- Email: [email protected]
Hallux Valgus Evaluation on MRI: Can Measurements Validated on Radiographs Be Used?
ARTICLE IN PRESS The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery j o u r n a l h o m e p a g e : w w w. j f a s . o r g
Original Research
Hallux Valgus Evaluation on MRI: Can Measurements Validated on Radiographs Be Used? Nathan Heineman, BA 1, Yin Xi, PhD 2, Lihua Zhang, MD 3, Riham Dessouky, MBBCh, Msc 4, Jed Hummel, MD 5, Justin Skweres, MD 5, Dane Wukich, MD 6, Avneesh Chhabra, MD 7 1Medical
Student, Department of Radiology, UT Southwestern Medical Center, Dallas, TX Professor, Department of Radiology, UT Southwestern Medical Center, Dallas, TX Professor, Department of Radiology, Peking University Third Hospital, Beijing, China 4 Assistant Professor, Department of Radiology, Faculty of Medicine, Zagazig University, Zagazig, Egypt 5Fellow, Department of Radiology, UT Southwestern Medical Center, Dallas, TX 6 Professor and Chairman, Department of Orthopaedic Surgery, UT Southwestern Medical Center, Dallas, TX 7Associate Professor and Musculoskeletal Division Chief, Department of Radiology, UT Southwestern Medical Center, Dallas, TX 2Assistant
3Associate
A R T I C L E
I N F O
Level of Clinical Evidence: 3
Keywords: bunion hallux valgus angle intermetatarsal angle magnetic resonance image radiograph
A B S T R A C T
Hallux valgus (HV) is a common deformity of the great toe affecting >23% of adults in the United States. The severity of the deformity is traditionally analyzed using radiographs to determine measurements such as the HV and intermetatarsal angles. We sought to determine the relationship between the radiographic and magnetic resonance imaging (MRI) measurements because this is not yet known. Two of us analyzed a series of 56 consecutive patients who had had radiographs and MRI performed on the same foot between April 27, 2015 and March 9, 2016 and who satisfied all other inclusion and exclusion criteria (age 18 to 100 years, no history of recent foot trauma, and no metal hardware in the foot). We found excellent interreader reliability (intraclass correlation 0.89 to 0.96) and intermodality agreement (intraclass correlation 0.83 to 0.91). The HV angle measured 15.0° ± 8.8° on the MRI scans and 13.8° ± 8.7° on the radiographs (mean difference −1.15° ± 3.89°), and the intermetatarsal angle was 9.0° ± 3.1° on the MRI scans and 8.8° ± 2.9° on the radiographs (mean difference −0.22° ± 2.10°). The HV measurements were reliable on both radiographs and MRI for the range of values tested. Small intermodality statistically significant differences in HV angle measurements were found; however, these might not be enough to be clinically significant. © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved.
Hallux valgus (HV), also known as a bunion, is a relatively common condition in the United States, affecting approximately 25% of adults aged 18 to 65 years and 36% of adults aged >65 years (1). The pathoanatomy is defined by medial deviation of the first metatarsal, lateral deviation of the hallux, and a prominent metatarsal head (2). This deformity is associated with an aching pain in the metatarsal head that can be sharply painful during ambulation and with shoe wear. It is also associated with a variety of painful and potentially disabling sequelae, including bursitis, cartilage degeneration, and nerve entrapment (3–5). In addition to the history and physical examination findings, a variety of radiologic findings are used to assess the severity of HV. These include the HV angle (HVA) and the intermetatarsal angle (IMA), among many others (6). The HVA and IMA have been shown to correlate best with Financial Disclosure: None reported. Conflict of Interest: A.C. is a consultant for ICON Medical. Address correspondence to: Avneesh Chhabra, MD, Department of Musculoskeletal Radiology, UT Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390-9178. E-mail address: [email protected] (A. Chhabra).
the magnitude of deformity (7). Although the angles have been well studied on radiographs, the magnetic resonance imaging (MRI) correlates of HV have not been described. MRI of the forefoot and midfoot has been increasingly used for the diagnosis of plantar plate tears, collateral ligament injuries, and cartilage assessment (8). This poses the question of whether radiographs would be necessary if the physician already planned to order an MRI scan. It is not known whether similar measurements used on the reference standard weightbearing radiographs can be used on non-weightbearing MRI scans for the diagnosis of HV and to classify its severity. If the measurements using MRI scans proved reliable, it might be unnecessary in certain cases to order radiographs and subject the patient to radiation. The primary goal of the present study was to assess the mean differences in the HVA and IMA when comparing radiographs and MRI scans in the same patient to determine whether similar measurements could be used reliably with MRI as are used on radiographs. A secondary goal was to establish the interreader reliability. We hypothesized that similar measurements could be used on MRI for the assessment, diagnosis, and grading of HV.
1067-2516/$ - see front matter © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved. https://doi.org/10.1053/j.jfas.2017.10.009
ARTICLE IN PRESS 2
N. Heineman et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Fig. 1. Measurement of hallux valgus angle (HVA) and intermetatarsal angle (IMA) on a radiograph and magnetic resonance image. Line a passes through the midpoint of the first metatarsal head and base, line b passes through the midpoint of the head and base of the first proximal phalanx, and line c passes through the midpoint of the second metatarsal head and base. The HVA measures the acute angle between lines a and b, and the IMA measures the acute angle between lines a and c.
Patients and Methods The institutional review board approved our retrospective study, and informed consent was waived. We used Primordial® software (San Mateo, CA) to retrospectively query our patient database and found 90 consecutive patients who had undergone MRI of the forefoot and midfoot between April 27, 2015 and March 9, 2016. Our inclusion criteria were age 18 to 100 years and weightbearing radiographs (anteroposterior and lateral) and non-weightbearing MRI scans of the same foot available, with both studies generated within 3 months. The exclusion criteria included recent foot trauma or orthopedic hardware (n = 19) and poor image quality (n = 15). The final sample was 56 patients who met the inclusion and exclusion criteria. For every patient, the HVA and IMA were measured by 2 of us (N.H., L.Z.) on weightbearing radiographs and non-weightbearing MRI scans for a total of 4 angles using Philips Intellispace® PACS software (Best, The Netherlands). The HVA is defined as the angle formed by the axis of the first metatarsal and the axis of the first proximal phalanx (Fig. 1). The IMA is defined as the angle formed by the axis of the first metatarsal and the axis of the second metatarsal (Fig. 1). The longitudinal axis (midline) was determined by connecting the centers of the bones as described by Miller (9). To ensure the quality of our data and provide a benchmark for the reproducibility of each measurement, 2 of us (N.H., medical student; L.Z., radiologist with 8 years’ experience) independently measured each angle on the radiographs and MRI scans in a separate setting. The readers were trained on an initial set of 6 cases from the sample under the direction of another senior radiologist (A.C.) with 20 years of radiology experience. The mean ± standard deviation were calculated for all variables. A 3-way mixed model was used to calculate the intraclass correlation coefficient, with 95% bootstrap confidence intervals (CIs) between readers and between the imaging modalities for HVA and IMA. Student’s t test was used to compare the 2 mean values, and a p value of < .05 was considered statistically significant. The mean differences between these values were calculated, and Bland-Altman plots were generated to quantify the magnitude and significance of any reported differences. We used both readers’ measurements to determine the mean HVA and IMA for each imaging modality.
Results
The mean angles and standard deviations for each reader for each imaging modality are reported in Table 3. The mean difference between the radiographs and MRI for the HVA was −1.15° ± 3.89° (p = .03). The mean difference between the radiographs and MRI for the IMA was −0.22° ± 2.10° (p = .43). This represented a small, but statistically significant, difference (p < .05) between the 2 imaging modalities for measuring the HVA but no significant difference for measuring the IMA. Bland-Altman plots (Figs. 2 and 3) were also generated, with 91% (51 of 56) of our data points within 2 standard deviations of the mean difference for the IMA and HVA (Table 4).
Table 1 Study cohort characteristics (N = 56 patients) Characteristic
n (%)
Sex Male Female Foot Left Right BMI (kg/m2) BMI group Normal or underweight (<25 kg/m2) Overweight (25 to 30 kg/m2) Obese (>30 kg/m2) NA Age (y) <65 >65
16 (28.6) 40 (71.4) 30 (53.6) 26 (46.4) 28.7 ± 7.0 19 (33.9) 15 (26.8) 20 (35.7) 2 (3.6) 54.2 ± 15.4 41 (73.2) 15 (26.8)
Abbreviations: BMI, body mass index; NA, not applicable.
The mean age of the study group was 54.2 ± 15.4 years. The mean body mass index was 28.7 ± 7.1 kg/m2, and 71.4% (40 of 56) were females. Of the 56 patients, 30 (53.6%) had undergone imaging of the left foot and 26 (46.4%) of the right foot (Table 1). The interreader reliability for the HVA was 0.96 (95% CI 0.94 to 0.98) and for the IMA was 0.89 (95% CI 0.8 to 0.95). The intermodality agreement (between the radiographs and MRI) using the same method was 0.91 (95% CI 0.89 to 0.93) for the HVA and 0.83 (95% CI 0.7 to 0.91) for the IMA (Table 2).
Table 2 Intraclass correlation coefficients Angle
Reader 1 Versus Reader 2
Radiograph Versus MRI
HVA IMA
0.96 (95% CI 0.94 to 0.98) 0.89 (95% CI 0.8 to 0.95)
0.91 (95% CI 0.89 to 0.93) 0.83 (95% CI 0.7 to 0.91)
Abbreviations: CI, confidence interval; HVA, hallux valgus angle; IMA, intermetatarsal angle; MRI, magnetic resonance imaging.
ARTICLE IN PRESS N. Heineman et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
3
Fig. 2. Bland-Altman plot of the hallux valgus angle (HVA). The y-axis represents the difference between the radiographic and magnetic resonance imaging measurements for each patient, and the x-axis represents the average of the same 2 data points. The solid line represents the mean difference between all patients, and the dashed lines define 2 standard deviations (SDs) to either side. Of the data points, 91% (n = 51) were within 2 SDs, with the data equally distributed to either side of the mean.
Fig. 3. Bland-Altman plot of the intermetatarsal angle (IMA). The y-axis represents the difference between the radiographic and magnetic resonance imaging measurements for each patient, and the x-axis represents the average of the same 2 data points. The solid line represents the mean difference between all patients, and the dashed lines define 2 standard deviations (SDs) to either side. Again, 91% (n = 51) of the data points were within 2 SDs, with the data equally distributed to either side of the mean. Table 3 Mean ± standard deviations of hallux valgus angles and intermetatarsal angles (N = 56 patients) Reader
1 2 Combined
HVA (°)
IMA (°)
Table 4 Mean differences in hallux valgus angle and intermetatarsal angle between imaging modalities (N = 56 patients)
Radiograph
MRI
Radiograph
MRI
Angle
15.8 ± 9.5 14.2 ± 8.6 15.0 ± 8.8
8.6 ± 3.1 9.0 ± 3.1 8.8 ± 2.9
9.32 ± 3.5 8.69 ± 3.5 9.0 ± 3.1
Mean Difference (Radiograph Minus MRI)
p Value
13.5 ± 8.9 14.13 ± 8.7 13.8 ± 8.7
HVA IMA
−1.15 ± 3.89 −0.22 ± 2.10
.0309 .4292
Abbreviations: HVA, hallux valgus angle; IMA, intermetatarsal angle; MRI, magnetic resonance imaging.
Abbreviations: HVA, hallux valgus angle; IMA, intermetatarsal angle; MRI, magnetic resonance imaging.
ARTICLE IN PRESS 4
N. Heineman et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Discussion Previous studies have shown the HVA and IMA to be strongly correlated with the severity of HV (7). These measurements play an important role in grading HV, planning for surgery, and assessing the postoperative success of the surgery. Both have been shown to have excellent interreader reliability on radiographs with an intraclass correlation for the HVA ranging from 0.89 to 0.99 and for the IMA ranging from 0.82 to 0.98 (7,10,11). We report these measurements using MRI scans and radiographs of the same patients. Our findings were consistent with the existing data, with excellent interreader reliability and intermodality reliability. Thus, both of these measurements are highly reproducible and can be taken on either modality. Because of the excellent interreader performance, we used both readers’ values when determining the mean HVA and IMA to calculate the mean differences. For the HVA, we found a small, but statistically significant, difference of −1.15 ± 3.89 (p = .03). Although a statistically significant difference, it might not be a clinically significant difference. HV is generally classified into categories (mild, moderate, severe) based on a broad ranges of angles, such that a difference of 1° would have little effect on the management of the disease. Moreover, 91% of the HVAs were within 2 standard deviations of the mean on the Bland-Altman plot, indicating that most of the data points were in statistical agreement. For the IMA, we found no statistically significant difference between the measurements (p = .43), and the intraclass correlation coefficient was slightly lower at 0.83. This is also in agreement with the existing data regarding radiographic measurements. No clinically significant difference was found between these 2 imaging modalities. Finally, 91% (n = 51) of the IMAs were within 2 standard deviations of the mean on the Bland-Altman plot, indicating that most of the data points were in statistical agreement. The key limitations of the present study included the retrospective nature of the study and that we did not evaluate the patients’ symptoms. On a review of the medical records, we found that most of these patients had undergone surgery for reasons other than a great toe abnormality, such as lesser toe plantar plate tear, Morton neuroma, and so forth. Therefore, a correlation with hallux pain or disability was not possible. Qualitative findings of cartilage assessment were also not assessed, because that will be the subject of future study. Finally, the
radiographs and MRI scans were not performed on the same day, although we ensured that both imaging studies had been obtained within 3 months of each other. In the future, correlation of the HVA and IMA on MRI with the soft tissue findings and malalignment of the sesamoid complex could generate important data regarding the severity and prognostication of HV. In conclusion, we found excellent interreader reliability for both the HVA and the IMA across both imaging modalities. No statistically significant difference was found between the weightbearing radiographs and non-weightbearing MRI scans when measuring the IMA, although the HVA measured ~1° smaller on the MRI scans. However, this difference is likely not large enough to be considered clinically significant. Therefore, over the range of values we tested, MRI can be considered a viable imaging modality for diagnosing and assessing the severity of HV. References 1. Nix S, Smith M, Vicenzino B. Prevalence of hallux valgus in the general population: a systematic review and meta-analysis. J Foot Ankle Res 3:21, 2010. 2. Wulker N, Mittag F. The treatment of hallux valgus. Dtsch Arztebl Int 109:857–867, 2012. 3. Haas M. Radiographic and biomechanical considerations of bunion surgery. In: Textbook of Bunion Surgery, pp. 23–62, edited by J Gerbert, T Sokoloff, Futura Publishing, New York, 1981. 4. Jahss M. Disorders of the hallux and first ray. In: Disorders of the Foot and Ankle: Medical and Surgical Management, pp. 943–1106, edited by M Jahss, WB Saunders, Philadelphia, 1991. 5. Rosen JS, Grady JF. Neuritic bunion syndrome. J Am Podiatr Med Assoc 76:641–644, 1986. 6. Coughlin MJ, Saltzman CL, Nunley JA, 2nd. Angular measurements in the evaluation of hallux valgus deformities: a report of the ad hoc committee of the American Orthopaedic Foot & Ankle Society on angular measurements. Foot Ankle Int 23:68–74, 2002. 7. Lee KM, Ahn S, Chung CY, Sung KH, Park MS. Reliability and relationship of radiographic measurements in hallux valgus. Clin Orthop Relat Res 470:2613–2621, 2012. 8. Ashman CJ, Klecker RJ, Yu JS. Forefoot pain involving the metatarsal region: differential diagnosis with MR imaging. Radiographics 21:1425–1440, 2001. 9. Miller JW. Distal first metatarsal displacement osteotomy: its place in the schema of bunion surgery. J Bone Joint Surg Am 56:923–931, 1974. 10. Seo JH, Ahn JY, Boedijono D. Point-connecting measurements of the hallux valgus deformity: a new measurement and its clinical application. Yonsei Med J 57:741– 747, 2016. 11. Srivastava S, Chockalingam N, El Fakhri T. Radiographic angles in hallux valgus: comparison between manual and computer-assisted measurements. J Foot Ankle Surg 49:523–528, 2010.
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery j o u r n a l h o m e p a g e : w w w. j f a s . o r g
Original Research
Hallux Valgus Evaluation on MRI: Can Measurements Validated on Radiographs Be Used? Nathan Heineman, BA 1, Yin Xi, PhD 2, Lihua Zhang, MD 3, Riham Dessouky, MBBCh, Msc 4, Jed Hummel, MD 5, Justin Skweres, MD 5, Dane Wukich, MD 6, Avneesh Chhabra, MD 7 1Medical
Student, Department of Radiology, UT Southwestern Medical Center, Dallas, TX Professor, Department of Radiology, UT Southwestern Medical Center, Dallas, TX Professor, Department of Radiology, Peking University Third Hospital, Beijing, China 4 Assistant Professor, Department of Radiology, Faculty of Medicine, Zagazig University, Zagazig, Egypt 5Fellow, Department of Radiology, UT Southwestern Medical Center, Dallas, TX 6 Professor and Chairman, Department of Orthopaedic Surgery, UT Southwestern Medical Center, Dallas, TX 7Associate Professor and Musculoskeletal Division Chief, Department of Radiology, UT Southwestern Medical Center, Dallas, TX 2Assistant
3Associate
A R T I C L E
I N F O
Level of Clinical Evidence: 3
Keywords: bunion hallux valgus angle intermetatarsal angle magnetic resonance image radiograph
A B S T R A C T
Hallux valgus (HV) is a common deformity of the great toe affecting >23% of adults in the United States. The severity of the deformity is traditionally analyzed using radiographs to determine measurements such as the HV and intermetatarsal angles. We sought to determine the relationship between the radiographic and magnetic resonance imaging (MRI) measurements because this is not yet known. Two of us analyzed a series of 56 consecutive patients who had had radiographs and MRI performed on the same foot between April 27, 2015 and March 9, 2016 and who satisfied all other inclusion and exclusion criteria (age 18 to 100 years, no history of recent foot trauma, and no metal hardware in the foot). We found excellent interreader reliability (intraclass correlation 0.89 to 0.96) and intermodality agreement (intraclass correlation 0.83 to 0.91). The HV angle measured 15.0° ± 8.8° on the MRI scans and 13.8° ± 8.7° on the radiographs (mean difference −1.15° ± 3.89°), and the intermetatarsal angle was 9.0° ± 3.1° on the MRI scans and 8.8° ± 2.9° on the radiographs (mean difference −0.22° ± 2.10°). The HV measurements were reliable on both radiographs and MRI for the range of values tested. Small intermodality statistically significant differences in HV angle measurements were found; however, these might not be enough to be clinically significant. © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved.
Hallux valgus (HV), also known as a bunion, is a relatively common condition in the United States, affecting approximately 25% of adults aged 18 to 65 years and 36% of adults aged >65 years (1). The pathoanatomy is defined by medial deviation of the first metatarsal, lateral deviation of the hallux, and a prominent metatarsal head (2). This deformity is associated with an aching pain in the metatarsal head that can be sharply painful during ambulation and with shoe wear. It is also associated with a variety of painful and potentially disabling sequelae, including bursitis, cartilage degeneration, and nerve entrapment (3–5). In addition to the history and physical examination findings, a variety of radiologic findings are used to assess the severity of HV. These include the HV angle (HVA) and the intermetatarsal angle (IMA), among many others (6). The HVA and IMA have been shown to correlate best with Financial Disclosure: None reported. Conflict of Interest: A.C. is a consultant for ICON Medical. Address correspondence to: Avneesh Chhabra, MD, Department of Musculoskeletal Radiology, UT Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390-9178. E-mail address: [email protected] (A. Chhabra).
the magnitude of deformity (7). Although the angles have been well studied on radiographs, the magnetic resonance imaging (MRI) correlates of HV have not been described. MRI of the forefoot and midfoot has been increasingly used for the diagnosis of plantar plate tears, collateral ligament injuries, and cartilage assessment (8). This poses the question of whether radiographs would be necessary if the physician already planned to order an MRI scan. It is not known whether similar measurements used on the reference standard weightbearing radiographs can be used on non-weightbearing MRI scans for the diagnosis of HV and to classify its severity. If the measurements using MRI scans proved reliable, it might be unnecessary in certain cases to order radiographs and subject the patient to radiation. The primary goal of the present study was to assess the mean differences in the HVA and IMA when comparing radiographs and MRI scans in the same patient to determine whether similar measurements could be used reliably with MRI as are used on radiographs. A secondary goal was to establish the interreader reliability. We hypothesized that similar measurements could be used on MRI for the assessment, diagnosis, and grading of HV.
1067-2516/$ - see front matter © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved. https://doi.org/10.1053/j.jfas.2017.10.009
ARTICLE IN PRESS 2
N. Heineman et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Fig. 1. Measurement of hallux valgus angle (HVA) and intermetatarsal angle (IMA) on a radiograph and magnetic resonance image. Line a passes through the midpoint of the first metatarsal head and base, line b passes through the midpoint of the head and base of the first proximal phalanx, and line c passes through the midpoint of the second metatarsal head and base. The HVA measures the acute angle between lines a and b, and the IMA measures the acute angle between lines a and c.
Patients and Methods The institutional review board approved our retrospective study, and informed consent was waived. We used Primordial® software (San Mateo, CA) to retrospectively query our patient database and found 90 consecutive patients who had undergone MRI of the forefoot and midfoot between April 27, 2015 and March 9, 2016. Our inclusion criteria were age 18 to 100 years and weightbearing radiographs (anteroposterior and lateral) and non-weightbearing MRI scans of the same foot available, with both studies generated within 3 months. The exclusion criteria included recent foot trauma or orthopedic hardware (n = 19) and poor image quality (n = 15). The final sample was 56 patients who met the inclusion and exclusion criteria. For every patient, the HVA and IMA were measured by 2 of us (N.H., L.Z.) on weightbearing radiographs and non-weightbearing MRI scans for a total of 4 angles using Philips Intellispace® PACS software (Best, The Netherlands). The HVA is defined as the angle formed by the axis of the first metatarsal and the axis of the first proximal phalanx (Fig. 1). The IMA is defined as the angle formed by the axis of the first metatarsal and the axis of the second metatarsal (Fig. 1). The longitudinal axis (midline) was determined by connecting the centers of the bones as described by Miller (9). To ensure the quality of our data and provide a benchmark for the reproducibility of each measurement, 2 of us (N.H., medical student; L.Z., radiologist with 8 years’ experience) independently measured each angle on the radiographs and MRI scans in a separate setting. The readers were trained on an initial set of 6 cases from the sample under the direction of another senior radiologist (A.C.) with 20 years of radiology experience. The mean ± standard deviation were calculated for all variables. A 3-way mixed model was used to calculate the intraclass correlation coefficient, with 95% bootstrap confidence intervals (CIs) between readers and between the imaging modalities for HVA and IMA. Student’s t test was used to compare the 2 mean values, and a p value of < .05 was considered statistically significant. The mean differences between these values were calculated, and Bland-Altman plots were generated to quantify the magnitude and significance of any reported differences. We used both readers’ measurements to determine the mean HVA and IMA for each imaging modality.
Results
The mean angles and standard deviations for each reader for each imaging modality are reported in Table 3. The mean difference between the radiographs and MRI for the HVA was −1.15° ± 3.89° (p = .03). The mean difference between the radiographs and MRI for the IMA was −0.22° ± 2.10° (p = .43). This represented a small, but statistically significant, difference (p < .05) between the 2 imaging modalities for measuring the HVA but no significant difference for measuring the IMA. Bland-Altman plots (Figs. 2 and 3) were also generated, with 91% (51 of 56) of our data points within 2 standard deviations of the mean difference for the IMA and HVA (Table 4).
Table 1 Study cohort characteristics (N = 56 patients) Characteristic
n (%)
Sex Male Female Foot Left Right BMI (kg/m2) BMI group Normal or underweight (<25 kg/m2) Overweight (25 to 30 kg/m2) Obese (>30 kg/m2) NA Age (y) <65 >65
16 (28.6) 40 (71.4) 30 (53.6) 26 (46.4) 28.7 ± 7.0 19 (33.9) 15 (26.8) 20 (35.7) 2 (3.6) 54.2 ± 15.4 41 (73.2) 15 (26.8)
Abbreviations: BMI, body mass index; NA, not applicable.
The mean age of the study group was 54.2 ± 15.4 years. The mean body mass index was 28.7 ± 7.1 kg/m2, and 71.4% (40 of 56) were females. Of the 56 patients, 30 (53.6%) had undergone imaging of the left foot and 26 (46.4%) of the right foot (Table 1). The interreader reliability for the HVA was 0.96 (95% CI 0.94 to 0.98) and for the IMA was 0.89 (95% CI 0.8 to 0.95). The intermodality agreement (between the radiographs and MRI) using the same method was 0.91 (95% CI 0.89 to 0.93) for the HVA and 0.83 (95% CI 0.7 to 0.91) for the IMA (Table 2).
Table 2 Intraclass correlation coefficients Angle
Reader 1 Versus Reader 2
Radiograph Versus MRI
HVA IMA
0.96 (95% CI 0.94 to 0.98) 0.89 (95% CI 0.8 to 0.95)
0.91 (95% CI 0.89 to 0.93) 0.83 (95% CI 0.7 to 0.91)
Abbreviations: CI, confidence interval; HVA, hallux valgus angle; IMA, intermetatarsal angle; MRI, magnetic resonance imaging.
ARTICLE IN PRESS N. Heineman et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
3
Fig. 2. Bland-Altman plot of the hallux valgus angle (HVA). The y-axis represents the difference between the radiographic and magnetic resonance imaging measurements for each patient, and the x-axis represents the average of the same 2 data points. The solid line represents the mean difference between all patients, and the dashed lines define 2 standard deviations (SDs) to either side. Of the data points, 91% (n = 51) were within 2 SDs, with the data equally distributed to either side of the mean.
Fig. 3. Bland-Altman plot of the intermetatarsal angle (IMA). The y-axis represents the difference between the radiographic and magnetic resonance imaging measurements for each patient, and the x-axis represents the average of the same 2 data points. The solid line represents the mean difference between all patients, and the dashed lines define 2 standard deviations (SDs) to either side. Again, 91% (n = 51) of the data points were within 2 SDs, with the data equally distributed to either side of the mean. Table 3 Mean ± standard deviations of hallux valgus angles and intermetatarsal angles (N = 56 patients) Reader
1 2 Combined
HVA (°)
IMA (°)
Table 4 Mean differences in hallux valgus angle and intermetatarsal angle between imaging modalities (N = 56 patients)
Radiograph
MRI
Radiograph
MRI
Angle
15.8 ± 9.5 14.2 ± 8.6 15.0 ± 8.8
8.6 ± 3.1 9.0 ± 3.1 8.8 ± 2.9
9.32 ± 3.5 8.69 ± 3.5 9.0 ± 3.1
Mean Difference (Radiograph Minus MRI)
p Value
13.5 ± 8.9 14.13 ± 8.7 13.8 ± 8.7
HVA IMA
−1.15 ± 3.89 −0.22 ± 2.10
.0309 .4292
Abbreviations: HVA, hallux valgus angle; IMA, intermetatarsal angle; MRI, magnetic resonance imaging.
Abbreviations: HVA, hallux valgus angle; IMA, intermetatarsal angle; MRI, magnetic resonance imaging.
ARTICLE IN PRESS 4
N. Heineman et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Discussion Previous studies have shown the HVA and IMA to be strongly correlated with the severity of HV (7). These measurements play an important role in grading HV, planning for surgery, and assessing the postoperative success of the surgery. Both have been shown to have excellent interreader reliability on radiographs with an intraclass correlation for the HVA ranging from 0.89 to 0.99 and for the IMA ranging from 0.82 to 0.98 (7,10,11). We report these measurements using MRI scans and radiographs of the same patients. Our findings were consistent with the existing data, with excellent interreader reliability and intermodality reliability. Thus, both of these measurements are highly reproducible and can be taken on either modality. Because of the excellent interreader performance, we used both readers’ values when determining the mean HVA and IMA to calculate the mean differences. For the HVA, we found a small, but statistically significant, difference of −1.15 ± 3.89 (p = .03). Although a statistically significant difference, it might not be a clinically significant difference. HV is generally classified into categories (mild, moderate, severe) based on a broad ranges of angles, such that a difference of 1° would have little effect on the management of the disease. Moreover, 91% of the HVAs were within 2 standard deviations of the mean on the Bland-Altman plot, indicating that most of the data points were in statistical agreement. For the IMA, we found no statistically significant difference between the measurements (p = .43), and the intraclass correlation coefficient was slightly lower at 0.83. This is also in agreement with the existing data regarding radiographic measurements. No clinically significant difference was found between these 2 imaging modalities. Finally, 91% (n = 51) of the IMAs were within 2 standard deviations of the mean on the Bland-Altman plot, indicating that most of the data points were in statistical agreement. The key limitations of the present study included the retrospective nature of the study and that we did not evaluate the patients’ symptoms. On a review of the medical records, we found that most of these patients had undergone surgery for reasons other than a great toe abnormality, such as lesser toe plantar plate tear, Morton neuroma, and so forth. Therefore, a correlation with hallux pain or disability was not possible. Qualitative findings of cartilage assessment were also not assessed, because that will be the subject of future study. Finally, the
radiographs and MRI scans were not performed on the same day, although we ensured that both imaging studies had been obtained within 3 months of each other. In the future, correlation of the HVA and IMA on MRI with the soft tissue findings and malalignment of the sesamoid complex could generate important data regarding the severity and prognostication of HV. In conclusion, we found excellent interreader reliability for both the HVA and the IMA across both imaging modalities. No statistically significant difference was found between the weightbearing radiographs and non-weightbearing MRI scans when measuring the IMA, although the HVA measured ~1° smaller on the MRI scans. However, this difference is likely not large enough to be considered clinically significant. Therefore, over the range of values we tested, MRI can be considered a viable imaging modality for diagnosing and assessing the severity of HV. References 1. Nix S, Smith M, Vicenzino B. Prevalence of hallux valgus in the general population: a systematic review and meta-analysis. J Foot Ankle Res 3:21, 2010. 2. Wulker N, Mittag F. The treatment of hallux valgus. Dtsch Arztebl Int 109:857–867, 2012. 3. Haas M. Radiographic and biomechanical considerations of bunion surgery. In: Textbook of Bunion Surgery, pp. 23–62, edited by J Gerbert, T Sokoloff, Futura Publishing, New York, 1981. 4. Jahss M. Disorders of the hallux and first ray. In: Disorders of the Foot and Ankle: Medical and Surgical Management, pp. 943–1106, edited by M Jahss, WB Saunders, Philadelphia, 1991. 5. Rosen JS, Grady JF. Neuritic bunion syndrome. J Am Podiatr Med Assoc 76:641–644, 1986. 6. Coughlin MJ, Saltzman CL, Nunley JA, 2nd. Angular measurements in the evaluation of hallux valgus deformities: a report of the ad hoc committee of the American Orthopaedic Foot & Ankle Society on angular measurements. Foot Ankle Int 23:68–74, 2002. 7. Lee KM, Ahn S, Chung CY, Sung KH, Park MS. Reliability and relationship of radiographic measurements in hallux valgus. Clin Orthop Relat Res 470:2613–2621, 2012. 8. Ashman CJ, Klecker RJ, Yu JS. Forefoot pain involving the metatarsal region: differential diagnosis with MR imaging. Radiographics 21:1425–1440, 2001. 9. Miller JW. Distal first metatarsal displacement osteotomy: its place in the schema of bunion surgery. J Bone Joint Surg Am 56:923–931, 1974. 10. Seo JH, Ahn JY, Boedijono D. Point-connecting measurements of the hallux valgus deformity: a new measurement and its clinical application. Yonsei Med J 57:741– 747, 2016. 11. Srivastava S, Chockalingam N, El Fakhri T. Radiographic angles in hallux valgus: comparison between manual and computer-assisted measurements. J Foot Ankle Surg 49:523–528, 2010.