- Email: [email protected]
Influence of implant diameters on the integration of screw implants
Copyright 9 Munksgaard 1997
Int. ,L Oral Maxifioj2zc. Surg 1997; 26:141 148 Printed in Denmarlc . All rights reserved
[tlrertTatiana]Jeurn~7lo)
Oral&
Maxillofacial Surgery I S S N 0901-5027
Influence of implant diameters on the integration of screw
implants An experimental study in rabbits
C.-J. Ivanoff 1,2,3, L. Sennerby 1,2, C. Johansson 2, B. Rangert 4, U. Lekholm 1,2 ~Br&nemark Clinic, Public Dental Health Service and Faculty of Odontology, 2Department of Biomaterials/Handicap Research and Institute for Surgical Sciences, Faculty of Medicine, University of Gdteborg; 3Department of Oral and Maxillofacial Surgery, M61ndal Hospital, Mdlndal; 4Nobel Biocare AB, Gdteborg, Sweden
C.-J. Ivanoff L Sennerby, C. Johansson, B. Rangert, U. Lekholm: Influence of implant diameters on the integration of screw implants. An experimental study in rabbits. Int, J, Oral Maxillo/izc. Surg. 1997; 26." 141-148. 9 Munksgaard, 1997
Abstract. The influence of diameter on the integration of titanium screw-shaped implants was studied in the rabbit tibia by means of removal torque measurements and histomorphometry. Implants 3.0, 3.75, 5.0, and 6.0 mm in diameter and 6.0 mm long were inserted through one cortical layer in the tibial metaphyses of nine rabbits and allowed to heal for 12 weeks. The implants were then unscrewed with a torque gauge, and the peak torque required to shear off the implants was recorded. The histologic analysis in undemineralized ground sections comprised (1) a gross description of the implant sites and assessments of (2) the total implant length in bone and (3) in the cortical passage, as well as (4) the thickness of the cortical bone adjacent to the implants. From the removal torque values obtained and morphometric measurements, a mean shear stress value was calculated for each implant type. The biomechanical tests showed a statistically significant increase of removal torque with increasing implant diameter. The resistance to shear seemed to be determined by the implant surface in supportive cortical bone, whereas the newly formed bone at the periosteal and endosteal surfaces did not seem to have any supportive properties after 12 weeks. It is suggested that wide diameter implants may be used clinically to increase implant stability.
The clinical result of screw implants (Nobel Biocare) has been well documented for treatment of complete and partial edentulism~,3,11,13 2~ 22 26 so s9 Despite the documented high predictability and successes, complications and failures have been reported 11,13,2~ Poor bone quality and extreme jawbone resorption have been pointed out as risk factors. Methods to improve bone volume have consequently been presented, such as different grafting and augmentation procedures 5'4a. In situations of poor bone quality, prolonged unloaded healing periods have been suggested 14, an approach which has also been sup-
ported by experimental findings presented by JOHANSSON ~r ALBREKTSSON23, From the latter study performed in the rabbit tibia, the authors reported a gradtiaIly increasing implant removal torque and bone-to-implant contact within a 12-month period. The stability and resistance to shear forces of implants also seem to be dependent on the biomechanical properties of the bone in the implant interface, as reported by STEINEMANN et al. 37 and Ru~o D~ I~ZENDE • JOHANSSON34, while cortical bone seems to provide better stability and long-term integration than cancellous bone 15A8~36'43.
Key words: endosteal implants; screw implants; removal torque; diameter; biomechanic; bone support. Accepted for publication 28 August 1996
The surgical technique used may also be a source of error, negatively influencing the integration process. Overheating the bone and intraoperative implant mobility have been reported as
Table 1. Measurements (mm) of implants used in study Outer diameter 3.0 3.75 5.0 6.0
Inner diameter
Pitch height
2.46 3.15 4.13 4.92
0.5 0.6 0.8 1.0
1 42
I v a n o f f et al.
Fig. 1. Four different types of implants used
in study.
5.00 mm O 3~75mm O
~
6.00 mm O 3.00 mm O
Tibiadx Tibiasin Fig. 2. Diagram of implantation sites for four types of implants in tibial metaphyses. O=diameter.
two of the most frequent surgical errors 12,28. An intermittent drilling technique under profuse irrigation with
sterile saline has been advocated to avoid overheating the bone 1. Intraoperative implant mobility may be prevented by using a precise drilling procedure 1,a,~9. Another option investigated by several authors is to modify implant design, on both the micro- and macrolevel. A rough surface topography has been reported to increase osseointegration as compared to a smooth surface 7'16'31'4~ Hydroxyaparite-coated implants have been shown to result in faster bone integration in the short term, whereas they did not seem to influence the integration process positively in the long run 17. Screw implants seem to provide better initial stability and resistance to compression and tension stress forces 4 than cylindric implants6,! 6. In implant insertion, initial stability is of paramount importance ~. This can be accomplished either by engaging the mandibular lower cortex when operating between the mental foramina, or by engaging the inferior cortical plates of the maxillary sinuses and/or nasal cavities. In jaws not suitable for standard implants, and/or in situations where
primary stability of the fixture is not achievable, the use of a wider implant diameter has been suggested 27. Increased bone-to-implant contact and engagement of as much cortical bone as possible may thereby be obtained. This notion is supported by clinical data in that higher success rates are reported for wider implants (4.0 mm) in soft bone29, 39. No experimental studies have been reported, in which the results of different implant diameters were specifically assessed for their capacity to improve bone integration. Therefore, the objective of the present investigation was to evaluate the influence of different implant diameters on integration of titanium implants in the rabbit tibia. Material and methods Animals and anesthesia
Nine adult New Zealand White rabbits of both sexes, weighing 5-6 kg, were used in the study. The animals were kept in standard cages and fed ad libitum on conventional laboratory diet. The experiment was approve d by the animal ethics committee at the University of G6teborg, Sweden. During the surgical sessions,
Fig. 3. Contact radiographs of four different implant diameters taken after retrieval. Note that implants originally were inserted through one cortical layer only. a) 3.0 ram, b) 3.75 mm, c) 5.0 ram, d) 6.0 mm.
Implant integration 100"
80-
gg 6ot "~
9
40-
o
8 8
20, .
.
.
.
,
Implant diameter (mm) Fig. 4. Results from removal torque measurements. All values (n 9) are given for each implant diameter. Curve fit to all plotted data of second degree shows r value of 0.80.
the animals were anesthetized with intramuscular injections of fentanyl and fluanison (Hypnorm | Janssen, Brussels, Belgium) at a dose of 0.7 mg/kg body weight, and intraperitoneai injections of diazepam (Stesolid| Dumex, Copenhagen, Denmark) at a dose of 1.5 mg/kg body weight. Approximately 1 ml of local anesthesia (2% lidocaine with epinephrine
143
tively, were used in the investigation (Fig. 1). Details regarding outer diameter (as measured from the most prominent part of the external thread), inner diameter (as measured from the deepest part of the external thread), and pitch height are presented in Table 1. All implants were custom-made of c.p. titanium (Nobel Biocare AB, G6teborg, Sweden). They were designed with a standard hexagonal head, which was fitted to an ordinary Br~nemark System | fixture mount. In contrast to the 3.0- and 3.75-mm implants, the 5.0- and 6.0-mm diameter ones were not shaped with a shoulder within their neck portion (Fig. 1) but were, therefore, threaded all the way up to their top margin. All implants were cleaned ultrasonically in butanol and absolute ethanol for 10 min. in each solution and sterilized according to a routine protocol (Nobel Biocare AB, G6teborg, Sweden). Implant surgery and research protocol
12.5/zg/ml; Xylocain/Adrenalin | Astra AB, SOdert~tlje, Sweden) was also injected into the areas exposed to surgery. Implants
Implants (fixtures) 6.0 mm long and either 3.0, 3.75, 5.0, or 6.0 mm in diameter, respec-
4
All surgery was performed under sterile conditions in an animal operation theater. Both tibial metaphyses were chosen as experimental sites and were exposed by curved skin incisions via fascial periosteal flaps. In each tibia, preparations for two implant sites (n= 36) were made, by the technique described by ADELL et al. 1. A guide drill was first used to mark the implant site locations approximately 10 m m apart. The sites were then sequentially enlarged with twist drills as described below, except for the 3.75-mm sites, where a Br~memark System | pilot drill was used as well, by the standard technique: Implant diameters: 3.0, 3.75, 5.0, 6.0 mm; twist drill diameters: 2.0 followed by 2.4 mrri, 2.0 followed by 3.0 ram, 2.0, 3.0 followed by 4.3 mm, 2.0, 3.0 followed by 5.3 mm. Countersinking was omitted in all sites, which were threaded with conventional taps suiting the various implant diameters. The implant sites were divided into four groups, according to the type of implant inserted in the cranial cortical layer (Fig. 2):
Group A. right proximal tibia: 5.0 mm (n= 9)
Group B. right distal tibia: 3.75 mm @=9) Group C. left proximal tibia: 6.0 mm (n=
9
9)
4
Group D. left distal tibia: 3.0 mm @=9)
,,4 Fig. 5. Light microscopy of ground section
Fig. 6. Light microscopy of ground section of
of 6.0-mm-wide implant that could not be unscrewed with torque gauge manometer. Most of implant surface is in contact with surromlding bone. Newly formed bone on periosteal (P) and endosteal (E) surfaces can be distinguished from cortical bone (C). Bridge of bone (*) between medial cortex and implant surface is seen. B M = b o n e marrow, I=implant. Staining 1% toluidine blue/pyronin-G•
3.0-mm-wide implant. Morphology indicates separation (*) between implant surface and surrounding bone due to removal torque test. Crack can be seen in bone radiating from third thread (arrow). Newly formed periosteal (P) and endosteal bone (E) can be distinguished from cortical bone. Endosteal bone has been separated from original cortical bone (open arrow). I=implant. Staining 1% toluidine blue/pyronin-G•
Consequently, a total of 36 implants were inserted in one cortical layer (Fig. 3). To avoid bony overgrowth of the fixture heads, the 5.0- and 6.0-mm implants (lacking shoulders) were inserted so that approximately ~ 3 threads were exposed above the marginal bone level. On the other hand, the 3.0- and 3.75-mm implants, were inserted so that only the inferior conical part of the fixture head was in contact with the outer cortex of the tibial bone. The surgical wounds were closed in layers with single resorbable subcutaneous sutures and silk sutures in the skin. Postoperatively, all
1 44
l v a n o f f et a/. Table 2. Torque measurenrents together with shear-stress calculations (mean+_SD) Implants Parameters
3.0
Torque (N cm)
Fig. 7. Light microscopy of ground section showing fracture line (arrows) within bone in cortical passage radiating from thread from 3.75-ram-wide implant. Crack is filled with blood cells (*). I=implant. Staining 1% toluidine blue/pyronin-G• 25.
13.7+6.2
3.75 24.8•
6.0
Statistics *
62.2-+20.3
All significantly different (P-<0.05) *P-<0.05 vs 3.0 and 3.75
Shear stress (N/mm2), total bone
3.1-+1.4
3.3-+1.7
1.8• *
1.9-+0.6 *
Shear stress (N/ram2), cortical passage
5.0_+2.3
5.4•
4.5-+1.4
4.9+1.6 NS
t Wilcoxon's signed rank test.
the aforementioned surgical areas were re-exposed. In some animals, a bony overgrowth of the fixture head/shoulder was seen. In order not to bias the removal torque results, this bone was carefully removed with a scalpel. An electronic torque meter (Digital Torque Measurement, DTF-1, Crane Electronics, UK) was connected to all implants via a standard Brfinemark System | fixture mount, and assessment of the peak torque (N cm), required to loosen the implants, was performed. A mean torque value was calculated for each research group.
Specimen retrieval
Fig. 8. Light microscopy of ground section of 3.75-mm-wide implant. Implant has been separated from bone surrounding head of implant (*), but threads seem still to be in contact with bone. Numerous cracks (arrows) show in cortical bone, and newly formed endosteal bone (E) has been separated from original cortical bone (C). Periosteal (P) bone formation, which can be distinguished from original cortical bone, is evident. I=implant, BM=bone marrow. Staining 1% toluidine blue/pyronin-GX6.3.
5.0 32.4~-9.9
The animals were killed by an intravenous overdose of pentobarbital (Mebumal | ACO Lfikemedel, Solna, Sweden), given via an ear vein. The implants, together with surrounding bone and soft tissues, were removed en bloc and fixed by immersion in 4% buffered formalin solution. The obtained specimens were embedded in acrylic resin, and cut and ground in an Exakt sawing machine (Exakt Apparatebau, Nodersted, Germany) to about 10-/~m-thick sections 1~ The specimens were finally stained with 1% toluidine blue/ pyronin-G.
Histologic analyses The obtained sections were histologically analyzed in a Leitz microscope equipped with a Microvid | morphometric system connected to a PC. The equipment permitted direct quantitative calculations in the ground sections obtained a4. The histologic assessments involved (1) gross morphologic description of the bone, tissues surrounding the implants, (2) measurements of the total implant length in bone, and (3) the implant length in the cortical passage, and (4) measurement of the thickness of the cortical bone adjacent to the implant. A mean value for the parameters was calculated for each implant, as well as the mean for the four research groups. Because the implants had been fractured loose in order to measure the removal torque levels, the true degree of bone-to-implant contact could not be measured.
Calculations A mean shear stress value (N/mm 2) was calculated by the formula: T hzdsr, where T is the removal torque (N mm), d is the mean diameter of the implant (ram), r is the mean
Total implant length in bone 108"~
6"
///////A
*
animals were given antibiotics (Intencillin | 2250000 IE/5 ml; 0.1 ml/kg body weight; LEO, Helsingborg, Sweden) and analgetics (Temgesic| 0.05 mg/kg body weight; Reckitt and Coleman, USA) as single intramuscular injections. A postoperative healing period of 12 weeks was allowed for all rabbits.
Torque measurements At the end of the healing period, the animals were anesthetized as previously described, and
4-
0
9
3.0
I
"
3.75
l
I
5.0
6.0
Implant diameter (mm) Fig. 9. Results from morphometric measurements of total implant length in bone (mm• Statistically significant difference was shown between all groups except 3.0 and 3.75 mm.
Implant integration
145
Table 3. Morphometric measurements and calculations (mean_+SD) Implants Parameter
3.0
3.75
5.0
6.0
Statistics
Cortical thickness (mm)
1.46_+0.18
1.51_+0.29 1.22-+0.20*
1.32-+0.11 *
Implant length in bone (ram)
3.83_+0.79 *
4.00+0.48
7.46+3.08
* NS vs 3.75 mm
Implant length in cortical passage (mm)
2.37-+0.37 2.42_+0.34 2.17_+0.45 *
2.69-+0.29
* P<0.05 vs all groups
5.65+1.78
* P-<0.05 vs 3.0 and 3.75 mm
Wilcoxon's signed rank test.
length of the lever arm (mm), and s is the length of the implant in bone (mm). Two s values were used in the formula: (1) the total implant length in bone, and (2) the implant length in the cortical passage, respectively.
level for plastic deformation of their hexagonal head, which is estimated to be 90 N cm 25, and were consequently not loosened. These sites were ascribed the value of 90 N cm. A curve fit of the second degree showed an r value of 0.80.
Statistics
Wilcoxon's signed rank test was used for statistical analyses. Results Clinical findings
The animals recovered well after surgery, and the healing of the surgical sites was uneventful. All 36 fixtures were found stable at re-entry when tested with the torque manometer. Removal torque values
The mean removal torque values for the four groups were 13.7_+6.2, 24.8_+15.1, 32.4_+9.9, and 62.2+20.3 N cm for the 3.0-, 3.75-, 5.0-, and 6.0-mm-wide implants, respectively. There was a statistically significant difference between all groups (Table 2 and Fig. 4). Two implants in group C exceeded the critical
Histologic findings
The histologic examination revealed that all implants had initially engaged one cortical layer as desired. However, endosteal bone formation, which seemed to be more prominent for the 5.0- and 6.0-mm implants, had resulted in encapsulation of the threads in the bone marrow cavity to various degrees (Fig. 5). This bone had a lamellar appearance but could still be distinguished from the original cortical bone. Because the implants had been loosened, a separation could be observed in the bone-implant interface (Fig. 6), and fracture lines radiating from the implant thread peaks were observed (Fig. 7). Fracture lines were most often present between the original cortical bone and the newly formed endosteal bone, which had been ripped off and separated from the orig-
I m p l a n t l e n g t h in the cortical passage 10"
a=
6-
inal cortical bone (Fig. 8). Bone formation had also occurred at the periosteal surface to various extents (Figs. 5, 6 and 8). The mean values for the measured total implant length in bone were 3.83-+0.79, 4.00-+0.48, 5.65-+1.78, and 7.46_+3.08 m m for the 3.0-, 3.75-, 5.0-, and 6.0-mm implants, respectively (Table 3 and Fig. 9). The corresponding values for the implant length in the cortical passage were 2.37-+0.37, 2.42-+0.34, 2.17_+0.45, and 2.69_+0.29 mm, respectively (Table 3 and Fig. 10). The thickness of the cortical bone adjacent to the implants was found to be 1.46_+0.18, 1.51_+0.29, 1.22_+0.20, and 1.32_+0.11 m m for the 3.0-, 3.75-, 5.0-, and 6.0-mm implant sites (Table 3 and Fig. 11). There was a statistically significant difference between the distal sites (3.0 and 3.75 mm) and the proximal sites (5.0 and 6.0 ram). Calculations
F r o m the total implant length in bone, the mean shear stresses were calculated to be 3.1_+1.4, 3.3_+1.7, 1.8_+0.6, and 1.9_+0.6 N / m m 2 for groups D, B, A, and C, respectively (Table 2 and Fig. 12). The shear stress for the 5.0- and 6.0m m implants were significantly lower than for the others. The corresponding shear stresses, for the implant length in the cortical passage, were 5.0_+2.3, 5.4+2.9, 4.5_+1.4, and 4.9-+1.6 N / m m 2, respectively, with no significant differences between the different implants (Table 2 and Fig. 13).
4Discussion
9
3.0
I
I
3.75
!
5.0
"
!
6.0
Implant diameter (mm) Fig. 10. Results from morphometric measurements of implant length in cortical passage (mm-+SD). Implants of 5 mm had statistically significantly less implant length in cortical passage.
The present experimental study demonstrated an increased removal torque with larger implant diameters, a finding which should be expected from geometric effects. The larger radius gives a linear increase of the shear force lever arm as well as a linear increase of the surface area in contact with bone.
146
lvanofj'et al. Thickness of the cortical bone at the implant sites _
. ID
1" 9
9
I
I
3.0
I
3.75
5.0
9
6.0
Implant diameter (ram) Fig. ll. Results from morphometric measurements of thickness of cortical bone adjacent to implants (mm• Bone was statistically thinner at proximal sites where 5.0- and 6.0-mm implants had been inserted, than for 3.0- and 3.75-mmimplants.
Shear stresses / total implant length in bone
r
Z
111 5' 4'
(D
3-
r,t3
2-
10
a
3.0
!
3.75
I
5.0
6.0
Implant diameter (mm) Fig. 12. Results from calculations of shear stress based on total implant length in bone. Shear stress for 5.0- and 6.0-mm implants was statistically lower than for 3.0- and 3.75-mmimplants.
Therefore, the removal torque should be proportional to the square of the diameter, if the shear stress is constant. The second-degree fit to the removal torque vs. implant diameter (Fig. 4) indicates that these geometric effects were dominant (r=0.80). The calculated shear stress based on the total length of bone in contact with the implants (Table 2, Fig. 12) demonstrated a significantly lower level for the 5.0- and 6.0-ram implants, while the shear stress relative to the original cortical passage was constant without significant differences between the different implant diameters (Table 2, Fig. 13).
This seems to indicate that the newly formed bone around the implants did not contribute substantially to the shear force resistance, a supposition which is in accordance with SEYr,rERBY et al? 6, who concluded that the bone in the cortical passage gives the main support to the implants. It has been demonstrated 13,15,1s,56 that compact bone provides better stability and long-term integration of endosteal implants than cancellous bone. IVANOFFet al. ~8 found a correlation between removal torque and degree of bone contact for 3.75-mm implants penetrating one or two cortical layers.
IVANOFF et al./s calculated about twice as high shear stresses for 3.75-mm wide implants inserted in the rabbit tibia after 12 weeks; i.e., 8 10 N/mm 2 as compared to about 5 N/mm 2 in the present study. However, in that study, the calculations were based on the true bone-implant contact of unremoved implants, which was about 40 50% in the cortical passage. In the present study, the calculations were based on the implant length in the cortical passage. Assuming that the bone contact was also about 40-50% in the present study, the shear stresses would be around 10-12.5 N/mm 2, corresponding to the results of IVANOFF et al. ~8. RUBO DE REZENDE ~; JOHANSSON34 calculated shear stresses for 3.7-ram-wide titanium implants inserted in the rabbit tibia after 12 weeks of healing. On the basis of the implant length in the cortical bone, they found the mean shear stress to be 1.9 N/ram 2. However, a shear stress value of 14.8 N/ram 2 was calculated taking into account the bone implant contact measured on unremoved implants. Based on the results of the present study and earlier work, the shear stress in rabbit bone after a 12week healing seems to be around 10 15 N/mm 2 related to the true contact area in the cortical passage of the tibia. The histologic observations in the present study support the concept of "nonsupportive" bone, since the newly formed endosteal bone had been sheared off from the original cortex. The supportive properties of the newly formed bone are probably related to the degree of maturation and/or the shape of the bone. In the present study, the new bone formed a collar around the implant. This is why it probably was too weak to resist shear, in contrast to the bone in the cortical passage. It is possible that if a longer healing period had been allowed, the newly formed bone would also have supported the implant. According to RORERTSet al. 33, it takes 18 weeks for the rabbit bone around an implant to mature. In the present study, the newly formed bone appeared to be lamellar and well organized but could still be distinguished from the original bone after 12 weeks of healing due to its darker staining and different morphologic appearance. It is likely that a uniform orientation of collagen fibers and complete maturation of the newly formed bone are required for all the interfacial bone to support an implant. The removal torque test is a simple
Implant integration Shear stresses/cortical passage
54"
32s G'
I
3.0
I
3.75
I
5.0
6.0
Implant diameter (mm)
Fig. 13. Results from calculations of shear stress based on implant length in cortical passage. There were no statistical differences between implant diameters tested.
1 4"1
ZARB GA, ALBREKTSSONT, eds.: Tissueintegrated prostheses. Osscointegration in clinical dentistry. Chicago: Quintessence, 1985:211 32. 2. ADELL R, ERIKSSON B, LEKIIOLM U, BRXNEMARIrP-I, JEMTW.A long-term follow-up study of osseointegrated implants in the treatment of the totally edentulous jaw. Int J Oral Maxillofac Impl 1990: 5: 347 59. 3. ALBREKTSSONT, DAHLE, ERBOML, et al. Osscointegrated oral implants. A Swedish multicenter study of 8139 consecutively inserted Nobelpharma implants. J Periodontol 1988: 59:287 96. 4. BRANEMARK P-I, BREINE U, ADELL R, HANSSONBO, LINDSTROM J, OHLSSON Z~. intra-osseous anchorage of dental prostheses. I. Experimental studies. Scand J Plast Reconstr Surg 1969: 3:81 100. 5. BUSER D, DAHLIN C, SCHENK RK. Guided bone regeneration in implant dentistry. Chicago: Quintessence, 1994. 6. CARLSSON L, ROSTLUND T, ALBREKTSSON
method and has frequently been used to evaluate the tissue response to titanium and other materials in experimental7,9,16 18,23,24,32,34 37,40,41 and clinical studies 38'43. The test is a valuable method for evaluating the interracial strength of endosteal implants. As seen in the present study, the bone morphology differed when comparing the proximal with the distal part of the tibial metaphysis; ideally, test and control implants should have been inserted at identical sites. However, as also shown in the present study, if implant and bone geometric variations are taken into account, it seems possible to compare the integration of implants when they have different thread designs and/ or are placed in sites with different bone morphologies. In a clinical situation where poor bone density is present, engagement of the buccal and lingual cortical plates with wider implants is desirable in order to achieve good primary stability of the implants and promote integration 14. Therefore, wider implants could be useful in the molar region, especially above the inferior alveolar nerve, where bicortical anchorage cannot be accomplished 18. However, this recommendation should be limited to situations with sufficient buccolingual dimension. In fact, too large an implant may reduce the cortical support, especially around the neck of the implant, and consequently jeopardize the primary stability of the implant. With, p o o r bone density 29,39, clinical data indicate a higher success rate for wider implants (4.0 m m
in diameter) than for standard implants (3.75 m m in diameter). Promising clinical results have also been presented for implants 5.0 and 5.5 mm in diameter 27. However, the relationship between the clinical performance of an implant and the shear strength of the bone-implant interface is not known. The possible positive effect of wide-diameter implants has yet to be established in controlled clinical trials. In conclusion, the stability of screwshaped titanium implants, as measured with a removal torque test, seems to be determined by the implant surface area in supportive cortical bone and the lever arm relative to the axis of rotation. It is suggested that the use of wide-diameter implants may increase stability in the clinical situation because it involves more supportive cortical bone. Controlled prospective clinical trials are needed to study the influence of implant diameter, as related to the host-tissue anatomy.
Acknowledgements. This study was supported by grants from the Swedish Medical Research Council, the Swedish National Board for Industrial and Technical Development (NUTEK), the Sylvan Foundation, the Hjalmar Svensson Foundation, the Greta and Einar Asker Foundation, the Wilhelm and Martina Lundgren Foundation, the Swedish National Board of Health, the FRF Foundation, and Nobel Biocare AB. References 1. ADELL R, LEKHOLM U, BRANEMARKP-I.
Surgical procedure. In: BRANEMARKP-I,
B, ALBREKTSSONT, BR2,NEMARKP-I. Osseointegration of titanium implants. Acta Orthop Scand 1986: 57:285 9. 7. CARLSSONL, ROSTLUNDT, ALBREKTSSON B, ALBP,EKTSSON12 Removal torques for polished and rough titanium implants. Int J Oral Maxillofac Impl 1988: 3: 21-4. 8. CARLSSONL, R0STLUNDT, ALBREKTSSON B, ALBREKTSSONT. Implant fixation improved by close fit. Cylindrical implantbone interface studied in rabbits. Acta Orthop Scand 1988: 59: 272-5. 9. CARLSSONLV. On the development of a new concept for orthopaedic implant fixation. Thesis, University of G6teborg, Sweden, 1989. 10. DONATH K. Die Trenn-Dtinnschliff-Technik zur Herstellung histologischer Pr~iparate yon nicht schneidbaren Geweben und Materialen. Pr~iparator 1988: 34: 197 206. 11. ENGQUIST B, BERGENDAHLT, KALLUS T,
LINDEN U. A retrospective multicenter evaluation of osseointegrated implants supporting overdentures. Int J Oral Maxillofac Impl 1988: 2:129 34. 12. ERICSSONRA. Heat-induced bone tissue injury. An in vivo investigation of heat tolerance of bone tissue and temperature rise in the drilling of cortical bone. Thesis, University of G6teborg, Sweden, 1984. 13. FRIBERG B, JEMT T, LEKHOLM U. Early failures in 464l consecutively placed BrLnemark dental implants. A study from stage I surgery to the connection of completed prostheses. Int J Oral Maxillofac Impl 1991: 6: 142-6. 14. FRIBERGB. Bone quality evaluation during implant placement. Academic dissertation for the degree of master of dental science. University of G6teborg, G6teborg, Sweden, 1994. 15. FRmER6 B, SENNERBYL, ROOS J, LEKHOLMU. Identification of bone quality in conjunction with insertion of titanium im-
148
IvanoJfet al.
plants. A pilot study in jaw autopsy specimens. Clin Oral Impl Res 1995: 6: 213-19. 16. GOTFREDSEN K, NIMB L, HJORT1NGHANSEN E, JENSENJS, HOLM~NA. Histomorphometric and removal torque analysis for TiO2-blasted titanium implants. An experimental study on dogs. Clin Oral Impl Res 1992: 3: 77-84. 17. GOTTLANDERM. On hard tissue reactions to hydroxyapatite-coated titanium implants. Thesis, University of G6teborg, Sweden, 1994. 18. IVANOFFC-J, SENNERBYL, LEKHOLMg. Influence of mono-and bicortical anchorage on the integration of titanium implants. A study in the rabbit tibia. Int J Oral Maxillofac Surg 1996: 25: 229-35. 19. IVANOFFC-J, SENNERBYL, LEKHOLM U. Influence of initial implant mobility on the integration of titanium implants. An experimental study in rabbits. Clin Oral Impl Res 1996: 7:120 7. 20. JAFFIN RA, BERMAN CL. The excessive loss of Br~nemark fixtures in type IV bone. A 5-year analysis. J Periodontol 1991: 62: 2M. 21. JEMT T, LEKHOLM U, ADELL R. Osseointegrated implants in the treatment of partially edentulous patients: a preliminary study of 876 consecutively installed fixtures. Int J Oral Maxillofac Impl 1989: 4:211 7. 22. JEMT T, LEKHOLMU. Implant treatment in edentulous maxillae: a 5-year followup report on patients with different degrees of jaw resorption. Int J Oral Maxillofac Impl 1995: 10:303 11. 23. JOHANSSON C, ALBREKTSSON T. Integration of screw implants in the rabbit. A 1-year follow-up of removal torque of titanium implants. Int J Oral Maxillofac Impl 1987: 2:69 75. 24. JOHANSSON C. On tissue reactions to metal implants. Thesis, University of G6teborg, GOteborg, Sweden, 1991. 25. JORNI~USL. Personal communication. Nobel Biocare AB, G6teborg, Sweden, 1996. 26. LANEY WR, JEMT T, HARRIS D. et al. Osseointegrated implants for single-tooth replacement: progress report from a multicenter prospective study after 3
years. Int J Oral Maxillofac lmpl 1994: 9: 49-54. 27. LANGERB, LANGERL, HERRMANN1, JORN/~USL. The wide fixture. A solution for special bone situations and a rescue for the compromised implant. I. Int J Oral Maxillofac Impl 1993: 8:400 8. 28. LEKHOLMU, ADELL R, BR.I.NEMARKP-I. Complications. In: BRA.NEMARK P-I, ZARB GA, AEBREKTSSONT, eds.: Tissueintegrated prostheses. Osseointegration in clinical dentistry. Chicago: Quintessence 1985: 233M0. 29. LEKHOLM U. The Brgmemark implant technique. A standardized procedure under continuous development. In: LANEYWR, TOLMANDE, eds.: Tissue integration in oral, orthopedic and maxillofacial reconstruction. Chicago: Quintessence, 1992: 194-9. 30. LEKHOLMU, VAN STEENBERGHED, HERRMAN I, et al. Osseointegrated implants in the treatment of partially edentulous jaws: a prospective 5-year multicenter study. Int J Oral Maxillofac Impl 1994: 9: 627-35. 31. MANIATOPOULOSC, PILLIAR RM, SMITH DC. Threaded versus porous-surfaced designs for implant stabilization in boneendodontic implant model. J Biomed Mater Res 1986: 20:1309 33. 32. MORBERG P. On bone tissue reactions to acrylic cement. Thesis, University of G6teborg, Sweden, 1991. 33. ROBERTS WE, GARETTO LW, BREZNIAK N. Bone physiology and metabolism. In: MISCH CE, ed.: Contemporary implant dentistry. St. Louis, MO: Mosby Year Book, 1993:327 53. 34. RUBO DE REZENDE ML, JOHANSSONCB. Quantitative bone tissue response to commercially pure titanium implants. J Mater Sci: Mater Med, 1993: 4:233 9. 35. SEWNERBYL. On bone tissue response to titanium implants. Thesis, University of G6teborg, Sweden, 1991. 36. SENNERBYL, THOMSENP, ERICSSONL. A morphometric and biomechanic comparison of titanium implants inserted in rabbit cortical and cancellous bone. Int J Oral Maxillofac Impl 1992: 7: 62-71.
37. STEINEMANN SG, EULENBEROER J, MAEUSLI P-A, SCHROEDERA. Adhesion of bone to titanium. In: CHRISTEL P, MEUNIER A, LEE AJC, eds.: Biological and biomechanical performance of biomaterials. Amsterdam: Elsevier, 1986: 409-14. 38. TJELLSTROMA, JACOBSSONM, ALBREKTSSON"12Removal torque of osseointegrated craniofacial implants. A clinical study. Int J Oral Maxillofac Impl 1988: 3:287 9. 39. VAN STEENBERGHE D, LEKHOLM U, BOLENDER C, et al. The applicability of osseointegrated oral implants in the rehabilitation of partial edentulism: a prospective multicenter study on 558 fixtures. Int J Oral Maxillofac Impl 1990: 5: 272-81. 40. WENNERBERG A, ALBREKTSSON T, ANDERSSONB, KROLJJ. A histomorphometric and removal torque study of screwshaped titanium implants with three different surface topographies. Clin Oral Impl Res 1995: 6: 24-30. 41. WENNERBERG A. On surface roughness and implant incorporation. Thesis, University of G6teborg, Sweden, 1996. 42. WORTHINGTON P, BRANEMARKP-I. Advanced osseointegration surgery. Applications in the maxillofacial region. Chicago: Quintessence, i992. 43. YAMANAKAE, TJELLSTROMA, JACOBSSON M, ALBREKTSSON T. Long-term observations on removal torque of directly bone-anchored implants in man. In: YANAGIHARA N, SUZUKI J, eds.: Transplants and implants in otology. II. Proceedings of the Second International Symposium of Transplants and Implants in Otology. Matsyama, Ehime, Japan, 1-5 April 1991. Amsterdam: Kugler 1992.
Address:
Dr Carl-Johan Ivanoff Department of Oral and Maxillofacial Surgery Mdlndal Hospital S-431 80 M6lndal Sweden
Int. ,L Oral Maxifioj2zc. Surg 1997; 26:141 148 Printed in Denmarlc . All rights reserved
[tlrertTatiana]Jeurn~7lo)
Oral&
Maxillofacial Surgery I S S N 0901-5027
Influence of implant diameters on the integration of screw
implants An experimental study in rabbits
C.-J. Ivanoff 1,2,3, L. Sennerby 1,2, C. Johansson 2, B. Rangert 4, U. Lekholm 1,2 ~Br&nemark Clinic, Public Dental Health Service and Faculty of Odontology, 2Department of Biomaterials/Handicap Research and Institute for Surgical Sciences, Faculty of Medicine, University of Gdteborg; 3Department of Oral and Maxillofacial Surgery, M61ndal Hospital, Mdlndal; 4Nobel Biocare AB, Gdteborg, Sweden
C.-J. Ivanoff L Sennerby, C. Johansson, B. Rangert, U. Lekholm: Influence of implant diameters on the integration of screw implants. An experimental study in rabbits. Int, J, Oral Maxillo/izc. Surg. 1997; 26." 141-148. 9 Munksgaard, 1997
Abstract. The influence of diameter on the integration of titanium screw-shaped implants was studied in the rabbit tibia by means of removal torque measurements and histomorphometry. Implants 3.0, 3.75, 5.0, and 6.0 mm in diameter and 6.0 mm long were inserted through one cortical layer in the tibial metaphyses of nine rabbits and allowed to heal for 12 weeks. The implants were then unscrewed with a torque gauge, and the peak torque required to shear off the implants was recorded. The histologic analysis in undemineralized ground sections comprised (1) a gross description of the implant sites and assessments of (2) the total implant length in bone and (3) in the cortical passage, as well as (4) the thickness of the cortical bone adjacent to the implants. From the removal torque values obtained and morphometric measurements, a mean shear stress value was calculated for each implant type. The biomechanical tests showed a statistically significant increase of removal torque with increasing implant diameter. The resistance to shear seemed to be determined by the implant surface in supportive cortical bone, whereas the newly formed bone at the periosteal and endosteal surfaces did not seem to have any supportive properties after 12 weeks. It is suggested that wide diameter implants may be used clinically to increase implant stability.
The clinical result of screw implants (Nobel Biocare) has been well documented for treatment of complete and partial edentulism~,3,11,13 2~ 22 26 so s9 Despite the documented high predictability and successes, complications and failures have been reported 11,13,2~ Poor bone quality and extreme jawbone resorption have been pointed out as risk factors. Methods to improve bone volume have consequently been presented, such as different grafting and augmentation procedures 5'4a. In situations of poor bone quality, prolonged unloaded healing periods have been suggested 14, an approach which has also been sup-
ported by experimental findings presented by JOHANSSON ~r ALBREKTSSON23, From the latter study performed in the rabbit tibia, the authors reported a gradtiaIly increasing implant removal torque and bone-to-implant contact within a 12-month period. The stability and resistance to shear forces of implants also seem to be dependent on the biomechanical properties of the bone in the implant interface, as reported by STEINEMANN et al. 37 and Ru~o D~ I~ZENDE • JOHANSSON34, while cortical bone seems to provide better stability and long-term integration than cancellous bone 15A8~36'43.
Key words: endosteal implants; screw implants; removal torque; diameter; biomechanic; bone support. Accepted for publication 28 August 1996
The surgical technique used may also be a source of error, negatively influencing the integration process. Overheating the bone and intraoperative implant mobility have been reported as
Table 1. Measurements (mm) of implants used in study Outer diameter 3.0 3.75 5.0 6.0
Inner diameter
Pitch height
2.46 3.15 4.13 4.92
0.5 0.6 0.8 1.0
1 42
I v a n o f f et al.
Fig. 1. Four different types of implants used
in study.
5.00 mm O 3~75mm O
~
6.00 mm O 3.00 mm O
Tibiadx Tibiasin Fig. 2. Diagram of implantation sites for four types of implants in tibial metaphyses. O=diameter.
two of the most frequent surgical errors 12,28. An intermittent drilling technique under profuse irrigation with
sterile saline has been advocated to avoid overheating the bone 1. Intraoperative implant mobility may be prevented by using a precise drilling procedure 1,a,~9. Another option investigated by several authors is to modify implant design, on both the micro- and macrolevel. A rough surface topography has been reported to increase osseointegration as compared to a smooth surface 7'16'31'4~ Hydroxyaparite-coated implants have been shown to result in faster bone integration in the short term, whereas they did not seem to influence the integration process positively in the long run 17. Screw implants seem to provide better initial stability and resistance to compression and tension stress forces 4 than cylindric implants6,! 6. In implant insertion, initial stability is of paramount importance ~. This can be accomplished either by engaging the mandibular lower cortex when operating between the mental foramina, or by engaging the inferior cortical plates of the maxillary sinuses and/or nasal cavities. In jaws not suitable for standard implants, and/or in situations where
primary stability of the fixture is not achievable, the use of a wider implant diameter has been suggested 27. Increased bone-to-implant contact and engagement of as much cortical bone as possible may thereby be obtained. This notion is supported by clinical data in that higher success rates are reported for wider implants (4.0 mm) in soft bone29, 39. No experimental studies have been reported, in which the results of different implant diameters were specifically assessed for their capacity to improve bone integration. Therefore, the objective of the present investigation was to evaluate the influence of different implant diameters on integration of titanium implants in the rabbit tibia. Material and methods Animals and anesthesia
Nine adult New Zealand White rabbits of both sexes, weighing 5-6 kg, were used in the study. The animals were kept in standard cages and fed ad libitum on conventional laboratory diet. The experiment was approve d by the animal ethics committee at the University of G6teborg, Sweden. During the surgical sessions,
Fig. 3. Contact radiographs of four different implant diameters taken after retrieval. Note that implants originally were inserted through one cortical layer only. a) 3.0 ram, b) 3.75 mm, c) 5.0 ram, d) 6.0 mm.
Implant integration 100"
80-
gg 6ot "~
9
40-
o
8 8
20, .
.
.
.
,
Implant diameter (mm) Fig. 4. Results from removal torque measurements. All values (n 9) are given for each implant diameter. Curve fit to all plotted data of second degree shows r value of 0.80.
the animals were anesthetized with intramuscular injections of fentanyl and fluanison (Hypnorm | Janssen, Brussels, Belgium) at a dose of 0.7 mg/kg body weight, and intraperitoneai injections of diazepam (Stesolid| Dumex, Copenhagen, Denmark) at a dose of 1.5 mg/kg body weight. Approximately 1 ml of local anesthesia (2% lidocaine with epinephrine
143
tively, were used in the investigation (Fig. 1). Details regarding outer diameter (as measured from the most prominent part of the external thread), inner diameter (as measured from the deepest part of the external thread), and pitch height are presented in Table 1. All implants were custom-made of c.p. titanium (Nobel Biocare AB, G6teborg, Sweden). They were designed with a standard hexagonal head, which was fitted to an ordinary Br~nemark System | fixture mount. In contrast to the 3.0- and 3.75-mm implants, the 5.0- and 6.0-mm diameter ones were not shaped with a shoulder within their neck portion (Fig. 1) but were, therefore, threaded all the way up to their top margin. All implants were cleaned ultrasonically in butanol and absolute ethanol for 10 min. in each solution and sterilized according to a routine protocol (Nobel Biocare AB, G6teborg, Sweden). Implant surgery and research protocol
12.5/zg/ml; Xylocain/Adrenalin | Astra AB, SOdert~tlje, Sweden) was also injected into the areas exposed to surgery. Implants
Implants (fixtures) 6.0 mm long and either 3.0, 3.75, 5.0, or 6.0 mm in diameter, respec-
4
All surgery was performed under sterile conditions in an animal operation theater. Both tibial metaphyses were chosen as experimental sites and were exposed by curved skin incisions via fascial periosteal flaps. In each tibia, preparations for two implant sites (n= 36) were made, by the technique described by ADELL et al. 1. A guide drill was first used to mark the implant site locations approximately 10 m m apart. The sites were then sequentially enlarged with twist drills as described below, except for the 3.75-mm sites, where a Br~memark System | pilot drill was used as well, by the standard technique: Implant diameters: 3.0, 3.75, 5.0, 6.0 mm; twist drill diameters: 2.0 followed by 2.4 mrri, 2.0 followed by 3.0 ram, 2.0, 3.0 followed by 4.3 mm, 2.0, 3.0 followed by 5.3 mm. Countersinking was omitted in all sites, which were threaded with conventional taps suiting the various implant diameters. The implant sites were divided into four groups, according to the type of implant inserted in the cranial cortical layer (Fig. 2):
Group A. right proximal tibia: 5.0 mm (n= 9)
Group B. right distal tibia: 3.75 mm @=9) Group C. left proximal tibia: 6.0 mm (n=
9
9)
4
Group D. left distal tibia: 3.0 mm @=9)
,,4 Fig. 5. Light microscopy of ground section
Fig. 6. Light microscopy of ground section of
of 6.0-mm-wide implant that could not be unscrewed with torque gauge manometer. Most of implant surface is in contact with surromlding bone. Newly formed bone on periosteal (P) and endosteal (E) surfaces can be distinguished from cortical bone (C). Bridge of bone (*) between medial cortex and implant surface is seen. B M = b o n e marrow, I=implant. Staining 1% toluidine blue/pyronin-G•
3.0-mm-wide implant. Morphology indicates separation (*) between implant surface and surrounding bone due to removal torque test. Crack can be seen in bone radiating from third thread (arrow). Newly formed periosteal (P) and endosteal bone (E) can be distinguished from cortical bone. Endosteal bone has been separated from original cortical bone (open arrow). I=implant. Staining 1% toluidine blue/pyronin-G•
Consequently, a total of 36 implants were inserted in one cortical layer (Fig. 3). To avoid bony overgrowth of the fixture heads, the 5.0- and 6.0-mm implants (lacking shoulders) were inserted so that approximately ~ 3 threads were exposed above the marginal bone level. On the other hand, the 3.0- and 3.75-mm implants, were inserted so that only the inferior conical part of the fixture head was in contact with the outer cortex of the tibial bone. The surgical wounds were closed in layers with single resorbable subcutaneous sutures and silk sutures in the skin. Postoperatively, all
1 44
l v a n o f f et a/. Table 2. Torque measurenrents together with shear-stress calculations (mean+_SD) Implants Parameters
3.0
Torque (N cm)
Fig. 7. Light microscopy of ground section showing fracture line (arrows) within bone in cortical passage radiating from thread from 3.75-ram-wide implant. Crack is filled with blood cells (*). I=implant. Staining 1% toluidine blue/pyronin-G• 25.
13.7+6.2
3.75 24.8•
6.0
Statistics *
62.2-+20.3
All significantly different (P-<0.05) *P-<0.05 vs 3.0 and 3.75
Shear stress (N/mm2), total bone
3.1-+1.4
3.3-+1.7
1.8• *
1.9-+0.6 *
Shear stress (N/ram2), cortical passage
5.0_+2.3
5.4•
4.5-+1.4
4.9+1.6 NS
t Wilcoxon's signed rank test.
the aforementioned surgical areas were re-exposed. In some animals, a bony overgrowth of the fixture head/shoulder was seen. In order not to bias the removal torque results, this bone was carefully removed with a scalpel. An electronic torque meter (Digital Torque Measurement, DTF-1, Crane Electronics, UK) was connected to all implants via a standard Brfinemark System | fixture mount, and assessment of the peak torque (N cm), required to loosen the implants, was performed. A mean torque value was calculated for each research group.
Specimen retrieval
Fig. 8. Light microscopy of ground section of 3.75-mm-wide implant. Implant has been separated from bone surrounding head of implant (*), but threads seem still to be in contact with bone. Numerous cracks (arrows) show in cortical bone, and newly formed endosteal bone (E) has been separated from original cortical bone (C). Periosteal (P) bone formation, which can be distinguished from original cortical bone, is evident. I=implant, BM=bone marrow. Staining 1% toluidine blue/pyronin-GX6.3.
5.0 32.4~-9.9
The animals were killed by an intravenous overdose of pentobarbital (Mebumal | ACO Lfikemedel, Solna, Sweden), given via an ear vein. The implants, together with surrounding bone and soft tissues, were removed en bloc and fixed by immersion in 4% buffered formalin solution. The obtained specimens were embedded in acrylic resin, and cut and ground in an Exakt sawing machine (Exakt Apparatebau, Nodersted, Germany) to about 10-/~m-thick sections 1~ The specimens were finally stained with 1% toluidine blue/ pyronin-G.
Histologic analyses The obtained sections were histologically analyzed in a Leitz microscope equipped with a Microvid | morphometric system connected to a PC. The equipment permitted direct quantitative calculations in the ground sections obtained a4. The histologic assessments involved (1) gross morphologic description of the bone, tissues surrounding the implants, (2) measurements of the total implant length in bone, and (3) the implant length in the cortical passage, and (4) measurement of the thickness of the cortical bone adjacent to the implant. A mean value for the parameters was calculated for each implant, as well as the mean for the four research groups. Because the implants had been fractured loose in order to measure the removal torque levels, the true degree of bone-to-implant contact could not be measured.
Calculations A mean shear stress value (N/mm 2) was calculated by the formula: T hzdsr, where T is the removal torque (N mm), d is the mean diameter of the implant (ram), r is the mean
Total implant length in bone 108"~
6"
///////A
*
animals were given antibiotics (Intencillin | 2250000 IE/5 ml; 0.1 ml/kg body weight; LEO, Helsingborg, Sweden) and analgetics (Temgesic| 0.05 mg/kg body weight; Reckitt and Coleman, USA) as single intramuscular injections. A postoperative healing period of 12 weeks was allowed for all rabbits.
Torque measurements At the end of the healing period, the animals were anesthetized as previously described, and
4-
0
9
3.0
I
"
3.75
l
I
5.0
6.0
Implant diameter (mm) Fig. 9. Results from morphometric measurements of total implant length in bone (mm• Statistically significant difference was shown between all groups except 3.0 and 3.75 mm.
Implant integration
145
Table 3. Morphometric measurements and calculations (mean_+SD) Implants Parameter
3.0
3.75
5.0
6.0
Statistics
Cortical thickness (mm)
1.46_+0.18
1.51_+0.29 1.22-+0.20*
1.32-+0.11 *
Implant length in bone (ram)
3.83_+0.79 *
4.00+0.48
7.46+3.08
* NS vs 3.75 mm
Implant length in cortical passage (mm)
2.37-+0.37 2.42_+0.34 2.17_+0.45 *
2.69-+0.29
* P<0.05 vs all groups
5.65+1.78
* P-<0.05 vs 3.0 and 3.75 mm
Wilcoxon's signed rank test.
length of the lever arm (mm), and s is the length of the implant in bone (mm). Two s values were used in the formula: (1) the total implant length in bone, and (2) the implant length in the cortical passage, respectively.
level for plastic deformation of their hexagonal head, which is estimated to be 90 N cm 25, and were consequently not loosened. These sites were ascribed the value of 90 N cm. A curve fit of the second degree showed an r value of 0.80.
Statistics
Wilcoxon's signed rank test was used for statistical analyses. Results Clinical findings
The animals recovered well after surgery, and the healing of the surgical sites was uneventful. All 36 fixtures were found stable at re-entry when tested with the torque manometer. Removal torque values
The mean removal torque values for the four groups were 13.7_+6.2, 24.8_+15.1, 32.4_+9.9, and 62.2+20.3 N cm for the 3.0-, 3.75-, 5.0-, and 6.0-mm-wide implants, respectively. There was a statistically significant difference between all groups (Table 2 and Fig. 4). Two implants in group C exceeded the critical
Histologic findings
The histologic examination revealed that all implants had initially engaged one cortical layer as desired. However, endosteal bone formation, which seemed to be more prominent for the 5.0- and 6.0-mm implants, had resulted in encapsulation of the threads in the bone marrow cavity to various degrees (Fig. 5). This bone had a lamellar appearance but could still be distinguished from the original cortical bone. Because the implants had been loosened, a separation could be observed in the bone-implant interface (Fig. 6), and fracture lines radiating from the implant thread peaks were observed (Fig. 7). Fracture lines were most often present between the original cortical bone and the newly formed endosteal bone, which had been ripped off and separated from the orig-
I m p l a n t l e n g t h in the cortical passage 10"
a=
6-
inal cortical bone (Fig. 8). Bone formation had also occurred at the periosteal surface to various extents (Figs. 5, 6 and 8). The mean values for the measured total implant length in bone were 3.83-+0.79, 4.00-+0.48, 5.65-+1.78, and 7.46_+3.08 m m for the 3.0-, 3.75-, 5.0-, and 6.0-mm implants, respectively (Table 3 and Fig. 9). The corresponding values for the implant length in the cortical passage were 2.37-+0.37, 2.42-+0.34, 2.17_+0.45, and 2.69_+0.29 mm, respectively (Table 3 and Fig. 10). The thickness of the cortical bone adjacent to the implants was found to be 1.46_+0.18, 1.51_+0.29, 1.22_+0.20, and 1.32_+0.11 m m for the 3.0-, 3.75-, 5.0-, and 6.0-mm implant sites (Table 3 and Fig. 11). There was a statistically significant difference between the distal sites (3.0 and 3.75 mm) and the proximal sites (5.0 and 6.0 ram). Calculations
F r o m the total implant length in bone, the mean shear stresses were calculated to be 3.1_+1.4, 3.3_+1.7, 1.8_+0.6, and 1.9_+0.6 N / m m 2 for groups D, B, A, and C, respectively (Table 2 and Fig. 12). The shear stress for the 5.0- and 6.0m m implants were significantly lower than for the others. The corresponding shear stresses, for the implant length in the cortical passage, were 5.0_+2.3, 5.4+2.9, 4.5_+1.4, and 4.9-+1.6 N / m m 2, respectively, with no significant differences between the different implants (Table 2 and Fig. 13).
4Discussion
9
3.0
I
I
3.75
!
5.0
"
!
6.0
Implant diameter (mm) Fig. 10. Results from morphometric measurements of implant length in cortical passage (mm-+SD). Implants of 5 mm had statistically significantly less implant length in cortical passage.
The present experimental study demonstrated an increased removal torque with larger implant diameters, a finding which should be expected from geometric effects. The larger radius gives a linear increase of the shear force lever arm as well as a linear increase of the surface area in contact with bone.
146
lvanofj'et al. Thickness of the cortical bone at the implant sites _
. ID
1" 9
9
I
I
3.0
I
3.75
5.0
9
6.0
Implant diameter (ram) Fig. ll. Results from morphometric measurements of thickness of cortical bone adjacent to implants (mm• Bone was statistically thinner at proximal sites where 5.0- and 6.0-mm implants had been inserted, than for 3.0- and 3.75-mmimplants.
Shear stresses / total implant length in bone
r
Z
111 5' 4'
(D
3-
r,t3
2-
10
a
3.0
!
3.75
I
5.0
6.0
Implant diameter (mm) Fig. 12. Results from calculations of shear stress based on total implant length in bone. Shear stress for 5.0- and 6.0-mm implants was statistically lower than for 3.0- and 3.75-mmimplants.
Therefore, the removal torque should be proportional to the square of the diameter, if the shear stress is constant. The second-degree fit to the removal torque vs. implant diameter (Fig. 4) indicates that these geometric effects were dominant (r=0.80). The calculated shear stress based on the total length of bone in contact with the implants (Table 2, Fig. 12) demonstrated a significantly lower level for the 5.0- and 6.0-ram implants, while the shear stress relative to the original cortical passage was constant without significant differences between the different implant diameters (Table 2, Fig. 13).
This seems to indicate that the newly formed bone around the implants did not contribute substantially to the shear force resistance, a supposition which is in accordance with SEYr,rERBY et al? 6, who concluded that the bone in the cortical passage gives the main support to the implants. It has been demonstrated 13,15,1s,56 that compact bone provides better stability and long-term integration of endosteal implants than cancellous bone. IVANOFFet al. ~8 found a correlation between removal torque and degree of bone contact for 3.75-mm implants penetrating one or two cortical layers.
IVANOFF et al./s calculated about twice as high shear stresses for 3.75-mm wide implants inserted in the rabbit tibia after 12 weeks; i.e., 8 10 N/mm 2 as compared to about 5 N/mm 2 in the present study. However, in that study, the calculations were based on the true bone-implant contact of unremoved implants, which was about 40 50% in the cortical passage. In the present study, the calculations were based on the implant length in the cortical passage. Assuming that the bone contact was also about 40-50% in the present study, the shear stresses would be around 10-12.5 N/mm 2, corresponding to the results of IVANOFF et al. ~8. RUBO DE REZENDE ~; JOHANSSON34 calculated shear stresses for 3.7-ram-wide titanium implants inserted in the rabbit tibia after 12 weeks of healing. On the basis of the implant length in the cortical bone, they found the mean shear stress to be 1.9 N/ram 2. However, a shear stress value of 14.8 N/ram 2 was calculated taking into account the bone implant contact measured on unremoved implants. Based on the results of the present study and earlier work, the shear stress in rabbit bone after a 12week healing seems to be around 10 15 N/mm 2 related to the true contact area in the cortical passage of the tibia. The histologic observations in the present study support the concept of "nonsupportive" bone, since the newly formed endosteal bone had been sheared off from the original cortex. The supportive properties of the newly formed bone are probably related to the degree of maturation and/or the shape of the bone. In the present study, the new bone formed a collar around the implant. This is why it probably was too weak to resist shear, in contrast to the bone in the cortical passage. It is possible that if a longer healing period had been allowed, the newly formed bone would also have supported the implant. According to RORERTSet al. 33, it takes 18 weeks for the rabbit bone around an implant to mature. In the present study, the newly formed bone appeared to be lamellar and well organized but could still be distinguished from the original bone after 12 weeks of healing due to its darker staining and different morphologic appearance. It is likely that a uniform orientation of collagen fibers and complete maturation of the newly formed bone are required for all the interfacial bone to support an implant. The removal torque test is a simple
Implant integration Shear stresses/cortical passage
54"
32s G'
I
3.0
I
3.75
I
5.0
6.0
Implant diameter (mm)
Fig. 13. Results from calculations of shear stress based on implant length in cortical passage. There were no statistical differences between implant diameters tested.
1 4"1
ZARB GA, ALBREKTSSONT, eds.: Tissueintegrated prostheses. Osscointegration in clinical dentistry. Chicago: Quintessence, 1985:211 32. 2. ADELL R, ERIKSSON B, LEKIIOLM U, BRXNEMARIrP-I, JEMTW.A long-term follow-up study of osseointegrated implants in the treatment of the totally edentulous jaw. Int J Oral Maxillofac Impl 1990: 5: 347 59. 3. ALBREKTSSONT, DAHLE, ERBOML, et al. Osscointegrated oral implants. A Swedish multicenter study of 8139 consecutively inserted Nobelpharma implants. J Periodontol 1988: 59:287 96. 4. BRANEMARK P-I, BREINE U, ADELL R, HANSSONBO, LINDSTROM J, OHLSSON Z~. intra-osseous anchorage of dental prostheses. I. Experimental studies. Scand J Plast Reconstr Surg 1969: 3:81 100. 5. BUSER D, DAHLIN C, SCHENK RK. Guided bone regeneration in implant dentistry. Chicago: Quintessence, 1994. 6. CARLSSON L, ROSTLUND T, ALBREKTSSON
method and has frequently been used to evaluate the tissue response to titanium and other materials in experimental7,9,16 18,23,24,32,34 37,40,41 and clinical studies 38'43. The test is a valuable method for evaluating the interracial strength of endosteal implants. As seen in the present study, the bone morphology differed when comparing the proximal with the distal part of the tibial metaphysis; ideally, test and control implants should have been inserted at identical sites. However, as also shown in the present study, if implant and bone geometric variations are taken into account, it seems possible to compare the integration of implants when they have different thread designs and/ or are placed in sites with different bone morphologies. In a clinical situation where poor bone density is present, engagement of the buccal and lingual cortical plates with wider implants is desirable in order to achieve good primary stability of the implants and promote integration 14. Therefore, wider implants could be useful in the molar region, especially above the inferior alveolar nerve, where bicortical anchorage cannot be accomplished 18. However, this recommendation should be limited to situations with sufficient buccolingual dimension. In fact, too large an implant may reduce the cortical support, especially around the neck of the implant, and consequently jeopardize the primary stability of the implant. With, p o o r bone density 29,39, clinical data indicate a higher success rate for wider implants (4.0 m m
in diameter) than for standard implants (3.75 m m in diameter). Promising clinical results have also been presented for implants 5.0 and 5.5 mm in diameter 27. However, the relationship between the clinical performance of an implant and the shear strength of the bone-implant interface is not known. The possible positive effect of wide-diameter implants has yet to be established in controlled clinical trials. In conclusion, the stability of screwshaped titanium implants, as measured with a removal torque test, seems to be determined by the implant surface area in supportive cortical bone and the lever arm relative to the axis of rotation. It is suggested that the use of wide-diameter implants may increase stability in the clinical situation because it involves more supportive cortical bone. Controlled prospective clinical trials are needed to study the influence of implant diameter, as related to the host-tissue anatomy.
Acknowledgements. This study was supported by grants from the Swedish Medical Research Council, the Swedish National Board for Industrial and Technical Development (NUTEK), the Sylvan Foundation, the Hjalmar Svensson Foundation, the Greta and Einar Asker Foundation, the Wilhelm and Martina Lundgren Foundation, the Swedish National Board of Health, the FRF Foundation, and Nobel Biocare AB. References 1. ADELL R, LEKHOLM U, BRANEMARKP-I.
Surgical procedure. In: BRANEMARKP-I,
B, ALBREKTSSONT, BR2,NEMARKP-I. Osseointegration of titanium implants. Acta Orthop Scand 1986: 57:285 9. 7. CARLSSONL, ROSTLUNDT, ALBREKTSSON B, ALBP,EKTSSON12 Removal torques for polished and rough titanium implants. Int J Oral Maxillofac Impl 1988: 3: 21-4. 8. CARLSSONL, R0STLUNDT, ALBREKTSSON B, ALBREKTSSONT. Implant fixation improved by close fit. Cylindrical implantbone interface studied in rabbits. Acta Orthop Scand 1988: 59: 272-5. 9. CARLSSONLV. On the development of a new concept for orthopaedic implant fixation. Thesis, University of G6teborg, Sweden, 1989. 10. DONATH K. Die Trenn-Dtinnschliff-Technik zur Herstellung histologischer Pr~iparate yon nicht schneidbaren Geweben und Materialen. Pr~iparator 1988: 34: 197 206. 11. ENGQUIST B, BERGENDAHLT, KALLUS T,
LINDEN U. A retrospective multicenter evaluation of osseointegrated implants supporting overdentures. Int J Oral Maxillofac Impl 1988: 2:129 34. 12. ERICSSONRA. Heat-induced bone tissue injury. An in vivo investigation of heat tolerance of bone tissue and temperature rise in the drilling of cortical bone. Thesis, University of G6teborg, Sweden, 1984. 13. FRIBERG B, JEMT T, LEKHOLM U. Early failures in 464l consecutively placed BrLnemark dental implants. A study from stage I surgery to the connection of completed prostheses. Int J Oral Maxillofac Impl 1991: 6: 142-6. 14. FRIBERGB. Bone quality evaluation during implant placement. Academic dissertation for the degree of master of dental science. University of G6teborg, G6teborg, Sweden, 1994. 15. FRmER6 B, SENNERBYL, ROOS J, LEKHOLMU. Identification of bone quality in conjunction with insertion of titanium im-
148
IvanoJfet al.
plants. A pilot study in jaw autopsy specimens. Clin Oral Impl Res 1995: 6: 213-19. 16. GOTFREDSEN K, NIMB L, HJORT1NGHANSEN E, JENSENJS, HOLM~NA. Histomorphometric and removal torque analysis for TiO2-blasted titanium implants. An experimental study on dogs. Clin Oral Impl Res 1992: 3: 77-84. 17. GOTTLANDERM. On hard tissue reactions to hydroxyapatite-coated titanium implants. Thesis, University of G6teborg, Sweden, 1994. 18. IVANOFFC-J, SENNERBYL, LEKHOLMg. Influence of mono-and bicortical anchorage on the integration of titanium implants. A study in the rabbit tibia. Int J Oral Maxillofac Surg 1996: 25: 229-35. 19. IVANOFFC-J, SENNERBYL, LEKHOLM U. Influence of initial implant mobility on the integration of titanium implants. An experimental study in rabbits. Clin Oral Impl Res 1996: 7:120 7. 20. JAFFIN RA, BERMAN CL. The excessive loss of Br~nemark fixtures in type IV bone. A 5-year analysis. J Periodontol 1991: 62: 2M. 21. JEMT T, LEKHOLM U, ADELL R. Osseointegrated implants in the treatment of partially edentulous patients: a preliminary study of 876 consecutively installed fixtures. Int J Oral Maxillofac Impl 1989: 4:211 7. 22. JEMT T, LEKHOLMU. Implant treatment in edentulous maxillae: a 5-year followup report on patients with different degrees of jaw resorption. Int J Oral Maxillofac Impl 1995: 10:303 11. 23. JOHANSSON C, ALBREKTSSON T. Integration of screw implants in the rabbit. A 1-year follow-up of removal torque of titanium implants. Int J Oral Maxillofac Impl 1987: 2:69 75. 24. JOHANSSON C. On tissue reactions to metal implants. Thesis, University of G6teborg, GOteborg, Sweden, 1991. 25. JORNI~USL. Personal communication. Nobel Biocare AB, G6teborg, Sweden, 1996. 26. LANEY WR, JEMT T, HARRIS D. et al. Osseointegrated implants for single-tooth replacement: progress report from a multicenter prospective study after 3
years. Int J Oral Maxillofac lmpl 1994: 9: 49-54. 27. LANGERB, LANGERL, HERRMANN1, JORN/~USL. The wide fixture. A solution for special bone situations and a rescue for the compromised implant. I. Int J Oral Maxillofac Impl 1993: 8:400 8. 28. LEKHOLMU, ADELL R, BR.I.NEMARKP-I. Complications. In: BRA.NEMARK P-I, ZARB GA, AEBREKTSSONT, eds.: Tissueintegrated prostheses. Osseointegration in clinical dentistry. Chicago: Quintessence 1985: 233M0. 29. LEKHOLM U. The Brgmemark implant technique. A standardized procedure under continuous development. In: LANEYWR, TOLMANDE, eds.: Tissue integration in oral, orthopedic and maxillofacial reconstruction. Chicago: Quintessence, 1992: 194-9. 30. LEKHOLMU, VAN STEENBERGHED, HERRMAN I, et al. Osseointegrated implants in the treatment of partially edentulous jaws: a prospective 5-year multicenter study. Int J Oral Maxillofac Impl 1994: 9: 627-35. 31. MANIATOPOULOSC, PILLIAR RM, SMITH DC. Threaded versus porous-surfaced designs for implant stabilization in boneendodontic implant model. J Biomed Mater Res 1986: 20:1309 33. 32. MORBERG P. On bone tissue reactions to acrylic cement. Thesis, University of G6teborg, Sweden, 1991. 33. ROBERTS WE, GARETTO LW, BREZNIAK N. Bone physiology and metabolism. In: MISCH CE, ed.: Contemporary implant dentistry. St. Louis, MO: Mosby Year Book, 1993:327 53. 34. RUBO DE REZENDE ML, JOHANSSONCB. Quantitative bone tissue response to commercially pure titanium implants. J Mater Sci: Mater Med, 1993: 4:233 9. 35. SEWNERBYL. On bone tissue response to titanium implants. Thesis, University of G6teborg, Sweden, 1991. 36. SENNERBYL, THOMSENP, ERICSSONL. A morphometric and biomechanic comparison of titanium implants inserted in rabbit cortical and cancellous bone. Int J Oral Maxillofac Impl 1992: 7: 62-71.
37. STEINEMANN SG, EULENBEROER J, MAEUSLI P-A, SCHROEDERA. Adhesion of bone to titanium. In: CHRISTEL P, MEUNIER A, LEE AJC, eds.: Biological and biomechanical performance of biomaterials. Amsterdam: Elsevier, 1986: 409-14. 38. TJELLSTROMA, JACOBSSONM, ALBREKTSSON"12Removal torque of osseointegrated craniofacial implants. A clinical study. Int J Oral Maxillofac Impl 1988: 3:287 9. 39. VAN STEENBERGHE D, LEKHOLM U, BOLENDER C, et al. The applicability of osseointegrated oral implants in the rehabilitation of partial edentulism: a prospective multicenter study on 558 fixtures. Int J Oral Maxillofac Impl 1990: 5: 272-81. 40. WENNERBERG A, ALBREKTSSON T, ANDERSSONB, KROLJJ. A histomorphometric and removal torque study of screwshaped titanium implants with three different surface topographies. Clin Oral Impl Res 1995: 6: 24-30. 41. WENNERBERG A. On surface roughness and implant incorporation. Thesis, University of G6teborg, Sweden, 1996. 42. WORTHINGTON P, BRANEMARKP-I. Advanced osseointegration surgery. Applications in the maxillofacial region. Chicago: Quintessence, i992. 43. YAMANAKAE, TJELLSTROMA, JACOBSSON M, ALBREKTSSON T. Long-term observations on removal torque of directly bone-anchored implants in man. In: YANAGIHARA N, SUZUKI J, eds.: Transplants and implants in otology. II. Proceedings of the Second International Symposium of Transplants and Implants in Otology. Matsyama, Ehime, Japan, 1-5 April 1991. Amsterdam: Kugler 1992.
Address:
Dr Carl-Johan Ivanoff Department of Oral and Maxillofacial Surgery Mdlndal Hospital S-431 80 M6lndal Sweden