Jean-Claude Hoeffel

Fibrous tumours in children: imaging features of a heterogeneous group of disorders

Background. Fibrous tumours are predominantly soft tissue lesions which are relatively frequent in childhood but are little known. Imaging is often used in the evaluation of these tumours but their characteristics, particularly on US or MRI, have not been studied systematically. Objectives. To provide an overview of the clinical and imaging features of the different disorders, and to correlate them with the currently used classification schemes. Material and methods. Twenty-five patients with fibrous tumours were evaluated retrospectively. Clinical histories were studied for the histopathological diagnosis, age, signs and symptoms at presentation, mode of therapy and follow-up where available. Imaging findings were analysed for the following variables: number, location, size, margin and architecture of soft tissue and/or visceral lesions and the presence and pattern of osseous involvement. Comparison with the available literature was performed. Results. The following tumour types were encountered: desmoid fibromatosis (n = 9), myofibromatosis (n = 7), fibromatosis colli (n = 2), congenital-infantile fibrosarcoma (n = 2), adult-type fibrosarcoma (n = 2), fibrous hamartoma of infancy (n = 1), angiofibroma (n = 1) and hyaline fibromatosis (n = 1). Conclusions. While some tumours were non-specific in their clinical and radiological manifestation, others such as myofibromatosis, fibromatosis colli, fibrous hamartoma of infancy and angiofibroma exhibited a characteristic pattern which allowed a diagnosis to be made even without histology.

Metastases to bone in medulloblastoma

Medulloblastoma is a relatively common intracranial neoplasm in childhood. Its extraneural spread was, until recently, thought to be a rare occurrence. Metastases are most commonly to bone. Five patients with medulloblastoma metastatic to bone are presented, and findings are compared to those of previous reports. Two of the five cases showed patchy extensive osteosclerosis of the pelvis and/or proximal femora. One case had concurrent lymph node involvement. In patients with past or present medulloblastoma and bone pain, metastases to bone should be excluded. Medulloblastoma metastatic to bone is a rare cause of extensive osteosclerosis.

About one case of vertebral chondroblastoma

The authors report the case of a 9-year-old child which suffered from a chondroblastoma of the cervical spine. He had clinical signs of superior mediastinum compression with inflammation of the respiratory upper air-way. Surgery was performed. Diagnosis was established by the pathologist. The recurrence one year later was treated again by surgery and also by radiotherapy. Another recurrence four years later was fatal. The difficulties of the diagnosis and the appearance of recurrences in some cases of chondroblastoma are reviewed in the literature.

Complications of lumbar puncture with injection of hydrosoluble material.

We report two cases of severe disorders after spinal puncture with injection of hydrosoluble material. The first case concerned a 36-year-old woman with intrathecal injection of 125 ml of hydrocortisone acetate. An intracranial occipital hematoma developed. The second case concerned a 26-year-old man with intrathecal injection of contrast media and hydrocortisone. A chemical meningitis occurred. In both cases the natural course was favorable. Both complications are well known but rare. A review of the literature is made with description of the mechanisms. Preventive therapeutic measures are reported.

Mandibulo-acral dysplasia.

Abstract We report on a 7 year-old-girl with mandibulo-acral dysplasia. When she was 3 years of age it mimicked scleroderma because of skin atrophy and later on a Hutchinson-Gilford progeria syndrome (HGP). Acro-mandibular dysplasia was diagnosed because of facial hypoplasia and mandibular hypoplasia. The bilateral proximal mid-humeral notch seen in this case is unusual.

Anterior synovial recess of the hip: how to assess a pathological condition?

Sir, We write with regard to the recent article by Rohrschneider et al. [1] which reviewed the sonographic appearance of the normal hip joint and established a normal value for the neck-capsule distance (NCD) in healthy children. We agree with the authors concerning the value of the sagittal scan plane along the femoral neck and iliofemoral ligament previously described by Marchal et al. [2]. The maximum distance from the femoral neck to the anterior limit of the capsule (NCD-A), including the synovial reflection in the inferior recess and the width of the capsule, is preferred to the neck to posterior limit of capsule distance (NCD-P) because of poor sonographic differentiation of the dorsal limit of the capsule. In our experience, when using a 7MHz transducer, a sharp capsule outline with bright interfaces is adequately displayed if careful attention is paid to strict orientation of the transducer to produce an ultrasound beam perpendicular to the capsule surface. With the hip in neutral or slight external rotation, accurate, reproducible measurements of NCD-A and NCD-P may be obtained midway between the growth plate of the femoral head and the insertion of the capsule on to the intertrochanteric line [3]. We have performed over 200 examinations in children aged 4 months–14 years with normal or painful hips. In the normal hip (n = 125), the capsule was highly echogenic with a concave orientation, while the synovial membrane was either of low or slightly high echogenicity. The dorsal limit of the capsule was always apparent. The NCD-P remains quite constant, with a mean of 3.1 ± 0.7 (SD) mm (range 1.5– 4.2 mm). There was no correlation with the child’s age. Rohrschneider et al. [1] reported a large capsule thickness range (0.2– 5.5 mm), and our results confirm these data (mean 1.9 mm, range 0.4–4.7 mm). Capsule thickness and age of the child were also correlated (r = 0.70). It is important to note that in addition to the longitudinally oriented fibres of the capsule, there is also a circular group of fibres around the neck (the zona orbicularis) that result in macroscopically evident focal thickening. The capsule is further strengthened by surrounding ligaments such as the iliofemoral ligament. In the painful hip, it was always possible to differentiate between the capsule and the underlying synovial recess. The NCD-P was in all cases very easy to assess. Distinction between hypoechoic synovial proliferation and effusion was not possible in most cases unless a dynamic compression study was performed. Because synovial membrane is the primary site of inflammation, quantification of the amount of synovial proliferation and joint effusion with NCD-P is essential for the diagnosis and follow-up of common pathological conditions of the hip such as transient synovitis, juvenile chronic arthritis and septic arthritis. NCD-P is better suited for the evaluation of these inflammatory changes. The sensitivity is not impaired by the presence of capsular thickening. NCD-A below the limit of 7.5 mm in children above 100 cm in height, as introduced by Rohrschneider et al. [1], may obscure obvious synovitis (Fig. 1). We believe NCD-P is more reliable than NCD-A for the evaluation and followup of the painful hip. A NCD-P greater than 3.5 mm at any age has to be considered abnormal.

Coarctation of the aorta: A call for early detection.

BACKGROUND: Early repair of coarctation of the aorta (COA) is associated with few perioperative complications and better long-term outcome. Therefore, early detection and treatment of COA patients is extremely desirable. The aim of this study was to review our referral system, the effectiveness of neonatal screening examination, and orient physicians about this abnormality, the mode and age of presentation, differences in presentation between cases with isolated COA, and cases associated with other cardiac lesions. PATIENTS AND METHODS: This was a retrospective study of 61 inpatients admitted to our hospital between January 1989 and December 1996, who were found to have COA. Referral data was analyzed and compared to hospital cardiac evaluation findings. RESULTS: All the patients were referred after being symptomatic, but there was no suspicion of COA in any of the cases. One of the patients was referred by a pediatrician from a private clinic, and the rest by hospital pediatricians. Femoral pulse and cardiac murmur were commented upon in only three of the referral letters. Systolic pressure gradient (SPG) between upper and lower limbs and systolic hypertension were not commented on in any of the referral letters. SPG >/=10 mm Hg and systolic hypertension were found in 100% and 58% of the patients, respectively, upon evaluation in our center. SPG in patients with COA associated with other cardiac lesions was significantly lower than in patients with isolated COA (P=0.02). CONCLUSION: Increased awareness in our primary health physicians of the importance of the neonatal screening examination and of measuring blood pressure in the limbs to detect COA early is needed. Timing of the neonatal screening examination between the third day and the third week is recommended. Systolic pressure gradient is a reliable method to detect COA, and in the lower significant range (>/=10 mm Hg) is associated with complex cardiac lesion rather than isolated COA.

Inflammatory pseudotumour of the lung

Sir, I would like to make some comments on the article by Hadimeri [1]. Inflammatory pseudotumour, or plasma cell granuloma, is uncommon, but not rare: Twenty-two cases were reported in 1990 [2], 19 involving the lung. Nine occurred in children: six under 10 years and three between 12 and 14 years. Matsubara [3] divided the clinicopathologic findings in 32 cases of inflammatory pseudotumour of the lung into three groups: organizing pneumonia type (44 %), fibrous histiocytoma type (44%) and lymphoplasmatic type (12 %). Two cases of the fibrous histiocytoma type occurred in children. Endobronchial growth was predominant in 12 % of cases described by Matsubara [3] and in a few other isolated cases in the literature. Ten years ago we saw a case similar to that reported by Hadimeri but with endobronchial growth. A 13-year-old boy had suffered from fever for several weeks. He was pale and had lost weight. Chest radiographs revealed a large well-marginated area of opacity in the right lung. Bronchoscopy showed a tumour obstructing the origin of the right middle lobe bronchus. CT of the chest confirmed the expansile mass in the right hemi~ thorax. Biopsy confirmed an inflammatory pseudotumour. Haemoptysis appeared and, as the pulmonary opacity did not decrease, surgery was performed. Pneumonectomy was necessary but made difficult by cardiac arrest during surgery. The postoperative course was uneventful. The pathologist confirmed the diagnosis of inflammatory pseudotumour with no sign of malignancy. Inflammatory pseudotumours of the lung have the appearance of a mass and are the most common type of lung tumour in childhood. Some are aggressive lesions with bilateral pulmonary involvement and recurrence. Maier [4] reported the case of a 7-year-old girl with a plasma cell granuloma of the lung evolving into a malignant fibrous histiocytoma that recurred after lobectomy in 1952 and required pneumonectomy in 1953, at which time it had extended into the mediastinum at the pulmonary hilus. A metastatic nodule was removed from within the diaphragmatic musculature. Thirty-three years after the most recent operation the patient had no problems.