Intraoperative use of impact microindentation to assess distal radius bone quality

In patients with diminished bone mass and quality, it can be difficult to determine if the bone has sufficient strength to support fixation hardware. Loss of fixation and screw perforation occur in up to 16% and 20%, respectively, of intraarticular proximal humerus fractures treated with locking plate fixation,¹ and osteoporosis is a known risk factor for such losses of humeral reduction.^2–4 With respect to the fixation of distal radius (DR) fractures or Colles’ fractures, radial shortening after volar plating has been negatively associated with areal bone mineral density (aBMD) of the femoral neck in elderly females.² Intuitively, bone mass and quality should factor into operative decision making, but there is a paucity of data regarding their clinical utility, particularly intraoperatively.³ 

Dual energy x-ray absorptiometry (DXA) is the current gold standard for assessing bone mass. Although utilized for osteoporosis screening, DXA-derived aBMD measurements lack specificity in predicting a patient’s risk of low-energy, fragility fracture.⁴ A disproportionate increase in fracture risk with age exists beyond the age-related decline in bone mass, likely due to the inability to measure other factors affecting the fracture resistance of bone.^5–8 Thus, the Fracture Risk Assessment Tool (FRAX) has replaced the osteoporosis threshold of a T-score less than −2.5, the number of standard deviations below normal aBMD, in assessing fracture risk because it incorporates clinical risk factors and aBMD.^9,10 Both DXA and FRAX underestimate fracture risk in certain populations, such as diabetics.¹¹ Although current evidence suggests that diminished bone quality rather than decreased aBMD is responsible for the increased fracture risk in patients with type 2 diabetes, there are limited modalities available for characterizing the inherent quality of a person’s bone tissue.¹² 

Rather than using radiography to evaluate fracture risk and bone quality, reference point indentation (RPI) directly measures bone tissue’s resistance to deformation, thereby characterizing the mechanical behavior of a bone without destructive mechanical tests.^13–17 A successor to the BioDent™ cyclic RPI device, the OsteoProbe (Active Life Scientific, Inc., Santa Barbara, CA), is a handheld instrument for measuring the indentation depth of a micrometer-sized stainless steel probe tip.^5,16 By engaging periosteal bone at 10 N and then delivering a single impact force of ∼40 N, the OsteoProbe records the maximum penetration depth of a stainless steel, conical–spherical tip into cortical bone.¹⁶ The probe introduces a microindentation into the bone. The more easily this microindentation is induced, the deeper the probe indents the bone. This indentation distance increase (IDI) is indexed to the IDI acquired from a reference material immediately after the bone IDI measurements. The OsteoProbe-based measurement known as bone material strength index (BMSi) is 100 times the IDI of reference material divided by the IDI of the patient’s bone, with a lower BMSi indicative of a deeper probe indent.⁵ The reference material is a manufacturer-issued block of polymethylmethacrylate (PMMA). The OsteoProbe does not require a reference probe for periosteum penetration and is well tolerated by patients, making it convenient for clinical use.^16–18 Although theoretically promising, there is conflicting evidence for the clinical utility of impact microindentation for assessing osteoporosis.¹⁹ Some studies demonstrate a significant association between low BMSi and fracture incidence,^6,13,20–22 while others do not;^23–27 for example, BMSi of the mid-tibia diaphysis was lower in patients with fragility fractures as compared to non-fracture, control patients with similar aBMD, suggesting that impact microindentation may be able to capture properties of bone not detected by DXA.⁶ However, other studies found that BMSi is associated with aBMD but not with prevalent fractures.⁷ BMSi was significantly lower for postmenopausal women with a distal radius fracture than age-matched women without a fracture, but the lower BMSi among women experiencing a hip fracture was not strictly significant when compared to the same control group.⁸ 

While used in controlled research environments, the OsteoProbe has not yet been implemented in an intraoperative setting. Thus, the objective of this study was to determine the feasibility of the intraoperative usage of the commercially available microindentation instrument known as the OsteoProbe^® for directly assessing bone quality. A secondary aim was to determine whether an intraoperative measurement of indentation resistance at the distal radius could discriminate between high-energy fractures of non-osteoporotic bone and low-energy fractures of osteoporotic bone.

This single-institution, cross-sectional study was conducted at a level 1 trauma center with an orthopedic hand and upper extremity center. Enrollment occurred between July 2015 and May 2018. Inclusion criteria were patients 18 years of age or older with a unilateral distal radial metaphysis fracture requiring open reduction and internal fixation with volar locked plating. Exclusion criteria included known risk factors of pathologic fractures, patients who had received osteoporosis treatment either within the last 5 years or lasting longer than 5 years, patients with type 1 diabetes mellitus, patients with any other bone disease, patients with a history of cancer, abnormal serum calcium, or chronic steroid use, patients who were unable to undergo a DXA scan, and patients with radial shaft fractures.

Based on clinical history, patients were stratified into two groups: high-energy fractures of presumably otherwise normal, non-osteoporotic bone (e.g., motor vehicle collision) and low-energy fractures of presumably osteoporotic bone (e.g., fall from a chair or standing height onto an outstretched hand).

Institutional review board approval was obtained prior to conducting the study. Informed consent, as well as HIPAA consent, was obtained from each subject prior to their participation in study activities.

The OsteoProbe (Active Life Scientific, Inc., Santa Barbara, CA) microindentation instrument was used intraoperatively to measure local indentation resistance (BMSi) of the distal radius. Two primary surgeons performed all of the intraoperative microindentation measurements throughout the study. The OsteoProbe was attached to a portable laptop computer in the operating room (Fig. 1).

The distal radius was exposed with a standard volar radial approach. Prior to fracture stabilization with the volar plate, the patient’s arm was secured to a Strickland hand table and further stabilized by the assistant surgeon to minimize wrist movement during the indentation procedure. A new and sterilized probe tip was placed on the OsteoProbe instrument, which was then covered with a 15.24 × 121.92-cm² (6 × 48-in.²) sterilized plastic sleeve (Cone Instruments, Caledonia, MI; Fig. 2).

After bringing the instrument into the sterile field, a small hole was cut in the sleeve corner, allowing the OsteoProbe tip to exit the sleeve. The sleeve was secured to the OsteoProbe using a strip of Ioban surgical drape (3M, St. Paul, MN).

The area of indentation was located ∼5 mm proximal to the distal radial fracture site and between the one-third distal and ultra-distal regions. Two rows of five indentation sites, located ∼2 mm apart, were marked using a sterile surgical marking pen [Fig. 3(a)].

The OsteoProbe tip was then positioned perpendicular to the exposed bone, and the handle was slowly pushed down until ∼10 N of force was achieved, the force required to trigger the impact load. This process was repeated for the ten marked indentation sites [Fig. 3(b)]. If a measurement failed to register, extra indents were performed immediately adjacent to the original indent. Based on the conical shape transition from the spherical tip to the probe’s main body, the indent size was ∼300 μm in both diameter and depth. Following indentation, the fracture was stabilized using a volar locking plate. As demonstrated in the intraoperative radiographs, the volar locking plate extended proximal to the indentation sites, thereby protecting the indentation sites [Figs. 4(a) and 4(b)].

Using the same probe tip, the block of reference material was indented five times [Fig. 3(c)]. The tip was subsequently discarded. As determined by the instrument software, BMSi at each site was the harmonic mean of the indentation depth measurements or IDIs acquired from the block times 100 divided by IDI measurements of the bone. For statistical analysis, we used the mean of the 10 BMSi measurements per case.

Patients were seen post-operatively in the clinic based on the standard of care guidelines. This typically corresponded to visits at 8–10 days, 1, 2, 3, 4, 6, 8, 12, and 16 weeks post-operatively, although these were not mandated dates. All cases were reviewed by the primary orthopedic surgeon for the assessment of fracture fixation and healing.

Areal bone mineral density of the hip, femoral neck, and lumbar spine (L1–L4) as well as the contralateral ultra- and one-third distal radius and ulna were measured using a DXA scanner (Lunar iDXA, GE Healthcare, Chicago, IL) at 1–8 weeks post-surgery. The ultradistal region of the forearm is adjacent to the radial end plate and is 15 mm in length. The one-third distal region is 20 mm in length and is one-third of the distance between the ulnar styloid and the olecranon. Standard measurements of aBMD (g/cm²) were provided by the manufacturer’s software.

Differences in the bone measurements between the low-energy and high-energy fracture groups were tested for significance using the two-tailed Mann–Whitney test at a confidence level of 95% (alpha = 0.05). Fisher’s exact test of a 2 × 2 contingency table (also two-tailed) was used to determine whether differences in demographics between the two groups were significant. The Anderson–Darlington test (alpha = 0.05) was used to determine if the ten indents per case demonstrated normal distribution.

Twenty-eight patients were included in the study analysis (Fig. 5, Table I) with 17 in the low-energy fracture group.

As expected, the low-energy fracture group was significantly older than the high-energy fracture group (p = 0.0003). The body mass index did not vary between these two fracture groups (p = 0.2216). There were more women in the low-energy fracture group and more men in the high-energy fracture group. In the low-energy group, there were 15 females and two males; six right and 11 left wrists were involved with five AO type A2, three type A3, one type B3, six type C1, and two type C3 fractures. In the high energy group, there were seven males and four females; three right and eight left wrists involved with three AO type A2, one type A3, one type B3, five type C1, and one type C3 fractures. There were no significant differences in fracture type between the groups (Table I). There were four patients (three females and one male) with type two diabetes mellitus in the low-energy fracture group, with an average age of 62.1 ± 10.2 years. Average hemoglobin A1c closest to the date of the surgery was 7.2% ± 2%. The duration of type 2 diabetes diagnoses ranged from 9 to 15 years.

Intraoperative usage of the OsteoProbe added an estimated maximum of 10 min of operative time to the typical distal radius volar plate fixation procedure. All fractures healed without loss of fixation and there were no cases of infection or delayed healing. There were two low-energy fractures and one high-energy fracture, all with an AO type C3 fracture that healed with a slight intra-articular step-off. The patients were followed for an average of 11.5 weeks (range: 3.1–20.1 weeks). Of the two patients who did not complete at least a six-week follow-up, there were no signs of infection or fracture healing complications at their three-week follow-up visits.

There was no significant difference in BMSi between the low- and high-energy fracture groups (Fig. 6).

There was also no significant difference in BMSi between the low- and high-energy fracture groups when the four patients with type 2 diabetes were excluded from the low-energy fracture group (p = 0.4331). The median (first quartile, third quartile) of the low-energy group was 76.3 (69.8, 79.7) with all patients included and 76.3 (69.8, 79.8) excluding the four with type 2 diabetes, whereas the median (interquartile range) of the high-energy group was 73.3 (62.7, 78.2). Among the 28 cases, BMSi of the ten indents did not pass the normality test for two patients, one from each fracture group (Fig. 7). Only one case required extra indents (three additional), leaving 27 cases in which ten consecutive BMSi measurements from bone were registered by the software.

There were no significant differences in aBMD at the lumbar spine or the ultra- and one-third distal radius between the low- and high-energy fracture groups (Table II), although the forearm measurements trended toward being lower in the low-energy fracture group. The aBMD measurements at the femoral neck and hip were 15.2% and 16.5% lower in the low-energy compared to the high-energy fracture group, but the difference was only significant for hip aBMD (Table II). DXA scans were unable to be obtained for four patients due to a loss of follow-up. One additional patient was unable to undergo a DXA scan of the lumbar spine region due to a back brace.

This study investigated the intraoperative usage of an impact microindentation instrument that was developed for the assessment of bone quality at the tibia mid-diaphysis but applied to the distal radius near a fracture site because intraoperative bone quality assessment is currently limited. The majority of orthopedic surgeons solely rely on the haptic assessment of exposed bone through manipulation with their hands and instruments.²⁸ However, this technique is highly subjective and dependent upon experience.²⁸ Clearly, the development of a direct, quick method to assess bone quality could be revolutionary for intraoperative decision making and warrants ongoing research. Given the relative ease of use and design for the detection of poor bone matrix quality, the OsteoProbe was able to provide measurements of indentation depth prior to volar plate fixation of distal radius fractures without complications.

Although there was no significant difference in intraoperative BMSi of the distal radius between the low- and high-energy fracture groups, bone quality assessment during fracture fixation is feasible. When using the OsteoProbe intraoperatively, further clinical research is necessary to delineate the utility and best practices for performing the indentation procedure. OsteoProbe indentation-based measurements at the tibia mid-diaphysis are permissible in Europe (CE Mark) and the US (FDA) without an investigational device exemption. The role of microindentation in characterizing osseous mechanical behavior and diagnosing osteoporosis continues to be investigated,^19,29 and this study demonstrates its safety and feasibility for intraoperative usage. Because radiographic methods cannot detect minute changes or differences in the ultrastructure of bone tissue, it remains important to develop in vivo techniques to evaluate bone quality in real-time.^19,29

In addition to clinical safety, intraoperative usage of impact microindentation instruments requires consideration of potential logistical challenges. In this study, the use of the OsteoProbe technique did not substantially impact procedure length, with an estimated maximum of only ten additional minutes of operative time. Logistical considerations include the necessity of trained personnel to set up the instrument and operate the software on a laptop computer outside of the sterile field. An assistant surgeon or other operating room staff member is also required to appropriately stabilize the forearm during the indentation procedure. This is critical because any bone movement during the impact of the probe tip affects the measurement. When comparing the BMSi measurements from the left distal radius of this study (n = 19), regardless of fracture energy, to BMSi measurements also from the left distal radius (n = 10 elderly female and male donors) in our previous cadaver study,³⁰ there were no differences between in vivo intraoperative measurements and ex vivo laboratory measurements (Fig. 8) where the bone was rigidly secured in a vice.

This suggests adequate stability of the patient’s radius during the intraoperative impact indentation procedure. It is possible, however, that the fracture altered the local material properties of the adjacent radial bone such that the generation of microdamage during the overloading event could lower the resistance of cortical bone to microindentation. If this was the case, our intraoperative measurements of BMSi were not sensitive enough to detect such an effect.

Acquiring ten micro-indentation measurements to determine the subject’s mean BMSi is the current recommendation for the tibia mid-diaphysis,⁵ and acquiring ten such intraoperative measurements from the distal radius was not difficult, with only one case requiring extra indents to obtain ten useable measurements. However, for 2 of the 28 cases (see Fig. 8 for the distribution), the ten BMSi values did not follow a Gaussian distribution (i.e., failed the normality test). As with measurements from the tibia mid-shaft, this study indicates that the mean of ten indents is suitable to assess BMSi of a subject’s distal radius.

Study limitations include limited patient enrollment, primarily due to ineligibility (Fig. 5). The follow-up time-period may not have been sufficient to detect all potential healing complications, such as late hardware complications. The low- and high-energy fracture patients were not age-matched, which could influence results since the increasing age is an independent risk factor for fragility fracture. In one of the largest cohorts to date, BMSi measurements at the tibia mid-diaphysis were weakly correlated with age (Pearson’s r = −0.131, p = 0.014).¹³ In a similar study involving women in Spain and Norway, BMSi did not correlate with age.¹⁴ This study was composed of adult individuals and cannot comment on the utility of BMSi in the pediatric population. In this study, there was an unequal distribution of males and females within the low and high-energy fracture groups. There are no known sex-related differences in BMSi.^6,22 There was no significant difference in BMI between the low- and high-energy fracture groups. Although BMI is known to impact aBMD as measured by DXA,^31,32 a prior study in the literature found that BMSi differences are independent of BMI.⁸ As noted in the Sec. III, there were four patients with type 2 diabetes mellitus in the low-energy fracture group. Post-menopausal women with type 2 diabetes have been found to have decreased BMSi,^33,34 which was inversely proportional to the duration of their diabetes, as compared to matched controls.³⁵ In our study population, additional statistical analyses did not demonstrate any differing or contradictory results when these four patients were excluded, suggesting that this was not a major limitation in our overall findings. The study population was also predominantly composed of Caucasian individuals, limiting generalizability to more diverse patient populations. As mentioned previously, the OsteoProbe exerts a single impact force of ∼40 N and has primarily been used on the tibia. BMSi of the tibia mid-diaphysis was significantly higher (absolute difference = 9.9%) than BMSi of the distal radius in our previous cadaver study.³⁰ Perhaps a lower impact force would be more appropriate for the distal radius, allowing for better discerning nuanced differences between the non-osteoporotic and osteoporotic bone in this location. There were no significant differences in AO fracture types between the two groups, suggesting that this was not a confounding factor in this analysis. The fractures were retrospectively classified as low- and high-energy based on clinical history alone, so it is possible that some fractures were misclassified.

We anticipate that impact microindentation or other direct in vivo bone assessments could become important in predicting fracture risk and determining appropriate fracture stabilization methods because the inherent bone quality, not just the amount of bone, is important to fracture resistance.³⁶ The suboptimal ability of FRAX or DXA to predict fractures and the dearth of methods for intraoperative bone quality assessment are well documented. In this study, aBMD was lower in patients who had experienced low-energy fractures, but only at the hip, again indicating the variability and limitations of this method. The lower hip aBMD may simply reflect the older number of women in the low-energy group than in the high energy group as bone mass declines with age. Clinical techniques to characterize the tissue-level properties of cortical bone would be a fundamental shift in the assessment and treatment of fractures of the suboptimal quality bone. Moreover, opportunistic use of such techniques during surgery could also help prevent future fragility fractures by identifying and treating high-risk patients. If these measurements can be obtained during a procedure that is already indicated for fracture treatment, it may avoid the cost and time of additional post-operative diagnostics such as a DXA scan. This study demonstrates the feasibility and safety of intraoperative usage of an impact microindentation instrument. There were no complications associated with the BMSi measurements, and the additional time to acquire the measurements was minimal. However, the findings also emphasize the importance of further research to elucidate the potential utility of real-time assessment of bone quality in surgical decision making.

We thank Ms. Julie Shelton for her dedicated coordination and organized oversight of study procedures, as well as her gracious assistance with the preparation of this manuscript.

This study received in-kind support from Active Life Scientific, Inc., in the form of one OsteoProbe Research Use Only device. This work was also supported, in part, by the Vanderbilt CTSA grantfrom the National Center for Advancing Translational Sciences (NCATS)/National Institutes of Health (NIH) under Grant No. UL1 TR002243, the NCATS/NIH for utilization of REDCap under Award No. UL1 TR000445, and the Department of Veterans Affairs (VA), Veterans Health Administration, Office of Research and Development under Grant No. 1I01BX004297. The study contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Center for Advancing Translational Sciences, the National Institutes of Health, or the VA.

The authors have no conflicts to disclose.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical stands. Informed consent was obtained from all individual participants included in the study.

The data that support the findings of this study are available from the corresponding author upon reasonable request.