THIRTY-THREE CHILDREN and one adult with sacral agenesis (SA) were studied by computed tomographic myelography and/or magnetic resonance imaging and were monitored for a mean period of 4.7 years. Four children had the OEIS (concurrent omphalocele, cloacal exstrophy, imperforate anus, and spinal deformities) complex, and three others had VATER (vertebral abnormality, anal imperforation, tracheoesophageal fistula, and renal-radial anomalies) syndrome. All patients shared some of the characteristic features of SA, namely, a short, intergluteal cleft, flattened buttocks, narrow hips, distal leg atrophy, and talipes deformities. Neurologically, lumbosacral sensation was much better preserved than the motor functions, and urinary and bowel symptoms were universal. The level of the vertebral aplasia was correlated with the motor but not with the sensory level. The important neuroimaging findings of SA were as follows: 1) 12 patients (35%) had nonstenotic, tapered narrowing of the caudal bony canal, and 2 patients had hyperostosis indenting the caudal thecal sac; 2) 16 patients (47%) had nonstenotic, tapered narrowing and shortening of the dural sac, but 3 patients (9%) had true, symptomatic dural stenosis, in which the cauda equina was severely constricted by a pencil-sized caudal dural sac; 3) the coni could be divided into those ending above the L1 vertebral body (Group 1, 14 patients) and those ending below L1 (Group 2, 20 patients). Thirteen of 14 Group 1 coni were club or wedge-shaped, terminating abruptly at T11 or T12, as if the normal tip was missing. All 20 Group 2 coni were tethered: 13 were tethered by a thick filum; 2 were extremely elongated and had a terminal hydromyelia; 3 were terminal myelocystoceles; and 2 were tethered by a transitional lipoma. High blunt coni were highly correlated with high (severe) sacral malformations (sacrum ending at S1), but low-lying tethered coni were highly correlated with low sacral malformations (S2 or lower pieces present). In conclusion, 1) dural stenosis and tethering lesions are much more common in patients with SA than was previously thought; 2) paradoxically, tethered coni are found more commonly in cases of mild versus severe SA; 3) neurological deterioration is frequently associated with tethered coni and in dural stenosis but not with blunt, short conuses, which may be the “norm” in SA; 4) all tethering lesions should be excised to release the cord, and dural stenosis should be treated by decompressive duraplasty; 5) magnetic resonance imaging screening should be ordered for patients with characteristic features or x-ray evidence of SA, anorectal, and genitourinary anomalies or other complex malformations of the caudal region. It is also postulated that the teratogenic target of SA is the caudal eminence after closure of the caudal neuropore and that various conus anomalies are due to embryological errors occurring at different stages of secondary neurulation.
Department of Neurosurgery, Children's Hospital of Pittsburgh, and University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
Reprint requests: Dachling Pang, MD, FRCS(C), FACS, Department of Neurosurgery, Children's Hospital of Pittsburgh, 3705 Fifth Avenue, Pittsburgh, PA 15213.
Received, July 13, 1992. Accepted, November 24, 1992.