Twelve consecutive subjects who met the inclusion criteria with idiopathic clubfoot deformity were recruited from two institutions: St. Louis Children’s Hospital (six patients) and Saint Louis Shriners Hospital (six patients). All but one patient had isolated unilateral clubfoot deformity. The other lower extremity was without congenital deformities except in one patient who had an additional anomaly of a congenital vertical talus on the right side and a clubfoot deformity on the left. All patients were treated with serial casting followed by either a percutaneous tendoachilles tenotomy or an open Z-lengthening of the Achilles tendon. No patient had extensive soft tissue release surgery for clubfoot treatment. Specific exclusion criteria related to MRA imaging included age younger than 6 years, history of claustrophobia, implanted or accidental exposure to metal fragments, and pregnancy. Specific implanted metal fragments included a pacemaker, defibrillator, neurostimulator, artificial heart valve, and cerebral aneurysm clip. One of 12 scanned patients was excluded later owing to motion artifact on the MRA. Thus, 11 patients were included in the final analysis. The mean age of the 11 patients was 16.5 years (range, 6–60 years). There were eight male patients (73%) and three female patients (27%). A positive family history of clubfoot was documented in six patients. Seven (64%) of the 12 clubfeet were right-sided (Table ). The minimum followup was 60 months (mean, 85 months; range, 50–108 months). Our Human Research Protection Committee approved the study, and written informed consent was obtained for all individuals.
All 11 patients were examined by one author (MBD) who specializes in clubfoot evaluation and treatment of children and adults. Eight of the 11 included patients were treated by one surgeon (MBD) from birth using the Ponseti method of serial manipulations and castings, a percutaneous tendoachilles tenotomy, and foot abduction bracing for 3 years [31
]. These eight patients had Grade III deformities, using the classification of Dimeglio et al., at the onset of treatment, indicating moderate to severe clubfoot [3
]. Three patients experienced a relapse of the clubfoot deformity. In each case, the recurrent clubfoot deformity was treated successfully with repeat casting and no additional surgery. The remaining three patients, who were adults at the time they were seen at our institution, all had their initial clubfoot treatment done elsewhere as infants with serial manipulations and castings, open Z-lengthening of the Achilles tendon, and foot abduction bracing for 2 years. The three adult patients were evaluated at the time of the study by one author (MBD) with physical and radiographic examinations of the feet and lower extremities. Medical records and photographs from birth and early clubfoot treatment also were reviewed for the three patients to ensure an accurate diagnosis. Severity classification was not available for the three adult patients, although clinical records made notation of severe, rigid deformities in all three.
MRA was obtained from 5 to 59 years after completion of initial treatment. No patient was sedated for the purpose of MRA and patients were excluded if they could not lie still for the duration of the MRA study. Noncontrast-enhanced MRA was performed on a 1.5-T system (Magnetom Avanto; Siemens Medical Solutions USA, Inc, Malvern, PA, USA). Coronal scout images, fat-suppressed T2-weighted images, and NATIVE sequences were obtained from the tibial plateau to the ankle. NATIVE is a 3-D angiographic technique that allows acquisition of noncontrast-enhanced images with high resolution using a short scan time. Noncontrast-enhanced arterial phases were acquired by assessing inherent differences in signal between fast flowing blood during the systolic phase and the slower blood during the diastolic phase of the cardiac cycle. An ECG or pulse triggering is used to synchronize the data acquisition to the cardiac cycle. The difference between images acquired during fast- and slow-flow phases resulted in the angiographic images. A phase-contrast flow quantification with retrospective gating was used to determine trigger times for fast- (systolic) and slow-flow (diastolic) phases before the 3-D acquisition. The 3-D sequence then was acquired with a base resolution of 256, 30 cm field of view, 1.5 mm thickness, TR of 234, TE of 17 ms, bandwidth of 977 Hz/pixel, 2.3 ms echo spacing, centric encoding, trigger delay of 0/600 for slow flow, and trigger delay of 280/300 ms for fast flow. A phase array peripheral extremity surface coil was used with parallel imaging with an acceleration factor of two. The total scan time for imaging was less than 30 minutes. A T1-weighted MR image set was acquired for the same anatomic coverage. This image set was used as a complementary image to the NATIVE sequence to evaluate muscle mass and fat.
Images were transferred electronically to a workstation for analysis. The vascular structures were analyzed independently by two experienced readers (MS, SS) who were blinded to the clinical history (ie, knowledge of which extremity was abnormal). A consensus reading was performed in instances of disagreement. The arterial pattern in the clubfoot limb was compared with that of the contralateral limb. The presence or absence of the anterior tibial, posterior tibial, peroneal, and dorsalis pedis arteries was recorded. The MRA was considered normal when the anterior tibial, posterior tibial, and peroneal arteries were of equal caliber at the ankle. An artery was considered abnormal if it was not detectable (ie, absent) or small compared with the opposite side (ie, hypoplastic).
Volume measurements of subcutaneous fat, muscle, and total mass were made of the affected and unaffected limbs by one observer (LJM) who was trained in volumetric analysis (Fig. ). Volume measurements were made at the widest calf diameter in both legs. Measurements were acquired using semiautomated image segmentation software implemented in the Analyze software system (Mayo Clinic Foundation, Biomedical Imaging Resource, Rochester, MN, USA). Tibial and fibular lengths were measured from the articular surfaces of each bone.
Fig. 1 A coronal MR image of both lower legs of a 10-year-old boy with right-sided congenital clubfoot deformity is shown. Cross sections of both limbs were taken at the solid line for volumetric analysis of muscle and fat composition. The solid line indicates (more ...)
Continuous variables are presented as the mean and SD, and categorical variables are presented as the frequency (percentage). We determined differences in volumetric fat and muscle mass and tibia and fibula lengths on MR images between affected and unaffected extremities using paired t test analyses. Statistical analysis was performed using the SPSS® software package (SPSS Inc, Chicago, IL, USA).