Surgical resection is a mainstay of treatment for pediatric low-grade glioma (LGG) within all current therapy algorithms, yet associated morbidity is scarcely reported. As supratentorial midline (SML) interventions are particularly challenging, we investigated the frequency of neurosurgical complications/new impairments aiming to identify their risk factors. Records were retrospectively analyzed from 318 patients with SML-LGG from successive German multicenter LGG studies, undergoing surgery between May 1998 and June 2020. Exactly 537 operations (230 resections, 167 biopsies, 140 nontumor procedures) were performed in 318 patients (54% male, median age: 7.6 years at diagnosis, 9.5 years at operation, 11% NF1, 42.5% optic pathway glioma). Surgical mortality rate was 0.93%. Applying the Drake classification, postoperative surgical morbidity was observed following 254/537 (47.3%) and medical morbidity following 97/537 (18.1%) patients with a 40.1% 30-day persistence rate for newly developed neurological deficits (65/162). Neuroendocrine impairment affected 53/318 patients (16.7%), visual deterioration 34/318 (10.7%). Postsurgical morbidity was associated with patient age <3 years at operation, tumor volume ≥80 cm3, presence of hydrocephalus, complete resection, surgery in centers with less than median reported tumor-related procedures and during the earlier study period between 1998 and 2006, while the neurosurgical approach, tumor location, NF1 status or previous nonsurgical treatment were not. Neurosurgery-associated morbidity was frequent in pediatric patients with SML-LGG undergoing surgery in the German LGG-studies. We identified patient- and institution-associated factors that may increase the risk for complications. We advocate that local multidisciplinary teams consider the planned extent of resection and surgical skills.
Introduction
Neurosurgery is considered the mainstay of treatment for pediatric low-grade glioma (LGG); the extent of resection determines subsequent stratification in current treatment protocols. Yet, surgical radicality must be balanced against the risks of complications that may affect long-term quality of life. We investigated whether this consideration impacted surgical resection patterns over time for patients of the German LGG studies.
Patients and Methods
Four thousand two hundred and seventy pediatric patients from three successive LGG studies (median age at diagnosis 7.6 years, neurofibromatosis (NF1) 14.7%) were grouped into 5 consecutive time intervals (TI1-5) for date of diagnosis and analyzed for timing and extent of first surgery with respect to tumor site, histology, NF1-status, sex, and age.
Results
The fraction of radiological LGG diagnoses increased over time (TI1 12.6%; TI5 21.7%), while the extent of the first neurosurgical intervention (3440/4270) showed a reduced fraction of complete/subtotal and an increase of partial resections from TI1 to TI5. Binary logistic regression analysis for the first intervention within the first year following diagnosis confirmed the temporal trends (p < 0.001) and the link with tumor site for each extent of resection (p < 0.001). Higher age is related to more complete resections in the cerebellum and cerebral hemispheres.
Conclusions
The declining extent of surgical resections over time was unrelated to patient characteristics. It paralleled the evolution of comprehensive treatment algorithms; thus, it may reflect alignment of surgical practice to recommendations in respect to age, tumor site, and NF1-status integrated as such into current treatment guidelines. Further investigations are needed to understand how planning, performance, or tumor characteristics impact achieving surgical goals.
