LUMBOPERITONEAL SHUNTING - A RETROSPECTIVE STUDY IN THE PEDIATRIC POPULATION

THERE IS A shortage of data concerning the long-term follow-up of patients with lumboperitoneal (LP) shunts, especially in the pediatric population. A retrospective study of 143 patients who underwent LP shunting between 1974 and 1991 was therefore performed. The mean age at the time of shunt insertion was 3.3 years (range, 18 d to 17.8 yr), and the indication for shunting was: hydrocephalus (81%), cerebrospinal fluid fistula (12%), and pseudotumor cerebri (7%). The mean follow-up time was 5.7 years (range, 5 d to 17.5 yr), and during this period, there were five deaths of which one was shunt related (2.5 yr post-shunt insertion). Of the types of LP shunt used during the study period, the T-tube shunt (101 patients) fared significantly better (P= 0.003) than the percutaneous type (42 patients), and the overall survival characteristics for the T-tube shunt approximated those seen for ventriculoperitoneal shunts, with a 50% probability of remaining free of malfunctions for 5 years. A high rate of migration (19%) was partially responsible for the poor performance of the percutaneous-type shunts. By the end of the study, 40 patients (28%) had been converted to ventricular shunts, and this rate of conversion was similar for both shunt types. LP shunts have certain advantages over other forms of cerebrospinal fluid diversion and were successfully used for various clinical conditions during this study. However, LP shunting is associated with some unique complications, and there was evidence of scoliosis in 14%, back stiffness in 13.7%, back pain in 10%, sciatica in 10%, neurological changes in the lower limbs in 6%, and symptomatic tonsillar herniation in 4.2% of the study patients. As we have recently reported, the incidence of hindbrain herniation may be as high as 70% in asymptomatic patients with LP shunts; we suggest that these unique complications be borne in mind at the time of shunt insertion and at subsequent follow-up.