Partially threaded headless screw may benefit adequate interfragmentary compression and reduced driving torque for small bone fixation

Abstract

Headless compression screws (HCSs) are commonly used to fixate small bones and articular fractures. Understanding the biomechanical efficacy of different HCS designs can help surgeons make proper interfragmentary compression when a specific implant is chosen. HCSs with three different central shaft designs (unthreaded, fully threaded, and partially threaded) were studied: the Herbert–Whipple, Mini-Acutrak 2, and headless reduction (HLR). Polyurethane foam blocks were machined with a simulated fracture gap of 0.5 mm and set onto a custom-made jig to simultaneously measure compression force and driving torque during screw insertion. The maximal achievable compression forces and driving torques recorded were 47.4 ± 0.9 N and 145.11 ± 1.65 N mm for the HLR, 50.98 ± 1.29 N and 152.62 ± 2.83 N mm for the Mini-Acutrak 2, and 19.33 ± 1.0 N and 33.4 ± 2.2 N mm for the Herbert–Whipple. Overall, the compression force of the Mini-Acutrak 2 and HLR increased with the torque. Unlike the other screws, the Herbert–Whipple’s driving torque increased while the compression force decreased after peak compression force was achieved. The partially threaded shaft design (HLR) demonstrated equivalent biomechanical advantage with the Mini-Acutrak 2 in interfragmentary compression. The HCSs with cone-shaped proximal ends (HLR and Mini-Acutrak 2) maintained their compression force during over-fastening, whereas the unthreaded central shaft of the Herbert–Whipple screw caused it to lose compression force.

Introduction

Headless compression screws (HCSs) have been used to successfully treat scaphoid, radial head, and capitellum fractures and osteotomies of the tarsal bones. With the evolution of the HCSs, several types of implant are now available. Contemporary HCSs can be divided according to shaft style into unthreaded, fully threaded, or partially threaded shaft designs. The first HCS, the Herbert compression screw (Zimmer, Warsaw, IN, USA) was introduced in the early 1980s ¹ and consists of a threadless central shaft with threads of different pitch at either end (unthreaded shaft; Figure 1(a)). The Mini-Acutrak 2 screw (Acumed, Beaverton, OR, USA), a second-generation HCS, has a fully threaded variable-pitch design (fully threaded shaft; Figure 1(b)). The headless reduction (HLR) screw (A Plus Biotechnology Co., Ltd., New Taipei City, Taiwan) has a fully threaded design, but two thread runouts are included in the middle shaft (partially threaded shaft; Figure 1(c)).

Figure 1.

Types of HCS tested in this study, (a) Herbert–Whipple, (b) Mini-Acutrak 2, and (c) HLR screw. HCS: headless compression screw; HLR: headless reduction.

The Mini-Acutrak 2 and the HLR include design modifications intended to improve biomechanical properties. The Mini-Acutrak 2 was reported to exhibit higher ability and sustainability in interfragment compression than other commercialized unthreaded shaft HCS types. ² Its fully threaded design generates a large thread-to-bone contact area. However, the increased purchase afforded by the greater thread surface area makes removal more difficult if required. Furthermore, great torque is applied during the insertion of the Mini-Acutrak 2. ² In dense bone, the required driving torque makes installation difficult. ³ The HLR was developed to overcome the shortcomings of the Mini-Acutrak 2 screw. The thread runouts along the shaft reduce the required driving torque and thereby facilitate operation. The HLR is relatively new, and it remains unclear whether the HLR benefits reduced driving torque while maintaining adequate fragment compression force.

The objectives of this study were to measure the interfragmentary compression and fastening torque generated across a simulated fracture made in synthetic bone block by three HCSs with differently threaded shafts.

Materials and methods

Screws

Three cannulated HCS types were studied: the Herbert–Whipple screw (Zimmer, Warsaw, IN, USA), the Mini-Acutrak 2 (Acumed, Beaverton, OR, USA), and the HLR (A Plus Biotechnology Co., Ltd., New Taipei City, Taiwan). All screws have a nominal diameter of 3.6 mm, a cannulation diameter of 1.1 mm, and a length of 30 mm. Detailed dimensions of each screw type are shown in Table 1. For each screw type, five tests were carried out to compensate for potential experimental errors.

Table 1.

Dimensions of all screw types.

Screw type	Location	Thread diameter (mm)	Thread pitch	Length (mm)	Cannulation diameter (mm)	Number of thread
Herbert–Whipple	Trailing	3.8	1.0	30	1.1	5
	Leading	3.0	1.2			6
Mini-Acutrak 2	Trailing	3.6	0.8–1.1	30	1.1	29
	Leading	3.4	1.1–1.25
HLR	Trailing	3.6	0.6–1.0	30	1.1	25
	Leading	3.3	1.0–1.3

Synthetic bones

The high variability in the density and elastic modulus of cancellous bone affects the biomechanical testing results. ⁴ It has been well established that the synthetic foam materials produce relatively less intra- and inter-specimen variability when compared with the cadaver bone. ⁵ The foam block has consistent material properties similar to the human cancellous bone. Solid polyurethane foams are widely used as an ideal medium for mimicking human cancellous bone, which has been confirmed by the American Society for Testing and Materials (ASTM F1839-08) ⁶ as a standard material for testing orthopedic devices and instruments. In this work, we used polyurethane foam (Sawbones®, Vashon, WA, density 0.16 g/cc, compression modulus 58 MPa) as a substitute for osteoporotic cancellous bone as is common in experimental investigations. ^{2,7 –9}

Specimen preparation

The 40-mm synthetic bone blocks were transversely cut into 20-mm-thick pieces by computer numerical control milling machine to simulate fracture fragments and then shaped to conform to the experimental jig. Four rods within the setup allowed both bone fragments to slide freely as compression was applied. During screw insertion, the driving torque was transferred from the four rods to a torque cell (Model TCF-0.5N; NIPPON TOKUSHU SOKKI, Tokyo, Japan). A washer load cell (Model LTH300; FUTEK Advanced Sensor Technology, CA, USA) was sandwiched between the two bone fragments to measure the compression force (Figure 2). A small gap representing the fracture was set at 0.5 mm, because the compression force generated by a fully threaded HCS mainly depends on the number of threads that cross the bone fragments. ² In total, 30 pairs of bone fragments were fabricated using polyurethane foam blocks.

Figure 2.

Experimental setup to measure interfragmentary compression force and driving torque during HCS insertion. HCS: headless compression screw.

Testing procedure

Insertion of each screw type was performed in accordance with the manufacturer’s surgical guideline for small fragment fixation. Each paired fragment was connected with an appropriate guide pin through a predrilled central hole. The foam was then drilled using drill bits corresponding to the screws tested. The screws were inserted manually by an experienced surgeon. The compression force and the driving torque were measured during screw insertion. A time delay of 10 s between intervals was used to ensure stable measurements. After the screw head was flush with the bone surface (defined as zero revolution), fastening continued until the compression force reached a steady state (no significant change) to study whether the compression force is lost during over-fastening. The experiment was conducted five times for each HCS type, for a total of 15 tests.

Statistical analysis

The maximum compression force and the driving torque for each screw type were estimated as the mean and standard deviation. One-way analysis of variance was used for comparison of the maximum compression forces and the driving torques among different screw types. A p value of <0.05 indicated statistical significance. This study used the statistical package Microsoft Excel and its statistical software (Microsoft Corporation, Redmond, WA, USA).

Results

HLR

The interfragmentary compression and driving torque gradually increased with insertion of the HLR (Figure 3). At the recommended stage (zero revolution), the compression force was 45.83 ± 1.29 N. After over-fastening, the compression force reached a maximum value of 47.4 ± 0.9 N. The fastening torque did not significantly change due to over-fastening and averaged 145.11 ± 1.65 Nmm.

Figure 3.

Compression force against revolution and driving torque against revolution of the HLR screw. HLR: headless reduction.

Mini-Acutrak 2

Similar to the HLR, an overall trend of increased compression force during screw implantation was observed with the Mini-Acutrak 2. The compression force reached a maximum level of 50.98 ± 1.29 N after over-fastening (Figure 4). However, the fastening torque (152.62 ± 2.83 N mm) was higher than that of the HLR (p = 0.075).

Figure 4.

Compression force against revolution and driving torque against revolution of the Mini-Acutrak 2 screw.

Herbert–Whipple

Among the three screw types, the maximum compression force was the lowest for the Herbert–Whipple screw (19.33 ± 1.0 N; Figure 5). Unlike the other two screws, the maximum compression force was achieved at the recommended insertion depth. With over-fastening, the torque increased, but the interfragmentary compression decreased.

Figure 5.

Compression force against revolution and driving torque against revolution of the Herbert–Whipple screw.

Discussion

Scaphoid, radial head, and capitellum fractures and osteotomies of the tarsal bones can be internally repaired using small- and mini-fragment compression screws. Greater thread pitch on the leading (distal) end of the screw than on the trailing (proximal) end creates interfragment compression by drawing the bone fragments together. Since the introduction of the first-generation HCS, the Herbert–Whipple screw, several more HCSs have become commercially available. Generally, HCSs can be divided into three types: unthreaded shaft, fully threaded variable-pitch, and partially threaded shaft designs. The essential biomechanical attributes of HCSs are interfragmentary compression and driving torque. This study aimed to measure and compare the compression force and driving torque generated during screw insertion.

The new-generation HCSs, the Mini-Acutrak 2 and HLR, outperformed the Herbert–Whipple screw. The Mini-Acutrak 2 generated the highest compression force (50.98 ± 1.29 N), and no reduction was observed due to over-fastening. This result was attributed to the fully threaded, taper, and variable-pitch design. This finding is consistent with recent reports by Assari et al. ² and Hart et al. ¹⁰ The maximum achievable compression force of the HLR (47.4 ± 0.9 N) was slightly less than that of the Mini-Acutrak 2, but the difference was not statistically significant. Relatively, the required driving torque was lower for the HLR (145.11 ± 1.65 N mm) than for the Mini-Acutrak 2 (152.62 ± 2.83 N mm) which may be a result of the unthreaded regions. Compared with the fully threaded Mini-Acutrak 2 (29 threads), the two thread runouts of the HLR (25 threads) decreased the contact area with bone, resulting in reduced torque. This is favorable for surgical operation because the risk of rotating the proximal fragment and losing the reduction is greater when high insertion torque is required to advance the screw. Furthermore, in a dense bone, the required driving force may make installation difficult. ¹¹

Assari et al. ² reported a slightly lower compression force of the Mini-Acutrak 2 (45.4 ± 0.8 N) on the same foam material used in this study. This could be because they used a shorter screw (24 mm). Hart et al. ¹⁰ reported considerably higher compression of the Mini-Acutrak 2 (104 ± 15 N). It can be explained that the polyurethane foam used in their study was denser (0.32 g/cc) and approximated scaphoid cancellous bone of a young adult. Moreover, a 1-mm layer of denser foam (0.64 g/cc) was laminated to the surface to represent cortical bone. We did not consider a cortical layer because the recommended insertion depth is 2 mm below the cortical surface of the bone. ¹²

The Herbert–Whipple screw had the lowest compression force of 19.33 ± 1.0 N and generated peak compression when the implant was fully buried in the foam. This result was expected because of the threadless central shaft area of this screw. The interfragmentary compression generated by the Herbert–Whipple was caused by the greater pitch on the leading (distal) end of the screw than the trailing (proximal) end. As the leading thread of the Herbert–Whipple crossed the fracture gap in the foam, no compression was generated because the trailing thread did not engage the bone. For this shanked screw design, the generation of interfragmentary compression force depends on the pitch difference after the insertion of the trailing thread into the bone. The driving torque was significantly lower for the Herbert–Whipple than the Mini-Acutrak 2 and the HLR, which can be attributed to the smaller contact area with the bone. Furthermore, it was noted that the Herbert–Whipple’s driving torque continued to increase with over-fastening, but compression simultaneously decreased. This can be explained by the fact that the trailing threads cross the fracture gap at a deeper level of insertion. Relatively, the HLR and Mini-Acutrak 2 both have a conically shaped proximal end, thus maintaining the compression during extra turns. Fastening torque is considered as the only sensible tactile reference for the surgeon while inserting the screw. Intuitively, fastening torque directly correlates with compression force owing to the increased friction between bone and implant. This study showed that the torque can be increasing, which gives the impression of more compression force (the Mini-Acutrak 2 and the HLR), while for the Herbert–Whipple screw the compression force may be decreasing (Figure 5). When using a shanked HCS, fastening torque can be considered as a misleading measure of the compression force.

Our study has some limitations. Although the mechanical properties of the synthetic bones matched with those of human cancellous bone, differences exist in the structure of the two materials. Specifically, synthetic bones have an almost uniform pore size, whereas human cancellous bone has a complex anatomical structure. This can affect the compression efficacy and fastening torque of the screws. The results of this study were based on a single-density synthetic bone; however, the biomechanical performance of the screws changes with different bone density levels. ⁸ The tests were conducted in synthetic bones with a perfect fracture gap simulated by parallel planes. These were required to enable replicable and reliable testing. Finally, most screw loosening cases could be attributed to physiological cyclic loading. Further evaluation of interfragmentary compression that simulates the cyclic loading of screws under physiological situations is necessary.

Conclusion

The Mini-Acutrak 2 screw and the HLR screw exhibited similar interfragmentary compression force. They can be considered biomechanically equivalent and more effective than the Herbert–Whipple screw. The Herbert–Whipple screw showed a reduction in compression force and a simultaneous increase in torque during over-fastening. This indicates that the hand feel of torque may not be an appropriate proxy for compression when the Herbert–Whipple screw is used.