 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1997 The Journal of Bone and Joint Surgery, Inc.
Current Concepts Review - Hemihypertrophy. Concepts and Controversies* , CLEVELAND, OHIOInvestigation performed at Case Western Reserve University, Rainbow Babies and Childrens Hospital, University Hospitals of Cleveland, Cleveland
Hemihypertrophy is best defined as asymmetry between the right and left sides of the body to a greater degree than can be attributed to normal variation. This asymmetry often appears as a difference in the size or length of the extremities, but it may also involve the head, trunk, and internal organs. It is often extremely difficult to distinguish hemihypertrophy from normal variation. However, this distinction is very important because young children who have hemihypertrophy are at increased risk of malignant intra-abdominal tumors, such as Wilms tumor or hepatoblastoma9,21,22,24,27,29,31,47,60. The increased risk of neoplasia has led to the controversial recommendation of frequent abdominal ultrasound screening designed to detect the tumor at the earliest possible stage2,12,16,25. The orthopaedic literature regarding hemihypertrophy has focused primarily on the treatment of discrepancy between the lengths of the lower limbs and has only superficially acknowledged link between hemihypertrophy and cancer in children4,37,38,45. Even the most recent pediatric and pediatric orthopaedic textbooks mention this association only briefly if at all8,42,59. The objectives of the present review were to describe the clinical features of hemihypertrophy and its related syndromes, to focus on the relationship between hemihypertrophy and malignant tumors in children, and to discuss the controversy surrounding the frequent abdominal ultrasound screening that is recommended for this group of patients.
Hemihypertrophy occurs when a limb or one side of the body enlarges asymmetrically compared with the contralateral limb or side beyond the limits of normal variation. This enlargement may affect only the soft tissues, but it ordinarily involves both the soft tissues and the underlying skeleton. Although this condition is often referred to as hemihypertrophy, it is frequently not possible to determine if the enlargement is due to an increase in the size of cells (hypertrophy) or to an increase in the number of cells of normal size (hyperplasia)57. Isolated or non-syndromic hemihypertrophy must be differentiated from overgrowth that occurs as part of a recognized clinical syndrome affecting other organ systems. Although this distinction is important because the presence of a syndrome may raise unique treatment issues, it also has serious prognostic implications. As will be discussed, children who have hemihypertrophy associated with Beckwith-Wiedemann syndrome are at increased risk for embryonal tumors compared with children who have non-syndromic hemihypertrophy. In this review, the term hemihypertrophy will be used to denote non-syndromic overgrowth of a limb, unless otherwise specified.
Prevalence
Etiology
Clinical Features Non-syndromic hemihypertrophy is associated with many anomalies that do not fit into any recognized clinical syndrome. Genito-urinary abnormalities are particularly common and include inguinal hernias, renal cysts, cryptorchidism, sponge kidney, and horse-shoe kidney13,28,57,60. A compensatory scoliosis may also develop as a result of pelvic obliquity associated with the discrepancy between the lengths of the lower limbs. This spinal curvature is not a structural deformity, and it resolves once the limb lengths are equalized operatively or with the use of orthotics. Cutaneous and vascular lesions are not associated with non-syndromic hemihypertrophy and, when present, indicate a generalized overgrowth syndrome, such as neurofibromatosis, Klippel-Trénaunay-Weber syndrome, Proteus syndrome, or Beckwith-Wiedemann syndrome57.
Classification Congenital hemihypertrophy may be classified as total or limited. Children who have total hemihypertrophy have involvement of all organ systems, including the ipsilateral paired organs, whereas those who have limited hemihypertrophy have only muscular, vascular, skeletal, or neurological involvement. The limited forms of hemihypertrophy are also described according to the extent of somatic involvement. Such forms include classic hemihypertrophy (involvement of the ipsilateral upper and lower limbs) as well as the form in which the overgrowth is segmental (limited to an upper or lower limb), facial (involving the head, face, or a portion thereof), or crossed (involving the contralateral upper and lower limbs)4,61. Hemihypertrophy may also be classified as isolated (non-syndromic hemihypertrophy) or as part of a recognized clinical syndrome (syndromic hemihypertrophy). We prefer the term non-syndromic hemihypertrophy to isolated hemihypertrophy because children who have this form of the disease may have other clinical anomalies that do not fit into any currently recognized syndrome.
Diagnosis Although hemihypertrophy is easily detected in its severe form, smaller discrepancies in limb length and circumference may not be clinically apparent in the early stages of development. It is nonetheless important to distinguish between pathological overgrowth and normal variations in growth as soon as possible, given the association between hemihypertrophy and malignant tumors in children. This distinction is often difficult. In the normal population, the upper and lower extremities may differ in length and circumference by as much as one to two centimeters compared with the contralateral limbs1. In a series of 1000 United States Army recruits studied by Rush and Steiner53 in 1946, only 23 per cent were found to have lower extremities of equal length; the mean discrepancy for the remaining 77 per cent was 0.75 centimeter. Assuming a mean lower-limb length of eighty-two centimeters at the age of eighteen years1, this 0.75-centimeter discrepancy constitutes a mean difference in limb lengths of 0.9 per cent, with use of the data of Pappas and Nehme45 as a reference. Fifteen per cent of the recruits were noted to have a discrepancy of 1.0 centimeter or more. Again assuming a mean length of eighty-two centimeters at the age of eighteen years, this constitutes a normal mean difference of 1.2 per cent or more. The right eye and right ear may be located a few millimeters higher than those on the left, and the nipples are rarely located at the same level or at the same distance from the midline51. Given the extent of asymmetry in normal individuals, it may not be possible to distinguish mild cases of hemihypertrophy from normal variation. The magnitude of the growth discrepancy must also be interpreted with respect to the age and size of the child. With use of a threshold of a 5 per cent difference to define abnormal asymmetry, as suggested by Pappas and Nehme45, as well as the data of Anderson et al.1 regarding mean lengths of the femur and tibia at each age, this translates into a discrepancy between the lengths of the lower limbs of 1.3 centimeters at the age of one year, 2.3 centimeters at the age of five years, 3.2 centimeters at the age of ten years, and 4.1 centimeters at the age of eighteen years or maturity. In an unpublished survey cited by Pappas and Nehme, 376 normal children in the Growth Study of Children's Hospital in Boston had a mean discrepancy of 0.4 centimeter (1.7 per cent) at the age of one year, which increased to a mean of 0.8 centimeter (1.3 per cent) at the age of ten years and a mean of 1.1 centimeters (1.3 per cent) at maturity. Therefore, a one-centimeter discrepancy between the lengths of the lower extremities of an adolescent may fall within the normal distribution of limb-length discrepancy (1.2 per cent), but it would be in the range of pathological overgrowth (4.2 per cent) in an infant. In some cases, it may also be difficult to distinguish whether the larger extremity has hypertrophied or whether the smaller extremity has atrophied (hemihypotrophy or hemiatrophy)4. As far as we know, there has been no reported association between hemiatrophy and the development of embryonal tumors; therefore, this distinction has serious prognostic implications. In these situations, comparing the height of the child standing with that of the child sitting and comparing the lengths of the right and left femora and tibiae with normal standards may reveal whether the abnormal limb is hypertrophic or hypotrophic1. To our knowledge, however, this technique has not been clinically validated. Radiographic studies, such as scanograms and orthoroentgenograms, are used for more precise measurement of skeletal lengths than is possible with clinical examination alone.
Differential Diagnosis
Beckwith-Wiedemann Syndrome The diagnosis of Beckwith-Wiedemann syndrome is made on the basis of clinical findings alone. Elliott and Maher18 suggested that macroglossia, prenatal or postnatal overgrowth, and defects of the anterior aspect of the chest wall are major diagnostic criteria, which are supported by minor features of ear creases or pits, flamed-shaped nevi on the face, hypoglycemia, enlarged kidneys, and hemihypertrophy. Birth weight may be an indication as children who have Beckwith-Wiedemann syndrome typically have a birth weight that is in greater than the ninetieth percentile. Therefore, a careful physical examination should be performed when Beckwith-Wiedemann syndrome is suspected. Laboratory testing, such as for levels of serum glucose and calcium in newborns, can also be supportive of the diagnosis. Standard cytogenetic analysis may also be performed to examine the chromosome 11p.15 region for deletions and translocations, which have been seen in some patients18,62. Hemihypertrophy occurs in approximately 13 per cent of patients who have Beckwith-Wiedemann syndrome: forty-eight (12 per cent) of 388 children in a study by Sotelo-Avila et al.56 and twenty-two (13 per cent) of 174 children in a study by Wiedemann66 had hemihypertrophy. Hypoglycemia develops in the first few days of life in as many as one-third of infants who have Beckwith-Wiedemann syndrome, and it may result in neurological sequelae if it is not anticipated. Developmental delay and mental retardation occur rarely and may be a result of unrecognized or uncontrolled hypoglycemia. Girls are as likely to be affected as boys. The cause of Beckwith-Wiedemann syndrome and other overgrowth syndromes is currently under active investigation. While most cases of Beckwith-Wiedemann syndrome are sporadic, there is clear evidence that the underlying cause is genetic. Evidence to support an autosomal dominant transmission has been reported in several families. Elliott et al.19 found Beckwith-Wiedemann syndrome to be familial in five (7 per cent) of seventy-five children. Genetic linkage studies have mapped the familial form to a specific region of chromosome 11, further supporting the underlying genetic etiology of Beckwith-Wiedemann syndrome35,50. Translocations and deletions of chromosome 11 have also been identified in some patients44,62. Of considerable interest is the observation that the gene encoding IGF-II (insulin-like growth factor-II) is located in the same region of chromosome 11 as the genetic locus linked with Beckwith-Wiedemann syndrome. This raises the possibility of involvement of the IGF-II gene in the pathogenesis of Beckwith-Wiedemann syndrome and hemihypertrophy. Overexpression of IGF-II may be important in the etiology of Beckwith-Wiedemann syndrome because of its role in stimulating the growth of many diverse tissues during embryonic and fetal development64. Dysregulation of growth factor control of cell growth may also be important in both the development of hemihypertrophy in Beckwith-Wiedemann syndrome and the etiology of embryonal tumors14,63. Patients who have Beckwith-Wiedemann syndrome in conjunction with hemihypertrophy are at increased risk for embryonal tumors compared with those who have Beckwith-Wiedemann syndrome without hemihypertrophy33,55,56,64,66. Wiedemann66 reported a tumor in twenty-nine (7 per cent) of 388 patients who had Beckwith-Wiedemann syndrome. Twenty-four per cent (twelve) of the forty-nine patients who had hemihypertrophy associated with Beckwith-Wiedemann syndrome had a tumor compared with 5 per cent (seventeen) of the 339 patients who did not have hemihypertrophy. In a series of 174 children who had Beckwith-Wiedemann syndrome56, neoplasms developed in six (27 per cent) of twenty-two children who had hemihypertrophy compared with fourteen (9 per cent) of 152 children who did not have hemihypertrophy. The risk of a tumor does not appear to be related to the severity of the phenotypical expression of the syndrome. The most common neoplasm is a Wilms tumor. In a study by Sotelo-Avila and Gooch55, six of seventeen children who had a malignant tumor had a Wilms tumor. Others have reported similar findings56. Hepatoblastoma, adrenal cortical carcinoma, and rhabdomyosarcoma account for most of the remaining tumors; however, many other malignant tumors have been reported in association with hemihypertrophy and Beckwith-Wiedemann syndrome, including lymphoma, gonadoblastoma, brain-stem astrocytoma, ganglioneuroma, and carcinoid tumor of the appendix13,55.
Neurofibromatosis
Klippel-Trénaunay-Weber Syndrome
Proteus Syndrome
Hemihypertrophy is characterized by abrogation of normal cellular growth-control mechanisms, resulting in regional overgrowth of many types of tissues. Although the pathways regulating symmetry of the body and normal cellular growth control may be distinct, there is the possibility that the same abnormality that allows the overgrowth to occur also predisposes these patients to the formation of tumors55. For example, the current hypothesis regarding neoplastic transformation is that a single mutation in a growth-control pathway is insufficient to cause a malignant tumor. However, the acquisition of multiple sequential mutations within the same pathway may result in neoplastic transformation34. Thus, it is not surprising that children who have hemihypertrophy are also at increased risk for malignant tumors. The most common neoplasm associated with hemihypertrophy is Wilms tumor, followed by adrenal carcinoma and hepatoblastoma21,23,46,47. To our knowledge, the prevalence of neoplasia in children who have non-syndromic hemihypertrophy has never been studied in a large prospective study. The initial epidemiological studies that indicated an association between certain malignant tumors and hemihypertrophy were based on the prevalence of hemihypertrophy in children in whom an embryonal tumor had already been diagnosed22,40,49. Miller et al.41 retrospectively reviewed the hospital charts of 440 children who had a diagnosis of Wilms tumor between 1940 and 1963 and identified three children (0.7 per cent) who had hemihypertrophy. In a study of 225 patients who had a Wilms tumor, Fraumeni et al.22 identified only seven patients (3 per cent) who had hemihypertrophy. A 1976 survey by the National Wilms' Tumor Study Group identified hemihypertrophy in sixteen (3 per cent) of 547 children who had a Wilms tumor49. Unfortunately, these studies have often been misquoted as representing the prevalence of tumors in children who have hemihypertrophy. The best available data concerning the prevalence of neoplasia in children who have hemihypertrophy come from two separate sources. The first source was a retrospective survey by Parker and Skalko46 of 860,000 admissions to pediatric hospitals; ten children were noted to have total congenital hemihypertrophy on physical examination. The criteria for establishing the diagnosis of hemihypertrophy were not given, nor was mention made of other forms of hemihypertrophy. Two of these ten patients had an adrenal tumor, and Beckwith-Wiedemann syndrome was suspected in one of the two. The second source is a frequently cited abstract by Hoyme et al.31 from 1987. Those authors initially studied the cases of thirty-seven children with non-syndromic hemihypertrophy who were followed beyond the age of five years. Again, the diagnosis was made by physical examination, although no criteria were given. No tumors developed in these children. A second group of 104 children who had non-syndromic hemihypertrophy, which included the original thirty-seven patients, was then assembled from multiple centers in the United States and Canada, and these children were also followed beyond the age of five years. A tumor developed in four (4 per cent) of these children. In neither of these studies was the number of patients sufficient to allow firm conclusions regarding the risk of neoplasia in children who have non-syndromic hemihypertrophy. Whether the true prevalence is more or less than 3 per cent must therefore be determined by larger prospective studies.
The link between hemihypertrophy and embryonal tumors led Beckwith7 and others2,3,11,12,54,64,66 to recommend that serial abdominal ultrasound screening be performed every three to six months on children who have hemihypertrophy, in order to detect these tumors at the earliest possible stage. The same recommendation applies to children who have Beckwith-Wiedemann syndrome and those who have both Beckwith-Wiedemann syndrome and hemihypertrophy64. Most associated tumors arise in the abdomen and can be easily detected on abdominal sonography. Sonography is readily available and can be performed frequently without exposure of the patient to radiation or the need for sedation. The rationale for screening is intuitive. An effective screening program should lead to early diagnosis, early diagnosis should allow early treatment of the tumor, and early treatment should lead to a reduction in morbidity and mortality. The frequency of the examinations is dictated by the extremely rapid growth of these tumors. Several investigators have reported Wilms tumors developing within six months after a normal screening sonogram was made2,3. At our institution, a hepatoblastoma presented as a large palpable mass in a nine-month-old child eight months after a normal screening sonogram was made. Screening may be discontinued as a child matures, as the risk of malignant tumors decreases with age. In a report from the National Wilms' Tumor Study Group, Breslow et al.10 noted that 3041 (95 per cent) of 3201 children who had a Wilms tumor were diagnosed before the age of ten years. However, no consensus has been reached in the literature regarding the point at which screening may be discontinued12. Wiedemann66 originally recommended abdominal ultrasound screening every three months until the age of three years and then every six months with no suggested end point. Weng et al.64 recommended ultrasound examinations every three months until the age of seven years and then every six months until the completion of skeletal growth. A consensus of participants at the First International Conference on Molecular and Clinical Genetics of Childhood Renal Tumors suggested that ultrasound examination be performed every three months until the age of seven years, followed by physical examinations of the abdomen every six months until the completion of growth12. Unfortunately, these recommendations regarding screening are not well known outside of the genetics and pediatric literature, have not been rigorously tested in a prospective clinical trial, and are controversial. Frequent abdominal sonography, while non-invasive, requires an extensive commitment of financial resources, time, and effort by a family. It is therefore of critical importance to determine whether screening programs can detect the tumors earlier than is possible with careful physical examination and whether earlier detection can lead to a better clinical outcome for the patient. To date, we know of no data demonstrating that patients who have had serial abdominal ultrasound screening have better rates of survival than those who have not. In fact, a Wilms tumor with favorable histological characteristics has an excellent prognosis even when it is detected at a late stage. The National Wilms' Tumor Study Group26 found that twelve of thirteen children who had a stage-II, III, or IV Wilms tumor with focal anaplasia were alive and had not had a recurrence four years after treatment. In a recent retrospective study of forty-seven children who had a Wilms tumor associated with aniridia, Beckwith-Wiedemann syndrome, or hemihypertrophy, it was concluded that screening is of unproved value in the detection of Wilms tumor at an earlier stage16. However, seven of the fourteen children who had hemihypertrophy had the diagnosis of hemihypertrophy made at the same time as the diagnosis of Wilms tumor, and screening was performed in only one of the remaining seven patients. Although that study16 raises important questions about the value of screening, it is difficult to interpret because of the small number of children who had hemihypertrophy. Given the low prevalence of hemihypertrophy and the favorable prognosis for Wilms tumor, only a long-term, multinational, prospective study can achieve the statistical power necessary to prove or disprove the clinical efficacy of screening16. It is possible that patients who have a hepatoblastoma or an adrenal cortical carcinoma are more likely to benefit from early detection than are those who have a Wilms tumor, as the results of treatment of these tumors at a late stage are not as favorable. However, in Wiedemann's series66, six (19 per cent) of the tumors that developed in the thirty-two patients who had Beckwith-Wiedemann syndrome with hemihypertrophy arose in an extra-abdominal location. Sotelo-Avila and Gooch55 reported similar results. Patients who have an extra-abdominal tumor cannot benefit from abdominal screening. A screening program can also only benefit patients who are identified in advance as being at risk for a tumor. As Beckwith-Wiedemann syndrome is often diagnosed at birth, patients who have the disorder are amenable to screening for tumors. However, several retrospective studies16,25 have demonstrated that hemihypertrophy is not diagnosed until after a tumor has been identified in approximately one-half of patients who have Beckwith-Wiedemann syndrome. The National Wilms' Tumor Study Group25 reported that the diagnosis of hemihypertrophy was made less than one month before detection of a tumor in fifty (68 per cent) of seventy-four patients. This observation may reflect the inherent difficulty in distinguishing pathological overgrowth from normal growth variation in some patients, as was discussed previously.
Although the relationship between hemihypertrophy and embryonal tumors is well recognized, serious questions remain about how best to distinguish hemihypertrophy from normal variation in growth and how best to screen for associated malignant tumors. Currently, there are no objective criteria for distinguishing hemihypertrophy from normal variation in children. In many patients, the diagnosis of hemihypertrophy is not made until after a tumor has been recognized. These patients clearly cannot benefit from a cancer-screening program. Also, it is yet to be proved that serial abdominal ultrasound screening improves the survival of patients in whom hemihypertrophy has been diagnosed in the absence of a tumor. Despite these severe limitations, serial abdominal ultrasound screening continues to be recommended for children who have hemihypertrophy in an attempt to achieve the earliest possible diagnosis and treatment of these life-threatening embryonal tumors. Larger prospective studies of children who have hemihypertrophy are required to determine whether the expense of time, effort, and resources is justified. There are limited data available to help the practicing orthopaedic surgeon to determine the proper approach for a child who has hemihypertrophy. Until prospective clinical trials resolve this dilemma, we recommend the following guidelines. The primary role of the orthopaedist in the management of a child who has hemihypertrophy should be to confirm the diagnosis, help to distinguish non-syndromic from syndromic forms of this disorder, and treat the musculoskeletal problems related to the discrepancy between the lengths of the lower extremities. Children who have non-syndromic hemihypertrophy or hemihypertrophy associated with Beckwith-Wiedemann syndrome should be screened for the development of malignant intra-abdominal tumors. The orthopaedic surgeon should communicate with the primary-care physician about the increased risk of an embryonal tumor in these children. The primary-care physician may then want to refer the child to a clinical geneticist, who can more expertly evaluate the child for the presence of Beckwith-Wiedemann syndrome or other generalized overgrowth syndromes and better inform the parents about the advantages and disadvantages of serial abdominal ultrasound screening. If a genetics consultation is not locally available, the orthopaedic surgeon and the pediatrician should work together to ensure that the child receives serial abdominal examinations and ultrasound screening of the abdomen at appropriate intervals. This requires either making a relatively empirical decision about the frequency and duration of the screening program or contacting experts in their region to determine what screening recommendations are currently being followed. Currently, the frequency and duration of these abdominal ultrasound-screening examinations are widely variable, ranging from no screening at all to examination every three to six months until skeletal maturity.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Introduction
Hemihypertrophy
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |