SciELO - Scientific Electronic Library Online

vol.54 issue6Use of corticosteroids and the outcome of infants with bronchopulmonary dysplasiaBowel preparation for colonoscopy: comparison of mannitol and sodium phosphate. Results of a prospective randomized study author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista do Hospital das Clínicas

On-line version ISSN 1678-9903

Rev. Hosp. Clin. vol.54 n.6 São Paulo Nov./Dec. 1999 




Raul Bolliger Neto, João D. M. B. Alvarenga Rossi and Tomaz Puga Leivas



BOLLIGER Neto, R. - Experimental determination of bone cortex holding power of orthopedic screw. Rev. Hosp. Clín. Fac. Med. S. Paulo 54 (6): 181-186, 1999.


SUMMARY: Cylindrical specimens of bone measuring 15 mm in diameter were obtained from the lateral cortical layer of 10 pairs of femurs and tibias. A central hole 3.2 mm in diameter was drilled in each specimen. The hole was tapped, and a 4.5 mm cortical bone screw was inserted from the outer surface. The montage was submitted to push-out testing up to a complete strip of the bone threads. The cortical thickness and rupture load were measured, and the shear stress was calculated. The results were grouped according to the bone segment from which the specimen was obtained. The results showed that bone cortex screw holding power is dependent on the bone site. Additionally, the diaphyseal cortical bone tissue is both quantitatively and qualitatively more resistant to screw extraction than the metaphyseal tissue.


DESCRIPTORS: Bone and bones. Biomechanics. Bone screw. Tension resistance.



Several traditional methods for internal osteosynthesis use screws or plates to fix bone fragments. With the development of new surgical techniques and new implant materials, various investigations have been performed to improve different osteosynthesis methods. The effects of screw size and implant techniques of orthopedic screws in cortical bones on the shearing resistance have been investigated for this purpose.

The current literature reports that required screw extraction strength is linear to bone thickness.3 Extraction strength is dependent on implant location site,2 and it is remarkably dependent on the screw diameter.6 In vivo studies have shown that extraction resistance increases up to 1.5 to 1.9 times at the 6th post-implant week, then it decreases to 1.2 to 1.6 times at the 12th post-implant week.6 This reduction is thought to be caused by movement-induced bone resorption around the screw7. Uhthoff has shown in histology studies that the screw-to-bone surface contact is greater with self-tapping screws as opposed to non-self-tapping screws8. Despite this difference, the bone holding power for these different screws was not statistically significant3,9.

Reports in the literature recommend the use of wide screws4 with a core diameter of at least 3 mm1.

Investigations of screw holding power should control for the following variables: the bone into which the screw is implanted, the variability of the bone tissue at the implant location, and bone mechanical properties.

The objective of this study was to analyze the variations in the mechanical resistance of the metaphyseal and diaphyseal cortical bone tissue in the femur and tibia to push-out forces on an orthopedic screw.



Cylindrical cortical bone samples (15 mm diameter) were obtained from the lateral surface of 10 pairs of femurs and tibias. The bone samples were cut so their longitudinal axis was perpendicular to the bone surface. The bone samples were sequentially carved from the distal femur metaphysis and from the proximal tibia metaphysis towards the diaphysis of both bones. A distance of 25 mm was set between two consecutive bone samples. Tibia and femur bicondylar diameters were recorded, and the distance from the longitudinal axis to the distal femur articular surface and proximal tibia articular surface were also recorded (Fig. 1).



A bench drilling machine was used to drill a longitudinally-centered 3.2 mm diameter hole in each bone sample. The cortical bone thickness at the level of the hole was measured with a micrometer. The internal surface of the hole was fitted with a 4.5 mm diameter tap. A cortical spherical-head AO kind screw (Arbeitsgemeinschaft für den Ostheosynthesenfragen) with an internal six-faced slot, with external and core diameter of 4.5 mm and 3.0 mm, respectively, and having a 1.75 mm long thread and screw length of 35 mm was employed. The screw head was on the external cortical surface.

This testing unit was inserted into a metallic tube with a slightly conical internal surface (tube one) with the screw head facing down. Testing unit placement inside tube one served to prevent cortical layer rupture during push-out testing. A second tube (tube two) was placed upwards inside tube one to support the testing unit from the external surface, leaving a free passage for the screw. A third tube (tube three) was placed from the top, surrounding the screw end. This entire montage was centered under the loading anvil of a Kratos 5002 model machine with a load CCT of 10 TF (Fig. 2). A push-out load at 20 mm/minute was applied on the screw end. A 790 BBC Goerz Metrawatt Servogor graphic recorder plotted the maximal shearing resistance or maximal cortical layer stripping force.



Shearing tension was calculated for each testing unit using the following formula:

Shearing tension = rupture load/ [(thickness) x (screw external diameter) x p].

The testing units were positioned taking into account the distances from the articular surface to their centres. In order to overcome size differences among the bones, relative distances, calculated using the expression below, were used.

Relative distance = [(real distance) x (average bicondylar diameter)]/(bone bicondylar diameter)]

(obs. average bicondylar diameter for femora or for tibiae separately).

Each bone was divided in five distinct 5 cm segments corrected for the articular surface. Each segment was numbered (1 to 5) from the first segment that was marked at 5 cm from the articular surface. The epiphysial segments were not included (Table 1).



The testing units were grouped according to the bone of origin and according to the relative distance from the articular surface.

Cortical layer thickness, rupture load (extraction force for maximal shearing resistance) and shearing tension from the left and right femur and tibia segments were analyzed.

Student's t test and Tukey's variance analysis test were employed. A significance level of 5% (p<0.05) was chosen for all tests.



Average thickness, rupture load, and shearing tension and respective standard deviation values are shown in Tables 2, 3, and 4 as well as Figs 3, 4, 5, and 6.















Student's t test analysis for the three parameters for the respective segments on both sides was significant only for shearing tension in segment one on the left and right femur (p<0.03).



The results of this investigation are strictly concerned with the immediate post-operative period when the stability of the montage depends exclusively on mechanical factors. Schatzker et al6 reported an increase in mechanical resistance to extraction of the screw-bone system in the weeks following implantation.

The application of a relative distance as an indicator of the position of the bone sample was necessary to permit comparison between bones of different lengths. The arbitrary division of the bones included in the analysis into five segments numbered from the metaphysis to the diaphysis permitted the statistical comparison between different bone segments with an adequate number of samples in each segment.

The decision to select the lateral surface of the bone cortex as a source for the bone samples was made because this surface is usually chosen for the site of osteosynthesis procedures with plates and screws.

More specifically, the lateral surface of the femur has an easier surgical access, and the lateral surface of the tibia has adequate muscle coverage.

The variation recorded for the cortical thickness, rupture load, and shearing tension between the segments was statistically significant. It shows that there is a progressive increase in the measures from segment 1 through segment 3. This increase tends to attenuate or stabilize for segments 3 through 5 (Tables 1, 2, and 3 and Figs 3, 4, 5, and 6).

The right- and left-side segments did not show a statistically significant difference for the measures analyzed, except for the shearing tension on segment 1 of the femur. This difference is possibly due to a minor technical problem in this segment where the thickness of the cortical layer is on the average smaller than two steps of the screw thread.

These data compares with other reports in the literature (Table 5). The values reported by Lyon et al.4 (108 to 158 kgf) and Schatzker et al.6 (218 kgf) regarding the rupture load for the femur fall between the values obtained in this investigation (65 to 420 kgf). However, the data of Schatzker et al.6 were obtained in two cortical layers.



Regarding the tibia rupture load, the values obtained in this investigation (42-172 kgf) encompass the values yielded by Lyon et al.4 (108 to 170 kgf) measured in single bone cortex.

The femur shearing force range determined in this investigation (1.5 to 4.0 kgf/mm2) is greater than that reported by Lyon et al.4 (2.5 to 3.5 kgf/mm2). Both investigations were performed with a single bone cortex.

Schatzker et al.6 and Vangsness et al.8 recorded 6.17 kgf/mm2 and 4.68kgf/mm2 respectively for two cortical layers. Yet, Koraniy et al3 obtained 2.82 kgf/mm2 for femur shearing tension. The authors recorded 1.2 to 2.7 kgf/mm2 for tibia shearing, which was lower than the data reported by Lyon et al4 for a single bone cortex (2.60 to 3.90 kgf/mm2).

Hughes and Jordan2 have demonstrated that the shearing tension does not depend on the pilot hole as long as the hole diameter is smaller than the external screw diameter. Shearing tension does not depend on the amount of material between the screw threads either. Shearing tension reflects the intrinsic quality of the material holding the inserted screw.

We showed both an increase in cortex thickness and an increase in resistance to shearing, moving from the metaphysis to the diaphysis.

Regarding weight bearing, the thinness and reduced resistance presented by the cortical metaphyseal bone is counterbalanced by greater bone diameter in this location. The greater diameter of the metaphysis is responsible for a larger inertial momentum of the bone. The spongy nature of the epiphyseal and metaphyseal bone segments indicates that weight bearing is distributed in the local trabeculae. As the weight bearing is transferred from the metaphysis to the diaphysis, weight bearing becomes more dependent on the cortical layer, which, in turn, becomes thicker and more resistant. These functional considerations are in keeping with the finding that the closer to the diaphysis, the thicker and more resistant the cortical layer becomes.



The cortical layer thickness, rupture load, and shearing tension all progressively increased from the metaphysis to the diaphysis. This progressively increasing pattern corresponds to quantitative and qualitative changes in bone diameter, cortex thickness, and character of the local trabeculae along the femur and tibia.





BOLLIGER Neto R e col. ¾ Determinação experimental da retenção do parafuso ortopédico em cortex ósseo. Rev. Hosp. Clín. Fac. Med. S. Paulo 54 (6):181-186, 1999.


Foram retirados corpos de prova cilíndricos de 15 mm de diâmetro da camada cortical da face lateral de 10 pares de fêmures e tíbias oriundos de peças anatômicas. No centro de cada um destes foi feito um orifício de 3,2 mm e nele inserido, a partir da superfície externa, um parafuso cortical de 4,5 mm de diâmetro após rosqueamento com macho. Este conjunto foi submetido a um ensaio mecânico no qual o parafuso foi extraído. A espessura da camada cortical e a carga de ruptura foram medidas e a tensão de cisalhamento foi calculada. Os resultados foram agrupados de acordo com o segmento do osso de onde provinham. Os ensaios mostraram que a retenção cortical do parafuso varia ao longo do osso e que o tecido ósseo da córtex diafisária é tanto quantitativamente quanto qualitativamente mais resistente à extração do parafuso que o tecido metafisário.


DESCRITORES: Osso e ossos. Biomecânica. Parafusos ósseos. Resistência à tração.



1. ANSEL RH & SCALES JT - A study of some factors which affect the strength of screws and their insertion and holding power in bone. J Biomech 1968; 1: 279-302.         [ Links ]

2. HUGHES AW & JORDAN BA - The mechanical properties of surgical bone screws and some aspects of insertions practice. Injury 1972; 4: 25-38.         [ Links ]

3. KORANYI E, BOWMAN CE, KNECHT CD & JANSSEN M - The holding power of orthopedic screws in bone. Clin Orthop 1970; 72: 283-286.         [ Links ]

4. LYON WF, COCHRAN Jr & SMITH L - Actual holding power of various screws in bone. An Surg 1941; 114-367.         [ Links ]

5. MÜLLER ME, ALLGÖWER M, SCHNEIDER R et al. - Manual de Osteosíntesis: Técnica A.O. Berlin, Springer, 1980.         [ Links ]

6. SCHATZKER J, SANDERSON R & MURNAGHAM JP - The holding power of orthopedic screws in bone. Clin Orthop 1975; 108: 115-126.         [ Links ]

7. SCHATZKER J, HORNE JG & SUMNER-SMITH G - The effect of movement in the holding power of screws in bone. Clin Orthop 1975; 111: 257-263.         [ Links ]

8. UHTHOFF HK - Mechanical factors influencing the holding power of screws in compact bone. J Bone Joint Surg 1973; 55B: 633-639.         [ Links ]

9. VANGSNESS Jr CT, CARTER DR & FRANKEL V - In vitro evaluation of the loosening characteristics of self-tapped and non self-tapped cortical bone screws. Clin Orthop 1981; 157: 279-286.         [ Links ]



Received for publication on the 07/12/98

From the Institutes of Orthopedics and Traumatology, University of São Paulo School of Medicine, São Paulo ¾ Brazil.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License