Interfragmentary surface area as an index of comminution energy: proof of concept in a bone fracture surrogate

https://doi.org/10.1016/S0021-9290(01)00214-7Get rights and content

Abstract

Fracture mechanics theory postulates a monotonic relationship between energy absorption and fracture surface generation. We hypothesized that this relationship was demonstrable to the point that, on a continuous scale, comminuted fractures created with disparate levels of energy delivery could be discriminated. Using a bone fracture surrogate in conjunction with digital image analysis of CT fracture data, we measured the surface area freed by controlled, discrete fracture simulations. Prior to these simulations, the reproducibility of the digital image analysis algorithm was validated with repeated measurements by two different operators. The parametric fracture series results showed a statistically significant difference in measured de novo surface area between four specimen groups, over a range of input energies from 1.4×1010–9.1×1010 J/m3 (or 12.5–80.2 J/specimen). The results of this study provide confirmation that comminution severity can indeed be measured on a continuous scale, based on energy absorption (another clinically meaningful index).

Introduction

Although to date never explicitly demonstrated, there is an intuitively apparent relationship between energy absorption and the degree of comminution in bony fractures. This association is evident throughout the orthopaedic trauma literature, where references are commonly made to high-energy versus low-energy injuries (Teeny and Wiss, 1993; Helfet et al., 1994). The clinical intuition that ascribes injuries with large numbers of fragments to high-energy accidents is corroborated from fracture mechanics principles, in which the energy dissipated in crack propagation and the liberated interfragmentary surface area are directly correlated. For over twenty years, fracture mechanics principles have been applied to bone in the investigation of several of the loading modes and conditions which result in catastrophic failure (Melvin, 1993). This concept, that the necessary work-cost to yield two equal break surfaces is proportional to the surface size, dates back to the classical hypotheses formulated by von Rittinger in 1867 (von Rittinger, 1867). Griffith (1920) demonstrated experimentally that in brittle materials the external work (Ws) necessary to create a crack of projected area (A) in a material continuum equals twice the product of the projected area and the surface energy (γf) of the material. For elasto-plastic behavior, the critical strain energy release rate, Gc, a measure of the material's intrinsic fracture toughness, is proportional to the surface area.

Current bone fracture classification systems, the traditional means for stratifying injury severity and for choosing treatment protocol, are far from ideal, particularly for the case of high-energy injuries. Martin et al. (2000) found that these classification systems failed to perform reliably and reproducibly for comminuted fractures, even under optimal circumstances when fracture cases were reviewed by expert orthopaedic traumatologists who were adhering to strictly stipulated definitions. Their noted “disappointing” level of rater agreement was most pronounced in characterizing qualitative fracture features, especially the energy and stability of the injury. Part of the problem lies in the subjectivity inherent in assessing severity of comminution as a categorical variable (Swiontkowski et al., 1997). Injury severity occurs on a continuum from low-energy to high-energy. Therefore, a continuous measure of injury severity, rather than current categorical fracture classification, would be desirable. The desirability of a less ambiguous scheme is underscored, for example, in Hutson and Zych's (2000) discussion of treatment options for distal femur fractures, where it is plainly stated that a particular salvage technique is “recommended only for the salvage of severely comminuted and open fractures.” Objectively, then, how might one define the term “severely comminuted”?

A continuous measure for comminution severity predicated on energy delivery has its foundation in a concept with which clinicians are already conversant. The presence of many small fragments obviously influences the technical demand of fracture repair, but in addition, the energy transmitted through the joint is a strong determinant of post-traumatic osteoarthritis. The conventional wisdom among orthopaedists is that the greater the degree of energy absorbed by the joint, the higher the degree of cartilage injury and other soft tissue trauma, the higher the risk of complications, and the greater the likelihood of subsequent joint degeneration (Teeny and Wiss, 1993). In experimental settings, these clinical findings are substantiated by the fact that high-energy delivery to cartilage and other soft tissues has lead to demonstrable morphological, biochemical, and biomechanical tissue changes (Radin et al., 1984). In fact, chondrocyte viability decreases in direct proportion to increasing energy delivery (Jeffrey et al., 1995). The latent clinical ramifications of this include chronic pain, a circumspect prognosis, and poor or decreased function.

Computed tomography (CT) provides the opportunity for quantitative analysis of interfragmentary surface area on a continuous scale. Inference of energy absorption from interfragmentary surface area would provide a heretofore-lacking basis for objectively quantifying comminution. Toward that goal, image analysis procedures are here reported to assess the accuracy with which the surface area of fragments can typically be measured for such purpose, using geometrically regular fiducial objects. Next, variable energy impactions were conducted of a novel, homogeneous, brittle polymer with bone-like fragmentation properties (Beardsley et al., 2000). Using a clinical CT protocol to image the resulting fragment batches, we tested the hypotheses that (a) fragment sets from replicate impacts having the same energy absorption would have similar interfragmentary area, and (b) de novo fragment surface area would increase linearly with energy absorption.

Section snippets

Methods

In clinical CT scans, fragment versus background signal intensity is highly variable among patients and among individual fragments. Therefore, using a constant threshold value throughout a whole CT section may prove inadequate (Sumner et al., 1989) to reliably identify fragment borders. To automate the task of surface area measurement, a new algorithm (Fig. 1) was therefore developed for fragment-specific segmentation. This was implemented with a custom-written program utilizing PV-Wave command

Results

In evaluating algorithm performance as a function of ΔGS with the fiducial test cubes, region “growth” followed the sigmoid pattern evidenced in Fig. 2. There was a sharp step at the onset (low values of ΔGS), where parts of the cross-section far from the edge were rapidly registered as part of the region. Approaching the margins of the object, the plots of grayscale difference versus region size tended to plateau. The length and slope of this plateau differed from fragment to fragment.

Discussion

To our knowledge, this study represents the first time that fracture event energy has been objectively assessed from post hoc fragment geometry. This relationship appears to hold potential for application to skeletal trauma. CT data have long been widely utilized for quantifying bone geometry in orthopaedics (Genant et al., 1980). Wolfe and Katz (1995), for example, used CT imaging to document the percentage involvement of the articular surface in proximal interphalangeal joint impaction

Acknowledgments

Financial support provided by NIH grant #AR46601, an Arthritis Foundation Research Grant, and EBI Inc. Mr. William Lack assisted in specimen preparation.

References (25)

  • Cintala, M.J., Horz, F., 1987. The effects of impact velocity of the evolution of experimental regoliths. In Lunar and...
  • Currey, J.D., 1984. The Mechanical Adaptations of Bones. Princeton University Press, Princeton, NJ, pp....
  • Cited by (17)

    • The issues and complexities of establishing methodologies to differentiate between vertical and horizontal impact mechanisms in the analysis of skeletal trauma: An introductory femoral test

      2021, Forensic Science International
      Citation Excerpt :

      Comminution is generally seen as a feature of high velocity impacts [1]. Loading differences should result in variable outcomes [39], reflected (predominantly) in fragmentary differences [40,41], thus the inclusion of fragmentation number accounts for the interaction with the femoral midshaft (used as a proxy for overall femoral strength) [40–42]. The measurements for the Fracture Severity Index (FSI) are as listed below.

    • A new classification of impacted proximal humerus fractures based on the morpho-volumetric evaluation of humeral head bone loss with a 3D model

      2020, Journal of Shoulder and Elbow Surgery
      Citation Excerpt :

      By defining a comminution degree of a bony part, it is possible to evaluate the CV condition from another point of view. In their studies, Beardsley et al3,4 related the bone comminution to the energy of the trauma and used it as a measurement of articular injury severity. Higher energy input led to a greater number of fragments that were also smaller.

    • Computational techniques for the assessment of fracture repair

      2014, Injury
      Citation Excerpt :

      Despite this intuitive appreciation, however, there has historically been no means to objectively quantify the fracture severity based on fracture energy. Recent progress in this area has built on fracture mechanics theory [35], which formally relates the fracture-liberated (de novo) surface area in a brittle solid, such as bone, to the energy of the fracture. CT provides a source from which to extract the de novo surface area, using computational image analysis.

    View all citing articles on Scopus
    View full text