Abstract
Purpose
To investigate whether atrophy of the leg muscles present in congenital clubfoot (CCF) is primitive or secondary to treatment of the deformity.
Methods
Magnetic resonance imaging (MRI) of both legs was taken in three cohorts of patients with unilateral congenital clubfoot (UCCF): eight untreated newborns (age range 10 days to 2 weeks); eight children who had been treated with the Ponseti method (age range 2–4 years); eight adults whose deformity had been corrected by manipulation and casting according to Ponseti, followed by a limited posterior release performed at age 2–3 months (age range 19–23 years). All of the treated patients wore a brace until 3 years of age. Muscles were measured on transverse MRI scans of both legs taken midway between the articular surface of the knee and the articular surface of the ankle, using a computer program (AutoCAD 2002 LT). The same program was used to measure leg muscles in the histologic cross sections of the legs of two fetuses with UCCF, spontaneously aborted at 13 and 19 weeks of gestation, respectively. Measurements of the whole cross section of the leg (total leg volume: TLV), of the muscular tissue (muscular tissue volume: MTV), and of the adipose tissue (adipose tissue volume: ATV) of the tibia, fibula, and of the other soft tissues (tendons, nerves, and vessels) were taken by using an interactive image analyzer (IAS 2000, Delta System, Milan, Italy).
Results
Marked atrophy of the leg muscles on the clubfoot side was found in both fetuses and untreated newborns, with a percentage ratio of MTV between the normal and the affected leg of 1.3 and 1.5, respectively. Leg muscle atrophy increased with growth, and the percentage ratio of MTV between the normal and the affected leg was, respectively, 1.8 and 2 in treated children and adults. On the other hand, fatty tissue tended to increase relatively from birth to adulthood, but it could not compensate for the progressive muscular atrophy. As a result, the difference in TLV tended to increase from childhood to adulthood.
Conclusions
Our study shows that leg muscular atrophy is a primitive pathological component of CCF which is already present in the early stages of fetal CCF development and in newborns before starting treatment. Muscular atrophy increases with the patient's age, suggesting a mechanism of muscle growth impairment as a possible pathogenic factor of CCF.
Introduction
Several pathologic studies in fetuses and stillborns with congenital clubfoot (CCF) have shown atrophy and shortening of the leg muscles, with both the triceps surae and tibialis posterior being mostly affected. As a consequence of muscle belly shortening, their tendons are relatively longer than normal [1–7].
Atrophy of the musculature of the leg has also been described in clinical studies on CCF, as mainly evident in unilateral cases [8–11]. In spite of this common clinical finding, no study has, so far, correlated atrophy of the leg musculature in patients with CCF with atrophy of the same muscles present in fetuses. Moreover, no author has, so far, investigated leg musculature in babies with CCF before starting treatment.
Magnetic resonance imaging (MRI) is a very useful tool to study soft tissues, but, until now, MRI studies in CCF have aimed to assess the effect of treatment on both the shape and the relationships of the osteocartilaginous anlagen of the clubfoot in immature patients [12–14]. In our study, we used MRI to investigate leg muscles in unilateral congenital clubfoot (UCCF) cases of untreated babies and in children and adults who had been treated soon after birth. MRI findings were compared to the histopathologic features of the leg musculature in fetuses with UCCF. The aim of our study was to ascertain whether atrophy of the leg musculature in CCF is a primitive pathologic component of this congenital deformity or secondary to treatment.
Materials and methods
Two fetuses with UCCF were obtained from spontaneous abortion at 13 and 19 weeks of gestation, respectively. Both legs of each fetus were disarticulated at the knee joint, fixed in 10% buffered formaldehyde, and embedded in paraffin. Sections that were 7 μm thick cut in the transverse plane were stained with hematoxylin and eosin.
Twenty-four patients with UCCF had an MRI examination of both legs; eight patients were babies (age range 10 days to 2 weeks), eight were children (age range 2–4 years), and eight were adults (age range 19–23 years). All of them had a deformity graded at diagnosis as Manes group 3 [15]. Six babies were male and two female; five children were male and three female; five adults were male and three female. The eight babies had had no treatment. The eight children had been treated by manipulation and casting, followed by subcutaneous heel cord tenotomy, according to the Ponseti method [16]. In the eight adult patients, manipulation and casting according to Ponseti was followed by a limited posterior release, consisting in Z-lengthening of the Achilles tendon and posterior capsulotomy of the ankle joint [11]. All of the patients had an average of five plaster casts applied (range 4–6), and all of them were wearing or had worn a brace until 3 years of age at the time of the MRI examination. All of the patients had started their treatment within the second week of life, and none of them had had any previous treatment. No patient had had a recurrence of the deformity at the time of the MRI examination.
At the physical examination of babies with UCCF, no difference in calf circumference could be observed between their legs. In the 2–4-year-old children, a slight difference was evident that accounted for an average of 5 mm (range 4 mm to 1 cm), whereas in the adults, the difference in calf circumference accounted for an average of 3 cm (range 2.8–4 cm).
In the babies and children, the MRI examination was performed under general anesthesia, which was induced and maintained with Sevoflurane 2–6%. Premedication was done with Midazolam p.o. at the dose of 0.05 mg/kg of body weight. All MRI scans were obtained with a conventional T1-weighted spin echo sequence (echotime, 1 ms; repletion time, 400 ms) using 4-mm cross sectional and coronal images with no intersection interval. A 256 × 256 matrix and two excitations were used. The magnetic resonance images were converted into digital pictures for analysis on a computer.
Five serial cross sections of both legs of each fetus and five serial cross scans of both legs of each patient were selected for the study. The images selected were taken midway between the articular surface of the knee and the articular surface of the ankle (Fig. 1).
Magnetic resonance imaging (MRI) coronal scan of both legs of a 20-year-old girl with left congenital clubfoot (CCF). MRI cross scans of the patients’ legs were taken midway (solid line) between the knee and the ankle articular surface (lower dotted line). Histologic cross sections of the fetuses’ legs were cut at a similar level
Measurements of the whole cross sectional area of the leg (total leg volume: TLV), of the muscular tissue (muscular tissue volume: MTV), and of the adipose tissue (adipose tissue volume: ATV) of the tibia, fibula, and of the other soft tissues (tendons, nerves, and vessels) were taken by using an interactive image analyzer (IAS 2000, Delta System, Milan, Italy, info@deltacontrols.it).
The software Auto Cad 2002 LT elaborated by Microsoft Corporation (italyedu@microsoft.com) in association with ACIS: Spatial Technology, Inc. (info@spatialtechnologies.ca), and licensed by the Microsoft Corporation, calculated automatically the percentage volume of each tissue within the whole cross section area of the leg. Then, the ratio of the volume of the single tissue and the whole cross section area of the leg between the normal and the affected leg was also calculated.
All of the values reported are the average value of the measurements of the TLV and of the volume of each tissue calculated on either the histological sections of the fetuses or on the MRI scans of the 24 patients, taken at the same level.
The investigation protocol was submitted to the ethical committee of our University Hospital. Approval was given contingent on two comments and two conditions: (1) the study was considered feasible because it might yield new data to improve knowledge of the unknown pathogenesis of congenital clubfoot; (2) the chief anesthesiologist of our hospital, who is a member of the ethical committee, judged the MRI sedation protocol to be absolutely safe for both newborns and children; (3) all of the babies and children had to be carefully evaluated by a pediatrician, a pediatric cardiologist, and an anesthesiologist before sedation for the MRI examination; (4) informed consent had to be given by the parents after receiving a detailed description of the research protocol—they had to know in advance that MRI was not mandatory for treatment, but it was very important to highlight the mechanisms causing the congenital deformity.
A statistical analysis, using the exact Fisher's test [17], was performed on the percentage values obtained by computer measurement. The data are considered to be statistically significant when P < 0.05.
Results
Fetuses with unilateral congenital clubfoot (UCCF)
The serial cross sections of the affected leg showed atrophy of all of the muscles in comparison to the normal leg. Triceps surae, tibialis posterior, and the other muscles of the posterior compartment of the leg were more atrophic than those of both the anterolateral and the lateral compartments.
Atrophy was particularly evident in the lower third of the leg where, in normal conditions, tendons are represented more than muscles. No morphological abnormalities were observed in the muscle fibers (Fig. 2). The percentage of MTV, ATV, tibia, fibula, and other soft tissue volumes in the normal and the affected leg, the ratio of each single tissue between the normal and the affected leg, as well as the ratio of the TLV between the normal and the affected leg, are reported in Figs. 2 and 6.
In the cross sections of the legs of the two fetuses, muscular tissue was decreased on the affected side (32.88 vs. 41.59%), but the difference was not statistically significant (P < 0.412).
Patients with unilateral congenital clubfoot (UCCF)
In all our patients, both untreated babies and treated children and adults, MRI of the legs showed atrophy of all of the muscles on the affected side. The muscles mostly affected were the triceps surae, the tibialis posterior, and the other muscles of the posterior compartment of the leg. The muscles of both the anterolateral and the lateral compartments of the leg were also affected, but to a lesser extent. The subcutaneous fat pad was thicker than normal in the leg of the clubfoot side, and fat was also present in the interstitial spaces of the atrophic muscles. The ratio of the TLV between the normal and the affected leg, the percentage of MTV, ATV, tibia, fibula, and other soft tissue volumes in the normal and the affected leg, as well as the ratio of each single tissue between the normal and the affected leg in untreated babies and in treated children and adults, are reported in Figs. 3, 4, 5, and 6. In the MRI cross scans of the legs of all of the patients with UCCF, the percentages of muscular tissue were lower on the clubfoot side than on the normal side, whereas the adipose tissue percentages were higher on the clubfoot side than on the normal side. The difference in MTV between the normal and the affected leg was statistically significant in both children (P < 0.024) and adults (P < 0.005), but not in newborns (P < 0.105).
In unilateral congenital clubfoot, leg circumference discrepancy increases with growth, as shown by the ratio of the total leg volumes (TLVs, whole cross sectional areas) between the normal and the affected leg of the fetuses, newborns, children, and adults. Atrophy of the musculature of the affected leg also increases with growth, as shown by the ratio of the muscular tissue volume (MTV) between the normal and the affected leg from fetal age to adulthood. In the affected leg, a compensatory fat hypertrophy is also observed during growth, as shown by the ratio of the adipose tissue volume (ATV) between the affected and the normal leg
Discussion
As far as we know, no author has correlated leg muscle atrophy described in fetuses and stillborns with CCF [1–7] to leg muscle atrophy reported in clinical studies on CCF. It has always been debated whether calf atrophy is a primitive component of the congenital deformity or, rather, a consequence of the prolonged treatment to which patients with CCF are subjected, especially in severe cases [8–11]. Aronson and Puskarich [18] first showed, in a clinical study on patients with UCCF, that there is no correlation between the amount of calf atrophy on the clubfoot side and the time of immobilization in a plaster cast, up to a period of 2 years.
In our study, we confirmed leg muscle atrophy in fetuses, but we also found leg muscle atrophy in eight newborns with severe UCCF before starting treatment. We believe that this is a clear-cut demonstration that leg muscle atrophy is a basic defect in CCF rather than the consequence of treatment.
We carried out a quantitative analysis of MTV using a sophisticated computer software package. In fetuses and newborns, the MTV of the affected leg was, respectively, 9 and 15% less than in the normal leg, but this difference, although considerable, was not statistically significant. The MTV ratio between the normal and the affected leg tended to increase with growth, and the difference became statistically significant in both children and adults.
At the physical examination, no difference in calf thickness between the normal and the affected leg could be observed in either fetuses or newborns. The relative increase of ATV in the affected leg of both fetuses and newborns may explain this finding. However, the compensatory effect of fat hypertrophy was not able to counterbalance the decreased MTV of the affected leg in children and adults. In fact, nutritional studies have shown that, in children's limbs, fat deposition increases steeply during the first year of life, and then it becomes more or less constant, with a small peak during puberty in males [19, 20].
The combination of decreased MTV and the lack of TLV compensation by increased ATV resulted in calf thickness discrepancy in both children and adults. Thus, the ratio of TLVs between the normal and the affected leg increased to 1.07 in children and to 1.4 in adults. However, the MTV difference between the clubfoot and control leg was not statistically significant in fetuses and newborns, but was statistically significant in the treated children and adults. Thus, atrophy might have been worsened by treatment, although Aronson and Puskarich [18] found that plaster cast and/or brace immobilization for up to 2 years has no important influence on leg wasting in congenital clubfoot.
Our fetuses were autopsy specimens, and for that reason, it was not possible to perform histochemical or electron microscopy investigations on their musculature. However, at light microscopy, no evidence of muscle fiber degeneration could be detected, thus, suggesting a neuromuscular disease as a possible pathogenetic factor in CCF. On the other hand, this hypothesis has been ruled out by several studies performed on biopsy specimens of patients with CCF, where the only recurrent microscopic finding was a relative increase of type I fibers in the musculature of the affected leg [21–27]. Therefore, according to Gray and Katz [22], wasting of leg musculature in CCF can only be due to a reduction in the number of muscle fibers.
Bechtol and Mossman [1] and Flinchum [2] first suggested the hypothesis that foot deformity in CCF might be caused by the “failure of muscle growth to keep pace with bone growth.” Furthermore, to explain muscle imbalance [28, 29], we could speculate that, in CCF, some muscle groups grow up less than others.
The mechanisms of muscle growth are still poorly understood. It is well known that growth mechanisms include an increase in both thickness and length of the muscle belly. Growth implies the differentiation of myoblasts into new muscle fibers that may arrange themselves either in parallel, during radial growth, or in series, during longitudinal growth. These mechanisms seem to be regulated by different growth factors: radial muscle growth may occur at any place along the muscle belly, whereas longitudinal growth takes place mainly at the myotendinous junctions [30–32]. In CCF, both mechanisms of muscle growth might be impaired, although, according to our findings and to the pathologic findings of previous authors, longitudinal growth seems to be mostly impaired at the distal myotendinous junction. As a consequence, tendons on the clubfoot side appear to be longer than on the normal side [2, 5–7].
CCF is a genetic disorder [33]. The impairment of muscle growth could be caused genetically by a different expression of the pathologic gene varying from one case to another.
In our study, only severe clubfeet graded as Manes group 3 were included. It would be interesting to perform the same study in a clubfoot population of varying severity, including also Manes group 1 and 2 cases. In fact, it is a common observation in clinical practice that mild cases present less leg atrophy than severe cases [18], but no study on a large cohort of patients has, so far, demonstrated a correlation between leg muscle atrophy and clinical CCF severity. A new classification of CCF severity might be proposed, based on the amount of leg muscle atrophy measured either by MRI or by sonography.
One of the main limitations of our study is that the three cohorts of patients were each different from the other two, instead of having one single cohort of patients followed prospectively in years from birth to adulthood. Another limitation is the lack of a control group between treated and untreated children or adults to further elucidate the effect of treatment on the MTV, TLV, etc. However, all of our patients had a similar grading of their CCF severity, and treatment, too, was uniform. We plan to perform future prospective studies on a single cohort of patients followed from birth to adulthood.