**4. Technical tips**

The purpose of this chapter is to summarize tips used to face some technical challenges in CP hip surgery.

First, we cannot plan a procedure correctly without an accurate previous evaluation. There are several moments (in CP cases) where we can easily have a false perception of reality, and some of those points are discussed below.

Second, the surgery planning is the moment where we try to imagine the surgical approaches and the technical steps we will have to fulfill and what eventual orthopedic materials and respective ancillary equipment we will need to achieve our osteosynthesis. Sometimes, even the enterprises who sell orthopedic material have their own technique manuals not adapted to the specificity of this kind of "extreme" deformities. This is also the reason for the following explanations.

Third, the main goal in hip surgery, is a correct and, if possible, concentric hip reduction; as so, a good femoral osteosynthesis is not, *per se*, enough if final hip reduction is not satisfying. Taking care to shorten femur and to stabilize the femoral head in correct position (acetabulum "cleansing", iliopsoas distal section, acetabuloplasty, and/or capsulorrhaphy), without any residual stress in abduction, are mandatory to have a successful procedure.

These are the technical highlights that usually can raise some theoretical and practical discussion.

#### **4.1. Correct evaluation of hip deformities**

cast/splint—NB—Dega Osteotomy is to consider, even with closed triradiate cartilage [14], if very shallow acetabulum and reduction is not enough after capsuloplasty, when

• 12–14 Y-O (completely mature) sub-luxated/dislocated hip(s), with Reimer's Index >**70%**, **NOT reducible** after adductor tenotomy alone and with arthritic deformity.

○ (First option) adductor tenotomy + medial hamstrings tenotomy + iliopsoas tenotomy + open reduction (queiloplasty) + varus and shortening proximal femoral osteotomy + reduction capsuloplasty + > 1 month abduction cast/splint—NB—you can also consider Dega Osteotomy even with closed triradiate cartilage if very shallow acetabu-

○ (Second option, after painful failure of previous attempt of hip reduction) "salvage procedure", per example, Mc Hale procedure: it is a 90° proximal femoral valgus osteotomy with suture of distal ligamentum teres to lesser trochanter to avoid subsequent

**2.** For hip internal rotation (sometimes associated to knee flexed and foot equinus varus, in

• Concerning only the hip, in this chapter: sub-trochanteric external rotation osteotomy with plate (leaving ~30° passive internal rotation) in children or diaphyseal external rotation osteotomy with a nail in adolescents, in order to permit early standing. Usually, intertrochanteric osteotomy is proposed [7, 8], but, in walking patients, there is a risk of an further femoral head subluxation because of the continuous action of hip rotators muscles, and namely iliopsoas; after post-operative healing, its spastic action continues, forcing hip internal rotation, contributing for further dislocation. I believe that immediate sub-trochanter osteotomy (instead of intertrochanteric) is safer in the long-term evolution.

The purpose of this chapter is to summarize tips used to face some technical challenges in CP

First, we cannot plan a procedure correctly without an accurate previous evaluation. There are several moments (in CP cases) where we can easily have a false perception of reality, and

Second, the surgery planning is the moment where we try to imagine the surgical approaches and the technical steps we will have to fulfill and what eventual orthopedic materials and respective ancillary equipment we will need to achieve our osteosynthesis. Sometimes, even the enterprises who sell orthopedic material have their own technique manuals not adapted to the specificity of this kind of "extreme" deformities. This is also the reason for the following

you accept higher risk of complications.

femoral uprising.

142 Cerebral Palsy - Clinical and Therapeutic Aspects

diplegic patients)

**4. Technical tips**

some of those points are discussed below.

hip surgery.

explanations.

lum and reduction is not enough after capsuloplasty OR

After clinical evaluation, X-ray is fundamental. However, generally the patient is awake and information is biased by malposition of patient and spasticity. Or, if we do not have access to EOS® technology [9] or an eventual 3D CT scan, a correct understanding of real bony deformities, will be achieved only by X-ray evaluation under anesthesia. After this, we can have a clear idea of the real initial state, and we can adapt our planning to reach the final state desired.

A hyper-lordosis can preclude correct evaluation of acetabulum parameters, or an excessive femur rotation can increase an impression of femoral valgus (**Figure 4a** and **b**).

Another practical and important question is the real evaluation of valgus deformity when the physis have a long period of progressive deformation. Instead of being in the continuity of the femoral neck, it is deformed in valgus, sometimes adding about 10° more in evaluation (**Figure 5a** and **b**). This notion will be important for the amount of needed correction and for the correct perception of hip reduction.

**Figure 4.** (a, b) In AP view, the cervico-diaphyseal angle is measured as a 158° valgus, but if we correct rotation of the thigh, achieving a true femoral neck AP view (the great trochanter physis is clearly viewed), we find a 134° real angle the proximal femoral varization will be less important than planned.

**Figure 5.** (a) dislocated hip; the cervical line doesn't correspond to "head cervical" line, because of head progressive deformity in valgus, (b) post-operative reduced hip; when head is correctly reduced, we observe a bigger "true" cervicodiaphyseal angle than classically measured and an "infra leveling" of the Shenton's line.

#### **4.2. Calculation of the final angle in femoral osteotomy and bone osteosynthesis**

As mentioned previously, when we plan a proximal femoral osteotomy, the goal for the inner question of the femur is to reach a final state where the cervico-diaphyseal angle (or head-cervicodiaphyseal angle, when femoral head is in valgus) is about 120° in the AP view, with a correction of rotation that permits about 30° of hip internal rotation and an eventual femur shortening, depending on the previous amount of femur uprising associated with dislocation. Sometimes, when we have a flexed hip, we can incorporate a "deflexion" procedure, adding an extension component at the osteotomy site (with a corresponding slope for the plate positioning).

regarding the point of entrance on lateral femur), and *guide wire angle* is the angle of entrance

**Figure 6.** Planning varus osteotomy with a transfer; we measure the angle between initial (a) and the final (b) correct position we pretend, and it corresponds to the amount of varus needed, (c) planning varus osteotomy with a simple rotation of the X ray: we measure also the angle between initial and final correct position we pretend; it corresponds to

Hip Surgery in Cerebral Palsy

145

http://dx.doi.org/10.5772/intechopen.80442

A third strategy can be employed to insert the guide wire and is based on the following technical assumptions: when we use the equipment pointer to insert the guide wire in cases of significant valgus (>140–150°), the off-set provoked by the thickness of the metal piece gives us a worse result than planned (the osteotomy will be too far, causing the plate to be too lateral); another assumption is that soft tissues can interfere in the guide wire and pointer orientation when we have very "vertical" orientation of the guide wire and a huge femoral internal rotation; even so, if we know that we want a final head-cervico-diaphyseal of about 120°, and that there is an angle of about 10° between cervical line and real head cervical line (when femoral head is deformed in valgus). So, if we insert the guide wire by free hand, parallel with the femoral neck, and if we use a 110° plate, we will reach a final angle of 120°. Given the real constraints of soft tissue during surgery, it is easier to find the correct point of

Before we make the osteotomy, we have to make marks on the femur, so that we can have clear indicators when we achieve the final state after osteosynthesis. To avoid rotation errors, we can mark the bone, inserting Kirschner wires, above and distally to the osteotomy line (sufficiently far to not interfere with the plate placement), or marking the bone, for example, with a vertical superficial line made with a chisel or a saw, long enough, because of eventual femur shortening at the osteotomy site, once more, far from the location where the plate is

in femur which should be adjusted using the equipment pointer.

the amount of varus.

entrance for the wire using this technique (**Figure 7a** and **b**).

There are several materials we use to achieve a correct osteosynthesis—straight plates, angled blade plates, angled screw plates, Altdorf plates, and others. Age, weight, bone density, need of external contention (splints, casts), and surgeon experience, impact the implant choice for each patient.

For the correction of femoral proximal valgus, we have to calculate the amount of varization we need. There are several ways to calculate it: we can use a goniometer and an X-ray transfer (always remembering that correction is made more accurately after evaluation under anesthesia), measuring the angle between initial valgus and final planned position, just rotating image/transfer (**Figure 6a**–**c**). We can also plan the angle of entrance of our guide wire (when we use an angled plate for our osteosynthesis), using the following formula:

```
 initial angle − final (pretended) angle + plate angle = guide wire angle
```
In this formula, the *initial angle* is the real valgus of the patient, the *final angle* is about 120°, the *plate angle* depends on angle availability in the equipment we are using (and decision

**Figure 6.** Planning varus osteotomy with a transfer; we measure the angle between initial (a) and the final (b) correct position we pretend, and it corresponds to the amount of varus needed, (c) planning varus osteotomy with a simple rotation of the X ray: we measure also the angle between initial and final correct position we pretend; it corresponds to the amount of varus.

**4.2. Calculation of the final angle in femoral osteotomy and bone osteosynthesis**

diaphyseal angle than classically measured and an "infra leveling" of the Shenton's line.

144 Cerebral Palsy - Clinical and Therapeutic Aspects

component at the osteotomy site (with a corresponding slope for the plate positioning).

we use an angled plate for our osteosynthesis), using the following formula:

initial angle − final (pretended) angle + plate angle = guide wire angle

each patient.

As mentioned previously, when we plan a proximal femoral osteotomy, the goal for the inner question of the femur is to reach a final state where the cervico-diaphyseal angle (or head-cervicodiaphyseal angle, when femoral head is in valgus) is about 120° in the AP view, with a correction of rotation that permits about 30° of hip internal rotation and an eventual femur shortening, depending on the previous amount of femur uprising associated with dislocation. Sometimes, when we have a flexed hip, we can incorporate a "deflexion" procedure, adding an extension

**Figure 5.** (a) dislocated hip; the cervical line doesn't correspond to "head cervical" line, because of head progressive deformity in valgus, (b) post-operative reduced hip; when head is correctly reduced, we observe a bigger "true" cervico-

There are several materials we use to achieve a correct osteosynthesis—straight plates, angled blade plates, angled screw plates, Altdorf plates, and others. Age, weight, bone density, need of external contention (splints, casts), and surgeon experience, impact the implant choice for

For the correction of femoral proximal valgus, we have to calculate the amount of varization we need. There are several ways to calculate it: we can use a goniometer and an X-ray transfer (always remembering that correction is made more accurately after evaluation under anesthesia), measuring the angle between initial valgus and final planned position, just rotating image/transfer (**Figure 6a**–**c**). We can also plan the angle of entrance of our guide wire (when

In this formula, the *initial angle* is the real valgus of the patient, the *final angle* is about 120°, the *plate angle* depends on angle availability in the equipment we are using (and decision regarding the point of entrance on lateral femur), and *guide wire angle* is the angle of entrance in femur which should be adjusted using the equipment pointer.

A third strategy can be employed to insert the guide wire and is based on the following technical assumptions: when we use the equipment pointer to insert the guide wire in cases of significant valgus (>140–150°), the off-set provoked by the thickness of the metal piece gives us a worse result than planned (the osteotomy will be too far, causing the plate to be too lateral); another assumption is that soft tissues can interfere in the guide wire and pointer orientation when we have very "vertical" orientation of the guide wire and a huge femoral internal rotation; even so, if we know that we want a final head-cervico-diaphyseal of about 120°, and that there is an angle of about 10° between cervical line and real head cervical line (when femoral head is deformed in valgus). So, if we insert the guide wire by free hand, parallel with the femoral neck, and if we use a 110° plate, we will reach a final angle of 120°. Given the real constraints of soft tissue during surgery, it is easier to find the correct point of entrance for the wire using this technique (**Figure 7a** and **b**).

Before we make the osteotomy, we have to make marks on the femur, so that we can have clear indicators when we achieve the final state after osteosynthesis. To avoid rotation errors, we can mark the bone, inserting Kirschner wires, above and distally to the osteotomy line (sufficiently far to not interfere with the plate placement), or marking the bone, for example, with a vertical superficial line made with a chisel or a saw, long enough, because of eventual femur shortening at the osteotomy site, once more, far from the location where the plate is

a 100° [upper line (**Figure 8a**–**c, f**)] or a 110° [lower line (**Figure 8a, d**–**f**]) angled plate. The difference of the two entry points implies an osteotomy almost 1 cm more distal in the second option. If we are not aware of this issue, it is easy to make an osteotomy that is too distal which negatively affects hip's Pauwels' balance [15]. In the latter option, we risk obtaining a "Shepherd's crook"-like femur instead of a balanced hip (**Figure 8d**–**f**). It is not a rule to choose plates with lower angles because, sometimes, they reach the lower femoral neck or even the calcar and that is not our goal. In consequence, it is important to choose the best

Hip Surgery in Cerebral Palsy

147

http://dx.doi.org/10.5772/intechopen.80442

When we face a situation where the proximal femur lateral cortex prevents correct apposition of the plate, provoking the cited femur "shepherd's crook"-like deformity, we can cut the lateral beak, so that the plate can join completely its osseous "bed". The problem is that we will have less bone apposition between proximal and distal fragments (**Figure 9a**–**c**). It is advisable

**Figure 9.** The proximal femur lateral cortex prevents correct apposition of the plate (a), provoking a femur "shepherd's crook"- like deformity; we can cut the lateral beak (b), so that the plate can join completely its osseous "bed" (c).

**Figure 10.** Femur shortening is regulated until null tension on the femoral head remains on abduction, after hip

reduction; if some tension remains, the proximal femur should be more shortened.

angulation on a case by case basis.

to preserve periosteum and to join autograft, if possible.

**Figure 7.** (a, b) Profile and front guide wire correct insertion.

going to be screwed. At the site of the osteotomy, we try to detach and preserve periosteum so that it contributes to a good consolidation, in particular when the initial valgus is very important and respective correction implies some significant off-set of the proximal femur and risk of late pseudarthrosis. In these cases, the adjunction of an auto graft seems wise.

Now we discuss the entry point for the plate; as we can see in **Figure 8a**, we can imagine different entry points in function of the angulation of the plate. For example, we have planned

**Figure 8.** (a) Different entry points of the guide–wire, in function of the angulation of the plate, (b, c) more proximal femur osteotomy and achievement of a final balanced result, (d, e) slight more distal osteotomy and achievement of a final "Shepherd's crook", (f) comparison of anatomic final results in function of the osteotomy level.

a 100° [upper line (**Figure 8a**–**c, f**)] or a 110° [lower line (**Figure 8a, d**–**f**]) angled plate. The difference of the two entry points implies an osteotomy almost 1 cm more distal in the second option. If we are not aware of this issue, it is easy to make an osteotomy that is too distal which negatively affects hip's Pauwels' balance [15]. In the latter option, we risk obtaining a "Shepherd's crook"-like femur instead of a balanced hip (**Figure 8d**–**f**). It is not a rule to choose plates with lower angles because, sometimes, they reach the lower femoral neck or even the calcar and that is not our goal. In consequence, it is important to choose the best angulation on a case by case basis.

When we face a situation where the proximal femur lateral cortex prevents correct apposition of the plate, provoking the cited femur "shepherd's crook"-like deformity, we can cut the lateral beak, so that the plate can join completely its osseous "bed". The problem is that we will have less bone apposition between proximal and distal fragments (**Figure 9a**–**c**). It is advisable to preserve periosteum and to join autograft, if possible.

going to be screwed. At the site of the osteotomy, we try to detach and preserve periosteum so that it contributes to a good consolidation, in particular when the initial valgus is very important and respective correction implies some significant off-set of the proximal femur and risk

Now we discuss the entry point for the plate; as we can see in **Figure 8a**, we can imagine different entry points in function of the angulation of the plate. For example, we have planned

**Figure 8.** (a) Different entry points of the guide–wire, in function of the angulation of the plate, (b, c) more proximal femur osteotomy and achievement of a final balanced result, (d, e) slight more distal osteotomy and achievement of a

final "Shepherd's crook", (f) comparison of anatomic final results in function of the osteotomy level.

of late pseudarthrosis. In these cases, the adjunction of an auto graft seems wise.

**Figure 7.** (a, b) Profile and front guide wire correct insertion.

146 Cerebral Palsy - Clinical and Therapeutic Aspects

**Figure 9.** The proximal femur lateral cortex prevents correct apposition of the plate (a), provoking a femur "shepherd's crook"- like deformity; we can cut the lateral beak (b), so that the plate can join completely its osseous "bed" (c).

**Figure 10.** Femur shortening is regulated until null tension on the femoral head remains on abduction, after hip reduction; if some tension remains, the proximal femur should be more shortened.

**Figure 11.** Femur shortening, resection pieces–They can be used as auto-graft in supra-acetabular pelvic osteotomy or in the femur osteotomy site.

Before final screwing of the plate to distal fragment, we must be sure that hip muscle tension does not stress hip reduction during abduction. The shortening is made until null tension on the femoral head remains on abduction (**Figure 10**). If some stress remains, the femoral head has a tendency to upraise in acetabulum, or even dislocate. Sometimes we can make an important shortening, as much as 2–3 cm (**Figure 11**).

#### **4.3. Correct concentric hip reduction**

The steps used for hip reduction depend on reducibility of the femoral head; in classic spastic cases, we make an adductor (and ischio tibialis) tenotomy, and after iliopsoas tenotomy we can test if the femoral head is reducible or not; if it is difficult, we begin the proximal femoral osteotomy step of the procedure, so that we can free at maximum all of the proximal femur. With forceps, we can handle the femoral neck and trochanter and we can test reducibility and eventual stability of the head in acetabulum. We must be aware of a false sensations of reducibility, for example, in cases of head valgus deformity, which can fool us (we tend to follow the neck alignment visually and not the head location in R-ray). In these situations, to be sure that reduction is achieved, we have an "infra-leveling" of Shenton's line (**Figure 12**) because the femoral head is well centered and the neck is below the head level, and not in continuity with it. If the head is not totally reduced, we have to open the joint capsule and cleanse all the obstacles to total reduction with special care to transverse ligament excision and freeing of lower joint capsule, permitting complete lowering of the femoral head.

**5. Conclusions**

important and challenging concern.

with a wide array of comprehensive solutions.

discontinuity of the Shenton's line(that is "infra-leveled").

Cerebral palsy is a particularly complex field of medical knowledge, where clinical experience is probably more important than in other fields. For that reason, trying to teach and to share some "shortcuts" about CP "thinking" with new generations of health professionals is an

**Figure 13.** In this case, a bilateral Dega pelvic osteotomy was added to a femoral VDR osteotomy.

**Figure 12.** When femoral head is deformed in valgus, and when hip reduction is correctly achieved, we verify a

Hip Surgery in Cerebral Palsy

149

http://dx.doi.org/10.5772/intechopen.80442

The goal of this chapter is to facilitate hip evaluation and decision making for surgeons who deal with these complex problems in CP patients. Practical clinical situations were presented

As mentioned before, and depending on age (presence or not of the triradiate cartilage), depending also on anatomic profile of acetabulum, and of femoral head reducibility, we decide if we add an acetabular step [acetabuloplasty, (**Figure 13**)], reorientation procedure, shelf) or a capsulorrhaphy (cf. Section 3). A dynamic X-ray can be made at the end of the procedure to test the stability of hip reduction. The external immobilization depends on age and osteosynthesis stability. In younger children, we use pelvi-podalic casts and in older children and adolescents we use bi-cruro-podalic casts, for a period of 4–6 weeks.

**Figure 12.** When femoral head is deformed in valgus, and when hip reduction is correctly achieved, we verify a discontinuity of the Shenton's line(that is "infra-leveled").

**Figure 13.** In this case, a bilateral Dega pelvic osteotomy was added to a femoral VDR osteotomy.

#### **5. Conclusions**

Before final screwing of the plate to distal fragment, we must be sure that hip muscle tension does not stress hip reduction during abduction. The shortening is made until null tension on the femoral head remains on abduction (**Figure 10**). If some stress remains, the femoral head has a tendency to upraise in acetabulum, or even dislocate. Sometimes we can make an

**Figure 11.** Femur shortening, resection pieces–They can be used as auto-graft in supra-acetabular pelvic osteotomy or

The steps used for hip reduction depend on reducibility of the femoral head; in classic spastic cases, we make an adductor (and ischio tibialis) tenotomy, and after iliopsoas tenotomy we can test if the femoral head is reducible or not; if it is difficult, we begin the proximal femoral osteotomy step of the procedure, so that we can free at maximum all of the proximal femur. With forceps, we can handle the femoral neck and trochanter and we can test reducibility and eventual stability of the head in acetabulum. We must be aware of a false sensations of reducibility, for example, in cases of head valgus deformity, which can fool us (we tend to follow the neck alignment visually and not the head location in R-ray). In these situations, to be sure that reduction is achieved, we have an "infra-leveling" of Shenton's line (**Figure 12**) because the femoral head is well centered and the neck is below the head level, and not in continuity with it. If the head is not totally reduced, we have to open the joint capsule and cleanse all the obstacles to total reduction with special care to transverse ligament excision and freeing of lower joint capsule, permitting complete lowering of the

As mentioned before, and depending on age (presence or not of the triradiate cartilage), depending also on anatomic profile of acetabulum, and of femoral head reducibility, we decide if we add an acetabular step [acetabuloplasty, (**Figure 13**)], reorientation procedure, shelf) or a capsulorrhaphy (cf. Section 3). A dynamic X-ray can be made at the end of the procedure to test the stability of hip reduction. The external immobilization depends on age and osteosynthesis stability. In younger children, we use pelvi-podalic casts and in older children

and adolescents we use bi-cruro-podalic casts, for a period of 4–6 weeks.

important shortening, as much as 2–3 cm (**Figure 11**).

**4.3. Correct concentric hip reduction**

in the femur osteotomy site.

148 Cerebral Palsy - Clinical and Therapeutic Aspects

femoral head.

Cerebral palsy is a particularly complex field of medical knowledge, where clinical experience is probably more important than in other fields. For that reason, trying to teach and to share some "shortcuts" about CP "thinking" with new generations of health professionals is an important and challenging concern.

The goal of this chapter is to facilitate hip evaluation and decision making for surgeons who deal with these complex problems in CP patients. Practical clinical situations were presented with a wide array of comprehensive solutions.

I hope this summary of experience and reflections about the subject will be useful for interested readers.

[6] Hägglund SA, Pedersen H, Rodby-Bousquet E, Wagner P, Westbom L. Prevention of dislocation of the hip in children with cerebral palsy–20-year results of a population-based

Hip Surgery in Cerebral Palsy

151

http://dx.doi.org/10.5772/intechopen.80442

[7] Canavese F, Deslandes J, editors. Orthopedic Management of Children with Cerebral Palsy: A Comprehensive Approach. New York: Nova Science Publishers Inc; 2015. ISBN

[8] Valencia F. Management of hip deformities in cerebral palsy. Orthopedic Clinics of

[9] Melhem E, Assi A, El Rachkidi A, Ghanem I. EOS® biplanar X-ray imaging: Concept, developments, benefits, and limitations. The Journal of Children's Orthopaedics.

[10] Patel J, Shapiro F. Simultaneous progression patterns of scoliosis, pelvic obliquity, and hip subluxation/dislocation in non-ambulatory neuromuscular patients: An approach to deformity documentation. The Journal of Children's Orthopaedics. 2015;**9**:345-356. DOI:

[11] Senaran H, Shah SA, Glutting JJ, Dabney KW, Miller F. The associated effects of untreated unilateral hip dislocation in cerebral palsy scoliosis. Journal of Pediatric Orthopaedics.

[12] Campagnolo J, Ovídio J, Carvalho M, Jerónimo R, Guarda E. Faut-il opérer la hanche controlatérale réduite chez l'enfant paralysé cérébral à hanche luxée? Révision d'une série de 9 cas. Conférence abstract, Revue de Chirurgie Orthopédique et Traumatologique

[13] Robb J, Brunner R. A Dega-type osteotomy after closure of the triradiate cartilage in non-walking patients with severe cerebral palsy. The Journal of Bone and Joint Surgery

[14] Al-Ghadir M, Masquijo JJ, Guerra LA, Willis B. Combined femoral and pelvic osteotomies versus femoral osteotomy alone in the treatment of hip dysplasia in children with cerebral palsy. Journal of Pediatric Orthopaedics. 2009;**29**(7):779-783. DOI: 10.1097/

[15] Erceg M. The influence of femoral head shift on hip biomechanics: Additional parameters accounted. International Orthopaedics (SICOT). 2009;**33**:95-100. DOI: 10.1007/

prevention program. The Bone and Joint Journal. 2014;**96-B**:1546-1552

North America. 2010;**41**:549-559. DOI: 10.1016/j.ocl.2010.07.002

2016;**10**:1-14. DOI: 10.1007/s11832-016-0713-0

978-1-63483-349-3 (E-book)

10.1007/s11832-015-0683-7

2013;**99**(7):320. SOFCOT 2013, Paris

[British]. 2006;**88-B**:933-937

BPO.0b013e3181b76968

s00264-008-0544-9

2006;**26**(6):769-772

### **Acknowledgements**

This chapter has benefited from the experience and teaching of Professor José Salis Amaral. I would also like to express my gratitude to my surgical team at Hospital Dona Estefânia. A special thanks to my wife, Camille, who shares this commitment to science and to patients, and finally, my thoughts to all my family.
