O

Ca

MSa

hb

ARA

KMHOP3

1

msoo

1

Orthopaedics & Traumatology: Surgery & Research 103 (2017) 245–250

Available online at

ScienceDirectwww.sciencedirect.com

riginal article

an three-dimensional patient-specific cutting guides be used tochieve optimal correction for high tibial osteotomy? Pilot study

. Muniera, M. Donneza,b, M. Olliviera, X. Flechera, P. Chabranda, J.-N. Argensona,. Parrattea,∗

UMR CNRS 787/AMU, institut du mouvement et de l’appareil locomoteur, groupe interdisciplinaire en biomécanique, hôpital Sainte-Marguerite, centreospitalo-universitaire Sud, boulevard de Sainte-Marguerite, 13009 Marseille, FranceNewclip Technics, 44115 Haute-Goulaine, France

a r t i c l e i n f o

rticle history:eceived 6 May 2016ccepted 30 November 2016

eywords:edial tibiofemoral osteoarthritisigh tibial osteotomypening-wedge osteotomyatient-specific surgical guidesD printing

a b s t r a c t

Introduction: Treatment of medial tibiofemoral osteoarthritis with a high-tibial osteotomy (HTO) is mosteffective when the optimal angular correction is achieved. However, conventional instrumentation islimited when multiplanar correction is needed.Hypothesis: Use of patient-specific cutting guides (PSCGs) for HTO provides an accurate correction (dif-ference < 2◦) relative to the preoperative planning.Materials and methods: Between February 2014 and February 2015, 10 patients (mean age: 46 years[range: 31–59]; grade 1 or 2 osteoarthritis in Ahlbäck’s classification) were included prospectively in thisreliability and safety study. All patients were operated using the same medial opening-wedge osteotomytechnique. Preoperative planning was based on long-leg radiographs and CT scans with 3D reconstruction.The PSGCs were used to align the osteotomy cut and position the screw holes for the plate. The desiredcorrection was achieved in the three planes when the holes on the plate were aligned with the holesdrilled based on the PSCG. Preoperatively, the mean HKA angle was 171.9◦ (range: 166–179◦), the meanproximal tibial angle was 87◦ (86–88◦) and the mean tibial slope was 7.8◦ (1–22◦). The postoperativecorrection was compared to the planned correction using 3D CT scan transformations. Intraoperativeand postoperative complications were assessed at a minimum follow-up of 1 year.Results: The procedure was successfully carried out in all patients with the PSCGs. On postoperative long-leg radiographs, the mean HKA was 182.3◦ (180–185◦); on the CT scan, the mean tibial mechanical anglewas 94◦ (90–98◦) and the mean tibial slope was 7.1◦ (4–11◦). In 19 out of 20 postoperative HKA and slopemeasurements, the difference between the planned and achieved correction was < 2◦ based on the 3Danalysis of the three planes in space; in the other case, the slope was 13◦ instead of the planned 10◦.The intra-class correlation coefficients between the postoperative and planned parameters were 0.98

[0.92–0.99] for the HKA and 0.96 [0.79–0.99] for the tibial slope. There were no surgical site infections;one patient had a postoperative hematoma that resolved spontaneously.Discussion: The results of this study showed that use of PSCGs in HTO procedures helps to achieve optimalcorrection in a safe and reliable manner.

spec

Level of evidence: IV – Pro

. Introduction

In patients with moderate tibiofemoral osteoarthritis and legisalignment [1], a high-tibial osteotomy (HTO) is an ideal con-

ervative treatment [2]. However, the angular correction must beptimal in all three planes for the osteotomy to be effective. Radiol-gy studies have revealed that an under-correction is common, and

∗ Corresponding author.E-mail address: sebastien@parratte.fr (S. Parratte).

http://dx.doi.org/10.1016/j.otsr.2016.11.020877-0568/© 2017 Elsevier Masson SAS. All rights reserved.

tive cohort study.© 2017 Elsevier Masson SAS. All rights reserved.

that it is more significant with opening-wedge than closing-wedgeosteotomies [3,4].

Computer-assisted surgery can solve this problem, but it is notdevoid of errors and complications [5]. The recent introduction ofpatient-specific cutting guides (PSCGs) has raised the possibility ofmaking instrumentation specific to each patient, which could resultin more accurate correction of the bone deformity.

The objectives of this study were:

• to evaluate the reliability of the method by comparing theachieved angular correction to the planned correction;

246 M. Munier et al. / Orthopaedics & Traumatology:

Table 1Patient population.

Patient series

Side Right (2)/Left (8)Preoperative HKA 172.8 (167.1–179.5)Preoperative tibial slope 8.8 (1.8–22.4)Preoperative mMPTA 87.1 (86–88.8)Ahlbäck stage 2 (1–2)Age 46 (31–59)BMI 28.6 (24.2–39.4)

B

•

f2p

2

Ptr(ri

1m(

uwpd

tendon.

Ftac

MI: body mass index.

to evaluate the intra- and postoperative complications specific tothis method.

We hypothesized that performing a tibial osteotomy with PSCGsollowing three-dimensional (3D) planning is accurate (within

degrees) without increasing the intra- and postoperative com-lications.

. Material and methods

All the patients undergoing medial opening wedge HTO withSCGs between February 2014 and February 2015 were enrolled inhis single-center, prospective, observational study. Inclusion crite-ia were: patient < 60 years of age with medial knee osteoarthritisAhlbäck ≤ 2 [6,7]) with significant genu varum. Exclusion crite-ia: advanced osteoarthritis (Ahlbäck ≥ 3) and artefacts that wouldnterfere with obtaining a high-quality CT scan (Table 1).

Preoperatively, the mean HKA angle was 171.9◦ (range66–179◦), the mean tibial slope was 7.5◦ (1.8–22.4◦) and theean mechanical medial proximal tibial angle (mMPTA) was 86.7◦

86–88◦).The first 10 osteotomies performed by two different surgeons

sing this technique during the enrollment period were analyzedith a minimum follow-up of 1 year. Patient consent was collected

reoperatively after they were informed of the procedure, in accor-ance with the principles of the Helsinki declaration.

ig. 1. Two-dimensional planning with measurements of the tibial slope, HKA angle andhe angle between the anterior tibial cortex and the axis of the medial tibial plateau on as the intersection of two femoral head diameters, the middle of the line joining the tibia: the mMPTA is the angle between the line joining the center of the ankle with the cente

Surgery & Research 103 (2017) 245–250

2.1. Two-dimensional preoperative planning

Preoperatively, all patients had standard X-ray views taken(weight bearing long-leg radiographs, Schuss view, sky-line view,A/P and lateral, stress views) and underwent a CT scan according toa protocol developed collaboratively with the imaging department.The surgeon took measurements and filled out an order form for theengineer that specified the correction objectives in the frontal andsagittal planes through variations in the hip knee ankle (HKA) angleand the tibial slope (Fig. 1).

2.2. Virtual HTO

The CT scan protocol consisted of one acquisition centered onthe femoral head, one on the knee that captured the distal femurand 15 cm of the proximal tibia, and one centered over the ankle.The slice thickness was 0.625 mm for the knee and 2 mm for thehip and ankle (GE Light Speed VCT64).

The position of the resection plane and the frontal and sagi-ttal corrections were confirmed by the surgeon. The HTO modelwas used to virtually position the Activmotion plate (NewclipTechnics

®, Haute-Goulaine, France) on the tibia while following

the manufacturer’s recommendations for ideal positioning on theanteromedial side of the tibia (Fig. 2). The design of the PSCGstook into account the resection plane and the position of the screwtunnels relative to the virtual positioning of the plate.

2.3. Surgical technique

The PSCGs were secured to the bone with two pins and thenfluoroscopy done to confirm the orientation of the osteotomy cut(Fig. 3). The six holes needed for the plate were drilled before per-forming the osteotomy. The osteotomy was performed with thePSCGs in place. The saw blade was guided by a specific portion ofthe PSCG and then the proximal portion of the guide was removedto finish the osteotomy in the same plane or two planes, depend-ing on the planned correction and the position of the patellar

The lateral cortex was weakened with a drill bit to preserve alateral hinge and then the osteotomy cut was gradually openedwith a lamina-spreader distractor until the pre-drilled screw holes

mechanical medial proximal tibial angle (mMPTA): a: the tibial slope is defined by lateral view of the knee; b: the HKA is measured on long-leg standing radiographsl intercondylar eminence and the middle of the articular surface of the distal tibia;r of the knee and the axis of the tibial plateau on long-leg standing radiographs.

M. Munier et al. / Orthopaedics & Traumatology: Surgery & Research 103 (2017) 245–250 247

Fig. 2. Three-dimensional planning. Performing the virtual osteotomy.

F holes

w

wtpocF

2

a

ig. 3. A. Osteotomy with oscillating saw after the guide has been secured and the

ith the plate. D. Fluoroscopy view.

ere aligned with the holes in the plate (ActivmotionTM and self-apping-screws, Newclip TechnicsTM, Haute-Goulaine, France). Thelate was secured using six screws of the size chosen during the pre-perative planning and then the bone defect filled with injectablealcium phosphate cement (Quickset, Graftys

®, Aix-en-Provence,

rance).

.4. Postoperative radiological and clinical follow-up

All patients were reviewed at 1 month postoperative to evalu-te any early complications, then at 3 months for regular follow-up

drilled. B. Verification of the osteotomy cut. C. Making sure the screw holes line up

with radiographs (long-leg standing, A/P and lateral) and CT scan.With CentricityTM software (GE), the X-ray views were used tocompare the resulting radiographic parameters (HKA, tibial slope,mMTPA) to those planned preoperatively (Fig. 4). The postopera-tive CT scan was used to perform a new 3D reconstruction of thetibia to calculate the mMTPA angle and tibial slope [8] and comparethem to the preoperative model (Fig. 5).

To verify the reproducibility and accuracy of the measured

angles, the analysis was repeated four times by a single observerand then again by a second independent observer to obtain theinter- and intra-observer intraclass correlation coefficient (ICC)(Table 2).

248 M. Munier et al. / Orthopaedics & Traumatology: Surgery & Research 103 (2017) 245–250

Fig. 4. Postoperative radiographic measurements (mMPTA, HKA and tibial slope).

Fig. 5. Preoperative (grey) and postoperative (white) 3D reconstruct

Table 2Inter- and intra-observer intraclass correlation coefficients (ICC).

Intra-observer ICC Inter-observer ICC

Mean HKA (min, max) 0.98 [0.95–0.99] 0.92 [0.85–0.97]Mean tibial slope (min, max) 0.98 [0.94–0.99] 0.91 [0.83–0.95]Mean mMPTA (min, max) 0.98 [0.94–0.99] 0.92 [0.86–0.96]

m

d

3

om

The difference with the target was measured using a 3D transfor-

MPTA: mechanical medial proximal tibial angle.

Patients were reviewed at 1 year to ensure there were noelayed complications.

. Results

The PSCGs were correctly positioned on the tibia in all 10 cases;nce the bone landmarks had been exposed, the various tabsatched the patient’s anatomy perfectly.

ions providing a visual comparison of tibial slope and mMPTA.

3.1. Two-dimensional analysis

Postoperatively, the mean HKA was 182.3◦ (180.3–185◦), the tib-ial slope was 6.6◦ (1–13.2◦) and the mMPTA was 94.3◦ (90.8–98.1)(Table 3). The differences from the target value requested by thesurgeon were 0.98 [0.92–0.99] for the HKA and 0.96 [0.79–0.99] forthe tibial slope.

In the coronal plane, none of the knees were more than 2◦ fromthe planned value. In the sagittal plane, one knee had more than 2◦

difference between the planned and measured tibial slope (Table 4).

3.2. Three-dimensional analysis

mation between the preoperative planned value and the correctionachieved. The results were comparable to those observed in 2D. Inthe coronal plane, the difference was always less than 2◦. In the

M. Munier et al. / Orthopaedics & Traumatology: Surgery & Research 103 (2017) 245–250 249

Table 3Results of radiographic analysis in frontal and sagittal planes.

Preoperative Postoperative Target ICC target vs postoperative

Mean HKA (min, max) 171.9◦ (166.1–179.5) 182.3◦ (180.3–185) 182.4◦ (182–184) 0.98 [0.92–0.99]Mean tibial slope (min, max) 7.5◦ (1.8–22.4) 6.6◦ (1–13.2) 6.3◦ (2–10) 0.96 [0.79–0.99]Mean mMTPA (min, max) 86.7◦ (84–88.8) 94.3◦ (90.8–98.1) – –

mMPTA: mechanical medial proximal tibial angle.

Table 42D comparison HKA and tibial slope, preoperative, postoperative and target.

Patient HKA (◦) Slope (◦)

Preoperative Target Achieved Difference Preoperative Target Achieved Difference

1 169.6 182 182.3 0.3 10 9 9.6 0.62 170.8 182 182.4 0.4 9 9 9.4 0.43 179.5 182 181.8 0.2 22.4 10 13.2 3.24 175 182 183.8 1.8 3.8 2 2.4 0.45 176 184 182.3 1.7 1.8 3 1 26 172 182 182.2 0.2 10 9 9.2 0.27 172 184 185 1 6 4 5.3 1.38 166.1 182 180.3 1.7 3 4 3 19 171.3 182 181.4 0.6 4.2 5 4.6 0.4

10 170.4 182 182.5 0.5 7.2 6 6.3 0.3

Table 53D measurements of mMPTA and tibial slope and comparison with planned values.

HKA variation onorder form (OF)

mMPTA Differenceof variation

Variation TSand order form

Tibial slope (TS) Differenceof variation

Preoperative Postoperative Variation Preoperative Postoperative Variation

Patient 1 12 85.7 96.9 11.2 0.8 −1 10.4 9.2 −1.2 0.2Patient 2 11 85.2 96.2 11 0 0 10.7 9.1 −1.6 1.6Patient 3 2 86.1 87.4 1.3 0.7 −12.4 21.8 12.6 −9.2 3.2Patient 4 9 84 92.2 8.2 0.8 −1.8 10.6 9.2 −1.4 0.4Patient 5 6 86.6 91.8 5.2 0.8 1.2 2.3 2 0.3 0.9Patient 6 11 87.2 96.8 9.6 1.4 −1 6.6 6.4 −0.2 0.8Patient 7 12 86.8 97.1 10.3 1.7 −2 6.4 5.1 −1.3 0.7Patient 8 16 82.3 96.4 14.1 1.9 1 3.2 3.3 0.1 0.9Patient 9 11 85.8 96.4 10.6 0.4 0.8 4.5 5.2 0.7 0.1Patient 10 12 85.1 96.1 11 1 −1.2 7.8 6.4 −1.4 0.2

Table 6Published studies of PSCGs in high tibial osteotomy.

Title Author Date Journal Number ofpatients

X-ray evaluation Reliabilitystudied

3D surgical printing cutting guides for open-wedgehigh tibial osteotomy: do it yourself

Pérez-Mananes et al. 2016 J Knee Surg 8 tibias X-rays No

Virtual 3D planning and patient specific surgical Victor et al. 2013 Bone Joint J 4 tibias Radiographs Yes

–

sl

3

rau

4

mf

guides for osteotomies around the kneeOur study –

agittal plane, one knee had 3.2◦ difference and all the others hadess than 2◦ (Table 5).

.3. Complications

One patient had a hematoma 1 month after the surgery thatesolved spontaneously. None of the patients had a deep infectionnd the osteotomy had healed in all patients at the 3-month follow-p.

. Discussion

Our hypothesis that use of PSCGs based on 3D planning for HTOakes the procedure more accurate without increasing the risks

or the patients was confirmed.

10 femurs CT in 1 patient– 10 tibias X-rays and CT

scanYes

The small sample size is the primary limitation of this study.However, this was a pilot study of reliability and safety, hence, thesmall number of patients enrolled. In addition, the study popu-lation is comparable to other published studies reporting on theuse of this technology (Table 6). The other studies included casesof tibial and femoral osteotomy and did not have systematic CTscan follow-up measurements. Secondly, our study had no controlgroup. This is related to the study objectives, which were to assessthe reliability of this technique relative to its planned correctionand to assess its safety, not to demonstrate that this system wasbetter than the conventional method. A comparative study will beneeded to determine whether this system is superior.

The primary objective of this study was to analyze the reliabil-ity of this technique by comparing the accuracy of the achievedcorrection relative to the planned correction. Since the intraop-erative methods used to verify the correction can be inaccurate,

2 ology:

wsofwhtg1oailktPHpctHPc

itLchtaGwloc

5

Pcpc

D

Z

T

i

[

[

[

[

[

[

[

[

[

[

[

[

50 M. Munier et al. / Orthopaedics & Traumat

e chose to carry out another set of CT reconstructions and mea-urements [9]. There are only two published studies on the usef PSCGs for HTO. Victor et al. [10] reported accuracy of 1◦ in therontal plane and 2◦ in the sagittal plane; however, the analysisas performed with radiographs only. Perez-Mananes et al. [11]ad similar results in the frontal plane with an accuracy of 2◦, buthe tibial slope was not measured, nor was CT performed. Sara-aglia et al. [12] reported 96% accuracy for a planned HKA angle of84◦ ± 2◦ in their study with computer-assisted surgery. The resultsf our study compare favorably to those of Saragaglia; however,

comparative study would be needed to show that one methods superior to the other. Moreover, navigation is more costly andengthens the operating time [12,13]. An opening wedge HTO isnown to increase the tibial slope [3,14–16]. With the PSCGs, theibial slope can be maintained close to the initial anatomical data.ublished radiological results often point to under-correction afterTO. This problem is particularly apparent with an opening-wedgerocedure as the under correction is about 2◦ more than withlosing wedge [3,4]. Marti et al. reported a 50% accuracy betweenhe correction achieved and the one planned after opening wedgeTO [17], which was vastly different from our study where use ofSCGs led to a reliable correction that was similar to the plannedorrection.

Our second objective was to evaluate the occurrence ofntra- and postoperative complications. Intraoperative complica-ions have rarely been reported in osteotomy studies [18–20].obenhoffer [21] reported a relatively high rate of postoperativeomplications (6%), mainly due to soft tissue problems (infection,ematoma). In our study, there was one case of postopera-ive hematoma that spontaneously resolved without infectionnd that did not delay the functional recovery. According toomoll [22], the typical postoperative complications of opening-edge HTO are delayed union and non-union, with resulting

oss of correction. There were no instances of delayed union inur study; the follow-up was too short to detect any loss oforrection.

. Conclusion

The results of this study evaluating the reliability and safety ofSCGs during HTO showed that this technique achieves an optimalorrection in a safe and reliable manner. A prospective study com-aring use of PSCGs with the standard technique will be needed toonfirm the advantages of this technology.

isclosure of interest

Jean-Noël Argenson is an educational consultant for Zimmer.Sébastien Parratte is an educational consultant for Newclip,

immer and Arthrex.

Mathias Donnez received a scholarship grant from Newclip

echnics.The other authors declare that they have no competing

nterest.

[

Surgery & Research 103 (2017) 245–250

References

[1] Flecher X, Parratte S, Aubaniac JM, Argenson JNA. A 12-28-year follow-up studyof closing wedge high tibial osteotomy. Clin Orthop 2006;452:91–6.

[2] Lobenhoffer P. [Importance of osteotomy around to the knee for medialgonarthritis. Indications, technique and results]. Orthop 2014;43:425–31.

[3] Nerhus TK, Ekeland A, Solberg G, Sivertsen EA, Madsen JE, Heir S. Radiologicaloutcomes in a randomized trial comparing opening wedge and closing wedgetechniques of high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 2015,http://dx.doi.org/10.1007/s00167-015-3817-z.

[4] Brouwer RW, Bierma-Zeinstra SMA, van Raaij TM, Verhaar JN. Osteotomy formedial compartment arthritis of the knee using a closing wedge or an openingwedge controlled by a Puddu plate. A 1-year randomised, controlled study. JBone Joint Surg Br 2006;88:1454–9.

[5] Bonutti P, Dethmers D, Stiehl JB. Case report: femoral shaft fractureresulting from femoral tracker placement in navigated TKA. Clin Orthop2008;466:1499–502.

[6] Ahlbäck S, Rydberg J. [X-ray classification and examination technics ingonarthrosis]. Läkartidningen 1980;77:2091–3, 2096.

[7] Ahlbäck S. Osteoarthrosis of the knee. A radiographic investigation. Acta RadiolDiagn (Stockh) 1968;Suppl. 277:7–72.

[8] Julliard R, Genin P, Weil G, Palmkrantz P. [The median functional slope of thetibia. Principle. Technique of measurement. Value. Interest]. Rev Chir OrthopRepar Appar Mot 1993;79:625–34.

[9] Billings A, Scott DF, Camargo MP, Hofmann AA. High tibial osteotomy with acalibrated osteotomy guide, rigid internal fixation, and early motion. Long-termfollow-up. J Bone Joint Surg Am 2000;82:70–9.

10] Victor J, Premanathan A. Virtual 3D planning and patient specific surgical guidesfor osteotomies around the knee: a feasibility and proof-of-concept study. BoneJt J 2013;95–B:153–8.

11] Pérez-Mananes R, Burró JA, Manaute JR, Rodriguez FC, Martín JV. 3D Surgicalprinting cutting guides for open-wedge high tibial osteotomy: do it yourself. JKnee Surg 2016;29(8):690–5.

12] Saragaglia D, Roberts J. Navigated osteotomies around the knee in 170 patientswith osteoarthritis secondary to genu varum. Orthopedics 2005;28:s1269–74.

13] Hankemeier S, Hufner T, Wang G, Kendoff D, Zeichen J, Zheng G, et al. Navigatedopen-wedge high tibial osteotomy: advantages and disadvantages comparedto the conventional technique in a cadaver study. Knee Surg Sports TraumatolArthrosc 2006;14:917–21.

14] Brouwer RW, Bierma-Zeinstra SMA, van Koeveringe AJ, Verhaar JN. Patel-lar height and the inclination of the tibial plateau after high tibialosteotomy. The open versus the closed-wedge technique. J Bone Joint Surg Br2005;87:1227–32.

15] Ducat A, Sariali E, Lebel B, Mertl P, Hernigou P, Flecher X, et al. Posterior tib-ial slope changes after opening- and closing-wedge high tibial osteotomy:a comparative prospective multicenter study. Orthop Traumatol Surg Res2012;98:68–74.

16] El-Azab H, Glabgly P, Paul J, Imhoff AB, Hinterwimmer S. Patellar height andposterior tibial slope after open- and closed-wedge high tibial osteotomy: aradiological study on 100 patients. Am J Sports Med 2010;38:323–9.

17] Marti CB, Gautier E, Wachtl SW, Jakob RP. Accuracy of frontal and sagittal planecorrection in open-wedge high tibial osteotomy. Arthrosc J Arthrosc Relat Surg2004;20:366–72.

18] Brinkman JM, Luites JWH, Wymenga AB, van Heerwaarden RJ. Early fullweight bearing is safe in open-wedge high tibial osteotomy. Acta Orthop2010;81:193–8.

19] Brosset T, Pasquier G, Migaud H, Gougeon F. Opening wedge high tibialosteotomy performed without filling the defect but with locking plate fixation(TomoFixTM) and early weight-bearing: prospective evaluation of bone union,precision and maintenance of correction in 51 cases. Orthop Traumatol SurgRes 2011;97:705–11.

20] Staubli AE, De Simoni C, Babst R, Lobenhoffer P. TomoFix: a new LCP-concept foropen wedge osteotomy of the medial proximal tibia – early results in 92 cases.Injury 2003;34(Suppl. 2):B55–62.

21] Floerkemeier S, Staubli AE, Schroeter S, Goldhahn S, Lobenhoffer P. Out-

come after high tibial open-wedge osteotomy: a retrospective evaluation of533 patients. Knee Surg Sports Traumatol Arthrosc 2013;21:170–80.

22] Gomoll AH. High tibial osteotomy for the treatment of unicompartmentalknee osteoarthritis: a review of the literature, indications, and technique. PhysSportsmed 2011;39:45–54.