International Orthopaedics (SICOT) (2014) 38:273–283

DOI 10.1007/s00264-013-2227-4

REVIEW ARTICLE

Total knee arthroplasty in the valgus knee

Roberto Rossi & Federica Rosso & Umberto Cottino &
Federico Dettoni & Davide Edoardo Bonasia &
Matteo Bruzzone

Received: 17 November 2013 / Accepted: 24 November 2013 / Published online: 24 December 2013
# Springer-Verlag Berlin Heidelberg 2013

Abstract Valgus knee deformity is a challenge in total knee to assess the best pre-operative planning and to evaluate how
arthroplasty (TKA) and it is observed in nearly 10 % of to choose the grade of constraint of the implant. We will also
patients undergoing TKA. The valgus deformity is sustained review the main approaches and surgical techniques both for
by anatomical variations divided into bone remodelling and bone cuts and soft tissue management. Finally, we will report
soft tissue contraction/elongation. Bone tissue variations con- on our experience and technique.
sist of lateral cartilage erosion, lateral condylar hypoplasia and
metaphyseal femur and tibial plateau remodelling. Soft tissue Keywords Knee . Valgus . Arthroplasty
variations are represented by tightening of lateral structures:
lateral collateral ligament, posterolateral capsule, popliteus
tendon, hamstring tendons, the lateral head of the gastrocne- Introduction
mius and iliotibial band. Complete pre-operative planning and
clinical examination are mandatory to manage bone deformi- Valgus knee deformity is a challenge in total knee arthroplasty
ties and soft tissue contractions/elongations and to decide if a (TKA). This deformity (defined as a valgus angle equal to or
higher constrained prosthesis is necessary. Two different ap- greater than 10°) is observed in nearly 10 % of patients
proaches have been described to perform TKA in a valgus undergoing TKA [1]. It can be congenital or secondary to
knee: the anteromedial approach and the anterolateral one. In osteoarthrosis, rheumatic diseases, post-traumatic arthritis and
valgus knee deformity bone cuts can be performed differently to an over-correction consequent to a valgus osteotomy [2].
in order to correct low-grade deformities and reduce great Excessive pre-operative malalignment predisposes to a greater
deformities. There is still debate in the literature on the se- risk of failure compared to well-aligned knees. For this reason
quence of lateral soft tissue release to achieve the best align- it is important to correct the deformity during surgery even if it
ment without any instability. The aim of this article is to does not completely eliminate the increased risk of failure (1.9
review the anatomical variations underlying a valgus knee, vs 0.5 %) [3]. The valgus deformity is sustained by anatomical
variations divided into bone tissue remodelling and soft tissue
R. Rossi (*) : F. Dettoni : M. Bruzzone contraction/elongation. Bone tissue variations consist of later-
AO Mauriziano Umberto I, Department of Orthopedics and al cartilage erosion, lateral condylar hypoplasia and
Traumatology, Largo Turati 62, 10129 Turin, Italy
e-mail: [email protected]
metaphyseal femur and tibial plateau remodelling. Soft tissue
variations are represented by tightening of lateral structures:
F. Rosso : U. Cottino lateral collateral ligament (LCL), posterolateral capsule
University of Study of Turin, Via Giuseppe Verdi, 8, 10100 Turin, (PLC), popliteus tendon (POP), hamstring tendons, the lateral
Italy
head of the gastrocnemius (LHG) and iliotibial band (ITB).
D. E. Bonasia Some authors also described a posterior cruciate ligament
AO CTO-M. Adelaide, Department of Orthopedics and (PCL) alteration in valgus knees, but in the literature its
Traumatology, Via Zuretti 29, 10126 Turin, Italy influence in maintaining the deformity is not universally
recognised. The described deformities can lead to a tibial
R. Rossi
SCDU Ortopedia e Traumatologia, AO Mauriziano Umberto 1, external rotation and to a patellar lateral subluxation tendency
Largo Filippo Turati 62, 10128 Turin, Italy [4]. Three grades of valgus deformity have been described (I,
274 International Orthopaedics (SICOT) (2014) 38:273–283

II, III) [1, 5]. In grade I the deviation is less than 10°, passively (c) Templating
correctable, with contracture of the lateral soft tissue but In the radiographic anteroposterior view of the knee, a
without elongation of the medial collateral ligament (MCL, template of bone cuts should be performed. A line is
80 % of cases). In grade II the axial deviation ranges between drawn on the tibial anatomical axis and then a perpen-
10 and 20°, the lateral structures are contracted and the MCL dicular one is drawn at the level of the lateral tibial
is elongated but functional (15 % of cases). Grade III defor- plateau. This will give the surgeon an indication for the
mity is present in the remaining 5 % of the patients; the axial tibial resection. The femoral anatomical axis is drawn
deformity is greater than 20°, the lateral structures are tight and then a second line with the desired amount of valgus
and the medial stabilisers are not functional. Valgus deformity, (usually 3°) is also drawn at the level of the intercondylar
for these reasons, is a challenge for the surgeon both for gap notch. Figure 2 shows this kind of planning.
balancing and implant constraint choice. Several different On the lateral radiograph the surgeon should investi-
surgical techniques have been described to perform TKA in gate the presence of osteophytes in the posterior capsule;
valgus deformity; the aim of this article is to give an overview furthermore, a lateral view can be used for sizing the
of the most common approaches and to present our choice. femoral component [1] and for locating the entry point of
the femoral canal.
(d) Selection of the implant
Pre-operative planning and implant selection The implant selection should be carried out pre-opera-
tively, based on the radiological and clinical evaluation.
(a) Radiographic planning There is an open debate in the literature between
In our experience mandatory pre-operative radio- posterior-stabilised (PS) and cruciate-retaining (CR) im-
graphs of the knee undergoing TKA are: weight- plants, also in valgus deformity. In valgus deformity, the
bearing anteroposterior, lateral, Rosenberg and Merchant PCL is often contracted and it may limit the deformity
views. These are useful for planning and bone stock correction [7]. Furthermore, it may be more difficult to
evaluation. Attention should be focused on lateral distal obtain the deformity correction with an intact PCL, since
femoral hypoplasia, posterior femoral condyle erosion the PCL is a secondary stabiliser [8, 9]. Besides, the PS
and metaphyseal remodelling both of the femur and tibia, design is more stable than a CR one because of the post-
which can lead to malalignment or malrotation of the cam mechanism and the PS design allows for greater
femoral component. The patellofemoral joint can be lateralisation of the femoral and tibial components, which
partially dislocated. In addition anteroposterior and lat- improves patellar tracking and minimises the necessity to
eral views are mandatory to evaluate the amount of perform a lateral retinacular release [1]. For these reasons
osseous resection needed to correct deformities without some authors suggested that it is simpler to substitute a
leading to knee instability. A weight-bearing long leg contracted PCL with a PS design than to stabilise it using a
view is fundamental for the evaluation of lower limb CR implant, recommending that a PS design be used in
alignment, deformity level and to plan the amount of valgus deformity [2]. Figures 3 and 4 show a case in which
correction [1, 6]. a PS implant has been used in a valgus deformity. There is
(b) Knee evaluation some concern about the association between planovalgus
The overall alignment should be assessed both in the foot and failure of a CR implant. As described in the
supine and weight-bearing positions, and the gait should literature, a careful foot examination is mandatory to check
be observed, in order to identify other dynamic instabil- for the presence of ipsilateral posterior tibial tendon insuf-
ities. Any sagittal deformity, such as fixed flexion con- ficiency, especially in patients with a valgus knee [10].
tracture or recurvatum, as well as any rotational defor- Figure 5 shows a case of valgus knee associated with a
mity, should be tested during the physical examination. planovalgus foot.
The knee should be evaluated for anteroposterior laxity, Another debated issue is the amount of constraint
range of motion (ROM), coronal and sagittal deformity, needed to balance a valgus knee. Careful pre-operative
and mediolateral instability. A crucial point is to assess if planning is essential to estimate the grade of constraint
valgus deformity is fixed or still reducible. If a fixed that will be necessary during surgery. Accurate physical
deformity is present, the lateral structure is really tight examination is mandatory to evaluate if a valgus defor-
and medial ligaments are partially continent. In these mity is fixed or reducible: if a fixed valgus deformity is
cases after the lateral soft tissue release, there may be a present, after lateral soft tissue release the knee may be
laxity requiring the use of a constrained prosthesis. If the unstable because both the medial and lateral structures
deformity is reducible, soft tissue release will be less will no longer be functional. In these cases the surgeon
invasive, and a standard unconstrained prosthesis could should plan to also have a semi-constrained prosthesis in
be sufficient [6]. Figure 1 shows a reducible valgus knee. the operating theatre, such as a constrained condylar one.
International Orthopaedics (SICOT) (2014) 38:273–283 275

Fig. 1 Clinical examination

showing a reducible valgus knee.
a Valgus knee. b Reduction of the
valgus during a varus stress

There is some agreement in the literature that the sur- controversial in the literature [13, 14]. In the past, if
geon should resist the temptation to use a higher arthritis and extra-articular deformity coexisted, si-
constrained prosthesis. In 2013 [11] Pang et al. reported multaneous corrective osteotomy of angular deformity
a retrospective review of patients with a type II valgus and TKA have been considered the correct treatment.
knee who underwent primary TKA. There were 50 This highly demanding procedure increases the risk of
patients in whom a constrained TKA was performed poor TKA result in the case of osteotomy failure. It is
and another 50 cases who received a PS or CR prosthe- important to consider that even in extra-articular de-
sis. They concluded that there were no statistically sig- formities alignment correction is possible in the ma-
nificant differences in anteroposterior and mediolateral jority of cases without extra-articular procedures.
stability between the two groups. Constrained TKA were Good pre-operative planning and templating, intra-
associated with more significant joint line changes com- articular bone resection and soft tissue balancing in
pared with the unconstrained ones. flexion and extension are necessary [15, 16].
On the other hand, in grade III valgus deformity, in
which the medial soft tissue is no longer functional with
certainty, a higher constrained prosthesis is mandatory to
achieve a stable knee [1]. Figures 6 and 7 show a case Approaches
of grade III valgus knee in which a higher constrained
prosthesis was implanted. Anteromedial approach
(e) Osteotomy in the valgus knee and TKA
Patients with genu valgum and isolated lateral com- The anteromedial approach is used by the majority of authors
partment osteoarthritis are candidates for distal femoral and there are no contraindications even in the valgus knee [1,
varus osteotomy. 5]. The patellar dislocation is usually easy with this approach,
Osteotomies are performed in order to delay TKA because of the combination of the valgus deformity and the
implant [12] and to correct extra-articular deformities lateralisation of the tibial tuberosity (TT) [2]. When a medial
leading to early arthritis [13]. When the implant of a approach is used in a valgus knee, the surgeon should be very
TKA cannot be delayed anymore, it can be more chal- careful with the MCL detachment: the release of the medial
lenging if an osteotomy has been previously per- structures should be minimised in the valgus deformities and
formed. The results of these procedures still remain limited to overhanging osteophytes.
276 International Orthopaedics (SICOT) (2014) 38:273–283

The main disadvantage of the medial approach is that it is

more difficult to reach the posterolateral corner during the
lateral soft tissue release. For this reason sometimes a tibial
tuberosity osteotomy (TTO) is necessary; different complica-
tions have been reported with this technique. Besides, patellar
vascular damage has been described when a medial
parapatellar approach is combined with a lateral release [17].
Some authors reported that when a medial approach is per-
formed in a valgus knee, the results have been inferior com-
pared to varus deformity [18].

Anterolateral approach

In 1991 Keblish [4] described an anterolateral approach for

TKA in valgus deformities. He described a long incision along
the lateral border of the quadriceps muscle, taking care to
leave 1 cm of the lateral retinaculum, from the junction
between the vastus lateralis and the quadriceps tendon to the
patella, through 50 % of the tendon. The patella is dislocated
medially. Some authors [19, 20] described a TTO also to
protect the extensor mechanism during patellar eversion with
good results [17]. According to Keblish it may be difficult to
close the lateral compartment after the deformity correction.
Two different tricks are described to facilitate lateral closure:
(1) approximation of the infrapatellar fat pad to the patellar
Fig. 2 Anteroposterior view of an X-ray showing our preoperative
ligament and (2) separation of the vastus lateralis from the
planning to evaluate the bone cuts. The femoral anatomical axis is drawn rectus femoris, followed by suturing together the two tendons
and then a second line with the desired amount of valgus (usually 3°) is in a staggered position [4].
also drawn at the level of the intercondylar notch According to different authors [21], the main advantage of
the lateral approach is a better visualisation of the tight lateral
tissues; besides, if a lateral retinaculum release is necessary,

Fig. 3 Patient with a grade II left

valgus knee. a Long leg view
showing the valgus deformity.
b Anteroposterior view. c Lateral
view showing posterior
osteophytes
International Orthopaedics (SICOT) (2014) 38:273–283 277

Fig. 4 Post-operative X-ray

of the same patient as in Fig. 3
showing a well-positioned PS
implant. a Post-operative
anteroposterior and lateral views.
b Anteroposterior long leg view at
3 months showing a well-aligned
left knee

the patellar vascularisation will be not compromised. On the medial parapatellar approach was used and in the second one a
other hand, potential disadvantages of this technique include lateral parapatellar approach without TTO. They found no
difficulty in patellar eversion, sometimes requiring TTO, and significant differences in ROM but better post-operative flex-
less familiarity of many surgeons with this technique [22]. ion in the second group. Nikolopoulos et al. [23] reported on
There is an open debate about which approach leads to two groups: in the first a lateral parapatellar approach com-
better results. Sekiya et al. [21] reported on clinical and bined with TTO had been performed and in the second one a
radiological results in two randomised groups of patients with standard medial approach had been used. They found no
a valgus deformity after performing a TKA. In the first group a statistically significant differences in terms of clinical results,

Fig. 5 Clinical examination and

long leg view of the same patient
as in Figs. 3 and 4 showing the
association between planovalgus
foot and valgus knee. a Clinical
examination of the planovalgus
foot. b Anteroposterior long leg
view showing the valgus knee
278 International Orthopaedics (SICOT) (2014) 38:273–283

Fig. 6 Patient with a grade III

bilateral valgus knee more
symptomatic to the right side.
a Clinical examination showing
bilateral valgus knee. b, c Long
leg anteroposterior X-ray view
showing bilateral valgus

but in the lateral approach group a valgus deviation occurred Bone cuts
in 9 % of the patients, compared to 32 % in the medial
approach one. A similar study has been published by In valgus knee deformity bone cuts can be performed differ-
Hirschmann et al. [24], but they concluded that long-term ently in order to correct low-grade deformities and reduce
studies are necessary to show whether there is any difference great deformities. In order to make the right cut the surgeon
in prosthesis longevity between both types of approach. should pay attention to valgus knee bony variables: lateral

Fig. 7 Post-operative X-ray

on anteroposterior and lateral
views showing the rotating hinge
prosthesis implanted in the
same patient as in Fig. 6.
a Anteroposterior view.
b Lateral view
International Orthopaedics (SICOT) (2014) 38:273–283 279

femoral condylar hypoplasia and asymmetric cartilage wear each release the surgeon should evaluate the alignment
on the tibial plateau. These characteristics can influence limb and the stability of the knee, in order to achieve a
alignment, component rotation and patellar tracking [1]. symmetrical rectangular extension and flexion gaps
with the spacer block in situ.
Tibia Krackow et al. [5] normally release the ITB and
LCL first in the type I valgus knee, followed by the
The cut has to be perpendicular to the tibial long axis. The POP and the PLC, when necessary.
difference between these cases and a standard TKA is the Ranawat et al. [1] described a stepwise technique in
amount of the resection in grade II and III valgus deformi- which the first structure to be released is the PCL.
ties. The resection should be from 6 to 8 mm in the medial Then they perform a PLC intra-articular release using
compartment and always has to be performed after having an electrocautery at the level of the tibial cut surface.
removed all the osteophytes, especially in the lateral side The ITB is released when necessary with multiple
of the tibial plateau. In cases of severe bony deformity of inside-out stab incisions, as well as the LCL; on the
the tibial plateau, it can occur that almost no bone is contrary, the POP is normally preserved.
resected on the lateral side to avoid medial over-resection Favorito et al. [2] described that more commonly
or malaligned cuts. the LCL is the tightest structure, so it is the first
structure to be released. The next release is the POP,
Femur followed by the PLC, the femoral insertion of the LHG
and, finally, the ITB can be considered.
Distal cut: It is useful for the femoral cut to reduce valgus Different authors tried to described a lateral struc-
degrees of resection from 5–7° to 3° in order to properly ture release sequence based on functional and anatom-
correct pre-operative deformity. Also at this level lateral con- ical consideratios. Most of them agree in describing
dyle distal femoral resection can be minimal (1–2 mm) or the POP as an important structure for rotational and
absent in severe valgus deformity of the distal femur. Femoral valgus stability in flexion.
resection should be no more than 10 mm in the medial condyle Whiteside [26] described a soft tissue release se-
(usually 7–8 mm). quence based on the functional effect of the single
Anteroposterior cuts: The surgeon has to pay attention to structures. In his opinion, a ligament attached to the
lateral condylar hypoplasia that can determine a great femur near the epicondyles, so near the axis through
intra-rotation of the components if a posterior reference is which the tibia rotates and the knee flexes and extends,
used. Considering this aspect, Arima et al. support the utility has an important role in flexion stability. On the other
of using the anteroposterior axis in order to give the proper hand, a ligament attached far away from the
femoral rotation in valgus anatomy [25]. In cases of severe epicondyle is more important for the extension knee
trochlear dysplasia, the Whiteside line can be extremely diffi- stability. Following this theory, the LCL and POP,
cult to identify: in these cases the epicondylar axis or parallel which are attached to the epicondyle, are important
to the tibial cut technique should be used to assess a correct lateral stabilisers in flexion; these two structures are
femoral rotation. appropriate to release for a knee that is tight in flexion.
On the contrary, the ITB and the PLC are important
knee stabilisers in extension, so they should be re-
Soft tissue management leased when the knee is tighter in extension. Whiteside
concluded that a standard lateral release sequence is
(a) Lateral soft tissue not applicable to all patients, but the surgeon should
In the valgus knee many lateral structures are retracted: release a different structure if the knee is tight in
the ITB, the PLC, the LCL, the POP and the LHG. flexion, in extension or during all ROM.
In knees with severe valgus deformity, the PCL may Krackow and Mihalko [27] published a cadaveric
also be retracted, and it cannot be retained if the surgeon study in which they studied the amount of correction
wants to achieve full correction of the deformity. achieved with each release step of two different se-
There is agreement in the literature that lateral structure quences, comparing it in flexion and extension and
release is necessary in valgus deformity, but there is an measuring any rotational changes of the tibia. The
open debate on which is the best sequence and the best sequences were: ITB, POP, LCL and LHG and LCL,
technique to perform those releases. POP, ITB and LHG. They evaluated the amount of
The releases should be performed with the knee in correction at 0, 45 and 90° of flexion. The results
extension and using lamina spreaders to check the showed that the greatest varus rotation occurred once
tension of the medial and lateral compartments. After all structures were released, with the LHG origin last in
Table 1 Results reported in the literature on TKA in valgus knee using different release techniques and sequences
280

Authors Knees Valgus Technique Implant Results Follow-up Conclusions

deformity

Krackow et al. 99+40 Types I and II Type I: lateral soft tissue release CR Knee Society post-operative knee Minimum Same result for types I and II
(1991) [5] Type II: medial capsular score was 87.6 (± 10.6) and mean 2 years
ligament tightening post-operative functional score
was 52.3
Whiteside (1999) [26] 231 12–45° Tight in flex. and ext.: CR 1 % tight only in flexion and extension 6 years No cases of clinical instability
LCL+POP; tight in ext.: (10 % required release of the PLC), occurred, and joint stability
ITB; tight in flex.: LCL+POP 12 % tight only in extension; no did not deteriorate with time
posterior capsule release higher constrained prosthesis; no
was done only when necessary post-operative instability at 6 years
follow-up
Krackow and Mihalko 6 cadaveric None LCL first; POP and ITB None Complete lateral structure release: None When severe valgus deformities
(1999) [27] knees to grade the release limited correction to varus; the are present, the LCL should be
lateral aspect of the flexion gap considered first for release and
opens more than the extension gap; the POP and ITB have to be
early LCL release provides a more used to grade the release
uniform release of the gap; rotational
changes are variable, 6° tibial external
rotation in full extension with LCL release
Brilhault et al. 13 >10° LCL advancement CCK KSS 32–88; functional score 45–73 6.5 years No post-operative tibiofemoral or
(2002) [33] patellar instability and satisfactory
stable alignment
Ranawat et al. (2005) [1] 42 10° Inside-out pie-crusting of the PS Knee Society clinical score improved 5 years Reproducible technique with
ITB; resection of the proximal from 30 to 93 points; mean functional good results
part of the tibia and distal score improved from 34 to 81 points;
part of the femur to provide mean ROM 110°; 3 patients underwent
a balanced, rectangular space revision; no cases of delayed instability
Boyer et al. (2009) [29] 63 >10° Lateral parapatellar approach, PS Knee score improved from 37 to 91, 7 years Good access to tight structures
ITB release, PLC and flexion from 113 to 117°, functional with lateral parapatellar approach;
gastrocnemius release score from 29.5 to 78.7 and pain most important release is the ITB
successively score from 0.8 46.9
Mullaji and Shetty 10 >10° LCL+POP release with sliding No complications 20 months Computer navigation allows precise
(2010) [35] osteotomy using computer measurement of the difference
navigation between medial and lateral gaps
as well as the limb alignment and
to determination of the effect of
sequential soft tissue releases on both
Bremer et al. 79 8–40° Distalisation of the insertion CR Oxford score improved from 22 points Good stability after procedure; no
(2012) [34] of the LCL and POP by sliding preoperatively to 45 points; 1 knee conversion to a semi-constrained
osteotomy of the lateral femoral revised for infection and 1 due to or constrained knee prosthesis
condyle non-union of tibial tubercle; 3 cases
with complications associated with
the procedure; all revised successfully
International Orthopaedics (SICOT) (2014) 38:273–283

CCK constrained condylar knee, KSS Knee Society score

International Orthopaedics (SICOT) (2014) 38:273–283 281

both groups. Moreover, the largest increase occurred and a residual medial laxity is poorly tolerated if a valgus
after the release of the LCL. They concluded that in deformity persists post-operatively. In these conditions
severe valgus deformities, the LCL should be consid- the authors suggested tightening the medial ligamentous
ered first for release; POP and ITB should be used to structures, particularly if the PCL has been retained. In
grade the release. 1990 Krackow et al. [36] described a tightening tech-
Three main techniques are described in the litera- nique in which a small bone plug with the attached
ture to perform lateral soft tissue release. insertion of the PCL and the PLC is removed from the
Different authors described a subperiosteal tibia and moved distally, securing it with transosseous
technique or a Z-plasty technique to perform the sutures. In this technique the MCL is tightened by mov-
lateral releases [26–28]. In the technique described ing a bone block distally with its tibial insertion. Other
by Krackow and Mihalko [27] the knee was fully techniques have been described. The advancement of the
extended and the ligaments were palpated to ascertain MCL from the epicondyle or a division and imbrication
which structures were tightest. Typically the ITB and in order to tighten it can be performed in conjunction
the LCL were released first with a subperiosteal tech- with the use of a constrained condylar prosthesis [2].
nique, followed by the POP and occasionally by the
PLC. The LHG was released only if a flexion contrac-
ture were present. In the technique described by
Complications
Whiteside [26] the LCL and the POP were released
directly from the bone attachments leaving the perios-
In the literature different main complications have been
teum intact. The ITB was released subcutaneously and
described [2]:
released extrasynovially.
Other authors [4, 29] used a lateral parapatellar
& Tibiofemoral instability (2–70 %)
approach and automatically released the ITB from
& Recurrent valgus deformity (4–38 %)
Gerdy’s tubercle.
& Poor post-operative ROM (1–20 %)
Finally, Ranawat’s “pie-crusting technique” has
& Wound problems (4–13 %)
been described for releasing the lateral structure with an
& Patellar stress fracture and osteonecrosis (1–12 %)
inside-out procedure. In this technique the tight lateral
& Patellar maltracking (2–10 %)
structures are palpated when the lamina spreaders are
& Peroneal nerve palsy (0.3–9.5 %) [37, 38]
inserted, and they are released performing multiple trans-
verse incisions with a no. 15 surgical blade [30, 31]. One
Correction of a valgus deformity places the peroneal
of the disadvantages of this technique is the potential risk
nerve at risk for indirect injury via traction, ischaemia or
of peroneal nerve lesion. Bruzzone et al. [32], in a ca-
both. Apart from indirect mechanisms, concern has been
daveric study, concluded that the nerve is at overall risk
expressed over the use of the Ranawat inside-out release,
during the release of the PLC, in the triangle defined by
which may place the nerve at risk for direct laceration-type
the POP, the tibial cut surface and the most posterior
injuries. Direct injury to the peroneal nerve is most likely to
fibres of the ITB (“danger zone”), but not during the
occur during the posterolateral release of the capsule carried
pie-crusting of the ITB (“safe zone”).
out at the level of the tibial cut using the Ranawat technique
An alternative technique for lateral structure release
than during ITB pie-crusting [32].
has been described by Brilhault et al. associated with a
lateral parapatellar approach [33]. A sliding osteotomy of
the femoral LCL and POP insertions is done and the
resulting bone block is mobilised and placed more dis-
tally. This procedure produces a rectangular space and
had great results in Bremer et al.’s study: no conversion
to semi-constrained or constrained knee prosthesis [34].
Mullaji and Shetty [35] described a similar technique in
which, after the release of the PLC and the ITB, they
performed a computer navigated lateral epicondylar
osteotomy, with a more accurate repositioning of the
epicondyle.
(b) Medial soft tissue
As described by Krackow et al. [5] in grade II valgus
deformities the MCL may not be completely functional, Fig. 8 Cadaveric anatomy showing the POP, PLC and ITB
282 International Orthopaedics (SICOT) (2014) 38:273–283

Results the anatomy of the lateral structures. The knee is then stressed
in varus/valgus to evaluate the stability in extension and, if
Table 1 summarises the results in the literature with different further releases are necessary, we restart the procedure. In
release sequences. accordance with Ranawat et al. [1] we try to preserve at least
one or two lateral stabilisers. If instability is detected after the
releases have been performed we switch to a condylar
Our technique constrained prosthesis. Once the knee is balanced in exten-
sion, the flexion gap can be evaluated and assessed. When the
In all cases we perform weight-bearing anteroposterior, later- extension and flexion gaps are equal, femoral chamfer cuts
al, Rosenberg, Merchant and long leg views to assess the can be made and the trial components can be tested. To assess
lower limb alignment and the distal femoral and proximal tibial component rotation and sizing, we use the posterolateral
tibial remodelling. We perform pre-operative planning using corner locked technique we previously described [39].
the technique we described in the previous section (Fig. 2). At this point it is important to evaluate patellar tracking
We routinely use a PS implant in type I deformity. In type II with the tourniquet deflated and the “no thumb” technique. If
deformity we decide the level of constraint on the operative the trial components are well positioned and a lateral release is
field, based on the integrity and functionality of the MCL. If still necessary, a pie-crusting release of the lateral retinaculum
there is a medial residual instability we do not perform a can be performed. At this point the knee is irrigated, the
medial tightening, but we prefer to switch to a higher tourniquet is inflated and the components are cemented in
constrained implant. In type III deformities we routinely use position. During the pressurisation cement excess is removed.
a condylar constrained or, in the most severe cases, a rotating After cement polymerisation, the knee is checked in all posi-
hinge prosthesis. tions and a drain is positioned intra-articularly. The capsule is
Antibiotic prophylaxis is routinely administered to the closed in flexion taking care to avoid patella baja or alta.
patients (depending on the hospital's protocols), and they are The post-operative protocol is standard with continuous
positioned supine, as in a routine implant. passive motion from day one and weight-bearing with
We routinely perform a medial parapatellar approach, be- crutches as tolerated from day two. Controls are clinical after
ing really careful not to excessively release medial structures, two months, radiographic and clinical at three months and
minimising medial dissection to fully visualise the tibia. We one year and successively every two years.
first perform the tibial cut, perpendicular to the anatomical
axis, removing the smallest possible bone amount, especially
from the lateral side. For the femoral cuts we reduce the valgus
Conflict of interest The authors declare that they have no conflict of
degree of resection to 3°, in order to avoid under-correction of
interest.
the deformity. We pay attention to correctly access the femoral
canal with the intramedullary rod: the entry point in a valgus
knee is usually more medial than in a standard knee.
Pre-operative radiographs are extremely useful to correctly
References
access the canal avoiding gross mistakes.
With regard to anteroposterior cuts, we take extreme care in
evaluating possible hypoplasia of the lateral femoral condyle 1. Ranawat AS, Ranawat CS, Elkus M, Rasquinha VJ, Rossi R,
Babhulkar S (2005) Total knee arthroplasty for severe valgus defor-
and we triple-check the posterior condylar reference cutting mity. J Bone Joint Surg Am 87 Suppl 1(Pt 2):271–284
block position with both Whiteside line and transepicondylar 2. Favorito PJ, Mihalko WM, Krackow KA (2002) Total knee
axis; moreover, with the cutting blocks in site, we further arthroplasty in the valgus knee. J Am Acad Orthop Surg 10(1):16–24
check the balancing in flexion before performing the cuts. 3. Ritter MA, Davis KE, Davis P, Farris A, Malinzak RA, Berend
ME, Meding JB (2013) Preoperative malalignment increases risk
We follow the “three-step technique” described by of failure after total knee arthroplasty. J Bone Joint Surg Am
Ranawat et al. to evaluate the lateral soft tissue release [1]. 95(2):126–131
When distal femoral and proximal tibial cuts are performed, 4. Keblish PA (1991) The lateral approach to the valgus knee. Surgical
the osteophytes have been removed and the PCL has been technique and analysis of 53 cases with over two-year follow-up
evaluation. Clin Orthop Relat Res 271:52–62
sacrificed, we put the knee in extension with lamina spreaders 5. Krackow KA, Jones MM, Teeny SM, Hungerford DS (1991)
to evaluate the extension gap. At this point we routinely Primary total knee arthroplasty in patients with fixed valgus defor-
perform a selective lateral soft tissue release. We palpate the mity. Clin Orthop Relat Res 273:9–18
lateral structure and we proceed with a pie-crusting release 6. Robbins GM, Masri BA, Garbuz DS, Duncan CP (2001)
Preoperative planning to prevent instability in total knee arthroplasty.
just of the structures that are really tight, starting from the Orthop Clin North Am 32(4):611–626
PLC, followed by the ITB. We normally preserve the POP 7. Krackow KA (1990) The technique of total knee arthroplasty. Mosby,
because of its importance in flexion stability. Figure 8 shows St. Louis
International Orthopaedics (SICOT) (2014) 38:273–283 283

8. Matsuda Y, Ishii Y, Noguchi H, Ishii R (2005) Varus-valgus balance parapatellar approach combined with tibial tubercle osteotomy.
and range of movement after total knee arthroplasty. J Bone Joint Knee Surg Sports Traumatol Arthrosc 19(11):1834–1842
Surg Br 87:804–808 24. Hirschmann MT, Hoffmann M, Krause R, Jenabzadeh RA, Arnold
9. Mihalko WM, Krackow KA (2000) Anatomic and biomechanical MP, Friederich NF (2010) Anterolateral approach with tibial tubercle
aspects of pie crusting posterolateral structures for valgus deformity osteotomy versus standard medial approach for primary total knee
correction in total knee arthroplasty: a cadaveric study. J Arthroplasty arthroplasty: does it matter? BMC Musculoskelet Disord 11:167
15:347–353 25. Arima J, Whiteside LA, McCarthy DS, White SE (1995) Femoral
10. Meding JB, Keating EM, Ritter MA, Faris PM, Berend ME, rotational alignment, based on the anteroposterior axis, in total knee
Malinzak RA (2005) The planovalgus foot: a harbinger of failure arthroplasty in a valgus knee. A technical note. J Bone Joint Surg Am
of posterior cruciate-retaining total knee replacement. J Bone Joint 77(9):1331–1334
Surg Am 87(Suppl 2):59–62 26. Whiteside LA (1999) Selective ligament release in total knee
11. Pang HN, Yeo SJ, Chong HC, Chin PL, Chia SL, Lo NN (2013) Joint arthroplasty of the knee in valgus. Clin Orthop Relat Res 367:
line changes and outcomes in constrained versus unconstrained total 130–140
knee arthroplasty for the type II valgus knee. Knee Surg Sports 27. Krackow KA, Mihalko WM (1999) Flexion-extension joint gap
Traumatol Arthrosc 21(10):2363–2369 changes after lateral structure release for valgus deformity correction
12. Dettoni F, Bonasia DE, Castoldi F, Bruzzone M, Blonna D, Rossi R in total knee arthroplasty: a cadaveric study. J Arthroplasty 14(8):
(2010) High tibial osteotomy versus unicompartmental knee 994–1004
arthroplasty for medial compartment arthrosis of the knee: a review 28. Engh GA (2003) The difficult knee: severe varus and valgus. Clin
of the literature. Iowa Orthop J 30:131–140 Orthop Relat Res 416:58–63
13. Windsor RE, Insall JN, Vince KG (1988) Technical considerations of 29. Boyer P, Boublil D, Magrino B, Massin P, Huten D, Guepar Group
total knee arthroplasty after proximal tibial osteotomy. J Bone Joint (2009) Total knee replacement in the fixed valgus deformity using a
Surg Am 70:547–555 lateral approach: role of the automatic iliotibial band release for a
14. Nelson CL, Saleh KJ, Kassim RA, Windsor R, Haas S, Laskin R, successful balancing. Int Orthop 33(6):1577–1583
Sculco T (2003) Total knee arthroplasty after varus osteotomy of the 30. Aglietti P, Lup D, Cuomo P, Baldini A, De Luca L (2007) Total knee
distal part of the femur. J Bone Joint Surg Am 85-A(6):1062–1065 arthroplasty using a pie-crusting technique for valgus deformity. Clin
15. Rajgopal A, Vasdev A, Dahiya V, Tyagi VC, Gupta H (2013) Total Orthop Relat Res 464:73–77
knee arthroplasty in extra articular deformities: a series of 36 knees. 31. Clarke HD, Fuchs R, Scuderi GR, Scott WN, Insall JN (2005)
Indian J Orthop 47(1):35–39 Clinical results in valgus total knee arthroplasty with the “pie
16. Xiao-Gang Z, Shahzad K, Li C (2012) One-stage total knee crust” technique of lateral soft tissue releases. J Arthroplasty 20(8):
arthroplasty for patients with osteoarthritis of the knee and extra- 1010–1014
articular deformity. Int Orthop 36(12):2457–2463 32. Bruzzone M, Ranawat A, Castoldi F, Dettoni F, Rossi P, Rossi R
17. Buechel FF (1990) A sequential three-step lateral release for (2010) The risk of direct peroneal nerve injury using the Ranawat
correcting fixed valgus knee deformities during total knee “inside-out” lateral release technique in valgus total knee
arthroplasty. Clin Orthop Relat Res 260:170–175 arthroplasty. J Arthroplasty 25(1):161–165
18. Apostolopoulos AP, Nikolopoulos DD, Polyzois I, Nakos A, 33. Brilhault J, Lautman S, Favard L, Burdin P (2002) Lateral femoral
Liarokapis S, Stefanakis G et al (2010) Total knee arthroplasty in sliding osteotomy lateral release in total knee arthroplasty for a fixed
severe valgus deformity: interest of combining a lateral approach valgus deformity. J Bone Joint Surg Br 84(8):1131–1137
with a tibial tubercle osteotomy. Orthop Traumatol Surg Res 96(7): 34. Bremer D, Orth BC, Fitzek JG, Knutsen A (2012) Briard’s sagittal
777–784 sliding osteotomy of the lateral condyle in total knee arthroplasty of
19. Whiteside LA (1993) Correction of ligament and bone defects in total the severe valgus knee. Oper Orthop Traumatol 24(2):95–108
arthroplasty of the severely valgus knee. Clin Orthop 288:234–245 35. Mullaji AB, Shetty GM (2010) Lateral epicondylar osteotomy using
20. Wolff AMN, Hungerford DS, Pepe CL (1991) The effect of extra- computer navigation in total knee arthroplasty for rigid valgus defor-
articular varus and valgus deformity on total knee arthroplasty. Clin mities. J Arthroplasty 25(1):166–169
Orthop 271:35–51 36. Krackow KA, Holtgrewe JL (1990) Experience with a new technique
21. Sekiya H, Takatoku K, Takada H, Sugimoto N, Hoshino Y (2012) for managing severely overcorrected valgus high tibial osteotomy at
Lateral approach is advantageous in total knee arthroplasty for valgus total knee arthroplasty. Clin Orthop Relat Res 258:213–224
deformed knee. Eur J Orthop Surg Traumatol 37. Nercessian OA, Ugwonali OF, Park S (2005) Peroneal nerve palsy
22. Fiddian NJ, Blakeway C, Kumar A (1998) Replacement arthroplasty after total knee arthroplasty. J Arthroplasty 20:1068–1073
of the valgus knee. A modified lateral capsular approach with repo- 38. Schinsky MF, Macaulay W, Parks ML et al (2001) Nerve injury after
sitioning of vastus lateralis. J Bone Joint Surg Br 80:859–861 primary total knee arthroplasty. J Arthroplasty 16:1048–1054
23. Nikolopoulos DD, Polyzois I, Apostolopoulos AP, Rossas C, 39. Rossi R, Bruzzone M, Bonasia DE, Marmotti A, Castoldi F (2010)
Moutsios-Rentzos A, Michos IV (2011) Total knee arthroplasty in Evaluation of tibial rotational alignment in total knee arthroplasty: a
severe valgus knee deformity: comparison of a standard medial cadaver study. Knee Surg Sports Traumatol Arthrosc 18(7):889–893