Correlation of 3D Joint Space Width From Weightbearing CT With Outcomes After Intra-articular Calcaneal Fracture

Abstract

Background:

Post-traumatic osteoarthritis (PTOA) of the subtalar joint is a serious, disabling, and frequent complication following intra-articular calcaneal fractures (IACFs). Using plain radiographs to assess the subtalar joint for PTOA is imprecise and insensitive, hindering progress toward improving treatment and assessing outcomes. This study explored how low-dose, weightbearing CT (WBCT) can be used to provide reliable, quantitative 3D measures of subtalar joint space width (JSW) following intra-articular calcaneal fracture (IACF) and correlated the 3D JSW with clinical outcomes.

Methods:

After IRB approval, twenty-one patients (15 male, age 28–70 years) who sustained IACFs and were treated with percutaneous surgical reduction underwent WBCT scans at follow-up visits 2–15 years (average 7.8 years) after surgical treatment. Subtalar joint 3D JSW was computed after a semi-automated protocol was used to segment the talus and calcaneus from the WBCT data. Mean and minimum 3D JSW were calculated and compared to Kellgren-Lawrence (KL) radiographic OA grade, Rand-36 Physical (PCS) and Mental (MCS) Component Scores, and VAS pain scores. Spearman’s rank correlation was used to detect strength of association between variables, with significance set at p<0.05.

Results:

Mean 3D JSW measured from WBCT for patients with IACFs ranged from 0.90–2.51 mm (1.66 ± 0.43 mm) over the entire subtalar joint. Intra- and inter-rater reliability for the WBCT-based JSW measurement technique were 0.95 (95%CI, 0.91–0.97) and 0.97 (95%CI, 0.95–0.98), respectively. Mean and minimum 3D JSW values correlated inversely with VAS pain scores and KL grade (p<0.05), particularly in central and posterior subtalar regions.

Conclusion:

WBCT-based methods were used to quantify the preservation/loss of JSW in patients with IACFs, enabling more accurate, definitive measures of subtalar PTOA. The results of this study demonstrate that WBCT can be utilized to objectively assess subtalar PTOA and help to better understand how arthritic changes affect actual patient experience.

Level of Evidence:

Prognostic, Level III

INTRODUCTION

Intra-articular calcaneal fractures (IACFs) frequently lead to post-traumatic osteoarthritis (PTOA) of the subtalar joint resulting in substantial pain, morbidity, and disability.^18,23 Disease burden of PTOA produces limitations in quality of life comparable to congestive heart failure or kidney disease.¹⁷ Despite interventions to surgically reduce displaced and fragmented articular surfaces, for some injuries, PTOA remains nearly inevitable in certain cases. The interactions of mechanical and biologic mechanisms that lead to PTOA following a displaced IACF are poorly understood. Early after injury and treatment, the development of PTOA is hard to predict, leaving case-specific prognoses largely speculative. A long-term clinical follow-up study following displaced IACFs has shown that despite surgical management, 29% of these patients develop PTOA.²²

Current methods for assessing PTOA of the subtalar joint are imprecise and insensitive to early joint space changes, substantially hindering accurate timely identification of the precise location and severity of PTOA. Long term follow-up is challenging, and early markers of PTOA are necessary to assess treatment effects. Weightbearing plain radiographs have been the primary imaging modality for classification systems that assess the joint for PTOA, such as the Kellgren-Lawrence (KL) Grading Scale (Table 1).^8,21,28,30 KL grading is subjective and has poor inter-observer reliability in other joints.^6,14,15 The complex geometry of the subtalar joint, with its multi-faceted surfaces and overlapping bone densities, makes evaluating this joint with plain radiographs particularly difficult.^16,27

Table 1.

Kellgren-Lawrence Classification of Osteoarthritis.

Grade of osteoarthritis	Description
0—none	No radiographic findings of osteoarthritis
1—doubtful	Doubtful narrowing of joint space and possible osteophytic lipping
2—minimal	Definite osteophytes, definite narrowing of joint space
3—moderate	Moderate multiple osteophytes, definite narrowing of joint space, some sclerosis and possible deformity of bone contour
4—severe	Large osteophytes, marked narrowing of joint space, severe sclerosis and definite deformity of bone contour

Compared to radiographs, imaging with CT facilitates more detailed and accurate assessment of the normal and pathologic anatomy of the foot and ankle. For calcaneal fractures, CT-based measures address limitations of plain radiographs by providing more details of the 3D fracture anatomy and of the subtalar joint, which may be indispensable in surgical planning and assessing post-operative reduction. CT acquires 3D images, providing visualization of the entire subtalar joint. The cost, radiation dose, and non-weightbearing positioning have limited the clinical use of CT scans for assessing subtalar PTOA to when fusion or other secondary reconstructions are planned.

Recently, low-dose weightbearing CT (WBCT) imaging of the subtalar joint in a standing position has become available, allowing for a comprehensive assessment of the functional relationships of articular surfaces and bony structures (Figure 1).^10,20 There is now increasing amount of literature describing the use of WBCT in patients with foot and ankle disorders.¹ A study by Colin, et al described the morphology of the subtalar joint in a healthy, asymptomatic cohort using WBCT, under typical physiological stress.⁵ WBCT has also been found to be a reliable and consistent way to assess varus/valgus configuration of the posterior facet of the subtalar joint in osteoarthritic ankles.⁹ Joint space width (JSW) measurements have been used to characterize progression of PTOA in other joints.^11,13 Given the challenges with plain radiographic imaging of the subtalar joint, WBCT may provide more specific, sensitive, and reliable information on the JSW changes that occur in the subtalar joint after IACF.

Figure 1.

Three multiplanar images of the subtalar joint demonstrate the 3D information provided by WBCT imaging.

The objective of this study was to utilize WBCT images to develop quantitative measures of the subtalar joint space to estimate cartilage preservation or loss following displaced IACF and to prove that these measurements could be reliably made. A secondary objective was to correlate the resulting JSW measures with patient outcomes and KL grading of PTOA. The overall hypothesis was that quantitative measures of regional and global joint space of the subtalar joint from WBCT can be reliably made. Furthermore, that these measures correlate with traditional classification of PTOA made on plain radiographs and with the clinical outcome of patients with IACFs. A technique to develop objective measures of subtalar joint preservation has potential to facilitate development of novel treatments for the prevention and treatment of PTOA, establishing WBCT as an imaging modality to detect and follow PTOA after IACF for research and for clinical care.

MATERIALS AND METHODS

Study Population

Following IRB approval, the medical record system was searched using the CPT code for patients treated for IACFs. Patients treated surgically with reduction and percutaneous screw fixation and who were a minimum of 2 years following fracture were identified. Patients qualified for inclusion if they were 18 years or older at the time of initial treatment, had immediate pre- and post-operative CT scans, had at least two years clinical follow-up, and had the ability to complete questionnaires. Sanders classification were determined on pre-operative CT scans. Patients were excluded if they were unable to return for follow-up, unable to complete a WBCT scan due to pain or disability, had complete osseous fusion of the subtalar joint, or previously underwent total ankle arthroplasty.

Eligible patients were contacted and invited to return for clinical examination and imaging of both the injured and contralateral foot and ankle. Follow-up time ranged between 2–10 years. Of the 45 patients who met the inclusion and exclusion criteria listed above, the first 21 who volunteered after they were contacted by the research team via mailings and phone call were recruited. An enrollment flowchart is presented in Figure 2. The expense of the imaging studies precluded recruiting a larger sample size. All participants completed an IRB-approved informed consent process. Patients were financially compensated to offset the time and expense of participating in a clinic visit for research purposes.

Figure 2.

Flowchart for inclusion of participants

Twenty-one patients (16 male, age 23–70 yrs) who sustained displaced IACFs treated with percutaneous surgical reduction underwent WBCT scans at follow-up visits 2–15 years (average 7.9 yrs) after surgical treatment (Table 2). Sanders classification determined on pre-operative CT scans included SII (52.3%), SIII (38%), SIV (9.5%).

Table 2.

Patient Characteristics.

	Average (SD)	Range
Age, y	51	23–70
Right/left, no.	10/11	—
Male/female, no.	15/6	—
Follow-up, y	7.9 (3.6)	2–15
VAS score	2.5 (2.2)	0–10
PCS	47.6 (9.0)	28.4–58.0
MCS	55.6 (7.4)	40.4–68.9
KL grade	KLO = 5, KLI = 7, KL2 = 3, KL3 = 4, KL4 = 2

Abbreviations: KL, Kellgren-Lawrence; MCS, RAND-36 Mental Component Summary; PCS, RAND-36 Physical Component Summary; VAS, visual analog scale.

Image Acquisition and JSW Measurements

At the time of clinic visit, radiographs of the calcaneus were acquired (Figure 3). The weightbearing axial (Harris) view was obtained with the subject standing on the film cassette and the beam angled 45 degrees to the floor. This allows the entire calcaneus to be visualized from the posterior tuberosity to the talocalcaneal joint. On the weightbearing lateral view, the articulation with the cuboid and talus are in profile. The Broden’s view was obtained with 10 degrees of cephalic angulation of the radiographic beam focused at the tip of the fibula, with the ankle rotated 45 degrees internally. This provides views of the posterior subtalar joint. PTOA was assessed on these radiographs. Three clinicians independently reviewed the weightbearing radiographs of the calcaneus and assigned a KL grade (Table 1) to the subtalar joint. Where there was disagreement in KL grade it was resolved to the rating recorded by the majority of raters. KL grade ≥ 2 met criteria for PTOA and was used to define two groups of patients; those with and without PTOA.

Figure 3.

Calcaneus radiograph series with, (a) weightbearing lateral, (b) Broden’s and (c) weightbearing axial (Harris) views of the calcaneus

For the WBCT, a commercial scanner (PedCAT, Curvebeam, LLC, Warrington, PA) was utilized (Figure 1). The scan was obtained with patients standing with the pelvis centered above their feet, torso and head vertical, looking straight ahead. A 3D dataset with isotropic spatial resolution of 0.37 mm per voxel and a 350 mm field of view was reconstructed from initial cone beam projection images and exported from the secure server to image analysis workstations.

A novel WBCT-based measurement technique to quantify subtalar PTOA was developed based upon prior experience in the knee.^24,26 Assessment of 3D JSW began with a semi-automated protocol to segment the talus and calcaneus bones from WBCT scans using custom MATLAB code (The MathWorks, Natick, MA) (Figure 4). The segmentations were performed by a graduate student who had over a year of joint segmentation experience and were reviewed by an expert with over 5 years of experience. Minor errors in the automated segmentations were identified and manually corrected using ITK-SNAP. Segmentations were exported as triangulated surface models to Geomagic Design X (3D Systems, Research Triangle Park, NC), where they were lightly smoothed to remove voxellation artifact. The 3D JSW measurements were computed as the normal distance from the center of each triangulated face on the talus subchondral bone to the opposing calcaneal subchondral bone (Figure 5b–c). In situations where step-offs and gaps left regions without an opposed surface, measurements were not recorded, as the measurement of JSW would be artificially high. This resulted in abrupt changes in the JSW measure that coincided with step-offs and gaps, although these were not the primary focus of this analysis. JSW measurements made in cartilage regions closely adjacent to these step-off and gaps captured joint space narrowing from cartilage loss in these areas.

Figure 4.

(a) Weightbearing sagittal CT scan of ankle (b) semi-automated segmentation of talus (green shading) tibia (yellow shading) and calcaneus (red shading) indicated.

Figure 5.

(a) Weightbearing coronal CT scan of ankle with semi-automated segmentation of talus (green shading) and calcaneus (red shading) indicated. (b) Digital surface models of the articulating talus and calcaneus. (c) A closeup view of the subtalar joint including a series of arrows showing the 3D JSW measurements, with the arrows colored according to JSW value.

These methods mapped 3D JSW at every location on the joint surface (Figure 6). This JSW map was then split into discrete regions using a semi-automated process in Geomagic Design X. Three equivalent segments along the curvature of the posterior facet in both the anterior to posterior and medial to lateral directions were defined, effectively dividing the surface into an anatomical 3×3 grid (i.e., 9 regions). Then the entire middle articular facet was treated as a 10^th region (Figure 7). The resulting 3D JSW maps contained hundreds of measurements in each anatomical region. To analyze these data, the mean and minimum JSW values were obtained in each region to provide singular representative metrics that were quantified in each region.

Figure 6.

Map demonstrating subtalar 3D Joint Space Width (JSW).

Figure 7.

Inferior view of the talus displaying subtalar joint divided into 10 parts: 1: Anterior Lateral; 2: Anterior Intermediate; 3: Anterior Medial; 4: Intermediate Lateral; 5: Intermediate Intermediate; 6: Intermediate Medial; 7: Posterior Lateral; 8: Posterior Intermediate; 9: Posterior Medial; 10: Middle Facet

Clinical Outcomes

Patients completed clinical outcome questionnaires at the time of follow-up, including Rand-36 Physical Component Score (PCS), Rand-36 Mental Component Score (MCS), and visual analog scale (VAS) pain scores. The RAND-36 is a self-administered health survey outcome measure to assess health-related quality of life.⁷ The RAND-36 health survey is frequently used in patients within a rehabilitation population and consists of several subscales: physical functioning, social functioning, role limitations due to physical health problems, role limitations due to personal or emotional problems, mental health, pain, vitality, and general health perceptions. The RAND-36 consists of the same 36 items as the 36-item short-form health survey (SF-36), but with a different scoring procedure.³² Pain was the primary clinical outcome score considered.²⁹

Statistical Methods

To assess intra-rater reliability and reproducibility of the cartilage measures, the semi-automated segmentations from WBCT scans and subsequent PTOA measurements for five randomly selected cases were recalculated by two of the initial observers independently 6 months after performing the first analysis and compared to their previous measurements. We define intra-rater reliability as a measurement of how consistent a rater is with respect to themselves. Inter-rater reliability is a measure of how consistent a rater is with respect to others, which was determined by comparing the JSW measurements between the raters. To assess agreement on the categorical KL grades, we used weighted kappa coefficients. To assess agreement on the WBCT-based JSW measurements (continuous variables), we used intraclass correlation coefficient (ICC).

To establish validity and clinical relevance of the WBCT-based JSW measurement technique, regional JSW measures were compared to clinical outcomes and plain radiographic measures of PTOA obtained at follow-up > 2 years from injury. The strength of association between mean and minimum JSW measures with KL Grade, VAS, and Rand-36 PCS and MCS was assessed with Spearman’s rank correlation test. We also ordered the mean and minimum JSW from the smallest to the largest with the outcomes above, using Spearman correlation test and an ordered values approach described by Buck et al.² This provided a more sensitive measure of disease progression due to the heterogeneity of the location of the narrowest JSW. Correlation coefficient values less than 0.5 are indicative of poor agreement, values between 0.5–0.75 indicate moderate agreement, values between 0.75–0.90 indicate good agreement, and values greater than 0.90 indicate excellent agreement. For inter- and intra-rater reliability of KL grade, Kendall’s τ_b was used for inter-rater reliability for KL grade, and weighted kappa for intra-rater reliability. Statistically significant level was set at p<0.05. All statistical analyses were completed using SAS software version 9.4 (SAS Institute Inc., Cary, NC).

RESULTS

Of the twenty-one patients included in the study who sustained displaced IACFs treated with percutaneous surgical reduction, nine of the 21 subtalar joints (42%) showed radiographic evidence of PTOA (KL grade ≥ 2). The KL grades on the follow-up weightbearing radiographs were KL0 (22.3%), KL1 (22.3%), KL2 (36.4%), KL3 (9.1%) and KL4 (9.1%). Mean 3D JSW measured from WBCT ranged from 0.90–2.51 mm (mean ± standard deviation: 1.66 ± 0.43 mm) over the entire subtalar joint.

Mean and minimum 3D JSW values and VAS pain scores shared a strong inverse correlation, particularly in the central and posterior subtalar regions, where the findings were statistically significant (Table 3). The mean and minimum JSW values also correlated inversely with KL grade (p=0.021). The global assessment of JSW using an ordered values approach revealed a strong inverse correlation with 3D JSW, VAS pain, and KL Grade (Table 4). Interestingly, there was no correlation between JSW and Rand-36 PCS or MCS, although several regional 3D JSW measures approached significance (p<0.1).

Table 3.

Correlation Coefficient of Mean and Minimum 3D JSW With Clinical Outcomes and KL Grade.^a

	KL grade	3D JSW	Ant med	Ant int	Ant lat	Int med	Int int	Int lat	Post med	Post int	Post lat	Middle facet
VAS score	0.50	Mean	−0.32	−0.34	−0.40	−0.47	−0.51	−0.47	−0.49	−0.67	−0.30	−0.22
		Min	−0.35	−0.27	−0.44	−0.49	−0.48	−0.47	−0.19	−0.50	−0.37	−0.53
PCS	−0.23	Mean	0.11	0.12	0.38	0.25	0.17	0.25	0.13	0.28	0.13	−0.01
		Min	0.36	−0.10	0.32	0.23	0.08	0.20	−0.07	0.28	0.06	−0.11
MCS	−0.05	Mean	−0.13	−0.03	−0.03	0.18	0.12	0.12	0.08	0.25	−0.05	−0.03
		Min	0.10	0.15	0.11	0.02	0.18	0.08	0.12	0.04	0.00	−0.12
KL grade	-	Mean	−0.36	−0.21	−0.30	−0.41	−0.59	−0.18	−0.61	−0.56	−0.23	0.06
		Min	−0.48	−0.10	−0.32	−0.60	−0.54	−0.32	−0.51	−0.37	−0.22	−0.20

Abbreviations: ant, anterior; int, intermediate; JSW, joint space width; KL, Kellgren-Lawrence; lat, lateral; MCS, RAND-36 Mental Component Score; med, medial; min, minimum; PCS, RAND-36 Physical Component Score; post, posterior; VAS, visual analog scale.

^aBoldface type indicates statistical significance (P < .05).

Table 4.

Correlation of 3D JSW With Clinical Outcomes and KL Grade: Ordered by Smallest JSW.^a

	KL grade		1	2	3	4	5	6	7	8	9	10
VAS score	0.50	Mean	−0.59	−0.54	−0.45	−0.54	−0.56	−0.57	−0.58	−0.52	−0.48	−0.35
		Min	−0.47	−0.61	−0.53	−0.56	−0.56	−0.53	−0.46	−0.61	−0.40	−0.39
PCS	−0.23	Mean	0.22	0.20	0.17	0.18	0.21	0.28	0.23	0.25	0.26	0.15
		Min	0.14	0.20	0.18	0.16	0.18	0.16	0.16	0.21	0.16	0.12
MCS	−0.05	Mean	0.08	0.12	−0.01	0.09	0.06	−0.01	0.03	0.10	0.21	0.03
		Min	0.02	0.04	0.08	0.01	0.01	0.11	0.16	0.22	0.25	0.19
KL grade	−	Mean	−0.58	−0.38	−0.47	−0.50	−0.52	−0.49	−0.44	−0.49	−0.48	0.36
		Min	−0.40	−0.48	−0.46	−0.50	−0.43	−0.52	−0.56	−0.54	−0.39	−0.33

Abbreviations: ant, anterior; int, intermediate; JSW, joint space width; KL, Kellgren-Lawrence; MCS, Rand-36 Mental Component Score; min, minimum; PCS, RAND-36 Physical Component Score; post, posterior; VAS, visual analog scale.

Note: The row I −10 represents the correlations with clinical outcomes after ordering patient joint space regions by smallest joint space width (eg l=each patient’s smallest joint space width region, 2 = each patient’s 2nd smallest joint space width region, etc.).

^aBoldface type indicates statistical significance (P < .05).

Inter- and intra-rater reliability for KL grade was moderate (0.58–0.72, p<0.01 and 0.55–0.57, p<0.001, respectively) (Table 5). In contrast, the inter- and intra-rater reliabilities for the WBCT-based JSW measurement technique were excellent, with an inter-rater reliability of 0.97 (95%CI, 0.95–0.98) and an intra-rater reliability of 0.95 (95%CI, 0.91–0.97). The greatest difference in mean JSW measured for inter-rater reliability was 0.23 mm, while the greatest difference for intra-rater reliability was 0.25 mm. The largest difference for minimum JSW measured was 0.28 mm and 0.19 mm for inter- and intra-rater reliability, respectively.

Table 5.

Inter- and Intrarater Reliability of the 3D JSW Measure Quantified by the Maximum Differences in the Mean and Minimum JSW Measurements.^a

Case	Maximum difference in mean JSW		Maximum difference in minimum JSW
	Interrater	Intrarater	Interrater	Intrarater
1	0.07	0.03	0.19	0.02
2	0.23	0.25	0.18	0.19
3	0.13	0.19	0.28	0.07
4	0.07	0.04	0.02	0.02
5	0.06	0.05	0.12	0.12
Average	0.11	0.11	0.15	0.08

^aData are presented in millimeters.

Abbreviation: JSW, joint space width

DISCUSSION

The challenges of accurately assessing the subtalar joint, with its complex, multi-faceted surfaces and overlapping bone densities, provides a compelling need for improved imaging capabilities. Plain radiographs and CT scans of intra-articular calcaneal fractures demonstrate disrupted articular surfaces and the degree of displacement or malalignment of articular fragments. Weightbearing radiographs are commonly utilized in the assessment of patients following IACF.²² However, these imaging modalities have not been used to quantify the preservation or loss of subtalar joint space. To better assess the development of PTOA, more accurate assessment of the subtalar joint is necessary. The purpose of this study was to assess the potential of WBCT to provide reliable, objective measures of maintained or lost subtalar JSW as an assessment of PTOA.

There are many potential advantages of WBCT scans when compared to more established imaging modalities like conventional CT, MRI, and combinations of weight-bearing radiographs with advanced imaging. Of these, it is the only imaging modality capable of evaluating the 3D anatomy of the subtalar joint space with the patient standing under normal weightbearing conditions. ^10,20 Other advantages of WBCT include fast image acquisition time, high-contrast resolution and spatial resolution, lower radiation dose, and a relatively small unit size with portable design, allowing close proximity to the clinical care area. Time spent acquiring images has previously been shown to be significantly lower for WBCT than for weightbearing radiographs or conventional CT scans.^19,20 Although radiation dosing was not directly measured in this study, prior studies have shown unilateral WBCT to have a comparable dose to 6 unilateral radiographs and 5.6% of unilateral foot and ankle CT.¹² Compared to a conventional ankle CT, low-dose WBCT has similar cost for patients and institutions.¹⁹ Total hospital billing for a unilateral weightbearing CT of the ankle is similar to a normal CT scan, $2529.00 at our institution. In comparison, a complete calcaneus radiograph series, including axial, lateral and Broden’s views, is $379.00 at our institution. Despite the similar costs at our institution for weightbearing and non-weightbearing CT scans, accounting for staff costs and time efficiency, WBCT has been reported as being up to 50 times more profitable per patient.¹⁹ Furthermore, WBCT has been found to have generally less capitalization cost than conventional CT scan technology.³ As compared to MRI, WBCT offers a clearer view of fixation instrumentation given the potential for image distortion that can occur with hardware present in MRI.³¹ Additionally, acquiring weightbearing scans offers improved evaluation of patterns of subtalar joint space loss both regionally and globally, which may be predictive of disease advancement.²⁵ These measurements were the focus of this study.

The measurement techniques used in this investigation were based on prior work that focused on WBCT detecting OA in the knee using a prototype system.^24,26 In that study, the sensitivity, accuracy and negative predictive value for detecting osteophytes were substantially higher with WBCT than radiographs. The joint surface area with 3D JSW<2.5mm had much better correlation with articular cartilage morphology evaluated on MRI than did radiographic JSW (r=0.84 vs. 0.66).²⁴ Similar results were found in this study. WBCT imaging of the subtalar joint provided more reliable assessment of PTOA following calcaneal fractures than plain radiographs.

Reliability

A primary aim of the study was to establish reliability of the novel measurement technique for detecting cartilage loss and PTOA and compare it to the current common practice using radiographs. The results of this study strongly suggest that WBCT is a more reliable technique than plain radiographs to assess PTOA in the subtalar joint. Kellgren and Lawrence found that using the KL grade for the tibiofemoral joint had higher inter-observer correlation coefficients than other joints.⁸ The KL grade for the subtalar joint has been shown to have low-to-moderate inter- and intra-rater reliability at best.^6,14,15 Despite this poor reliability, KL grade is frequently applied to the subtalar joint. In the absence of anything better accepted, it was chosen as the PTOA outcome metric to which we compared the WBCT-based technique.

In our study, the observers performed better than previously reported for KL reliability scores, demonstrating moderate inter- and intra-rater reliability. In comparison, the reliability of the WBCT-based JSW measures between repeated measurement sessions was almost perfect, as well as between raters. The greatest difference in mean or minimum JSW measurement for both inter- and intra-rater reliability was 0.28 mm, with the average difference for both reliability studies between the mean and min ranging from 0.08–0.15 mm. Since the isotropic voxel edge length was 0.37 mm, these results show the difference in measurements to be less than a voxel. Therefore, changes in our measures of JSW greater than the edge length of a voxel can confidently be attributed to true anatomical changes in the JSW.

Validity

As a secondary goal of this study, we assessed whether WBCT images used to quantify the preservation/loss of JSW in patients with IACF were correlated with clinical outcomes. We found that lower global 3D JSW values were strongly correlated with higher VAS pain scores, even in this relatively small cohort of patients. WBCT allows regional measures of 3D JSW, and several of these outperformed KL grade by better correlating with VAS pain, particularly in the central and posterior subtalar regions. A smaller 3D JSW may lead to higher patient pain scores secondary to locally accelerated cartilage degradation. The correlation of 3D JSW to clinical outcomes was stronger than for KL grade, suggesting that JSW represents PTOA clinical status more accurately than KL grade.

Limitations

There were several limitations to this study. At clinical follow-up, 42% of subtalar joints had evidence of PTOA. Given only a portion of participants demonstrated evidence of arthritic progression and there were no controls (or contralateral WBCT scans obtained), the study was susceptible to selection bias. Nevertheless, as this is a novel measurement technique, even a select cohort of patients and follow-up scans provided the ability to develop a technique, confirming its reliability and validity. Future development and expansion upon this image analysis technique and prospective studies with larger numbers of patients are necessary.

The quality of the plain weightbearing radiographs obtained in the study was variable. Although the nature of our radiographs represents the standard quality of radiographs obtained in the context of routine clinical care, rotated images may have impacted the reliability of reviewers to assess PTOA. The inconsistency in radiographs is consistent with other studies demonstrating that assessing subtalar joint space loss and severity of PTOA using radiographs is imprecise and insensitive, with variable reliability.^6,14,15 We did not utilize radiographs or WBCT to evaluate alignment, although this is potentially very important and the weightbearing position allows this to be the subject of further studies. While WBCT has potential to provide reliable 3D assessment of subtalar joint space, WBCT is not yet widely available, making its future clinical applicability unknown. It is currently unclear whether WBCT will be predominately a research tool, or whether wider availability will allow it to be incorporated in more routine clinical care.

Another limitation is the need for human effort to obtain accurate manual segmentations of the calcaneal and talar bone edges, due to the differences in bone density within regions of the subtalar joint. This was time-consuming, which limits the near-future clinical utility of 3D JSW measurements on WBCT and is part of the reason our study included a relatively small number of subjects. We anticipate that the time required for the JSW measurement technique will be significantly reduced in the future with the advent of tools like CubeVue Autometics (Curvebeam LLC)and this may lead to wider clinical use of JSW measures.

Lastly, the current study was designed to assess correlations between radiographic and 3D JSW by WBCT in a cross-sectional manner. In future studies, the new method should be applied on a longitudinal dataset of WBCT scans, to determine whether the proposed method demonstrates a loss or maintenance of subtalar joint space over time and whether loss of JSW can be detected earlier in the disease process. The current study was only a snapshot at relatively longer term follow up.

Future Research and Clinical Impact

Currently, little is known about the interaction of acute mechanical damage and chronic elevated contact stress and the biological processes that lead to PTOA. However, emerging research of therapeutic targets for PTOA prevention that can be applied before or concomitant with open reduction and internal fixation procedures holds promise for mitigating PTOA and preserving a patient’s quality of life.⁴The prolonged clinical course of a joint evolving to PTOA makes early indication of joint space loss with WBCT not only of potential clinical value, but important to evaluate the influence of candidate treatment intervention in research studies. Objectively measuring subtalar joint cartilage preservation or loss will enable research to assess the joint outcome relatively early after fracture. The high incidence of PTOA after calcaneal fracture makes it a good target to study novel biologic treatments for the prevention and treatment of PTOA.

WBCT also has the potential for clinical application as an improved assessment tool. Quick scan time, low radiation dose and weightbearing positioning are all advantages of WBCT over standard CT and plain radiographs. These advantages may be of use clinically in reconstruction, malunion, and certain postoperative malreductions. This study has significant potential to impact the preferred imaging of subtalar arthritis, correlate contact stress with subtalar PTOA development, and provide a novel framework to measure PTOA severity earlier in the disease process. However, we recognize that decreased cost and wider availability of WBCT units will be necessary for this technique to become a part of routine clinical practice.

CONCLUSION

The results of this study demonstrate that WBCT can be utilized to sensitively and reliably assess the subtalar JSW as a measure of PTOA and help to better understand how joint space changes affect patient clinical outcomes. By improving the early assessment of JSW through improved visualization of the subtalar joint, we may better understand the effects of surgical treatment in the prevention of PTOA and the natural progression of PTOA after intra-articular calcaneal fracture, accelerating scientific progress and advancing clinical care.