PAEDIATRIC SPLENECTOMY: THE JOHANNESBURG EXPERIENCE.

BACKGROUND: Splenectomy is uncommon in children and data on splenectomies in the South African paediatric population is sparse. A deeper understanding of the demographics, indications, techniques, and postoperative management of patients requiring splenectomy may improve care.

METHOD: Patient records for all splenectomies performed in children aged 0 to 16 years at Charlotte Maxeke Johannesburg Academic (CMJAH) and Chris Hani Baragwanath Academic Hospitals (CHBAH) between 2000 and 2015 were reviewed. Students' t- and Chi-squared tests were utilised for analytic statistics.

RESULTS: Mean age at surgery was 9.9 years with a range of 3 to 16 years. Overall, 91% of splenectomies were performed for haematological disorders with the remaining 9% performed for malignancy or trauma. Two-thirds (67%) of splenectomies were performed open and 33% laparoscopically. Mean postoperative length of stay (LOS) was significantly shorter in the laparoscopic cohort compared to the open cohort (4.5 vs 7.1 days, respectively; p < 0.05). Surgical complications were more common in the laparoscopic group (27%) compared to the open group (9%), as well as in those children older than the mean age at time of surgery. No partial splenectomies were performed. No cases of overwhelming post-splenectomy infection (OPSI) were recorded. In total, 20 patients (61%) were still alive, 3 had died, and 10 had been lost to follow up at study completion.

CONCLUSION: Local indications for splenectomy mirror those within international literature. Higher rates of postoperative complication in the open and laparoscopic groups in our series may be due to a learning curve in laparoscopic splenectomy, the low sample size, or the underlying state in which our patients present. Mean postoperative LOS was shorter in the laparoscopic than open group, but relatively longer for both groups than reported in the international literature. Laparoscopy is, as yet, not the preferred technique for splenectomy in our setting. All mortalities were due to progression of underlying disease and no cases of OPSI were recorded. The high rate of loss to follow up in this study is a significant barrier to accurate data collection and reporting.