Accuracy of Rod Contouring to Desired Angles With and Without a Template: Implications for Achieving Desired Spinal Alignment and Outcomes

Abstract

Study Design:

Biomechanical Study.

Objective:

The search for optimal spinal alignment has led to the development of sophisticated formulas and software for preoperative planning. However, preoperative plans are not always appropriately executed since rod contouring during surgery is often subjective and estimated by the surgeon. We aimed to assess whether rods contoured to specific angles with a French rod bender using a template guide will be more accurate than rods contoured without a template.

Methods:

Ten experienced spine surgeons were requested to contour two 125 × 5.5 mm Ti64 rods to 40°, 60° and 80° without templates and then 2 more rods using 2D metallic templates with the same angles. Rod angles were then measured for accuracy and compared.

Results:

Average angles for rods bent without a template to 40°, 60° and 80° were 60.2°, 78.9° and 97.5°, respectively. Without a template, rods were overbent by a mean of 18.9°. When using templates of 40°, 60° and 80°, mean bend angles were 41.5°, 59.1° and 78.7°, respectively, with an average underbend of 0.2°. Differences between the template and non-template groups for each target angle were all significant (p < 0.001).

Conclusions:

Without the template, surgeons tend to overbend rods compared to the desired angle, while surgeons improved markedly with a template guide. This tendency to overbend could have significant impact on patient outcomes and risk of proximal junctional failure and warrants further research to better enable surgeons to more accurately execute preoperative alignment plans.

Keywords

biomechanics proximal junctional kyphosis rod bending sagittal alignment spinal alignment spinal instrumentation spine surgery

Introduction

Achieving appropriate sagittal alignment is considered one of the most important objectives in spinal deformity surgery in order to help optimize patient-reported outcomes and to reduce the risks of junctional failure and need for revision surgery.^1-6 This has led surgeons to pursue defined spino-pelvic targets in order to achieve the desired postoperative alignment for each patient. Restoration of known parameters such as pelvic tilt (PT), sagittal vertical axis (SVA), T1-pelvic angle (TPA) and pelvic incidence to lumbar lordosis mismatch (PI-LL) has been associated with improved functional status and walking tolerance, as well as reduced risk of requiring revision surgery.^2,5,7-9 Hence, sophisticated planning tools and software have been developed to simplify preoperative assessment of alignment and enable surgeons to create specific alignment plans for each patient.^10-12

Spinal rod contouring and biomechanical properties are key components of posterior instrumentation outcomes.¹³ Achieving target overall realignment and desired regional sagittal angles is dependent on adequate contouring of the rods, reflecting the critical importance of the intraoperative rod bending step. However, despite the available sophisticated planning tools and alignment objectives, surgeons often lack the means of ensuring their plans are appropriately executed in the operating room and end up estimating rod angles by subjectively contouring rods intraoperatively. This can lead to both over and under-bending the rods once in the operating room, despite knowing in advance the desired alignment measures.

Numerous studies have reported that both under- and over-correcting lumbar lordosis can have a deleterious impact on spino-pelvic alignment and patient-reported outcomes.^1,5,10 Identifying significant differences between desired angles and the actual rod curvature estimated by the surgeon could carry significant implications for in vivo spinal deformity correction. The purpose of this study was to assess how accurately surgeons can bend rods to specified angles. We then assessed whether providing surgeons with a rod template reflecting the desired angle would improve their ability to achieve the requested angular bend.

Materials and Methods

Ten experienced spine surgeons were requested to contour two 125 × 5.5 mm Ti64 rods using a French rod bender to 40°, 60° and 80° without templates. The same surgeons were then asked to bend 2 additional rods using 2D metallic templates that had been bent to the corresponding requested angles. Rods were photographed overhead next to an object of known size (coin) to calibrate measurements, and photos were imported into Surgimap (Nemaris, New York, NY, USA) for measurements (Figure 1).¹⁴ The arc length of the rods was measured using the multi-line tool in Surgimap. The circle tool was used to overlay a circle of best fit to the center line of the bent rod and measure its radius. The angle of the bent rod was calculated using the formula s=rØ, where s is the arc length, r is the radius and Ø is the angle in radians converted to degrees. Rod angles were compared for average and absolute differences as well as error percentage between both groups. Descriptive statistics were performed using Minitab 17 (Minitab, State College, PA, USA), data was presented as mean value ± standard deviation, p-values were calculated using a 2-sample t-tests and p < 0.05 was considered to indicate statistical significance. Since this study was confined to biomechanical analyses without involvement of patients, patient material, or animals, Internal Review Board approval was not required and patient consent was not indicated.

Figure 1.

Rod photographed next to a coin for size calibration to facilitate angle measurement.

Results

A total of 120 rods were bent, one-half using 2D metallic templates of 40°, 60° and 80° as models and the other one-half were bent without a template. Mean angle values of the rods bent without a template to 40°, 60° and 80° were as follows: 60.2 ± 13.9° (range: 82.6°-37.8°), 78.9 ± 12.8° (range: 98.4°-56.8°), and 97.5 ± 18.6° (range: 134.2°-73°), respectively. Mean angle values of the rods bent using templates of 40°, 60° and 80° were as follows: 41.5 ± 2.5° (range: 47.2°-38.7°), 59.1 ± 2.3° (range: 65.1°-54.7°), and 78.7 ± 2.7° (range: 82.2°-72.5°), respectively (Table 1). For each of the desired angular bends, the differences between the achieved angular bends with and without a template were statistically significant (p < 0.001 for each, Figure 2).

Table 1.

Angular Measurements of Bent Rods With and Without a Template.

Target Angle	Template use	Average Bend Angle (°)	Standard Deviation (°)	Max Angle (°)	Min Angle (°)	Average Difference from Target Angle (°)	Error (%)
40°	With a Template	41.5	2.5	47.2	38.7	1.5	3.7
40°	Without a template	60.2	13.9	82.6	37.8	20.2	36.7
60°	With a Template	59.1	2.3	65.1	54.7	-0.9	1.4
60°	Without a template	78.9	12.8	98.4	56.8	18.9	31.4
80°	With a Template	78.7	2.7	82.2	72.5	-1.3	1.6
80°	Without a template	97.5	18.6	134.2	73.0	17.5	21.9

Figure 2.

Average bend angle results. Comparison of achieved angular bends with and without a template for the 40º, 60º, 80º and overall groups.

The average overbend without a template was 18.9 ± 15.1°, which is an average error of 30% (Table 1). In contrast, when using the templates there was a relatively small underbend that averaged −0.2 ± 2.5°, with an average error of only 2.3% (Figure 3). The angular bends achieved without the use of a template guide by surgeon are illustrated in Figure 4, and the average differences from the target angular bends achieved without and with the use of a template guide for each surgeon are graphically depicted in Figure 5.

Figure 3.

Average difference from target angle. Shown are mean differences from the target angle when surgeons were asked to bend rods to 40º, 60º or 80º with or without a template guide.

Figure 4.

Angular bends achieved without the use of a template guide for each surgeon. Data is grouped by the requested angular bend (40º, 60º or 80º), 2 points are shown for each surgeon for each angle, since each bend was performed in duplicate.

Figure 5.

Average differences from target angular bends achieved without (upper panel) and with (lower panel) the use of a template guide for each surgeon.

Discussion

Adult spinal deformity (ASD) is a common disease that has progressively increased over recent decades, partly due to a rise in the elderly population.¹⁵ Increasing prevalence of ASD has led to an upsurge in the number of spinal instrumentation procedures performed globally; and with them, greater attention has been directed at surgical outcome measures and goals. There is now substantial evidence that patient-reported outcomes measures correlate with sagittal alignment. Regardless of its etiology, sagittal malalignment implies alteration of spino-pelvic parameters that can compromise the ability to maintain an upright posture within the cone of economy, without significantly greater energy expenditure.^16,17 In particular, PI-LL mismatch, PT, and SVA are among the strongest radiographic predictors currently available that correlate with surgical outcomes, disability and quality of life.^1,3,18

Operative treatment goals for ASD include restoration of appropriate spino-pelvic alignment. Correction of lumbar lordosis is especially critical, since it has direct impact on SVA and mismatch reduction, as well as an indirect effect on PT, which serves as a compensatory measure for sagittal malalignment. To directly address the mismatch between PI and LL, patients often require osteotomies, soft tissue releases, interbody implants and instrumentation to gain sufficient lordosis. To accurately plan desired corrections, sagittal alignment analysis is of vital importance and many digital tools have been designed to aid surgeons when making preoperative decisions. Computer-based measurements have been shown to be faster, more precise and most surgeons at least recognize their importance when determining the degree of correction required to restore sagittal alignment.^10,12,19,20 Nonetheless, despite optimal preoperative planning, surgeons often lack the means of ensuring that the desired plan is appropriately executed in the operating room.

In the present study, when surgeons were asked to bend rods without a reference template, not only was there considerable variation in the angles, but nearly all of them tended to overbend the rods, especially as the target angles decreased. When aiming for 40°, 60° and 80° without the template there were mean differences from the target angles of 20.2°, 18.9° and 17.5°, respectively. These discrepancies between planned curvatures and the actual achieved angles might be a major contributing factor to reports in the literature in which in vivo correction not uncommonly differs from the preoperative alignment plans. In a study that included 161 patients with ASD, Moal and colleagues found that 48% of patients had inadequate lumbar lordosis improvement after surgery and only 23% of patients had complete global radiographic correction of the deformity.²¹ The arduous challenge to achieve desired angles could be explained, in part, by the difficulty in estimating rod curvatures intraoperatively, that when inadequate, can cause postoperative malalignment. Furthermore, accurate rod bending is not just critical for achieving appropriate lumbar lordosis, in procedures including the thoracic spine, there is also the need to contour the rod to facilitate achievement of appropriate thoracic kyphosis.

Proximal junctional kyphosis (PJK) is among the most common complications in patients who undergo long-segment spinal arthrodesis for ASD and arguably remains the biggest unsolved challenge with these procedures. PJK is associated with high rates of postoperative loss of sagittal alignment, pain, neurological deficit, instability and need for reintervention.^22-26 Although it is a complex, multi-factorial process not yet completely understood, there is evidence supporting a higher incidence of PJK in patients with significantly greater corrections in lumbar lordosis and sagittal vertical axis.^6,22 In a retrospective analysis of 679 ASD patients with fusions to the pelvis stratified by age group, Lafage et al assessed postoperative offsets from age-specific alignment norms and found that patients who developed PJK were significantly over-corrected when compared to age-adjusted sagittal alignment goals.⁶ If one extrapolates to a surgical scenario the degree of overcorrection seen when the rods were bent without a template in the present study, this could mean patients are often receiving rods with more lordosis than they may actually need. Since theoretically, achieving the correct regional sagittal contour is dependent on the correct angle of the rod, this means patients could end up with more postoperative lordosis than expected and therefore worse postoperative outcomes and higher rates of revision surgery. This is particularly relevant to the elderly population, who are more prone to overcorrection associated complications due to poorer bone quality, weaker soft-tissue support, and overall increased frailty.⁶ In contrast, when surgeons in our study were asked to bend the rods with the aid of templates, there was an average mean difference of just −0.2º from the target angle, which is only a 2.3% error.

Unless patient-specific rods are used, implanted rods are not in their originally straight manufactured form after in situ or French rod benders are used to contour them into the desired shape; causing defects or notches that may ultimately lower their fatigue life.^21,27,28 In fact, this “notch sensitivity” may help to explain why rod failure typically occurs at the screw-rod junction or at the apex of the curve where greater contouring has been done.^28,29 Our study revealed that nearly all participating surgeons overbent the rods when they did not use a template. In fact, there was an overall 18.9° overbend for all angles, which translates into a 30% angular error. This means rods were potentially subject to an unnecessary greater number of notches by the French bender to deliver the curvature deformation. When translating this to an in vivo environment, the more notching caused by overbending could affect the fatigue life of the rods and contribute to implant failure. In addition, a previous biomechanical study from Tang and colleagues³⁰ demonstrated that rods bent to greater angles had lower fatigue life, suggesting that rods bent to angles greater than necessary may needlessly decrease the fatigue life of the rod and increase the risk of rod fracture.

The importance of appropriate rod contouring is not confined to ASD corrective surgeries or to long-segment instrumentation. As with long-segment fusions, the alignment achieved with shorter-segment fusions is impacted by rod contouring in addition to the bony, disc, and soft tissue releases. Achieving adequate lumbar lordosis for shorter-segment fusions has been shown to have significant impact on the occurrence of adjacent segment disease. Rothenfluh and colleagues³¹ assessed 84 patients who had been treated with a short-segment (between 1 and 3 levels) posterior lumbar fusion. Patients with a postoperative PI-LL mismatch had a 10-fold greater risk of requiring revision surgery for adjacent segment disease.

Some limitations of this study should be noted. Rods were only bent in a single angular direction, however ASD patients often need rods to have both lordosis and kyphosis which could entail further discrepancies between desired contours and the actual rod curvature. Furthermore, surgeons were only asked to achieve a specific angle, but no information was given regarding the location of the rod or other specific visual cues that may be available intraoperatively; if surgeons were asked to bend rods to a certain degree of kyphosis versus lordosis, contouring estimates could change. However, this requires a larger sample and further research.

Conclusion

Obtaining adequate postoperative spinal alignment that matches preoperative planned alignment remains a challenge for surgeons. Analysis of angle difference between rods bent without a template versus rods bent with a template revealed significant improvement in accuracy when using the latter. Without the template, surgeons tended to overbend rods compared to the desired angle, while nearly all surgeons improved markedly with use of one. This overbending could have significant implications for patient outcomes and PJK risk and warrants further research to help enable surgeons to better execute preoperative alignment plans into the operative setting.

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article. CPA reports personal fees from Stryker, personal fees from Biomet Zimmer Spine, personal fees from DePuy Synthes, personal fees from Nuvasive, personal fees from Next Orthosurgical, personal fees from K2 M, personal fees from Medicrea, personal fees from DePuy Synthes, personal fees from Medtronic, personal fees from Medicrea, personal fees from K2 M, personal fees from Titan Spine, personal fees from DePuy Synthes, personal fees from ISSG, personal fees from Operative Neurosurgery, personal fees from SRS, personal fees from ISSG, personal fees from Global Spinal Analytics, personal fees from UCSF, outside the submitted work. CG reports consultancy fees from Stryker, Medtronic; serves on advisory boards for Stryker, Augmedics, and Medtronic; receives royalties from Stryker; holds stock in Augmedics; is on the board for National Spine Health Foundation; and receives institutional research funding from Medtronic. BD reports consultancy fees from Stryker. PK reports consultancy fees from Metronic; serves on advisory board for Medtronic, Stryker, and Depuy; receives royalties from Stryker and Globus. JG reports consultancy fees from Medtronic, Acuity, Stryker, NuVasive, and Mazor; received royalties from Acuity and NuVasive; serves on advisory board for Medtronic, Stryker, and National Spine Health Foundation; patent(s) with Medtronic; receives research support from Medtronic, Norton Healthcare, Intellirod Spine Inc, Pfizer, Texas Scottish Rite Hospital, Alan L. & Jacqueline B Stuart Spine Research, Cerpedics Inc, and Scoliosis Research Society; serves on speakers bureau for DePuy Synthes; has received honoraria from Pacira Pharmaceuticals, Baxter, Broadwater, and NASS; and is a reviewer for The Spine Journal and Spine Deformity. DD reports consultancy fees from SeaSpine, Medtronic, and Medicrea; receives royalties from SeaSpine, Astura Spine, and Medicrea; serves on advisory board for Stryker and Medtronic; receives educational/research support from Stryker and Medtronic (Medicrea). MBB reports consultancy fees from Stryker, Spineguard, SI-Bone, and Cousin; and received research grants from Medtronic, Stryker, Nuvasive, and SI-Bone. RL reports consultancy fees from Stryker and SI-Bone; receives royalties from Stryker; serves on speakers bureau for Stryker and SI-Bone; serves on advisory boards for Stryker and SI-Bone. CPB is an employee of Stryker. SB reports research support from Depuy Synthes, Stryker, Globus, Medtronic, SI Bone, Mirus, and ISSGF; reports consultancy fees from Stryker; receives royalties from Stryker and NuVasive; and holds stock in Carlsmed and Progenerative Medicine. JSS reports consultancy fees from Zimmer Biomet, Nuvasive, DePuy Synthes, Stryker, and Carlsmed; receives royalties from Zimmer Biomet, Nuvasive, and Thieme; holds stock in Alphatec and NuVasive; receives research funding to his institution from DePuy Synthes, International Spine Study Group Foundation (ISSGF), and AOSpine; receives fellowship grant funding to his institution from AOSpine; serves on the editorial boards of Journal of Neurosurgery Spine, Neurosurgery, Operative Neurosurgery, and Spine Deformity; and serves on Board of Directors of the Scoliosis Research Society. JPS and SC report no conflicts of interest.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study funded and completed through Stryker Spine. Please see following 2 pages for COI disclosures for each author.

ORCID iD

Justin S. Smith, MD, PhD

References

Glassman

Bridwell

Dimar

Horton

Berven

Schwab

. The impact of positive sagittal balance in adult spinal deformity. Spine (Phila Pa 1976). 2005;30(18):2024–2029.

Ryan

Protopsaltis

Ames

, et al. T1 pelvic angle (TPA) effectively evaluates sagittal deformity and assesses radiographical surgical outcomes longitudinally. Spine (Phila Pa 1976). 2014;39(15):1203–1210.

Schwab

Blondel

Bess

, et al. Radiographic spinopelvic parameters and disability in the setting of adult spinal deformity: a prospective multicenter analysis. Spine (Phila Pa 1976). 2013;38(13):E803–E812.

Klineberg

Schwab

Smith

Gupta

Lafage

Bess

. Sagittal spinal pelvic alignment. Neurosurg Clin N Am. 2013;24(2):157–162.

Lafage

Schwab

Patel

Hawkinson

Farcy

. Pelvic tilt and truncal inclination: two key radiographic parameters in the setting of adults with spinal deformity. Spine (Phila Pa 1976). 2009;34(17): E599–E606.

Lafage

Schwab

Glassman

, et al. Age-adjusted alignment goals have the potential to reduce PJK. Spine (Phila Pa 1976). 2017;42(17):1275–1282. doi:10.1097/BRS.0000000000002146

Sardi

Camacho

Diaz

Berbeo

. The Berbeo-Sardi angle (BSA): an innovative method to effectively estimate pelvic retroversion in anteroposterior Radiographs—a correlation with traditional parameters. Spine Deform. 2018;6(2):105–111.

Smith

Klineberg

Schwab

, et al. Change in classification grade by the SRS-Schwab adult spinal deformity classification predicts impact on health-related quality of life measures: prospective analysis of operative and non-operative treatment. Spine. 2013;38(19):1663–1671.

Schwab

Lafage

Patel

Farcy

. Sagittal plane considerations and the pelvis in the adult patient. Spine (Phila Pa 1976). 2009;34(17):1828–1833.

10.

Lafage

Ferrero

Henry

, et al. Validation of a new computer-assisted tool to measure spino-pelvic parameters. Spine J. 2015;15(12):2493–2502. doi:S1529-9430(15)01373-X [pii]

11.

Blondel

Schwab

Bess

, et al. Posterior global malalignment after osteotomy for sagittal plane deformity: it happens and here is why. Spine (Phila Pa 1976). 2013;38(7):394. doi:10.1097/BRS.0b013e3182872415 [doi]

12.

Mok

Berven

Diab

Hackbarth

Deviren

. Comparison of observer variation in conventional and three digital radiographic methods used in the evaluation of patients with adolescent idiopathic scoliosis. Spine (Phila Pa 1976). 2008;33(6):681–686. doi:10.1097/BRS.0b013e318166aa8d [doi]

13.

Cidambi

Glaser

Bastrom

Nunn

Ono

Newton

. Postoperative changes in spinal rod contour in adolescent idiopathic scoliosis: an in vivo deformation study. Spine (Phila Pa 1976). 2012;37(18):1566–1572. doi:10.1097/BRS.0b013e318252ccbe [doi].

14.

Akbar

Terran

Ames

Lafage

Schwab

. Use of Surgimap spine in sagittal plane analysis, osteotomy planning, and correction calculation. Neurosurg Clin N Am. 2013;24(2):163–172. doi:S1042-3680(12)00146-5 [pii]

15.

O’Lynnger

Zuckerman

Morone

Dewan

Vasquez-Castellanos

Cheng

. Trends for spine surgery for the elderly: Implications for access to healthcare in north America. Neurosurgery. 2015;77(Suppl 4):S136–142. doi:10.1227/NEU.0000000000000945 [doi]

16.

Dubousset

. Three-dimensional analysis of the scoliotic deformity. In: Weinstein

, ed. The Pediatric Spine: Principles and Practice. Raven Press; 1994:479–496.

17.

Ames

Smith

Scheer

, et al. Impact of spinopelvic alignment on decision making in deformity surgery in adults: a review. J Neurosurg Spine. 2012;16(6):547–564.

18.

Schwab

Smith

Biserni

Gamez

Farcy

Pagala

. Adult scoliosis: a quantitative radiographic and clinical analysis. Spine (Phila Pa 1976). 2002;27(4):387–392. doi:10.1097/00007632-200202150-00012 [doi]

19.

Ilharreborde

Steffen

Nectoux

, et al. Angle measurement reproducibility using EOS three-dimensional reconstructions in adolescent idiopathic scoliosis treated by posterior instrumentation. Spine (Phila Pa 1976). 2011;36(20):E1306–1313. doi:10.1097/BRS.0b013e3182293548 [doi]

20.

Ailon

Scheer

Lafage

, et al. Adult spinal deformity surgeons are unable to accurately predict postoperative spinal alignment using clinical judgment alone. Spine Deform. 2016;4(4):323–329. doi:S2212-134X(16)00039-3 [pii]

21.

Moal

Schwab

Ames

, et al. Radiographic outcomes of adult spinal deformity correction: a critical analysis of variability and failures across deformity patterns. Spine Deform. 2014;2(3):219–225. doi:S2212-134X(14)00006-9 [pii]

22.

Kim

Bridwell

Lenke

, et al. Patients with proximal junctional kyphosis requiring revision surgery have higher postoperative lumbar lordosis and larger sagittal balance corrections. Spine (Phila Pa 1976). 2014;39(9):E576–580. doi:10.1097/BRS.0000000000000246 [doi]

23.

Hyun

Lee

Park

Kim

Jahng

Kim

. Proximal junctional kyphosis and proximal junctional failure following adult spinal deformity surgery. Korean J Spine. 2017;14(4):126–132. doi:10.14245/kjs.2017.14.4.126 [doi]

24.

Kim

Iyer

. Proximal junctional kyphosis. J Am Acad Orthop Surg. 2016;24(5):318–326. doi:10.5435/JAAOS-D-14-00393 [doi]

25.

Glattes

Bridwell

Lenke

Kim

Rinella

Edwards

C II

. Proximal junctional kyphosis in adult spinal deformity following long instrumented posterior spinal fusion: Incidence, outcomes, and risk factor analysis. Spine (Phila Pa 1976). 2005;30(14):1643–1649. doi:00007632-200507150-00014 [pii]

26.

Kim

Bridwell

Lenke

Glattes

Rhim

Cheh

. Proximal junctional kyphosis in adult spinal deformity after segmental posterior spinal instrumentation and fusion: Minimum five-year follow-up. Spine (Phila Pa 1976). 2008;33(20):2179–2184. doi:10.1097/BRS.0b013e31817c0428 [doi]

27.

Lindsey

Deviren

Yeh

Puttlitz

. The effects of rod contouring on spinal construct fatigue strength. Spine (Phila Pa 1976). 2006;31(15):1680–1687. doi:00007632-200607010-00009 [pii]

28.

Yoshihara

. Rods in spinal surgery: a review of the literature. Spine J. 2013;13(10):1350–1358. doi:S1529-9430(13)00461-0 [pii]

29.

Nguyen

Buckley

Ames

Deviren

. The fatigue life of contoured cobalt chrome posterior spinal fusion rods. Proc Inst Mech Eng H. 2011;225(2):194–198. doi:10.1243/09544119JEIM763 [doi]

30.

Tang

Leasure

Smith

Buckley

Kondrashov

Ames

. Effect of severity of rod contour on posterior rod failure in the setting of lumbar pedicle subtraction osteotomy (PSO): a biomechanical study. Neurosurgery. 2013;72(2):276–282; discussion 283. doi:10.1227/NEU.0b013e31827ba066 [doi]

31.

Rothenfluh

Mueller

Rothenfluh

Min

. Pelvic incidence-lumbar lordosis mismatch predisposes to adjacent segment disease after lumbar spinal fusion. Eur Spine J. 2015;24(6):1251–1258. doi:10.1007/s00586-014-3454-0 [doi]