Radical resection of lumbosacral lipomas in children: the Great Ormond Street Hospital experience.

Abstract

In 2009, Pang described a radical resection technique for congenital lumbosacral lipomas, with lower long-term symptomatic re-tethering rates compared with partial resections and conservative management, and low surgical morbidity. We adopted this technique in 2011, and aim to describe our first results. In this monocentric retrospective audit, we included dorsal, transitional, chaotic, and caudal-type lumbosacral lipomas. Exclusion criteria were previously operated lipomas, pure filar lipomas, and concomitant major congenital anatomical urogenital/gastrointestinal abnormalities. Neuro-uro-orthopaedic status at presentation and at three months, one year and last postoperative follow-up, intraoperative electrophysiology, and extent of resection were collected. From January 2011 to September 2019, 91 patients were operated (median age 2y2m; 63 transitional; 14 caudal; 8 dorsal; 6 chaotic). Preoperatively, 67% were symptomatic. Preoperative and one-year postoperative rates of impaired ambulation (44% to 43%), hypoesthesia (8% to 5%), urodynamic/uroradiological abnormalities (49% to 37%), and foot/ankle deformities (8% to 5%) were comparable, whilst pain improved (25% to 5%) but catheterisation rates increased (21% to 36%). 23/92 (25%) suffered wound-related complications. 2/91 (2%) developed symptomatic re-tethering requiring second surgery. Mean cord/sac ratio was 0.47. 43% had > 20 mm3 residual fat, which improved with increasing surgical experience. Radical lipoma resection, guided by intraoperative neuromonitoring, with reconstruction of the neural placode and expansion duraplasty is technically feasible and results in low rates of late deterioration and re-tethering. Lipoma-type and pre-operative status are important outcome predictors. Operative risks are not insignificant. Future studies need to determine appropriate selection criteria for surgery.