dogs J Oral Maxillofac Surg 1993:51:879-884.

10. Piattelli A, Scarano A, Piattelli M, Vais E,Matarasso S. Hollow Implants retrievedfor fracture: A light and scanning electronmicroscope analysis of 4 cases. JPeriodontol 1998:09:18S-189.

11. Kay JF Calcium phosphate coatings fordental implants. Dent Clin North Am1992:36:1-18.

12. Biesbrock AR, Edgerton M. Evaluation ofthe clinical predictability of hydroxyap-atite-coated endosseous dental implantsA review of the literature. Int J OralMaxillofac Implants 1995;10:712-720,

13. Piattelli A, Trisi P. Microscopic and chem-ical analysis of bone-hydroxyapatite inter-face in a human retrieved implant. A casereport. J Periodontol 1993:64:906-909.

14 Rangert B, Krogh PH, Langer B, VanRoekel N. Bending overload and implantfracture A retrospective clinical analysis.Int J Oral Ma>;illofac Implants 1995:10:326-334

15 Steflik DE, Parr GR, Singh BB, Lake FT,Sisk AL, Howell FV, Shelton TW. Lightmicroscopic and scanning electron micro-scopic analyses of dental implants re-trieved from humans, J Oral Implantol1994:20:8-24,

16. ZarbG, Osseointegration: A requiem forthe periodontal ligament, Int JPeriodontics Restorative Dent 1991;1188-91.

17. Patterson EA, Johns RB. Theoreticalanalysis of the fatigue life of f xture screwsin osseointeg rated dental implants, Int JOral Maxillofac Implants 1992:7:2éH-33

18. SpiekermannH, Jansen VK, Richter EJ. A10-year follow-up study of IMZ and TPSimplants in the edentulous mandibleusing bar-retained overdentures. Int J OralMaxillofac Implants 1995;10:231-243.

The International Journal of Periodontics & Restorative Dentistry