Spontaneous neonatal gastric perforation.

Over a 7-year period (1990-1997) spontaneous gastric perforation was diagnosed in five neonates. The mean gestational age and birth weight were 33/40 weeks and 1.83 kg, respectively. All patients presented with severe abdominal distention and frank pneumoperitoneum on roentgenograms. All perforations were on the anterior wall of the greater curvature and were managed by prompt laparotomy and primary closure of the perforation. No gastrostomy was used, however, peritoneal drainage was used in all cases. There was no mortality or morbidity. An attempt to understand the possible etiology together with a review of the literature is presented.

Ambiguous genitalia: comparative role of pelvic ultrasonography and genitography.

Ambiguous genitalia represents a true medical and social emergency which needs a multi-disciplinary team approach for elucidation. The paediatric radiologist plays an important role in defining the genital anatomy which remains one of the most important factors in sex determination. Aiming to compare the predictive value of pelvic ultrasonography and genitography in sex determination in patients with ambiguous genitalia, we retrospectively reviewed the medical records of 69 patients, 53 females and 16 males, where both procedures were employed. In female pseudohermaphroditism, the presence of a uterus with or without vagina was predicted in 46 (86.8%) patients by ultrasound compared with 44 (83%) patients in whom a genitogram revealed a vagina +/- uterus. In six (11.3%) patients, a genitogram revealed a male-type urethra. The combination of ultrasound and genitogram, however, was more sensitive and predicted the presence of a uterus with or without vagina in 52 (98.1%) patients. In male pseudo-hermaphroditism, there was no false positive by ultrasound, and a genitogram revealed a male-type urethra in 12 (75%) patients. In conclusion, although real time pelvic ultrasonography is less invasive than genitography, its yield in elucidating genital anatomy is comparable. The combination of both procedures is more informative and has a better yield.

Eventration of the diaphragm in children.

Diaphragmatic plication is a safe, well tolerated procedure for congenital diaphragmatic eventration. It is the treatment of choice in neonates with respiratory distress following phrenic nerve injury. The abdominal approach is as good as the thoracic and may be safer in cases with central eventration and when the diagnosis of congenital diaphragmatic hernia cannot be excluded with confidence. However, in phrenic nerve paralysis the thoracic approach is superior in showing the anatomical distribution of the phrenic nerve. Experience with eight paediatric patients treated for diaphragmatic eventration is reported. Five had congenital eventration of the diaphragm and three had acquired eventration. All patients except one required surgical plication of the diaphragm. Three patients developed minor postoperative complications and all patients were alive and well on follow-up. It appears that diaphragmatic plication is a safe and well tolerated procedure for symptomatic diaphragmatic eventration.

Horseshoe lung: differential diagnosis.

A detailed radiologic and anatomic study of one horseshoe lung syndrome has already been published from this centre. A further case of horseshoe lung, which was also diagnosed prospectively by radiology, is described. Alongside, this is a set of a group of three conditions: we describe a right accessory lung with a peculiar feeding arterial branch originating from the left basal pulmonary artery. The bronchogram and oesophagogram of this case were normal. The other is a case of bridging bronchus where the left lower lobe bronchus originates from the right main stem bronchus. This was discovered at post mortem bronchogram performed on a 19 weeks foetus. Finally, in the dog, we found that the pulmonary angiogram and bronchogram display close similarity to human horseshoe lung though the dog has two separate lungs. The common denominator of all above cases is the presence of a pulmonary arterial branch or bronchus crossing the midline from the ipsi to the contralateral side. Therefore, it may be concluded "in contradiction to the common belief" that pulmonary angiography or bronchography alone is not sufficient for the diagnosis of horseshoe lung. When only one of these investigations is available, computerized tomography is necessary to show the isthmic lung tissue before the diagnosis of horseshoe lung is confirmed.