PEDIATRIC SURGERY FOR THE PRIMARY CARE PEDIATRICIAN, PART I
OUTPATIENT PEDIATRIC ORTHOPEDICS Common and Important Conditions Richard M. Schwend, MD, and Jennifer Geiger, MD
You can observe a lot by watching. YOGI BERRA
Approximately one third of medical problems in children are related to the musculoskeletal system. Most of these problems are common and can be precisely diagnosed. For these problems, nonoperative treatment or reassurance can be given by the pediatrician. Occasionally, a problem needs surgical treatment, but a precise diagnosis must be made. There is little agreement about what types of orthopedic problems a primary care pediatrician should understand in order to effectively care for children. Many pediatric residencies lack an organized teaching curriculum that effectively covers these topics or that includes a required pediatric orthopedic rotation. In this article the authors delineate pediatric orthopedic problems that: 1. Require recognition and urgent surgical treatment. 2. Are relatively common, but have different treatment options (observation, conservative treatment, and surgery) depending on their natural history. 3. Are common, need no surgical treatment, and are well managed by the pediatrician without referral. Whenever possible, the diagnosis should be made before a decision to refer is made. There are several reasons for this: 1. An accurate diagnosis allows the pediatrician to discuss the natural history of the condition properly. The parents can be reassured rather From State University of New York at Buffalo (RMS, JG); Department of Orthopedic Surgery, The Children’s Hospital of Buffalo (RMS), Buffalo, New York PEDIATRIC CLINICS OF NORTH AMERICA VOLUME 45 * NUMBER 4 AUGUST 1998
SCHWEND & GEIGER
than waiting to hear the same information from another physician. In particular, orthopedic problems are known to generate pressure from the parents to seek specialty consultation for reass~rance.~ Although some of these referrals are unavoidable, it is important to communicate to the specialist that the reason for the referral is for parental reassurance rather than for further work-up or treatment. 2. Referral to the wrong specialty can needlessly generate expensive tests and further delay in treatment or generate inappropriate For instance, an adolescent with back pain may have underlying school and home stress with depression and require further evaluation of these aspects of his or her life. A 5-year-old may have leukemic involvement of the spine. Referral of either of these children could delay diagnosis and treatment. A spinal cord tumor should be referred to the neurosurgeon, while vertebral osteomyelitis and discitis should be referred to the orthopedic surgeon. In each case, focused and timely referral depends on an accurate and specific diagnosis. 3. After a proper diagnosis, communication directly between the pediatrician and the appropriate specialist can often avoid an unnecessary referral, and avoid unnecessary tests. The authors reviewed their experience of referrals from primary care physicians over 1 year. The authors found that patients were usually referred to them for a symptom (i.e., hip pain) rather than for a specific diagnosis. Ninety-five percent of patients were eventually able to receive a diagnosis. Approximately 40% of these referrals involved problems that could have been managed by the referring physician since they did not require any further work-up or treatment, usually just reassurance or education. The authors found that it is useful to classify conditions as common or uncommon, and whether they require surgical or nonsurgical treatment (Table 1);many conditions fall in between. The following is a discussion of some of these more important or common conditions.
Table 1. COMMON ORTHOPEDIC CONDITIONS AND NEED FOR REFERRAL
Surgical Refer Immediately Common
SCFE Septic hip arthritis Displaced fractures
Open fractures Malignancies Compartment syndrome
Surgical or Nonsurgical Refer or Consult with Orthopedic Surgeon Scoliosis DDH Cerebral palsy Perthes disease Osteomyelitis Septic knee arthritis Spondylolisthesis Trigger thumb Benign Bone lesions Back pain in a young child
Nonsurgical Need Not Refer Unless Unusual Presentation or Functional Problem Develops Medial tibial torsion Idiopathic toe walking Medial femoral torsion Osgood-Schlatter disease Bow legs in a toddler Transient hip synovitis Anterior knee pain
A diagnosis is categorized by whether it is common or uncommon and requires surgical treatment or not. Some conditions such as slipped capital femoral epiphysis (SCFE), septic hip arthritis, or orthopedic malignancies require surgery and should be referred immediately to the orthopedic specialist. Nonsurgical conditions such as in-toeing from medial femoral or medial tibial torsion do not require surgery or even bracing and need no referral unless another diagnosis is suspected or serious functional problems have developed. Many conditions fall in between and require a thorough understanding of the natural history to determine appropriate evaluation, specific diagnosis, and timing of specialty referral.
OUTPATIENT PEDIATRIC ORTHOPEDICS
CONDITIONS REQUIRING URGENT ORTHOPEDIC REFERRAL Urgent referral is usually necessary for these conditions since there can be significant consequences if the diagnosis is missed. Among these, slipped capital femoral epiphysis and septic hip arthritis are the most common. Other problems to refer immediately include suspected malignancies, deep infections, and displaced intra-articular or physeal fractures. Slipped Capital Femoral Epiphysis Slipped capital femoral epiphysis (SCFE) is a relatively common hip problem, usually seen in the adolescent near the time of peak growth velocity. Displacement of the femoral epiphysis, either gradually or suddenly, from the femoral neck through the growth plate, occurs. A slipped epiphysis can later develop in the opposite hip in 36% of cases%and it does so more frequently in children presenting at a younger age.5*,80 Since obesity is present in at least one half of children with SCFE, mechanical stress during the growth spurt, when the physis is active, is an important cause of this condition. There is typically a narrow window of bone age during which the slip develops.50Boys are affected more than girls. There is a n increased frequency with endocrine disorders, especially hypothyroidism, panhypopituitarism, and renal disease. If the child is young or is unusually short, an endocrine disorder may be Because of its frequency, risk for degenerative arthritis, and the possibility of sudden instability and vascular damage, it is important to detect this condition. Clinical findings can vary, whether the slip is stable or unstable. In a stable slip, an obese adolescent boy may complain of activity related hip pain that is relieved by rest. As the pain worsens, there may be a mild limp after strenuous activity. These symptoms may last weeks or months, depending on the severity. Often the child has been seen by a physician and received a diagnosis of a pulled groin muscle. On physical examination there may be some pain with hip motion, restriction of flexion and, particularly, internal rotation while the hip is flexed to 90 degrees (Fig. 1).Limited internal rotation may be a subtle difference from the opposite side, or it may be significant with the examiner unable to internally rotate the hip even to a neutral rotation position. As the hip loses internal rotation, it develops obligatory external rotation, with an external foot progression angle while walking (Fig. 2). The diagnosis can be elusive since referral of the pain may be to the distal thigh, presenting as knee pain. All children who complain of knee pain should have the hip examined. Some children with an extremely externally rotated gait have foot pain as their presenting complaint. Occasionally, the patient with SCFE has minimal symptoms and no pain, making the diagnosis extremely By definition, the unstable slip does not allow the child to bear weight, even with crutches. This should be considered the equivalent of an acute, displaced fracture of the hip and should be treated urgently. The child is in extreme pain, and usually gives a history of a fall from a bicycle, or out of a tree; however, the fall may also be trivial if the physis had pre-existing weakness. Radiographs confirm the diagnosis and the patient must be urgently admitted to the hospital for treatment. The risk for avascular necrosis, even with immediate treatment, is as high as 47%.'' Radiographs are obtained with the patient in the supine position. Both hips need to be included on the anteroposterior (AP) and frog lateral pelvis films.
SCHWEND & GEIGER
Figure 1. Thirteen-year-old girl with stable left slipped capital femoral epiphysis. When the hip is flexed to 90 degrees, internal hip rotation is limited to neutral.
The opposite hip serves as a comparison; however, because of the chance of having the condition bilaterally, both hips may show changes. Unstable slips may be severely displaced, making the diagnosis straightforward (Fig. 3A). The more stable slips often show only very subtle changes on the frog lateral view, and the AP pelvic radiograph may appear normal. The natural history of untreated stable slipped epiphysis is variable. It can heal undisplaced, can slowly progress, or can acutely progress to become an unstable slip. A stable slipped epiphysis has little risk for avascular necrosis. If it is minimally displaced, the risk for arthritis later in life is negligible, whereas if the slipped epiphysis is allowed to progressively slip further, osteoarthritis is possible.lXThe biggest risk is for the stable epiphysis to suddenly displace and become unstable. The femoral head may then develop avascular necrosis in 33% to 50% of cases. This is most likely from the acute nature and severity of the trauma, not from the treatment.34,h8 Once avascular necrosis occurs, gradual degenerative changes in the hip result. Often hip replacement is necessary in the affected adult.43 Treatment for the stable slip includes ensuring that the child stop walking as soon as the diagnosis is made. Children have been known to slip and fall just before surgery, thus converting a nondisplaced stable slip to an unstable slip. Because of the variable nature of untreated SCFE and the effectiveness and simplicity of early surgery, treatment is surgical. Once the diagnosis of a stable slipped epiphysis is made in a walking child, plans for surgery should be made without delay. The most commonly performed procedure is percutaneous pin
OUTPATIENT PEDIATRIC ORTHOPEDICS
Figure 2. Fourteen-year-old girl with left hip pain who has been having difficulty walking for the past 4 weeks. When she stands there is limited weight-bearing ability on the left lower extremity. The left knee and foot are turned outward and the left hip and knee are flexed. Radiograph showed a stable left slipped capital femoral epiphysis.
or screw fixation between the femoral neck and the center of the epiphysis. Often, one well-placed screw is sufficient to provide stability.39This procedure, performed with fluoroscopic guidance, usually involves a very small 0.5 inch incision. Treatment priorities for the unstable slip are to avoid avascular necrosis, avoid iatrogenic complications, prevent further slip, and correct the deformity.6 Juvenile patients are more likely to have bilateral involvement and, because physeal growth disturbances can result from screw fixation, smooth pins may be ; ~ e d .Very ~ ~ unstable slips may undergo a gentle reduction and screw fixatioh (Fig. 3B). ~" After the slip is pinned, one should be vigilant about the opposite hip, taking any symptoms seriously.49Children with endocrine disorders have an extremely high prevalence of bilateral involvement. For these children, the asymptomatic hip is often prophylactically pinned.R8Often the child can tell if similar symptoms are developing with the other hip. Much of the restricted motion that was present before surgery is regained within 6 months, more from relief of pain, spasm, and synovitis than from osseous r e m ~ d e l i n gAt . ~ ~longterm follow-up there is still some loss of internal rotation that minimally affects the foot progression angle. If the residual deformity is severe, osteotomy is a consideration. Slipped capital femoral epiphysis is a frequent and serious enough condition that all pediatricians need to be aware of this diagnostic possibility in the older child with pain in the lower extremity, limp, or inability to walk. The most
SCHWEND & GEIGER
Figure 3. A, Thirteen-year-old boy with a displaced right slipped capital femoral epiphysis. Note the complete displacement of the femoral metaphysis from the epiphysis and the proximal migration of the rnetaphysis. 6,Postoperative AP pelvic radiograph of the unstable right slipped capital femoral epiphysis that is noted in Figure 4. Reduction consisted of very gentle positioning of the femur onto the epiphysis and fixation was with a single percutaneous cannulated screw fixation. Radiographs should always be done on the opposite hip, even if the patient is asymptomatic.
common problem in dealing with this condition is failure to consider the diagnosis, particularly for the mild stable slipped epiphysis or for atypical presentations. Surgery at earlier stages is simple and effective. All children with a SCFE should be referred. CONDITIONS THAT DO NOT REQUIRE IMMEDIATE SURGERY
These conditions (developmental dysplasia of the hip, cerebral palsy, adolescent idiopathic scoliosis, and back pain) may require more involved diagnostic studies, therapy, bracing, or surgery depending on the severity and natural history. Before a treatment decision can be made, an understanding of the natural history of the untreated condition, of the condition treated conservatively, and the natural history of surgical intervention must be u n d e r s t o ~ dReferral .~~ is not usually urgent and depends on the severity of the condition and the pediatrician’s level of experience. Developmental Dysplasia of the Hip in the Infant
Developmental or congenital dysplasia of the hip (DDH) is a spectrum of conditions ranging from laxity to dislocation in the newborn, to end stage arthritis in the elderly. Most infants at risk for developmental dysplasia do not require treatment. However, for those infants who need to be treated, diagnosis should be accurate. The complications of avascular necrosis and inadequate
OUTPATIENT PEDIATRIC ORTHOPEDICS
reduction must be avoided, and the patient must be followed to maturity. Developmental hip dysplasia is the most frequent single cause of end stage hip disease in the aduk7The primary problem in developmental dysplasia is capsular and ligamentous laxity early after birth.” Breech presentation can also cause instability and dislocation. The incidence of this disorder in breech infants is 10fold higher than in babies delivered in vertex. Frank breech, with the knees extended, puts even more stress on the hip by tightening the hamstring muscles. The acetabulum has a normal shape at birth and becomes shallow and maldirected if the femoral head is not perfectly reduced?* In the first few weeks of life, the capsular laxity resolves and the hip spontaneously becomes more stable. Barlow found the prevalence of instability to decrease from 1/60 infants on the first day of life to 1.5/1000 by the end of the second month.12 This spontaneous improvement is assisted by the infant’s muscle activity and is prevented by the hip being held in an abnormal position. However, a dislocated hip does not necessarily improve as the laxity resolves. Without treatment it can become more stable but remain in a dislocated position. With time, the dislocated hip develops secondary deformities of muscle contraction, capsular elongation and narrowing, obstruction by the iliopsoas tendon, transverse acetabular ligament and ligamentum teres, acetabular mal-direction and dysplasia, femoral antetorsion, and femoral head dysplasia. Studies performed on rabbits have shown that the secondary deformity of acetabular dysplasia is a result of dislocation or instability and does not cause the dislocati0n.45 Hip laxity, acetabular dysplasia, subluxation, and dislocation each have their own natural history. Hip laxity is often present at birth. Although the femoral head is well positioned in the socket, with a stress maneuver (Barlow test), the examiner perceives a sensation of movement as the femoral head comes out of the socket. The Barlow test is performed by gently holding the hip with the opposite hand, thumb on the inner thigh and fingers on the greater trochanter. The hip is gently adducted and the examiner perceives a motion of the hip subluxating or dislocating. The Barlow maneuver should never be done forcibly or excessively since instability may be induced. These hips should be observed for persistent instability but they usually improve spontaneously during the first few weeks of life. Acetabular dysplasia is often associated with hip instability. As the laxity improves, the acetabular dysplasia generally improves, although more slowly over several months or years. By the time the child is walking, acetabular dysplasia improves much more slowly, even in a completely reduced hip. Because the bones do not remodel reliably after 4 years of age, dysplasia does not spontaneously improve in the school age child and corrective osteotomies are often needed. If the hip is subluxated, the acetabular dysplasia has insufficient corrective forces to remodel fully and requires treatment. Hip dislocation has a unique and varied natural history. In the newborn, a dislocated hip can improve spontaneously to become a normal hip, can become a subluxated hip or can develop acetabular dysplasia. However, this improvement is not at all predictable and the untreated dislocated hip can become fixed in a dislocated position. In the newborn, the hip is noted to be sitting in a dislocated position but can be gently pushed back into the acetabulum (positive Ortolani test). The Ortolani test gives more useful information than the Barlow test and is a safer maneuver to perform. By several months of age, the dislocation becomes more fixed and cannot reliably be re-positioned into the acetabulum. As the hip remains dislocated for several months, the iliopsoas, adductors, and hamstring muscles develop contractures and the hip develops limited abduction. Later in infancy, the Ortolani sign can disappear and the hip is then considered
SCHWEND & GEIGER
irreducible. Whereas the young infant can often be treated with simple nonoperative means, the walking age child, especially if the hip is irreducible, often requires closed reduction or operative treatment. Any child presenting with a delay in walking, especially beyond 18 months, should be investigated for a hip dislocation. The natural history of treated dislocations does not always lead to a successful outcome. Treatment begun late is much less likely to succeed.60 About half of infants and children needing closed reduction have acetabular dysplasia as However, very early diagnosis and treatment early in infancy will make reduction more [email protected]
' Universal ultrasound screening of infants has not become an accepted standard in the United States. The test is extremely sensitive to minor degrees of instability and acetabular morphology abnormalities. Not all facilities caring for newborns have the technical experience to perform ultrasound examinations or to provide appropriate follow-up and treatment when indicated. Furthermore, because of the excessive number of children treated unnecessarily, universal screening has not been proven to be cost effective in the United States. Selective screening of at-risk infants is becoming accepted in the United States. This includes those with a positive family history of hip dysplasia, history of breech presentation, congenital deformities, instability on examination (Ortolani positive), or other evidence of dysplasia, such as limited abduction or a positive Galeazzi sign. The Galeazzi sign is noted by placing the child's knees together and flexing the hips to 90 degrees and fully flexing the knees. A dislocated hip will appear to be the shorter limb with.the knee closer to the hip on the involved side. Ultrasound imaging is most useful in the first 5 months of life before the ossific nucleus becomes too large and obscures the view into the acetabulum. The ultrasound, when used for at-risk screening, is performed at 4 to 6 weeks of age. By this time, physiologic laxity has resolved and findings on the ultrasound are more meaningful. An experienced ultrasonographer should perform the examination to decrease the chance of both false-positive and falsenegative examinations. The ultrasound examination can determine whether the hip is: 1. Reduced and stable 2. Reduced and loose 3. Subluxated and unstable 4. Dislocated and partially reducible but unstable, or 5. Dislocated and irreducible.60 If ultrasound imaging is not available to screen the at-risk infant, an AP pelvic radiograph at about 4 to 5 months should be done. The absence of the ossific nucleus in the neonate makes radiographs of the hip difficult to interpret. For this reason, radiographs are not useful to diagnose hip dysplasia in the newborn and need not be ordered. If there is a question of limb deficiency such as seen in proximal femoral focal deficiency or a teratologic hip dislocation as seen in arthrogryposis or sacral agenesis, the radiograph can show important bony anatomy that the ultrasound may not demonstrate. By about 5 months of age, the ossific nucleus should have appeared and a radiograph can be much more useful. Asymmetric appearance of the ossific nucleus can be a subtle indicator of dysplasia. The essence of successful outcome is timely, accurate diagnosis and safe, conservative treatment. The basic principle is to achieve a stable, concentric reduction, while avoiding the complication of osteonecrosis and subluxation. The hip with some laxity (Barlow positive) can be observed for resolution of the laxity. An ultrasound examination can be done in several weeks to confirm that
OUTPATIENT PEDIATRIC ORTHOPEDICS
it has normalized. Hips with a positive Ortolani sign at birth, indicating that the hip is resting out of the socket, should be treated early. These hips are less likely to spontaneously improve and need treatment to help contain the hip in the socket. If there are other deformities, a positive family history, or if the condition is bilateral, it is even more important to start treatment early. Many neonates, up to 10% in some studies, are noted to have high pitched soft tissue clicks about the hip and knee. This is the most common reason an infant is referred to a hip screening clinic. Most of these clicks spontaneously resolve by the end of the first year. Bond recently prospectively examined and performed ultrasound imaging for 50 neonates whose hip click persisted beyond 3 months of age.I6 None of these infants were from a high-risk group and all had otherwise normal clinical examinations. All ultrasound examinations were normal. A soft tissue click is common, slowly spontaneously resolves, and is not in itself an indicator of hip dysplasia; therefore, ultrasound imaging and treatment are not necessary. It is important not to misdirect valuable resources by treating this normal condition. The Pavlik harness is usually used to treat a dislocated or subluxated hip (Ortolani positive) in the infant less than 6 months of age. This device encourages hip flexion above 100 degrees and allows passive abduction. The posterior straps are to prevent excessive adduction; they should not be tightened to force abduction. Each hip should not be abducted beyond 50 degrees due to the risk of osteonecrosis. The full amount of flexion should be gradually reached by tightening the anterior straps over the first week. Pavlik, the designer of the device, wrote: ”The harness does not shorten the time of treatment but it carries it out with greater safety. The prirjnary aim of the treatment is to achieve concentric reduction and to prevent avascular necrosis, which cripples the child for the whole of his life.”65 The timing of initiating the Pavlik harness and the severity of the dislocation have recently been shown to be extremely important. Traditionally, success rates for all types of hip dysplasia treated in the Pavlik harness range from 85% to 95%.38Mubarak reported a group of infants with frankly dislocated hips, and found that if treatment was started before 3 weeks of age there was a 63% success rate, whereas those hips treated after 3 weeks had only a 20% success.6o The Pavlik harness is much less successful if started after 7 weeks of age and when used for bilateral dislocations. If the infant failed in the Pavlik harness, there was a 50% chance of needing an open reduction. A multicenter European study found better success and fewer complications in the less severely dislocated hips.= Vascular necrosis of the femoral head was twice as likely if treatment began after 3 months of age. Conservative treatment is best begun as soon as possible after birth, before secondary deformities develop and while there is still enough biologic plasticity in the hip. Osteonecrosis of the hip occurs in up to 22% of patients as a result of the Pavlik harness. Therefore, it should be used only when the benefits clearly outweigh its risk^.^^,^ It is important to ensure that the harness is not causing excessive pressure on the femoral head or on the posterior acetabulum. If the infant is uncomfortable in the harness, it should be adjusted or discontinued. One must be very careful to not overdiagnose and treat infants with a Pavlik harness for a condition that may spontaneously improve, because of the small but real risk of osteonecrosis with use of the harness. Contraindications to the Pavlik harness include infants over 6 months of age if reduction has not been achieved after 3 weeks of use, muscle imbalance (cerebral palsy, spina bifida), joint stiffness (arthrogryposis),or unusual ligamentous laxity as seen in EhlersDanlos syndrome.61
SCHWEND & GEIGER
In summary, there are several basic principles that should be followed in caring for children with developmental dysplasia of the hip. All newborns, the infant at each well-child visit, and all children who present with a significant delay in walking should be evaluated for DDH. Early diagnosis and treatment lead to better outcomes. Prevention is preferable to elaborate surgical reconstructive procedures. Overtreatment leads to complications and wasted resources for the family and the medical system. Treatment should avoid the complications of osteonecrosis and inadequate reduction. A child known to have DDH needs to be followed at least until skeletal maturity. For these reasons, the child with developmental dysplasia of the hip should be referred for orthopedic care.
Cerebral Palsy Cerebral palsy (CP) is a static disease of the central nervous system that involves the upper motor neurons and has its clinical manifestations in the periphery. Primary, secondary, and tertiary manifestations of cerebral palsy are e~ident.2~ The primary problems are balance, loss of selective motor control, strength, persistence of primitive reflexes, and spasticity. The secondary manifestations are the coping maneuvers the child uses to compensate for primary problems. Tertiary problems are the end result of primary and secondary abnormalities and include fixed bony deformities, scoliosis, and dislocated hips. Prevention of cerebral palsy is preferable. However, there will continue to be children with neuromuscular disease, and prevention of the secondary and tertiary manifestations is necessary for the involved child. Whenever a child with cerebral palsy is seen for €hefirst time, it is important to consider if correctable conditions may be causing the problem. Treatable disorders such as spinal cord tethering, syringomyelia, spinal cord tumors, and other lesions of the central nervous system should be excluded. Degenerative conditions, although not always treatable, are important to consider so that proper counseling and treatment plans can be offered to the parents. There are many simple classifications of cerebral palsy that help in the decisionmaking process: 1. Type of involvement-diplegia, hemiplegia, triplegia, or total involvement. 2. Type of tone and movement disorder-spastic or extrapyamidal. 3. Classification by walking ability-a convenient way to describe the child. It communicates various treatment priorities. Walking abilities include independent community walker, household walker, exercise or demonstration walker, and nonwalker. 4. The nonwalker-also called nonambulator, can be further divided into: A. able to sit independently B. uses one hand to sit C. needs to be propped to sit pr0perly.3~ Each type of cerebral palsy has its own natural history and treatment priorities. Treatment priorities must be established that are based on the child’s and family’s needs. These priorities include communication, family, friends, school, independence, general health, nutrition, sitting, and walking. Tremendous effort to get the child to walk do not make sense if some of the higher needs are not addressed. General principles of treatment to improve the child’s function should always start with the primary problems. Of the primary problems, treatment of spasticity has evolved rapidly over the past several years.
OUTPATIENT PEDIATRIC ORTHOPEDICS
Rhizotomy, baclofen, and botulinium toxin injections are now available as treatment options. Multidisciplinary clinics that include neurosurgery, orthopedics, physical medicine, and therapy are available to address problems of spasticity. This approach should complement other medical treatment that includes developmental and general pediatrics, neurology, and nutrition. Spastic diplegia can be seen in the premature infant with periventricular leukomalacia. The lower extremities are more involved than the upper. There is usually a delay in reaching all the motor milestones although most of these children walk by age 7. Not all are independent walkers and many rely on crutches or walkers for distance. Some children have sufficient involvement that a wheelchair is necessary. Primary problems of balance and poor selective control contribute to the need to use walking aids. Early in life, physical therapy is the standard treatment for spasticity, contractures, and general development. Ankle foot orthoses may be needed for an equinous gait. As the child reaches about 3 or 4 years of age attention should be given to evaluation of the spasticity. At this age, if spasticity is seriously affecting the gait, a multidisciplinary spasticity evaluation is quite helpful. Occasionally, only a few muscles are causing the majority of the problem and botox injections can be given to temporize for up to 6 months and thus delay more invasive treatmentjl More generalized spasticity can be treated by baclofen, either orally or by intrathecal pump inf~sion.~ A child with diplegia who was premature, has fair balance, selective motor control, strength, and no overriding primitive reflexes may be an ideal candidate for selective dorsal rhizotomy if spasticity is the major problem. Even with rhizotomy, secondary and tertiary problems of bony deformity may require orthopedic surgery?,58 If orthopedic surgery is required, all needed procedures are performed in 1 day, with postoperative physical therapyw Gait analysis is more frequently used in preoperative planning in the walking child, and can give very specific technical information of which muscles to lengthen or transfer and about bony realignment procedures.Z6, Spastic hemiplegia is more often seen in full-term infants. It can be very subtle and difficult to detect. It may be first noticed by an early hand preference, a delay in walking of several months, unilateral toe walking or posturing of the upper extremity, which is sometimes more obvious with running. Involved children generally walk by 2 years of age. Early treatment consists of physical and occupational therapy. Short bracing of the lower extremity may be needed. Gait analysis has demonstrated four different types of hemiplegia based on how proximal is the involvement in the lower extremityg1If surgery is recommended, it is usually performed by age 7. Lower extremity surgery generally involves correcting the secondary manifestations by muscle lengthenings and occasionally split transfer of the tibialis posterior muscle or the tibialis anterior. Upper extremity surgery can be very useful to the child who has an active interest in using the hand more but is impeded by poor position and c~ntractures.'~ We have occasionally performed upper extremity transfers to allow a child with triplegia better use of the hand, so that a walker can be grasped better. Patients with total involvement have markedly different treatment priorities. Although many can walk with assistive devices, their primary means of ambulation will be with a wheelchair. Spasticity may be an overriding factor impeding function and the child may be a candidate for treatment for spasticity treatment, although the functional indications are different than for the diplegic child. Early treatment is therapy, short leg bracing, and ensuring adequate seating. These children are at risk to have hip subluxation and dislocation, particularly if there
SCHWEND 81 GEIGER
is marked spasticity and contractures of the hips. Yearly or more frequent screening radiographs in children with hip contractures or spasticity is necessary. When hip spasticity is becoming a functional or seating problem or when there is radiographic evidence that the hip is at risk for dislocation, surgery is indicated. Earlier hip surgery allows for a relatively small procedure, whereas later surgery when the hip has dislocated is more involved!, 62 Scoliosis usually appears at a later age and often requires surgery by puberty.20Adequate preoperative nutrition is essential before undertaking any of these major operative procedures.10 Children with cerebral palsy have many priorities. Communication, family, friends, school, independence, general health, and nutrition often take priority over orthopedic concerns. Although pediatric orthopedic surgeons continue to manage the extremity and spine manifestations of cerebral palsy, the multidisciplinary spasticity team has been developed to evaluate and treat spasticity and its functional consequences. Adolescent idiopathic Scoliosis
Adolescent idiopathic scoliosis (AIS) is defined as a lateral curvature of the spine with rotation, occurring in the child over 11 years of age, for which no obvious cause is found. If a child is noted to have an underlying neurologic condition such as syringomyelia, then the scoliosis is a result of that condition and is not idiopathic. The frequency of scoliosis varies from 1.9% to 3%0,although for curves larger than 30 degrees the prevalence is only 0.3%.Girls are affected more often than boys. Although the etiology has ndt been determined, studies suggest a central nervous system cause.%,92 There is a family history of the condition about 30% of the time, with mothers of girls often having scoliosis. The importance of AIS relates to its frequency and the understandable desire to prevent progression to a severity that is cosmetically or psychologically unappealing or that could ultimately affect pulmonary function.29 Typically, an asymptomatic teenager is told after a school screening that she must see a physician because of possible scoliosis.If the family noted a curvature before she was 11 years old, this is juvenile scoliosis, not adolescent scoliosis. Occasionally back pain is present but typically not. It is necessary to inquire about maturity, Tanner stage, and menarche, since it is essential information in determining the natural history and the need for treatment. For the typical right thoracic curve, the right scapula is more prominent and there may be some asymmetry in the height of the shoulder. Thoracic curves with the apex or prominence to the left are unusual and may be a sign of an underlying 75 On the Adam's forward bend test there is elevation of one side abn~rmality.~~, of the chest, usually right, owing to the rotation. The scoliometer can be used to measure the degree of curvature; a curve of 7 degrees on the scoliometer correlates to a 20-degree radiographic curve by the Cobb method. This has been found to be the most effective cut-off to refer a patient in a screening program although larger curves can still be missed!* After determining clinically that a child has scoliosis, the next question must always be "what is the underlying etiology?" The pediatrician should then inquire about back pain, head or neck aches, bowel or bladder dysfunction, or weakness. Inquire about family history, which could give a clue to Friedreichs ataxia or neurofibromatosis. Since the extremities give an indication of integrity of the spinal-neuro axis, they should be examined in all cases of proven scoliosis. Remove the shoes and examine for calf atrophy, cavus foot, or asymmetry.
OUTPATIENT PEDIATRIC OIUHOPEDICS
If after screening there is a possibility that the curve is of significant magnitude (20 degrees) or if there is concern about an underlying spinal column abnormality, radiographs are obtained. The first radiograph done for scoliosis should always include a standing lateral film. The PA film will give an indication of maturity. Films taken in the morning can have 5 degrees less curvature than films taken in the e ~ e n i n g . ' Examine ~ , ~ ~ if the triradiate cartilage is still open and the degree of the Risser sign on the iliac crest apophysis. The role of magnetic resonance imaging in AIS is becoming better defined. In a prospective double-blinded study of 140 asymptomatic patients undergoing spinal fusion, a neurologic abnormality was detected in only four patients, one small syrinx and three Chiari I malformations. None of these patients needed neurosurgery. In a study of 98 patients with AIS undergoing magnetic resonance imaging (MRI) evaluation, all patients who needed neurosurgical treatment were discovered by physical examination, not the MRI.75,88 When conditions such as left thoracic scoliosis are studied prospectively in asymptomatic individuals, the prevalence of pathology is found to be much lower.59The MRI is best used to confirm a clinical diagnosis and not as a screening tool (Fig. 4). Since there are various degrees of severity in AIS and the condition varies with the physical maturity of the patient, it is important to understand the natural history when determining treatment. Natural history can be divided into untreated, conservatively treated, and surgically treated.31It can be further subdivided into the natural history of the child who is still growing and in the adult. In the growing child, the natural history is best determined by the physiological maturity and the degree of curvature. The Risser sign is a common method that measures the ileac crest apophysis on the standing anterior-posterior spine radiograph. Risser 0 indicates no ossification, and Risser 5 is fully mature and fused. Lonstein found that children with Risser sign 0 to 1 (immature) had a 68% risk of progressing 5 or more degrees if their curve was 20 to 29 degrees, whereas if they had Risser sign 2 to 4 the risk was only 23%. Lesser degrees of curvature (<20 degrees) were much less likely to progress, 22% for Curves detected after the immature child and 1.6%for the more mature the onset of maturity or after menarche are less likely to progress. Thoracic curves are often seen at a younger age and are more likely to progress. The child's growth spurt can cause curve progression, even after a successful posterior spinal fusion." Once the child reaches skeletal maturity, curve progression is much less likely. Skeletal maturity can be determined by no further increase in height on serial measurements, by Tanner staging, or radiographically by a Risser sign of 5, or by closed physes on a hand film. As a general guide, girls have their peak growth velocity at age 12, boys at age 14, and growth slows significantly 2 to 3 years after these peaks. Girls reach menarche at about age 13 and growth slows significantly 2 years later. Weinstein has shown that at maturity, thoracic curves over 50 degrees are likely to continue progression later in life and curves less than 30 degrees rarely progress. Despite worries about the adult ending with a life-threatening condition, mortality is not increased at 40 to 50 years of age for idiopathic curves, but is for juvenile and infantile curves.6'j The goals of conservative treatment are: 1. Identify any underlying diagnosis such as syringomyelia that may need treatment. 2. Identify which children are at risk for significant progression during their growing years.
SCHWEND & GEIGER
Figure 4. A, Seven-year-old boy presented with painful right thoracic scoliosis of 30 degrees. The neurologic examination was normal. The young age associated with back pain suggested an underlying abnormality. 6,MRI revealed a Chiari I malformation and a cervical thoracic syringomyelia, which was treated with decompression of the Chiari I malformation. The syringomyelia was not drained. C, One year after the neurosurgical procedure and after 20 hours per day of spine bracing the back pain has resolved and the curve has improved to 23 degrees.
OUTPATIENT PEDIATRIC ORTHOPEDICS
3. Prevent the curve from worsening to the degree that continued progression as an adult is likely. Only brace or cast treatment has been shown to improve the natural history of scoliosis. Electrical stimulation, manipulation, and therapy do not. Recent studies show conclusively that brace treatment works in the growing child. The Scoliosis Research Society (Chicago, IL) demonstrated that bracing of curves between 25 and 35 degrees has a 74% success rate compared to observation (34%) and electrical stimulation (33y0).~Lonstein has shown than any skeletally immature child presenting with a curve of 25 degrees should be immediately considered for brace treatment rather than waiting to show curve progression.54The brace has not been shown to work for curves less than 25 degrees or for children who refuse to wear it. It is contraindicated if there is thoracic lordosis. Bracing does not ultimately get rid of the curve; for the majority of children it maintains the curve at the same degrees as when it was started. If a premenarchal girl can have her 26-degree curve maintained at that degree until she is skeletally mature, she is very unlikely to progress after maturity. The more hours that the child wears the brace, the more effective are the results. A minimum of 20 hours/day is currently considered ideal. In a meta-analysis, the most effective conservative treatment (99% success) was the Milwaukee brace used 23 hours a days1 Surgical treatment is considered if the brace is no longer able to affect the natural history, and the curve is severe or progressive. Although surgery is used to prevent continued progression and ultimately pulmonary compromise, other indications including appearance, sRinal balance, and pain need to be considered, given some of the newer and safer surgical techniques. For any child who presents with scoliosis, the physician should always ask the question "Is this really adolescent idiopathic scoliosis?" If the child is juvenile, has an unusual curve, or has an underlying neurologic abnormality it may not be idiopathic and an underlying diagnosis should be investigated. Treatment decisions are always based on a thorough understanding of the natural history and a determination whether the treatment can improve on the expected outcome if left untreated. Children who have an underlying cause of the scoliosis or who are likely to need treatment should be referred.
Back Pain Back pain was the second most common spine complaint in this series. Like scoliosis, back pain is not a diagnosis, but a symptom. In a referral population, a specific diagnosis is more likely to be found than in the primary care setting. Hensinger reported his experience in a university orthopedic referral clinic of 100 children with back pain lasting for more than 2 years.33Eighty-four percent of the children had an underlying pathology. These conditions included spondylolysis, spondylolisthesis, kyphosis, scoliosis, infection, and tumor. The diagnosis is very dependent on the child's age. Whereas young children more often have an underlying diagnosis, older teenagers are more likely to have nonspecific back pain that is similar to the adult population. Several observations in the evaluation of children with back pain warrant extra concern.
1. The younger the child, the more likely there is an underlying pathology, particularly if the child is less than 5 years of age. Hensinger found that
SCHWEND &I GEIGER
the two most common causes of back pain in the young child were tumors and disc space infection.% 2. Any neurologic symptoms or findings, including bowel or bladder, are important clues of an underlying neurosurgical condition such as a spinal cord tumor, syringomyelia, ruptured disc or end-plate, tethered cord, or unrecognized spinal dysraphism. 3. Any stiffness to the spine suggests true pathology such as infection or tumor. 4. Associated deformity such as scoliosis or kyphosis, particularly if the curve is atypical or if there are neurologic findings, may indicate a hidden spinal cord or vertebral column pr~blern.~’ 5. Night pain is of concern since most mechanical strains and stresses are better at night, whereas tumors and infections tend to feel worse at night. Classically, lesions such as osteoid osteoma of the vertebral body can keep the child awake at night. Most mechanical pain resolves in several weeks, so pain lasting for more than 1 or 2 months is worrisome. 6. Any systemic symptoms such as fever, malaise, and weight loss suggest an underlying condition such as infection or leukemia. 7. Skin lesions such as caf6-au-hit spots of neurofibromatosis or hairy patch of spinal dysraphism. The history and clinical examination should be thorough and focused. Inquire about the onset of pain. Was it sudden and related to a stress or an injury suggesting spondylolysis or end plate avulsion, or was it more gradual. Location of pain can suggest Scheuermann kyphosis in the thoracic spine or spondylolysis if it is in the low lumbar spine. Radiatidn of pain down the limb can indicate a ruptured disc or an end plate avulsion. When examining the child, always have the gown open in the back and have the patient remove the pants and shoes so that the entire lower extremities can be properly examined. Examine for spinal balance, scoliosis, and excessive or unusual kyphosis. Ask the child to bend forward, backward, and side to side, looking for stiffness or spasm. Other diagnoses including pyelonephritis, gynecological problems, or sacroileac disorders should be excluded. We often see children with inflammation of the ileac crest apophysis, which can be confused with low back pain if it is over the posterior crest.%With the shoes off, a focused neurologic examination is performed. It is particularly important to visually inspect the legs and feet for calf thinness, caws feet, or claw toes, which could be clues to an underlying neurologic problem. The child’s age and worrisome findings should determine which tests, if any, need to be performed. An 18-year-old with a 1-week history of low back pain, no trauma, normal examination, and other social stress occurring may be observed without further involved testing. A 13-year-old gymnast with acute onset of back pain may need radiographs or even a SPECT scan to diagnose her posterior element stress reaction.% The 4-year-old with a 2-month history of thoracic back pain that is keeping her awake at night, fevers, and stiffness on examination needs a different type of investigation. For this child, spine films need to be obtained and looked specifically for disc space narrowing to suggest an infection. A sedimentation rate is also important if an infection or inflammatory condition is considered. A radionuclide bone scan is extremely useful for diagnosing a disc space infection in a young child. An MRI can give the same information and may show a spinal cord tumor that could have been missed on the bone scan. Although protocols have been published, it is still critical to use the history
OUTPATIENT PEDIATRIC ORTHOPEDICS
and physical examination to direct the next level of testing. Most children need no further testing than plain radiographs. However, further evaluation is always indicated if the history and physical examination suggest a particular diagnosis. In this circumstance, the laboratory and imaging studies are used to confirm the clinical suspicion and to better define the anatomy and pathology. Radiographs, sedimentation rate, and white blood count (WBC) in a child with no clear evidence of a particular pathology have been found sufficient to safely exclude significant underlying pathology.86However, the physician should continue observation until the symptoms resolve or additional findings occur. The natural history of back pain is best understood for each particular diagnosis. For example, the natural history of childhood disc space infections has recently been studied into adulthood and long-term difficulties were demonstrated to persist. Jansen, in a report of 35 patients, found that 15 still complained about backache, 26 had radiographic evidence of a blocked vertebrae, which is a fusion of two adjacent vertebra, and 28 had narrowing of the vertebral canal with marked restriction of extension.37 Despite attempts to reach a specific diagnosis, many children, particularly older adolescents, have nonspecific back pain. Back pain is quite frequent in adolescents, with 74% of school children having reported low back pain, usually lumbar in location, at some times in their lives? By 16 years of age, low back pain can be as frequent as in an adult.17Turner found that most children with nonspecific back pain are normal within a year.%However, 6 of 29 patients with this diagnosis had persistent back pain for more than 6 years. Psychologic factors also play a role in children repdrting nonspecific low back pain. In summary, back pain is common in the older adolescent but quite uncommon in young children. A thorough history and physical examination are the basis for further laboratory testing and imaging. While the young child usually requires more testing, one must resist the urge to use excessive tests or to overtreat the adolescent with typical low back pain. Children should be referred after making a specific diagnosis or if unexplained but worrisome findings are present. CONDITIONS RARELY REQUIRING REFERRAL OR TREATMENT
These common conditions include: transient hip synovitis, idiopathic toe walking, rotational problems, and Osgood-Schlatter disease. Parents are often worried, prompting unnecessary orthopedic referral for these conditions. It is important to accurately diagnose these conditions in order to reassure the family and avoid overutilization of resources. Transient Hip Synovitis
As the name indicates, transient synovitis of the hip is a self-limiting inflammatory condition, usually lasting about 10 days to several weeks. The importance of this common condition is for the physician to ensure that a more serious problem such as septic arthritis of the hip, osteomyelitis of the femur or acetabulum, or deep soft tissue abscess is not present. The condition is most common in the 3- to 8-year-old age range, whereas septic hip arthritis can affect all ages. Usually only one hip is affected, although there is ultrasound evidence that the other hip can have an asymptomatic effusion in up to 25% of cases.21
SCHWEND & GEIGER
The child may complain of hip pain and develop a limp. If a fever is present, it is mild. Gradually the pain and limp continue to the extent that the child may stop walking. On physical examination, there is no fever present or only a mild fever, and the child does not look ill. There is no warmth or swelling about the hip. Because of the effusion, the hip may be resting in a flexed position (Thomas test), with some abduction and external rotation. Motion of the hip is painful; however, the hip can usually be moved through some degree of motion. Radiographs are usually normal, although there can be evidence of mild joint space widening. The radiograph should be examined closely for lytic lesions that suggest osteomyelitis and for epiphyseal changes of Perthes’ disease. Although we continue to obtain a radiograph in a child with transient synovitis, it has been suggested that they Are not necessary in the typical case.s5A sedimentation rate is an extremely useful laboratory test to help distinguish transient synovitis from deep infections, such as septic hip arthritis or osteomyelitis. Ultrasound imaging has become a simple, quick, and reliable method of detecting the effusion typically seen in transient synovitis. The fluid is clear and shows none of the echo-shadows that can be present in the purulence of septic arthritis. The ultrasound imaging of a hip with transient synovitis shows an average of 8 to 10 mm of fluid between the femoral head and the hip capsule. A normal hip typically shows less than 4 mm of fluid.Z1,85 Ultrasound imaging is also helpful to distinguish Perthes’ disease which shows thickened synovium and articular cartilage distending the capsule, whereas transient synovitis shows mostly homogenous synovial fluid distending the capsule.24If a child is found to have transient synovitis, it is very unusual to later develop Perthes’ disease.= Radionuclide bone scanning is not indicated unless there is a probability of osteomyelitis. In one study using radionuclide scanning in transient synovitis, avascular changes in the head were present in 8% of cases, although only a fourth of these avascular cases later showed radiographic evidence of Perthes’ disea~e.~’ The authors have found that MIU is very useful in the evaluation of the hip in which osteomyelitis is likely. It shows the extent of the bony infection and demonstrates if there is an unrecognized soft tissue abscess. Generally, with typical clinical findings, clear fluid noted on ultrasound imaging, and normal or only slightly elevated ESR, a correct diagnosis of transient synovitis can be made. Treatment is conservative, including bedrest or activity restriction, and nonsteroidal anti-inflammatory medication. Hip aspiration is generally not needed unless the clinical and ultrasound findings suggest septic hip arthritis. If the hip joint is aspirated, the effusion may resolve somewhat qui~ker.8~ However, hip aspiration is usually not necessary. There are no long-term problems associated with transient synovitis. If septic hip arthritis is diagnosed, treatment is immediate surgery and appropriate antibiotic therapy? For the classic case of synovitis, referral to the orthopedic surgeon is generally not necessary, although a child with a suspected septic hip arthritis requires immediate referral. Idiopathic Toe Walking
Idiopathic toe walking is seen in young children, usually less than 4 years of age, and is thought to be a benign condition (Fig. 5). There is no underlying diagnosis such as cerebral palsy. The condition may be dynamic, with good passive range of motion, or have more fixed contractures. The prevalence is not well-described since many of these children are not seen by the physician. One retrospective study asking parents to recall if their child had been a toe walker
OUTPATIENT PEDIATRIC ORTHOPEDICS
Figure 5. Idiopathic toe walking. This 2-year-old girl had no perinatal problems and began walking independently at 12 months of age. She was able to walk normally when encouraged, but typically walked on her toes. With time, toe walking improved.
and by direct observation found a 25% occurrence. The origin is unclear. Many children are able to walk normally (heel-to-toe) when being observed, yet persist in toe walking when they stop concentrating on walking "correctly." Subtle neurologic abnormalities, particularly speech and language delays, are often present.z,77 Gait analysis has been able to distinguish between diplegic gait and the gait seen in idiopathic toe-walking. If the toe-walking is unilateral, consider another diagnosis. Toe-walking is not typically seen in the older child. The clinical evaluation includes a review of perinatal factors including prematurity, intrauterine growth retardation, or neonatal problems that could indicate mild cerebral palsy. As with any condition involving the neuromuscular system, inquire about speech and language, social development, fine motor, and gross motor development. Any delay in walking is noted. The child with idiopathic toe walking usually started walking on time, whereas spastic diplegia involves a delay, often by months or years. Children with spastic hemiplegia will toe walk on one side. Children with a shortened limb often toe walk on that side as a means to balance the gait. This is seen in conditions such as dislocated hip, congenital short femur, or posterior-medial tibia bowing. Other conditions giving unilateral toe walking include hemangioma of the gastrocnemius4" and fibromatosis near the sciatic nerve. Examine the child walking, paying attention to both the stance phase and the swing phase for each limb. The spine should be examined for evidence of dysraphism, the hips for stability and restricted motion, and the limbs for length. Determine if the deformity is dynamic or fixed. Check for spasticity or other evidence of upper motor neuron disease. Finally, in a boy, signs of Duchenne muscle dystrophy such as a Gower sign, proximal muscle weakness, or calf hypertrophy should be noted. Since the natural history of idiopathic toe walking is benign, treatment is conservative at first. If other diagnoses have been excluded, initial observation is appropriate. If the problem persists or if the child is having functional difficulty with falling or tripping, casting can be attempted for the rigid deformity and bracing for the child with a flexible deformity. As the child nears school age
SCHWEND & GEIGER
and continues to toe walk with a fixed deformity, surgical lengthening has been shown to be curative. In summary, this self-limiting condition can usually be distinguished from those more serious conditions by a thorough history and physical examination. Treatment is not always necessary unless the problem is persistent or if walking becomes difficult. Referral to the orthopedic surgeon is appropriate for such cases.
Rotational problems in children are the single most common reason for referral to the pediatric orthopedic surgeon. Although rarely requiring any bracing or surgical procedures, parents are often very anxious about their child and the possibility of future health problems should the condition be serious or not resolve. Rotational problems involve the transverse plane and include medial femoral torsion, medial tibial torsion and metatarsus adductus.= Rotational problems are quite prevalent, but at different ages. In the infant, especially if there has been some intrauterine crowding, the midfoot can be turned inward causing apparent intoeing. This is often present with internal tibial torsion, both being associated with intra-uterine constraints. As the infant begins to walk and loses some of the normal external hip rotation of infancy the internal tibial torsion becomes more apparent. The preschooler may show signs of medial femoral torsion, which is often hereditary. When examining the child with a rotational problem, the rotational profile is a useful means of classifying the child's condition. First, examine the child walking to see whether the feet are turning in from the straight ahead neutral position (negative)or turning out from neutral (positive).Record the amount in degrees. Most children toe outward with a positive-foot progression angle at a few degrees. Next, lay the child prone, knees together and flexed 90 degrees, and allow the hips to rotate inward (Fig. 6A). This gives a measure of femoral rotation. External rotation should be about the same as internal (Fig. 6B). Rotation markedly in one direction and limited in the other is abnormal. Next, examine the thigh foot angle with the hindfoot in a neutral position (Fig. 7). This angle is normally several degrees external as the child becomes older, but is rotated internally in the child who has medial tibia torsion. The foot is then examined for evidence of forefoot (metatarsus) adduction (Fig. 8). Internal tibial torsion, also called medial tibial torsion, is extremely common. When examining the child, always ask about the parents' concerns. Do they simply need reassurance or are they looking for treatment? Often treatment that was given in the past to other family members can influence the type of treatment the family is currently expecting. Since the untreated natural history of internal tibial torsion is one of gradual improvement, these interventions may have appeared successful to the family in the past. Ask what the child means by "toeing in or out," since they may be talking about a varus or valgus foot or metatarsus adductus. Occasionally a child with unilateral intoeing, particularly if there has been some delay in walking, will have mild hemiplegia as the cause. The intoeing in this case does not improve and the parents need information instead of reassurance. Many toddlers with intoeing also have physiologic tibia vara, since varus bowing is normally present until 18 months of age (Fig. 9A). Often the bowing is more apparent than real and is not present when examining the child lying with the knees facing forward (Figs. 9B and C). Rarely, the
OUTPATIENT PEDIATRIC ORTHOPEDICS
Figure 6. A, Three-year-old boy with excessive internal femoral rotation. The child is prone and both hips are allowed to rotate inward (legs outward) with gravity. Internal rotation is generally no more than 70 degrees. In this case the rotation was about 80 degrees for each hip. 6, Limited external rotation. Each hip is tested individually for external rotation. In this case external rotation was about 30 degrees. Normally internal and external rotation are about equal.
rotational problem is a sign of a dislocated hip, so the hips must always be examined. The natural history of medial tibia1 torsion is accelerated improvement in infancy followed by gradual, although not always complete, improvement during early childhood. Parents often complain that the toddler is tripping because
SCHWEND & GEIGER
Figure 7. Fourteen-month-old boy with internal rotated thigh foot angle, typical of internal tibia1 torsion. The hindfoot is held in a neutral position and the thigh axis is compared to the axis of the foot, in this case about 30 degrees.
Figure 8. Newborn with left foot calcaneovalgus and right foot metatarsus adductus. Both deformities in this case were flexible. The hip ultrasound was normal.
OUTPATIENT PEDIATRIC ORTHOPEDICS
Figure 9. A, Fourteen-month-old boy had mild bowing of his lower extremities and mild internal tibia torsion. He was referred for evaluation of bowlegs. His development was normal and walking had begun at 13 months. B,When supine, the hips had mild external rotation. Note the external position of the patella. There was apparent bowing of the lower extremities. C, With the knees facing anterior, the apparent bowing is not present, but the feet are noted to face inward. The diagnosis is internal tibia1 torsion. No treatment was given and the rotation had markedly improved by 2 years of age.
SCHWEND & GEIGER
of the deformity and are worried that a more severe injury may result. The child is developing walking ability at this age and it is difficult to separate what is caused from the torsion from that caused by normal development. There is no evidence that internal tibial torsion causes late arthritis. Rarely, a normal child continues to have tibial torsion of such severity that it is affecting function. Osteotomies may be indicated; however, the authors recommend that sufficient time pass until the child is 8- to 10-years-old to allow for maximum spontaneous correction. In a prospective study there was no evidence that Denis-Browne bracing changes the natural history compared to There is evidence from animal studies that the brace may be h a d by affecting the joint relationships rather than the bony torsion.” Therefore, it is best to avoid treating this benign condition but reassure the family about the expected natural history. Medial femoral torsion, also called internal femoral torsion, is an excessive amount of medial version of the relationship of the femoral neck to the distal femur. It tends to run in families and becomes most apparent during the preschool years. Normally children have a similar amount of internal and external hip rotation. If most of the hip rotation is medial and very little is lateral, this can cause some functional difficulty with walking. Children often “ W sit with knees together and hips internally rotated. This can be a comfortable position for the child with excessive medial torsion and is easy to balance in this position. It is uncertain if the W position worsens the torsion or is a result of the torsion. Medial femoral torsion gradually improves over time in at least 80% of children, but can still be seen in an adult who stands slightly pigeon t0ed.8~ Bracing, shoe inserts, and therapy have no proven benefit. There is no evidence that medial femoral torsion causes hip or knee arthritis. There is an association of external femoral torsion with later hip arthritis in the adult and slipped capital femoral epiphysis in the teenager. Some children with medial femoral torsion associated with external tibial torsion, termed miserable malalignmennt, have patella femoral pain, subluxation, or even dislocation. For these extremely rare circumstances of disability, femoral and tibia osteotomies have been occasionally used after 10 years of age. Rotational problems are important because they are common, are usually part of the normal developmental process, and can best be handled by taking the parents’ concerns seriously. Management includes accurate diagnosis to exclude other conditions, and reassurance and parent education rather than corrective devices. Referral or consultatjon to orthopedic surgery is appropriate for significant functional problems or if another diagnosis has been noted.
Osgood-SchlatterDisease Osgood-Schlatter disease is a common cause of anterior knee pain in the adolescent. It is one diagnosis in a spectrum of conditions that include SindingLarsen-Johansson disease of the inferior pole of the patella, plica syndrome, Hoffa’s disease of the anterior fat pad, and idiopathic anterior knee pain. It is a stress reaction of the insertion of the patella tendon into the tibia tubercle and not an underlying abnormality of 0ssification.9~The condition previously was more frequent in adolescent boys near the time of their growth spurt; however, girls are being seen more frequently with this diagnosis due to more active sports participation. Pathologically, there can be some fragmentation of the ossicles of the developing tibia tubercle or soft tissue involvement of the patella tendon insertion without obvious bony changes.* Often there is only a tendonitis of the patella tendon insertion and inflammation of the deep infrapatella [email protected]
OUTPATIENT PEDIATRIC ORTHOPEDICS
Clinically, a young teenager who is quite active in jumping sports such as basketball, complains of pain and swelling over the tibia1 tubercle. The condition may be unilateral or bilateral. The parents are often worried that a more serious problem, such as a tumor, is causing the firm and extremely tender bump. On examination there is an obviously swollen, tender, and occasionally warm area over the tibia tubercle. The hamstrings may be somewhat tight and there may be patella alta, both indicating high stresses and loads to the patella tendon. If the condition is unilateral or if the swelling is not directly over the tibia tubercle, other conditions such as osteofibrous dysplasia could be present. In these atypical cases, a radiograph helps to reassure the family and the physician. The radiograph can be normal or show fragmentation of the area of patella tendon insertion. However, ossicles superior and anterior to the tibia tubercle may be normal variants in an asymptomatic knee.15 The diagnosis is, therefore, clinical, with radiographs used only if another diagnosis needs to be excluded. The natural history of untreated Osgood-Schlatter disease is relatively benign, if allowed to heal and complications are avoided. Since unusual stress at a time of rapid growth is occurring, limiting the offending sports activity is usually sufficient to relieve the symptoms. Occasionally, the teenager may continue to be active despite the symptoms, and with a forceful jump a displaced tibia tubercle fracture may occur. Because of this possibility, the authors recommend that patients allow the lesion to heal. In a long-term natural history study of 50 patients, 76% had no residual limitation of activity, although 60% still had difficulty kneeling owing to the discomfort on the residual bump. Two groups of patients were noted, those with radiographic evidence of separated ossicles at the time of diagnosis who developed abnormal ossification at the time of follow-up, and those with no radiographic changes initially, who at follow-up tended to have no residual symptoms. Rarely, the involvement of the tibia tubercle is extensive enough to cause an anterior growth arrest, resulting in proximal tibia r e ~ u r v a t u m . ~ ~ As with any other stress-related injuries, recognition, reassurance, rest, and protection while spontaneous healing occurs is the most important aspect of treatment. Kneeling and squatting should be limited. Although sports do not need to be eliminated, they should be modified so that the symptoms are improved. Physical therapy is occasionally used if there is unusual hamstring tightness or patella alta. Because the natural history of all types of anterior knee pain tends to improve with time, nonoperative treatment is always recommended as initial treatment.74Since this is a stress injury, anti-inflammatory medications do not help to heal the lesion. If severe symptoms continue despite activity modification, immobilization can be useful. The radiographic abnormality, if present, does not need to heal in order for the symptoms to Surgical excision can be performed if the bump continues to cause discomfort with kneeling. Osgood-Schlatter disease is important to recognize as one of the relatively benign stress conditions causing anterior knee pain. Since fracture and later discomfort from the prominence can occur, conservative treatment of rest and protection are indicated. Typically, referral to the orthopedic surgeon is not necessary unless symptoms persist after treatment.
CONCLUSION It is important to make a precise diagnosis and to understand the natural history of pediatric orthopedic conditions. This allows the pediatrician to provide reassurance, treatment, or appropriate and timely referral. The most important
SCHWEND & GEIGER
requirement in making the correct diagnosis is to have a clear understanding of the diagnostic possibilities. Some conditions such as SCFE, septic hip arthritis, or orthopedic malignancies require surgery and should be referred immediately. Nonsurgical conditions, such as intoeing, do not require surgery or even bracing, and usually need no referral. Many conditions fall in between these extremes and require a thorough understanding of the natural history to determine appropriate treatment. Communicationbetween the pediatrician and the pediatric orthopedic surgeon can prevent unnecessary referrals, ensure referrals are made in a timely manner, and improve the care provided by both physicians. References 1. Abbott R, Johann-Murphy M, Shiminsk-Maher T, et a1 Selective dorsal rhizotomy: outcome and complications in treating spastic cerebral palsy. Neurosurgery 33:851857, 1993 2. Accardo P, Morrow J, Heaney MS, et al: Toe walking and language development. Clin Pediatr (Phila) 31:158-160, 1992 3. Albright AL, Barrow WB, Fasick MP, et al: Continuous intra-thecal baclofen infusion for spasticity of cerebral origin. JAMA 27:2475-2477, 1993 4. Aparicio G, Abril JC, Calvo E, et al: Radiologic study of patella height in Osgood!+&latter disease. 1 Pediatr Orthop 1763-66, 1997 5 Armstrong D D&tors’ perceptions of pressure from patients for referral. BMJ 302:11861188, 1991 6 Aronson DD, Loder RT: Treatment of the unstable (acute) slipped capital femoral epiphysis. Clin Orthop 32299-110, 1996 7. Aronson J: Osteoarthritis of the young adult hip: Etiology and treatments. In Lewis D, Anderson MD: Instructional Course Lectures, The American Academy of Orthopedic Surgeons, vol35. St. Louis, CV Mosby, 1986, pp 119-128 8. Bagg MR, Farber J, Miller F Long term follow-up of hip subluxation in cerebral palsy patients. J Pediatr Orthop 1332-36, 1993 9. Balague F, Skovron ML, Nordin M: Low back pain in schoolchildren: A study of familial and psychological factors. Spine 1(20):1265-1270, 1995 10. Banta JVSpinal deformities in cerebral palsy-surgical procedure. Orthopade 21:309315, 1992 11. Barlow DW, Staheli LT Effects of lateral rotation splinting on lower extremity bone growth An in vivo study in rabbits. J Pediatric Orthopedics 11:583-587, 1991 12. Barlow TG: Early diagnosis and treatment of congenital dislocation of the hip. J Bone Joint Surg Br 44B:292, 1962 13. Bearch WR, Strecker WB, Coe J, et a1 Use of the Green transfer in the treatment of patients with spastic cerebral palsy: 17 year experience. J Pediatr Orthop 11(6):731736, 1991 14. Beauchamp M, Labelle H, Grimard G, et a1 Diurnal variation of Cobb angle measurement in adolescent idiopathic scoliosis. Spine 18:1581-1583, 1993 15. Bloom RA, Gomori J: Ossicles anterior to the proximal tibia. Clinical Imaging 17137141, 1993 16. Bond CD, Hennrikus WL, DellaMaggiore E D Prospective evaluation of newborn softtissue hip “clicks” with ultrasound. J Pediatr Orthop 17:199-201, 1997 17. Burton AK Low back pain in children and adolescents: to treat or not. Bull Hosp Jt Dis 55:127, 1996 18. Carney BT, Weinstein SL Natural history of untreated chronic slipped capital femoral epiphysis. Clin Orthop 322:43-47, 1996 19. Coleman SS: Congenital dysplasia of the hip in the Navajo infant. Clin Orthop 56:179, 1968 20. Dias RC, Miller F, Dabney K, et a 1 Surgical correction of spinal deformity using a unit rod in children with cerebral palsy. J Pediatr Orthop 16734, 1996 21. Ehredorfer S, LeQuesne G, Penta M, et a1 Bilateral synovitis in symptomatic unilateral ’
OUTPATIENT PEDIATRIC ORTHOPEDICS
synovitis of the hip: An ultrasound study in 56 children. Acta Orthop Scand 67149152, 1996 22. Franks P, Clancy CM, Nutting PA: Sounding board: Gatekeeping revisited Protecting patients from over treatment. N Engl J Med 327424-429, 1992 23. Fujiok F, Terayama K, Sugimoto N, et al: Long-term results of congenital dislocation of the hip treated with the Pavlik harness. J Pediatr Orthop 15:747-752, 1995 24. Futami T, Kasahara Y, Suzuki S, et a1 Ultrasonography in transient synovitis and early Perthes disease. J Bone Joint Surg 73B635-639, 1991 25. Gage JR: Gait analysis in cerebral palsy. Oxford, Blackwell Scientific Publications Ltd. New York, Cambridge University Press, Mac Keith Press, 1991, pp 118-131 26. Gage J R Gait analysis: An essential tool in the treatment of cerebral palsy. Clin Orthop 288:126-134, 1993 27. Thompson G H Back Pain in children. J Bone Joint Surg Am 75A:928-938,1993 28. Gerbino PG, Micheli LJ: Back injuries in the young athlete. Clin Sports Med 14:571590, 1995 29. Goldberg MS, Mayo NE, Poitras B: The Ste-Justine adolescent idiopathic scoliosis cohort study, Part I1 Perception of health, self image and body image, and participation in physical activities. Spine 19:1562-1572, 1994 30. Grill F, Bensahel H, Canadell J, et al: The Pavlik harness in the treatment of congenital dislocation hip: Report on a multicenter study of the European Paediatric Orthopedic Society. J Pediatr Orthop 8:1-8, 1988 31. Hall-Personal Communication, Children’s Hospital, Boston, MA, 1992 32. Heinrich SD, Sharps C H Lower extremity torsional deformities in children: a prospective comparison of two treatment modalities. Orthopedics 14655-659,1991 33. Hensinger RN: Back pain in children. In Bradford DS, Hensinger RN (eds): The Pediatric Spine. New York, Thieme and Stratton, 1985, pp 41-62 34. Herman MJ, Dormans JP, Davidson RS, et al: Screw fixation of grade 111 slipped capital femoral epiphysis. Clin Orthop 3277-85, 1996 35. Hoffer MM: Basic considerations and classifications of cerebral palsy. AAOS Instructional Course Lectures 25:164-168,1976 36. Hurley JM, Betz RR, Loder RT, et al: Slipped capital femoral epiphysis: The prevalence of late contralateral slip. J Bone Joint Surg Am 78226230, 1996 37. Jansen BR, Hart W, Schreuder 0: Discitis in childhood: 12-35 year follow-up of 35 patients. Acta Orthop Scand 61:33-36, 1993 38. Kalamchi A, MacFarlane R The Pavlik Harness: Results in patients over three months of age. J Pediatr Orthop 2:3-8, 1982 39. Kibiloski LJ, Doane RM, Karol LA, et a1 Biomechanical analysis of single versus double screw fixation in slipped capital femoral epiphysis at physiological load levels. J Pediatr Orthop 14:627-630, 1994 40. Klemme WR, James P, Skinner SR: Latent onset unilateral toe-walking secondary to hemangioma of the gastrocnemius. J Pediatr Orthop 14:77?-775, 1994 41. Koman LA, Mooney JF, Smith B, et al: Management of cerebral palsy with botulium A toxin: Preliminary investigation. J Pediatr Orthop 13:489495, 1993 42. Korovessis PG, Stamatakis MV Prediction of scoliotic Cobb angle with the use of the scoliometer. Spine 21:1661-1666, 1996 43. Krahn TH, Canale ST, Bety JH, et al: Long term follow-up of patients with avascular necrosis after treatment of slipped capital femoral epiphysis. J Pediatr Orthop 13:1541.58, 1993 44. Krause BL, Williams JP, Catterall A: Natural history of Osgood-Schlatter disease. J Pediatr Orthop 106548,1990 45. Langenskiold A, Sarpio 0, Michelsson JE: Experimental dislocation of the hip in the rabbit. J Bone Joint Surg Br 44B:209-215, 1962 46. Ledwith CA, Fleisher GR Slipped capital femoral epiphysis without hip pain leads to missed diagnosis. Pediatrics 89660-662, 1992 47. Levy AR, Goldberg MS, Mayo NE, et a1 Reducing the lifetime risk of cancer from spinal radiographs among people with adolescent idiopathic scoliosis. Spine 21:15401547, 1996 48. Lindstrom JR, Ponsetti IV, Wenger DR Acetabular development after reduction in congenital dislocation of the hip. J Bone Joint Surg 61A:112-118, 1979
SCHWEND & GEIGER
49. Loder RT, Aronson DD, Greenfield M L The epidemiology of bilateral slipped capital femoral epiphysis: A study of children in Michigan. J Bone Joint Surg Am 75A311411147,1993
50. Loder RT, Farley FA, Herzenberg JE, et al: Narrow window of bone age in children with slipped capital femoral epiphyses. J Pediatr Orthop 13290-293,1993 51. Loder RT, Richards BS, Shapiro PS,et a1 Acute slipped capital femoral epiphysis: The importance of physeal stability. J Bone Joint Surg Am 75A:1134-1140,1993 52. Loder RT The demographics of slipped capital femoral epiphysis. An international multicenter study. Clin Orthop 322:8-27,1996 53. Lonstein JE, Carlson JM: The prediction of curve progression in untreated idiopathic scoliosis during growth. J Bone Joint Surg Am 66A1061-1071, 1984 54. Lonstein JE, Winter RB: The Milwaukee brace for the treatment of adolescent idiopathic scoliosis: A review of 1020 patients. J Bone Joint Surg Am 76:1207-1221, 1994 55. Lynch MC, Walsh HP: Tibia recurvatum as a complication of Osgood-Schlatter disease: A report of 2 cases. J Pediatr Orthop 11:543-544, 1991 56. Machida M, Dubousset J, Imamura Y: Pathogenesis of idiopathic scoliosis: SEPs in chickens with experimentally induced scoliosis and in patients with scoliosis. J Pediatr Orthop 14:329-335, 1994 57. Malvitz TA, Weinstein SL Closed reduction for congenital dysplasia of the hip. Fmctional and radiographic results after an average of thirty years. J Bone Joint Surg Am 76A.1777-1792,1994
58. Marty GR, Diaz LS, Gaebler-Spira D: Selective posterior rhizotomy and soft-tissue procedures for the treatment of cerebral palsy. J Bone Joint Surg Am R713-718,1995 59. Mejia EA, Hennrikus WL, Schwend RM, et a 1 A prospective evaluation of idiopathic left thoracic scoliosis with magnetic resonance imaging. J Pediatr Orthop 16354-358, 1996 60. Montgomery GB, Harding B, Harcke HT, et a1 Management of dislocated hips with Pavlik harness treatment and ultrasound monitoring. J Pediatr Orthop 17189-198,1997 61. Mubarak S, Garfin S, Vance R, et al: Pitfalls in the uge of the Pavlik harness for
treatment of congenital dysplasia, subluxation, and dislocation of the hip. J Bone Joint Surg Am 63A1239-1248,1981 62. Mubarak S, Valencia F, Wenger D One-stage correction of the spastic dislocated hip. Use of the pericapsular acetabuloplasty to improve coverage. J Bone Joint Surg Am 74~1347-1357, 1992 63. Nachemson AL, Peterson LE, Members of the Brace Study Group of the Scoliosis
64. 65. 66. 67.
68. 69. 70. 71. 72.
Research Society: Effectiveness of treatment with a brace in girls who have adolescent idiopathic scoliosis: A prospective, controlled study based on data from the Brace Study of the Scoliosis Research Society. J Bone Joint Surg Am 77A815-822, 1995 Nene AV, Evans GA, Patrick JH: Simultaneous multiple operations for spastic diplegia. Outcome and functional assessment of walking in eighteen patients. J Bone Joint Surg Br 75488-494, 1993 Pavlik A: The functional method of treatment using a harness with stirrups as the primary method of conservative therapy for infants with congenital dislocation of the hip. Clin Orthop 281:4, 1992 Pehrsson K, Larsson S, Oden A, et al: Long term follow-up of patients with untreated scoliosis. A study of mortality, causes of death and symptoms. Spine 171091-1096,1992 Ramirez N, Johnston CE, Brown RH: The prevalence of back pain in children who have idiopathic scoliosis. J Bone Joint Surgery 79:364-368, 1997 Rattey T, Piehl F, Wright JG: Acute slipped capital femoral epiphysis: Review of outcomes and rates of avascular necrosis. J Bone Joint Surg 78398402,1996 Rosenberg ZS, Kawelblum M, Cheung YY, et a1 Osgood-Schlatter lesion: Fracture or tendonitis? Scintigraphic, CT and MR imaging features. Radiology 185:853-858,1992 Rowe DE, Bemstein SM, Riddick MF, et a1 Meta-Analysis of the efficacy of nonoperative treatment for idiopathic scoliosis. J Bone Joint Surg Am 79:664-674, 1997 Royle SG, Galasko CS: The irritable hip. Scintigraphy in 192 children. Acta Orthop Scandanavia 63:25-28, 1992 Salter RB Etiology, pathogenesis and possible prevention of congenital dislocation of the hip. Can Med Assoc J 98933-945,1968
OUTPATIENT PEDIATRIC ORTHOPEDICS
73. Sanders JO, Herring JA, Browne RH: Posterior arthrodesis and instrumentation in the immature (Risser grade 0) spine in idiopathic scoliosis. J Bone Joint Surg Am 7739-45, 1995 74. Sandow MJ, Goodfellow J W The natural history of anterior knee pain in adolescents. J Bone Joint Surg Br 67B.36-38, 1985 75. Schwend Rh4, Hennrikus W, Hall JE, et a 1 Childhood scoliosis: Clinical indications for magnetic resonance imaging. J Bone Joint Surg Am 77A:46-53,1995 76. Segal LS, Davidson RS, Robertson WW, et a1 Growth disturbance of the proximal femur after pinning of juvenile slipped capital femoral epiphysis. J Pediatr Orthop 11:631-637, 1991 77. Shulman LH, Sala DA, Chu ML, et a 1 Developmental implications of idiopathic toe walking. J Pediatr 130541-546, 1997 78. Siege1 DB, Kasser JR, Sponseller P, et al: Slipped capital femoral epiphysis: A quantitative analysis of motion, gait and femoral remodeling after in situ fixation. J Bone Joint Surg Am 73659466,1991 79. Staheli LT: Rotational problems in children. J Bone Joint Surg Am 75A:939-949, 1993 80. Stasikelis PJ, Sullivan CM, Phillips WA, et a1 Slipped capital femoral epiphysis: Prediction of contralateral involvement. J Bone Joint Surg Am 78A:1149-1155, 1996 81. Sucato DJ, Schwend RM, Gillespie R Septic arthritis of the hip in children. Journal of the American Academy of Orthopaedic Surgeons 5:249-260,1997 82. Sutherland DH, Santi M, Abel MF: Treatment of stiff knee gait in cerebral palsy: A comparison by gait analysis of distal rectus femoris transfer vs proximal rectus release. J Pediatr Orthop 10433, 1990 83. Svenningsen S, Apalset K, Terjesen T, et al: Regression of femoral anteversion: A prospective study of intoeing children. Acta Orthop Scand 60170-173, 1989 84. Taussig G, Aufaure P, Phlliard D. Apophysitis of the ileac crest: A little known cause of low back pain and pelvic pain in children. Rev Chir Orthop Reparatrice Appar Mot 79:5841, 1993 85. Terjesan T, Osthus P: Ultrasound in the diagnosis and follow-up of transient synovitis of the hip. J Pediatr W h o p 11:608-613, 1991 86. Turner PG, Green JH, Galouko C S Back pain in childhood. Spine 148124314, 1989 87. Vierhout W, Knottnerus JA, Van Ooij, et al: Effectiveness of joint consultation sessions of general practicioners and orthopedic surgeons for locomotor-system disorders. Lancet 346:990-994, 1995 88. Wells D, King JD, Roe TF, et a 1 Review of slipped capital femoral epiphysis associated with endocrine disease. J Pediatr Orthop 13:610-614, 1993 89. Westeris U, Wongstrand H, Forsberg L The effect of arthrocentesis in transient synovitis of the hip in the child: A longitudinal sonographic study. J Pediatr Orthop 16:2429, 1996 90. Winter RB, Lonstein JE, Heifhoff KB, et a1 Magnetic resonance imaging evaluation of the adolescent patient with idiopathic scoliosis before spinal instrumentation and fusion: A prospective double blinded study of 140 patients. Spine 15(22):855-858, 1997 91. Winters TF, Gage JR, Hicks R Gait patterns in spastic hemiplegia in children and young adults. J Bone Joint Surg Am 69A:437441,1987 92. Wood LA, Haller RJ, Hansen PD, et a1 Decreased incidence of scoliosis in hearingimpaired children: Implications for a neurologic basis for idiopathic scoliosis. Spine 20776-780,1995 93. Yashar A, Loder RT, Hensinger RN: Determination of skeletal aEe in children with Osgood-Schlatter disease using radiographs of the knee. J PeYdiatr Orthop 1528301, 1995 v
Address reprint requests to: Richard M. Schwend, MD Department of Orthopedic Surgery Children’s Hospital 219 Bryant Street Buffalo, NY 14222