Research Article

Growth Modulation for Knee Coronal

Plane Deformities in Children With
Nutritional Rickets: A Prospective
Series With Treatment Algorithm

Abstract
Tamer A. EL-Sobky, MD
Purpose: To report prospectively the radioclinical outcome of guided
Shady Samir, MD
growth surgery for coronal plane deformities around the knee in
Mostafa M. Baraka, MD
young children with nutritional rickets on the intermediate term, to
Tamer A. Fayyad, MD assess the responsiveness of torsional deformities of the tibias to
Mahmoud A. Mahran, MD guided growth regarding function and objective clinical parameters,
Ahmad S. Aly, MD and to propose a treatment algorithm.
John Amen, MD Methods: Fifty children (male:female, 27:23) with knee coronal
Shady Mahmoud, MD plane deformities (knees:physes, 86:99), (varum:valgum, 51:35)
secondary to nutritional rickets were subjected to femoral and/or
tibial temporary hemiepiphysiodesis using a two-hole 8-plate. The
mean age at implantation was 3.8 6 1.5 years (range 2.5 to 5). The
mean follow-up was 2.8 years (range 2 to 4). All children received a
standing full-length AP radiographs of both lower limbs in neutral
rotation to measure the mechanical axis deviation, tibiofemoral
angle, and joint orientation angles. Tibial torsion was objectively
From the Division of Paediatric assessed by measuring the bimalleolar axis.
Orthopaedics, Department of Results: The radiologic measurements, tibiofemoral angle,
Orthopaedic Surgery, Faculty of
Medicine, Ain-Shams University, mechanical axis deviation, mechanical lateral distal femoral
Cairo, Egypt.
angle, medial proximal tibial angle, and Hilgenreiner-epiphyseal
Informed consent was obtained from angle, showed a highly statistically significant improvement (P #
all individual participants included in
the study. 0.001). Radiographic outcomes correlated with their clinical
JAAOS Glob Res Rev 2020;4: counterparts. The mean duration of correction of the mechanical
e19.00009
axis was 10.8 6 2.4 months (7 to 21). The mean follow-up for
DOI: 10.5435/ rebound of the deformity was 1.5 years (range 1 to 3).
JAAOSGlobal-D-19-00009
Conclusion: The radioclinical outcome is rewarding with a tolerable
Copyright © 2020 The Authors.
Published by Wolters Kluwer Health, complication profile. The mechanical complications were mostly
Inc. on behalf of the American
related to lengthy implant retainment encountered in severe
Academy of Orthopaedic Surgeons.
This is an open access article deformities. Internal tibial torsion seems profoundly responsive to
distributed under the Creative
Commons Attribution License 4.0
correction of coronal plane deformity. And, derotation osteotomies
(CCBY), which permits unrestricted are rarely justified. Our proposed algorithm may be used as a
use, distribution, and reproduction in
any medium, provided the original decision-taking guide for achieving the desired growth modulation
work is properly cited. in a more efficient manner.
Growth Modulation in Nutritional Rickets

G uided growth surgery for lower-

limb deformities took the place
of the time-honored corrective os-
been conducted to evaluate the per-
formance of guided growth in knee
deformities in nutritional rickets. We
when children possessed clinical skel-
etal deformities in association with
characteristic radiographic physeal-
teotomies. Its effectivity and safety postulate that physeal growth mod- metaphyseal changes and abnormali-
have been generally acknowledged.1 ulation surgery would be successful ties in laboratory bone mineral profile.
It is widely accepted that pediatric in restoring coronal alignment of the Rickets was considered healed based
orthopaedic surgeons are supposed knee. And, we hypothesize that tibial on the restoration of a normal phyeal
to use corrective osteotomies only as varus is a product of coronal and line and radiographs and normal serum
a salvage of a failure of guided transverse plane malalignment. Hence, levels of calcium, phosphorus, and
growth surgery.1,2 Temporary hemi- correction of tibial varus and internal alkaline phosphatase. Alkaline phos-
epiphysiodesis has been practiced torsion will occur simultaneously. The phatase is an important and afford-
for correction of angular knee objective of this study is to report able screening tool for nutritional
deformities in a wide spectrum of prospectively the clinical and radio- rickets25 We considered values above
etiologies including idiopathic,1,3,4 graphic outcome of guided growth the cutoff (450 IU/L) as abnormal.
osteochondrodysplasias, 5,6 post- surgery for coronal plane deformities Genu valgum and genu varum de-
traumatic,1 renal osteodystrophy,7 around the knee in young children with formities were included. The general
hypophosphatemic rickets, 8 and nutritional rickets on the intermediate inclusion criteria were children youn-
arthrogryposis,9 Temporary hemi- term. We present the largest and most ger than 7 years and healed nutritional
epiphysiodesis is done using sta- homogenous series of such patient rickets. We considered patients indi-
ples,10,11 a regular1,10-12 or hinged13 subset in the literature. In addition, we cated for surgery if they possessed one
tension band plate, percutaneous aim to assess the responsiveness of or more of the following criteria: (1)
transphyseal screw,4,14 and a one- torsional deformities of the tibias to the an overall coronal plane deformity
third tubular plate.15 Several clinical guided growth surgery with respect to “varus or valgus” $ 20° as per
and animal studies have shown that function and objective clinical param- tibiofemoral angle (TFA) that is per-
the above implants are equally eters and finally to propose a treatment sistent or worsening over the past
effective and safe in achieving guided algorithm to guide the decision-making 6 months, (2) a mechanical axis
growth with marginal differences process. deviation (MAD) bisecting the knee
related to complication rates, surgi- outside the central one-third of the
cal times, cost, and quality of transverse diameter of the proximal
life. 1,4,10,14-20 Contrastingly, the Methods tibial epiphysis, and (3) a clinically
response of the physis varies with symptomatic gait impairment in the
age, the nature of the disease, and We conducted a prospective inter- form of troublesome circumduction
nutrition and can be unpredict- ventional observational case study at gait in association with genu valgum
able.1,12,18,21,22 Nutritional rickets the authors’ institution from Febru- or massive intoeing in association
in young children remains a consid- ary 2015 to March 2017. During the with genu varum or frequent falls.
erable and prevalent public health study period, 50 children with knee Exclusions were (1) previous guided
problem worldwide including the coronal plane deformities secondary growth surgery or osteotomy on
developed countries.23-25 The ortho- to nutritional rickets fulfilled the the affected limb, (2) osteochon-
paedic manifestations of rickets can inclusion criteria and were subjected drodysplasias, endocrinopathies,
result in grave deformities of all to femoral and/or tibial temporary Blount disease, post-traumatic/infec-
extremities, gait abnormalities, and hemiepiphysiodesis using a two-hole tious deformities, and (3) all forms of
entail an increased risk of sustain- eight-plate. The patient and defor- resistant rickets. The study was
ing fractures.23-25 Nonetheless, no mity characteristics are displayed approved by our institution’s Ethical
exclusive or prospective studies have (Table 1). Rickets was diagnosed Committee of Scientific Research.

None of the following authors or any immediate family member has received anything of value from or has stock or stock options held in a
commercial company or institution related directly or indirectly to the subject of this article: Dr. EL-Sobky, Dr. Samir, Dr. Baraka,
Dr. Fayyad, Dr. Mahran, Dr. Aly, Amen, and Mahmoud.
Dr. EL-Sobky and Dr. Baraka conceived and designed the study. Dr. Baraka, Amen, and Mahmoud were particularly involved in literature search.
Otherwise, all authors contributed equally to acquisition, analysis, and interpretation of data. Dr. EL-Sobky and Dr. Baraka drafted the manuscript.
Dr. Samir, Dr. Fayyad, Dr. Mahran, Dr. Aly, Dr. Amen, and Dr. Mahmoud reviewed the final manuscript critically for important intellectual content.
All authors approved the final version and are responsible for the content. All parents consented to and were informed that data concerning this
study would be submitted for publication. The study was approved by our institution’s Ethical Committee of Scientific Research.

2 Journal of the American Academy of Orthopaedic Surgeons

 Tamer A. EL-Sobky, MD, et al

Surgical Technique Table 1

We approached the involved physis Patient and Deformity Characteristics
directly maintaining dissection ex-
Total participants 50 children
traperiosteal. We used fluoroscopy to
Age (yr) Mean: 3.8 6 1.5 (2.5-5)
avoid epiphyseal plate and joint line
Sex
injury while maintaining parallelism
Males 27 (54%)
of screws and their placement in mid-
Females 23 (46%)
lateral plane as much as possible. A
Side
thin Kirschner wire was inserted in the
central hole of the eight-plate to mark Left 11 (22%)
the physis. A 2.5-mm drill bit was used Right 3 (6%)
to insert a 3.5-mm solid screw. The Bilateral 36 (72%)
epiphyseal screw was inserted first. Deformity (no. of limbs/
knees)
The drill bit was advanced for only a
Varum 51 (59%)
few millimeters into epiphyseal bone,
and then, the screw was tightened till it Valgum 35 (41%)
fitted into the corresponding hole. Total 86 knees
Screw length was selected so as to No. of instrumented
physes
reach at least the midline of the
Femur 30
epiphysis.
Tibia 39
Biphyseal 30 (in 15 knees)
Outcome Measures Total 99
We conducted a full screening ortho- Mean follow-up (yr)a 2.8 years (range 2-4)
paedic examination with observational Mean follow-up for 1.5 years (range 1-3)
gait analysis. We have done a gonio- reboundb
metric measurement of the standing Mean duration of 10.8 6 2.41 (7-21)
TFA and the transmalleolar axis in the correction (mo)
prone position as an indicator of tibial a
The mean follow-up from implantation up to the last visit.
torsion. We obtained standing full- b
The mean follow-up from explantation to the last visit.
length AP radiographs of both lower
limbs in neutral rotation directly before
the index surgery. The standing long epiphysis. This measurement of MAD extremity was deemed clinically
radiograph was repeated once again would eliminate errors arising from satisfactory, standing full-length AP
once the limbs were clinically straight- magnification on radiographs and radiographs of both lower limbs
ened before plate removal. The radio- augment validity of results. Defor- were obtained. The end point for
graphic indices calculated were TFA, mity severity was graded into four plate removal was identified once the
MAD, mechanical lateral distal femoral zones by measuring the distance mechanical axis falls within the
angle (mLDFA), medial proximal tibial between mechanical axis and mid- central one-third of the tibial plateau
angle (MPTA), and Hilgenreiner- point of tibial epiphysis expressed as for both varus and valgus deform-
epiphyseal angle (HEA).26 The TFA percentage to 1/2 of width of tibial ities. For both varus and valgus de-
was defined as the angle between the epiphysis (Figure 1). We allowed formities, an accepted correction
following two lines. The first line is immediate weight-bearing according equated to a mechanical axis bi-
between the center of proximal diaph- to tolerance. Patients were observed secting the medial half of the central
ysis of the femur and the center of distal at weekly intervals in early postop- one-third of tibial plateau and an
diaphysis of the femur, and the second erative period until full functional optimal correction equated to a
line is between the center of proximal recovery and full range of motion mechanical axis bisecting the lateral
diaphysis of the tibia and the center of were achieved. Otherwise mothers half of the central one-third. We
distal diaphysis of the tibia. The MAD were instructed to do range-of- assessed the amelioration of tibial
was obtained by measuring the distance motion exercises at home. Patients torsion objectively by measuring the
between mechanical axis and midpoint were assessed clinically at 3-monthly bimalleolar axis. In addition, we
of tibial epiphysis expressed as per- intervals thereafter. When the coro- took history of frequent falls, child’s
centage to ½ of the width of tibial nal plane alignment of the lower ability to participate in normal daily

January 2020, Vol 4, No 1

Growth Modulation in Nutritional Rickets

Figure 1

A–D, Guided growth for a 4-year-old girl with unilateral genu valgum. A and B, Preoperative clinical and radiologic findings. Note
the mechanical axis is traversing zone 3. C and D, Radioclinical outcome at 3.6-year postimplant insertion. Note the mechanical
axis is optimally corrected, i.e., traversing the lateral half of the central one-third of the widest diameter of the proximal tibia.

activity, and more vigorous activities nonparametric variable between two was expressed as percentage of half
associated with young children. We study groups. Paired t-test was used plateau width. Negative values were
assessed rebound clinically because to assess the statistical significance of assigned for varus deformities, and
of logistic challenges. We defined the difference between two means positive values were assigned for valgus
rebound as a deviation of .10° from measured twice for the same study deformities. The preoperative MAD
the clinical TFA recorded right after group. was 2158.6 6 75.1 and 1146.9 6
implant removal. 76.7 for varus and valgus knees,
respectively. The values for the MAD
Results were as follows: 51 knees (59%)
Data Management and
Analysis acceptable alignment, 31 knees (36%)
The TFA, MAD, mLDFA, MPTA,
optimal alignment, and 4 knees (5%)
The collected data were coded and and HEA demonstrated a highly sta-
overcorrected. The overall complica-
tabulated using Statistical package tistically notable improvements at the
tion rate was small and tolerable and
for Social Science (SPSS 20 IBM). final follow-up visit (Table 2). These
not necessarily disease or technique-
Descriptive statistics included the radiographic improvements corre-
related (Table 4 and Figures 2–4).
mean, SD, and range for parametric lated positively with the clinical out-
numerical data, while median and in- comes. Valgus knees demonstrated a
terquartile range for nonparametric significantly high speed of correction
numerical data. Frequency and per- of TFA in contrast to varus knees (P = Discussion
centage of nonnumerical data were 0.034.) (Table 3). For both valgus
reported. Analytical statistics included and varus deformities, the speed of
the Student t-test was used to assess correction of TFA mounting to ,30° Surgical Indications and
the statistical significance of the dif- and $30° was 1.1° per month (in- Outcomes
ference between two study group terquartile range 0.5 to 1.7) and 2.5°/ Predictors of spontaneous resolution
means. The Mann-Whitney test (U month (interquartile range 2.1 to of persistent severe lower-limb de-
test) was used to assess the statistical 3.0), respectively. This difference was formities despite successful medical
significance of the difference of a highly significant (P , 0.001). MAD therapy of nutritional rickets in

4 Journal of the American Academy of Orthopaedic Surgeons

 Tamer A. EL-Sobky, MD, et al

Table 2
Radiographic Outcomes at the Final Follow-upa
Preoperative Postoperative Paired t-test
Factors Mean SD Mean SD P Value Sig.

Varum
Radiographic TFA 25.07 16.33 6.84 1.22 ,0.001 S
mLDFA 103.50 11.11 88.75 1.24 ,0.001 S
MPTA 79.50 7.86 87.95 0.96 ,0.001 S
HEA 37.67 9.16 23.86 5.72 ,0.001 S
Valgum
Radiographic TFA 26.12 7.61 8.03 2.90 ,0.001 S
mLDFA 80.76 7.71 88.56 2.58 ,0.001 S
MPTA 94.35 6.61 87.41 2.89 ,0.001 S
HEA 18.00 5.19 24.88 6.71 ,0.001 S

HEA = Hilgenreiner-epiphyseal angle, MAD = mechanical axis deviation, it is expressed as percentage of half plateau width (negative values are
assigned for varus deformities and positive values are assigned for valgus deformities), mLDFA = mechanical lateral distal femoral angle, MPTA =
medial proximal tibial angle, NS = non-significant; TFA = tibiofemoral angle
a
All measurements recorded in degrees.

Table 3
Duration of Plate Application and Speed of Correction as per Varus Versus Valgus Kneesa
Deformity
Varum Valgum
t-test
Factors Mean SD Mean SD P Value Sig.
(T)
Duration of correction (mo) 10.9 2.2 10.8 2.9 0.834 NS
TFA correction 14.50 5–24 18.00 12–25 0.230(M) NS
Rate of correctionb 1.10 0.4–2.25 1.65 1.1–2.3 0.034(M) S

TFA = tibiofemoral angle

a
All measurements recorded in degrees; T = t-test; M = Mann-Whitney test.
b
Rate of correction were measured by the amount of corrected TFA divided by the duration in months.

regard to patient and disease char- implemented in our study are fairly have derotated in conjugation with
acteristics are yet to be identified in similar to those proposed by the varus correction indicates that tibial
the literature. In that regard, age and above authors in regard to genu varus and internal torsion are
severity of deformity are crucial fac- varum.24 In addition, we based our strongly linked. The above natural
tors. Likewise, the exact indications indications on clinical parameters history study24 failed to address the
of surgical interventions currently such as symptomatic gait difficulties fate of internal tibial torsion known
used to correct angular knee de- and deformity severity. It is note- to be a close associate of varus. In
formities in these children remain worthy that this study24 was hospital- our study, the markedly symptom-
undetermined and mainly subject to based which makes it specifically atic tibial intoeing gait was used as
surgeon’s preference. Yet, a very prone to bias. In addition, the study one of the indicators of surgery in
recent prospective natural history underestimated the role of environ- addition to its use as an outcome
study suggested that varum above 4 mental and genetic predisposition measure. Because of the dynami-
years and 18° of valgum above 9 related to the pathoetiology of cally developing TFA in childhood,27
years usually do not correct and may nutritional rickets.23 The fact that all we considered any mechanical axis
require surgical intervention.24 The but two tibial varus deformities with traversing the central one-third of
indications for surgical intervention internal tibial torsion in our study the widest diameter of the tibial

January 2020, Vol 4, No 1

Growth Modulation in Nutritional Rickets

Table 4
Complications
Involved
Radioclinical Deformity Physes/
Complication Patients Setting (Knees) Hemiside Management

Overcorrection 4 (5%) Complication 2 valgus and 2 medial femoral Two femoral varus overcorrections were managed
attributed to poor 2 varus were by reimplantation of another plate in the
follow-up overcorrected contralateral hemiside of the epiphysis until the
into varus, and 2 desired correction was reached. The two tibial
lateral tibial valgus overcorrections were simply put under
physes were observation as overcorrection was mild with no
overcorrected functional limitations.
into valgus
Implant 8 (16%) All except one were 3 valgus and 3 medial femoral, Two of these patients (one physis each) achieved
disassemblya epiphyseal screw 11 varus 1 medial tibial, correction of the residual valgus and varus
loosening which and 11 lateral deformities through additional plating of the
were encountered tibial physes ipsilateral virgin medial tibial and lateral femoral
in limbs that were physes, respectively, to speed up correction. The
corrected through other 6 patients (13 physes) were followed up till
one physis only and full correction. Of the above 13 physes, 5 were
exhibited a TFA $ bilateral genu varum, and one patient with a
30°b windswept deformity had 3 instrumented physes.
Screw breakagea 1 (2%) Metaphyseal screw 1 varus Occurred in right The left knee progressed to an acceptable
breakage. lateral tibial correction of MAD. Both tibial implant constructs
Consequently, construct in a were removed, and the revision surgery was
absolutely no child with undertaken on the right knee only. The broken
growth modulation bilateral genu screw was exchanged with a new and longer one.
was noticed around varum.
that knee.
Additional 4 (5%) One tibia in a child 3 varus and 3 lateral tibial and 1 tibial derotation osteotomy. Tibial intoeing gait
surgery with bilateral genu 1 valgus 1 medial was markedly symptomatic to necessitate a
varum failed to femoral physisc derotation osteotomy. Also see 3 additional
derotate in surgeries under implant disassembly (2) and
conjugation with screw breakage (1).
full correction of
varus deformity.
Persistent 1 (2%) One tibia in a child 1 varus 1 lateral tibial Tibial intoeing gait was completely asymptomatic.
internal tibial with bilateral genu Patient was managed conservatively.
torsiond varum failed to
derotate in
conjugation with
full correction of
varus deformity.
Rebound 1 (2%) Rebound mounted to 1 valgus 1 medial femoral Rebound was associated with marked ligamentous
50% of the clinical collateral laxity and occurred shortly after implant
TFA recorded removal.
immediately after
implant removal.
Recurrence of 2 (4%) The two patients of 4 varus 4 lateral tibial They were subjected to medical treatment until
rachitic activity recurrence of radiographic and biochemical healing. This did
rachitic activity not affect the final outcome.
occurred during the
course of surgical
treatment and
before plate
removal.
Superficial 2 (4%) Infections were 1 varus and 1 medial femoral Infections eventually resolved after implant
infections superficial and 1 valgus and 1 lateral removal. Infections had no effect on overall
early onset. tibial correction.

MAD = mechanical axis deviation, TFA = tibiofemoral angle

a
Implant failure was subdivided into two categories namely implant disassembly or screw/plate breakage.
b
The only exception was a complete extrusion of the epiphyseal and metaphyseal screw of the medial tibial physis in the genu valgum component
of a windswept deformity. The ipsilateral femoral physis also suffered implant disassembly but to a lesser degree.
c
These were the physes that were originally instrumented. Additional surgery to speed up correction was done on two lateral femoral and one
medial tibial physes for varus and valgus knees, respectively, in addition to one derotation osteotomy.
d
The 2 knees with persistent internal tibial torsion occurred in 51 knees with genu varum. It is noteworthy that knees with genu valgum did not
originally exhibit internal tibial torsion.

6 Journal of the American Academy of Orthopaedic Surgeons

 Tamer A. EL-Sobky, MD, et al

Figure 2

A–D, Guided growth for a 5-year-old boy with bilateral genu valgum. A and B, Preoperative clinical and radiologic findings.
Note the mechanical axis is traversing zone 2 bilaterally. Because the deformity was predominantly femoral, the femoral
physis only was attacked. C and D, Radioclinical outcome at 1.9-year postimplant insertion. Note the mechanical axis of the
left limb traverses the lateral half of the central one-third of tibia, i.e., optimal correction. Yet, the mechanical axis of the right
limb traverses outside the central one-third (zone 1). Note the loosening backing up of the epiphyseal screws bilaterally. Both
implants were removed, and patient was scheduled for implantation of the right medial tibial physis to correct residual valgus.

epiphysis to be accepted. Yet, we tion to diverse surgical methodolo- agree with the above authors as to the
opted for a more valgus orientation of gies and implants.1,10,30,31 In turn, fundamental involvement of the epi-
the mechanical axis, ie, a mechanical this may undermine the capability of physeal screw in loosening. Yet, in our
axis traversing the lateral half of the such studies to provide valid and study, screw length was not an issue.
central one-third. Long-term studies generalizable conclusions. We pre- The incidence of screw loosening or
of guided growth around the knee sented the only homogeneous and implant disassembly in our study was
suggested that varus alignment of the prospective series on angular knee de- tolerable and mainly linked to degree
mechanical axis may be a precursor formities in children with nutritional of deformity and the overall duration
of rebound.28 The above argues for rickets. Various authors have reported of implant retainment from insertion
our stratification of radiographic satisfactory radioclinical outcomes in to removal. Furthermore, this compli-
outcomes. The rebound phenomena relatively small subsets of children with cation almost exclusively occurred in
has also been attributed to children nutritional rickets.1,31 This goes in line limbs where only one physis was im-
with a rapid rate of correction29 and with the overall outcome of our study. planted. This prompted us to upgrade
to the underlying pathology. 30 our treatment protocol and attack
The radiographic findings of our Complications and Evolved both physes i.e., tibial and femoral—if
study demonstrated that structural Surgical Strategy the deformity mounted to TFA of
realignment of the MAD after guided Various studies reported screw loos- $30° irrespective of the specific con-
growth surgery around the knee can ening and have related it to a variety of tribution of each physis to the overall
have a notable positive effect on the factors namely, osteopenic bone qual- MAD. This evolved strategy aimed at
hip posture in terms of measurement ity such as resistant rickets, short screw speeding the rate of correction thereby
improvement of the HEA. length, and proximity of epiphyseal averting screw loosening. In the mild
Most studies on guided growth screw start point to physis.2,9,32,33 In counterparts of limb deformity i.e., 20°
surgery were retrospective and to consequence, some authors recom- to 30°, we achieved correction from
some extent involved heterogeneous mended longer and or larger screws to the physis with greater contribution to
patient and disease cohorts in addi- overcome such complications.32,33 We deformity and neglected the other. We

January 2020, Vol 4, No 1

Growth Modulation in Nutritional Rickets

Figure 3 reported an unacceptably high rate

of implant failure (58%). Yet, they
defined implant failure as implant
migration or breakage.21 The studies
that reported screw breakage strictly
linked it with the metaphyseal screw
in Blount disease with a severe
deformity and high body mass index
and to some extent linked with young
age.1,11,18,21,22,34 In our study, we
encountered one screw breakage of
the right proximal medial tibial plate
in a young child with bilateral genu
varum. A large survey study22 reported
that screw breakage occurred almost
always in the metaphyseal screw-not
where the head meets the shank but
where the shank enters the lateral
cortex. This was precisely the case in
our only screw breakage case. The
A and B, Guided growth for a 4-year-old boy with severe wind-swept deformity. findings of the above-mentioned au-
A, Preoperative clinical appearance and (B) 1.7-year postoperative appearance thors1,11,18,21,22,34 and our study out-
immediately before implant removal shows optimal alignment.
comes point out that this screw
breakage complication is an inherent
believe that minor distortions (10° to physes had an overall time of implant characteristic of Blount disease and
15°) of the mLDFA or MPTA would be retainment from insertion to removal of its underlying pathoetiology. We
compensated as long as the mechanical greater than 2 years with two implants record one limb only with rebound
axis is restored. The above authors retained for 4.5 years each. Yet, the phenomena.
managed this complication by a variety authors failed to make any inferences
of methods namely screw removal, re- from such data.2 This observation lends
insertion of longer screws, conversion support to our explanation for screw Study Limitations
to an osteotomy, or simply observa- loosening that incriminates a long We acknowledge limitations of this
tion.2,32,33 In our 14 knees (16%) implant retainment period as a likely study. First, we relied on qualitative
which developed screw loosening, only causation. Our surgical strategy assessment (absent/present) of tibial
2 knees achieved correction of the comprised drilling the near cortex torsion. We did not measure tibial
residual valgus and varus deformities for a few millimeters and then con- torsion quantitatively per degrees of
through additional plating of the ipsi- tinuing screw insertion free handed bimalleolar axis deviation with respect
lateral virgin physis. In the remaining to maximize screw purchase. In to the frontal plane nor did we resort
12 knees, we simply chose to observe addition, we aimed at crossing the to CT or MRI due to logistic chal-
four of them until they achieved the end midpoint of the epiphysis and lenges. It is noteworthy that the liter-
point for implant removal, and the re- metaphysis at least. Longer screws ature reports mixed results in regard to
maining two were revised by guided were permissible as long as the far the accuracy of clinical bimalleolar
growth. We acknowledge that screw cortex was not violated. We esti- axis measurement as an indicator for
loosening undermines the capability of mated that this screw placement tibial torsion when measured against
the whole implant construct to provide strategy would safeguard against various standard reference tests.35,36
growth modulation and cause undue implant disassembly. The upgraded The satisfactory outcome of the parent-
delay in correction speed. Nevertheless, and final treatment algorithm is de- reported assessment regarding frequent
this observation strategy was successful picted (Figure 5). falls and activity of daily life provides
in achieving an accepted correction at On the other hand, the incidence of support for our qualitative measure-
the expense of lengthy implant retain- screw breakage ranged from three ment of the bimalleolar axis and cor-
ment. It is noteworthy that Masquijo physes (0.55%) in a series of 967 roborates our conclusions in regard to
et al2 in their series on failure of tension physes up to eight physes (26%) in a tibial torsion. Likewise, we assessed the
band plating recorded that three of nine series of 31 physes.1,11,34 Funk et al21 rebound phenomenon clinically only

8 Journal of the American Academy of Orthopaedic Surgeons

 Tamer A. EL-Sobky, MD, et al

Figure 4

A–D, Radiographic appearance for the same patient as in Figure 3. A and B, Radiographs at time of implant insertion. C and
D, Postoperative radiographic appearance (1.7 years). Despite massive implant disassembly especially of the tibial physis
on the right limb with the severe genu valgum, full clinical correction was achieved.

due to suboptimal resource settings. Figure 5

Future studies should consider using
more rigorous objective clinical and
radiologic tools to assess tibial der-
otation. We did not correlate the age
of the patient with the speed of cor-
rection as our patients belonged to a
homogenous age group (young chil-
dren , 5 years).

Conclusion
The radioclinical outcome of this
study demonstrates that guided
growth surgery in young children
with nutritional rickets is effective
with a tolerable complication profile
in the clinical setting described. Fur-
thermore, these results expand with
considerable certainty the range of
indications in which guided growth can
be practiced. Generally, tibial intoeing
exhibited derotation with coronal plane Final treatment algorithm. Surgery is indicated if the coronal tibiofemoral angle is
correction irrespective of plate/screw $ 20° that is persistent or progressive over the past 6 months. If residual tibial
torsion is deemed functionally symptomatic after reaching the desired correction
positioning and derotation osteotomies in the coronal plane, a supramalleolar derotation osteotomy can be
are not deemed necessary unless there simultaneously pursued with implant removal.
is a markedly troublesome gait and
function. Our proposed algorithm used as a decision-taking guide for Generally, this algorithm may also
for treatment of angular knee de- achieving the desired growth modu- have applicability to correction of
formities in rachitic children may be lation in a more efficient manner. coronal plane deformities caused by

January 2020, Vol 4, No 1

Growth Modulation in Nutritional Rickets

pathoetiologies other than rickets. The management of lower limb deformity in Madsen B: Hemiepiphysiodesis: Similar
hypophosphataemic rickets. J Child treatment time for tension-band plating
homogenous patient and disease char- Orthop 2017;11:298-305. and for stapling: A randomized clinical
acteristics allow for reasonable gener- trial on guided growth for idiopathic
9. Palocaren T, Thabet AM, Rogers K, et al: genu valgum. Acta Orthop 2013;84:
alizability of the conclusions. Natural Anterior distal femoral stapling for 202-206.
history studies are on great demand and correcting knee flexion contracture in
children with arthrogryposis–preliminary 21. Funk SS, Mignemi ME, Schoenecker JG,
may refine the indications of surgical
results. J Pediatr Orthop 2010;30:169-173. Lovejoy SA, Mencio GA, Martus JE:
intervention especially in regard to pa- Hemiepiphysiodesis implants for late-onset
10. Jelinek EM, Bittersohl B, Martiny F,
tient’s age and deformity severity by Scharfstädt A, Krauspe R, Westhoff B: The
tibia vara: A comparison of cost, surgical
success, and implant failure. J Pediatr
providing prognostic insights. 8-plate versus physeal stapling for Orthop 2016;36:29-35.
temporary hemiepiphyseodesis correcting
genu valgum and genu varum: A 22. Burghardt RD, Specht SC, Herzenberg JE:
retrospective analysis of thirty five patients. Mechanical failures of eight-plate guided
Acknowledgment Int Orthop 2012;36:599-605. growth system for temporary
hemiepiphysiodesis. J Pediatr Orthop
Compliance with ethical standards. 11. Wiemann JM IV, Tryon C, Szalay EA: 2010;30:594-597.
Physeal stapling versus 8-plate
Ethical approval: All procedures done hemiepiphysiodesis for guided correction of 23. Creo AL, Thacher TD, Pettifor JM, Strand
in studies involving human participants angular deformity about the knee. J Pediatr MA, Fischer PR: Nutritional rickets around
Orthop 2009;29:481-485. the world: An update. Paediatr Int Child
were in accordance with the ethical
Health 2017;37:84-98.
standards of the institutional research 12. Boero S, Michelis MB, Riganti S: Use of the
eight-plate for angular correction of knee 24. Prakash J, Mehtani A, Sud A, Reddy BK: Is
committee and with the 1964 Helsinki deformities due to idiopathic and surgery always indicated in rachitic coronal
Declaration and its later amendments pathologic physis: Initiating treatment knee deformities? Our experience in 198
or comparable ethical standards. The according to etiology. J Child Orthop 2011; knees. J Orthop Surg (Hong Kong) 2017;
5:209-216. 25:2309499017693532.
study was approved by the authors’
institutional review board. 13. Kadhim M, Gauthier L, Logan K, El- 25. El-Sakka A, Penon C, Hegazy A, Elbatrawy
Hawary R, Orlik B: Guided growth for S, Gobashy A, Moreira A: Evaluating bone
angular correction in children: A health in Egyptian children with forearm
comparison of two tension band plate fractures: A case control study. Int J Pediatr
References designs. J Pediatr Orthop B 2018;27:1-7. 2016;2016:7297092.
14. Bayhan IA, Karatas AF, Rogers KJ, Bowen 26. Cheema JI, Grissom LE, Harcke HT:
1. Danino B, Rödl R, Herzenberg JE, et al:
JR, Thacker MM: Comparing Radiographic characteristics of lower-
Guided growth: Preliminary results of a
percutaneous physeal epiphysiodesis and extremity bowing in children.
multinational study of 967 physes in 537
eight-plate epiphysiodesis for the treatment Radiographics 2003;23:871-880.
patients. J Child Orthop 2018;12:91-96.
of limb length discrepancy. J Pediatr
2. Masquijo JJ, Firth GB, Sepúlveda D: Failure Orthop 2017;37:323-327. 27. Mohd-Karim M, Sulaiman A, Munajat I,
of tension band plating: A case series. J Syurahbil A: Clinical measurement of the
Pediatr Orthop B 2017;26:449-453. 15. Böhm S, Krieg AH, Hefti F, Brunner R, tibio-femoral angle in Malay children.
Hasler CC, Gaston M: Growth guidance of Malays Orthop J 2015;9:9-12.
3. Zajonz D, Schumann E, Wojan M, et al: angular lower limb deformities using a one-
Treatment of genu valgum in children by third two-hole tubular plate. J Child 28. Farr S, Alrabai HM, Meizer E, Ganger R,
means of temporary hemiepiphysiodesis Orthop 2013;7:289-294. Radler C: Rebound of frontal plane
using eight-plates: Short-term findings. malalignment after tension band plating. J
BMC Musculoskelet Disord 2017;18:456. 16. Kumar A, Gaba S, Sud A, Mandlecha P, Pediatr Orthop 2018;38:365-369.
Goel L, Nayak M: Comparative study
4. Park H, Park M, Kim SM, Kim HW, Lee between staples and eight plate in the 29. Park SS, Kang S, Kim JY: Prediction of
DH: Hemiepiphysiodesis for idiopathic management of coronal plane deformities rebound phenomenon after removal of
genu valgum: Percutaneous transphyseal of the knee in skeletally immature children. hemiepiphyseal staples in patients with
screw versus tension-band plate. J Pediatr J Child Orthop 2016;10:429-437. idiopathic genu valgum deformity. Bone
Orthop 2018;38:325-330. Joint J 2016;98-B:1270-1275.
17. Noonan KJ, Halanski MA, Leiferman E,
5. Wood M, Davison JE, Cleary MA, Wilsman N: Growth retardation 30. Yang I, Gottliebsen M, Martinkevich P,
Eastwood DM: Guided growth surgery for (hemiepiphyseal stapling) and growth Schindeler A, Little DG: Guided growth:
genu valgum in mucopolysaccharidosis acceleration (periosteal resection) as a Current perspectives and future challenges.
type VI. Mol Genet Metab 2017;120:S141. method to improve guided growth in a JBJS Rev 2017;5:e1.
lamb model. J Pediatr Orthop 2016;36:
6. Yilmaz G, Oto M, Thabet AM, et al: 362-369. 31. Kulkarni RM, Ilyas Rushnaiwala FM,
Correction of lower extremity angular Kulkarni G, Negandhi R, Kulkarni MG,
deformities in skeletal dysplasia with 18. Heflin JA, Ford S, Stevens P: Guided Kulkarni SG: Correction of coronal plane
hemiepiphysiodesis: A preliminary report. J growth for tibia vara (Blount’s disease). deformities around the knee using a tension
Pediatr Orthop 2014;34:336-345. Medicine (Baltimore) 2016;95:e4951. band plate in children younger than 10
years. Indian J Orthop 2015;49:208-218.
7. Gigante C, Borgo A, Corradin M: 19. Sung KH, Chung CY, Lee KM, et al:
Correction of lower limb deformities in Determining the best treatment for coronal 32. Oda JE, Thacker MM: Distal tibial physeal
children with renal osteodystrophy by angular deformity of the knee joint in bridge: A complication from a tension band
guided growth technique. J Child Orthop growing children: A decision analysis. plate and screw construct. Report of a case.
2017;11:79-84. Biomed Res Int 2014;2014:603432. J Pediatr Orthop B 2013;22:259-263.

8. Horn A, Wright J, Bockenhauer D, Van’t 20. Gottliebsen M, Rahbek O, Hvid I, 33. Ballal MS, Bruce CE, Nayagam S: Correcting
Hoff W, Eastwood DM: The orthopaedic Davidsen M, Hellfritzsch MB, Møller- genu varum and genu valgum in children by

10 Journal of the American Academy of Orthopaedic Surgeons

 Tamer A. EL-Sobky, MD, et al

guided growth: Temporary for the treatment of Blount disease. J 36. Lee SH, Chung CY, Park MS, Choi IH,
hemiepiphysiodesis using tension band plates. Pediatr Orthop 2009;29:57-60. Cho TJ: Tibial torsion in cerebral palsy:
J Bone Joint Surg Br 2010;92:273-276. Validity and reliability of measurement.
35. Milner CE, Soames RW: A comparison of
34. Schroerlucke S, Bertrand S, Clapp J, Bundy four in vivo methods of measuring tibial Clin Orthop Relat Res 2009;467:
J, Gregg FO: Failure of Orthofix eight-plate torsion. J Anat 1998;193:139-44. 2098-2104.

January 2020, Vol 4, No 1