Integrating Overall Survival and Tumor Control Probability Models to Predict Local Progression After Brain Metastasis Radiosurgery.

Purpose: Stereotactic radiosurgery (SRS) for brain metastases is frequently prescribed to the maximum tolerated dose to minimize the probability of local progression. However, many patients die from extracranial disease prior to local progression and may not require maximally aggressive treatment. Recently, improvements in models of SRS tumor control probability (TCP) and overall survival (OS) have been made. We predicted that by combining models of OS and TCP, we could better predict the true risk of local progression after SRS than by using TCP modeling alone. Methods and Materials: Records of patients undergoing SRS at a single institution were reviewed retrospectively. Using established TCP and OS models, for each patient, the probability of 1-year survival [p(OS)] was calculated, as was the probability of 1-year local progression [p(LP)]) for each treated lesion. Joint-probability was used to combine the models [p(LP,OS)=p(LP)*p(OS)]. Analyses were conducted at the individual metastasis and whole-patient levels. Fine-Gray regression was used to model p(LP) or p(LP,OS) on the risk of local progression after SRS, with death as a competing risk. Results: At the patient level, 1-year local progression was 0.08 (95% CI, 0.03-0.15), median p(LP,OS) was 0.13 (95% CI, 0.07-0.2), and median p(LP) was 0.29 (95% CI, 0.22-0.38). At the metastasis level, 1-year local progression was 0.02 (95% CI, 0.01-0.04), median p(LP,OS) was 0.05 (95% CI, 0.02-0.07), and median p(LP) was 0.10 (95% CI, 0.07-0.13). p(LP,OS) was found to be significantly associated with the risk of local progression at the patient level (P = .048) and metastasis level (P = .007); however, p(LP) was not (P = .16 and P = .28, respectively). Conclusions: Simultaneous modeling of OS and TCP more accurately predicted local progression than TCP modeling alone. Better understanding which patients with brain metastases are at risk of local progression after SRS may help personalize treatment to minimize risk without sacrificing efficacy.

Ventriculopleural shunts in a pediatric population: a review of 170 consecutive patients.

OBJECTIVE: The authors sought to determine the outcome of using the pleural space as the terminus for ventricular CSF-diverting shunts in a pediatric population. METHODS: All ventriculopleural (VPl) shunt insertions or revisions done between 1978 and 2018 in patients at Children's Hospital Los Angeles were identified. Data recorded for analysis were age, sex, weight, etiology of hydrocephalus, previous shunt history, reason for VPl shunt insertion or conversion from a ventriculoperitoneal (VP) or ventriculoatrial (VA) shunt, valve type, nature of malfunction, presence of shunt infection or pleural effusion, and conversion to a different distal site. RESULTS: A total of 170 patients (mean age 14 ± 4 years) with a VPl shunt who were followed up for a mean of 57 ± 53 months were identified. The reasons for conversion to a VPl shunt for 167 patients were previous shunt infection in 57 (34%), multiple abdominal procedures in 44 (26%), inadequate absorption of CSF in 34 (20%), abdominal pseudocyst in 25 (15%), and obesity in 7 (4%). No VPl revisions were required in 97 (57%) patients. Of the 73 (43%) patients who did require revision, the most common reason was proximal obstruction in 32 (44%). The next most frequent complication was pleural effusion in 22 (30%) and included 3 patients with shunt infection. All 22 patients with a clinically significant pleural effusion required changing the distal end of the shunt from the pleural space. Pleural effusion was more likely to occur in VPl shunts without an antisiphon valve. Of the 29 children < 10 years old, 7 (24%) developed a pleural effusion requiring a revision of the distal catheter to outside the pleural space compared with 15 (11%) who were older (p = 0.049). There were 14 shunt infections with a rate of 4.2% per procedure and 8.2% per patient. CONCLUSIONS: VPl shunts in children younger than 10 years of age have a significantly higher rate of symptomatic pleural effusion, requiring revision of the shunt's terminus to a different location. VPl shunt complication rates are similar to those of VP shunts. The technical difficulty of inserting a VPl shunt is comparable to that of a VP shunt. In a patient older than 10 years, all else being equal, the authors recommend that the distal end of a shunt be placed into the pleural space rather than the right atrium if the peritoneal cavity is not suitable.

Usefulness of postoperative ventriculography and intracranial pressure monitoring following endoscopic third ventriculostomy.

PURPOSE: The authors sought to determine whether the insertion of an external ventricular drain (EVD) at the time of surgery to monitor intracranial pressure (ICP) and ventriculography done within the first day following an endoscopic third ventriculostomy (ETV) is of benefit in postoperative patient management. METHODS: Following IRB approval, ETV procedures done by the senior author between January 1, 2007, and December 31, 2016, were reviewed. Included in a consecutive fashion were all patients who underwent an ETV with placement of an external ventricular drain (EVD) that was preceded preoperatively by an MRI or CT study and followed by a contrast CT ventriculogram within the first postoperative day. RESULTS: Identified were 72 patients who met the above criteria; however, technical ventriculography failure occurred in 4 (6%) and were eliminated from the analysis. Of the remaining 68 patients, contrasted CSF was seen in the basal cisterns/subarachnoid spaces (SAS) in 66 (97%) indicating a patent ETV and absent in 2 (3%) indicating a non-patent ETV. Of the 66 patients with a patent ETV, 34 (52%) patients were discharged on postoperative days 1 (8), 2 (13), and 3 (13) as their ICPs were not elevated and their clinical symptoms normal. EVDs placed at the time of the ETV recorded raised ICP > 20 cm H2O in 17/68 (25%) patients for 1 or more days, all of whom had a patent ETV. Because of persistently elevated ICP requiring CSF drainage for control, 4 of these patients were shunted on postoperative days 5, 6, 6, and 10 and 3 with prolonged elevated ICP for 5, 6, and 11 days postoperatively were not shunted as their ICP and symptoms progressively normalized. The remaining 17/68 (25%) patients did not have a postoperative ICP > 20 cm H2O; 14 were discharged after resolution of symptoms and other clinical factors, 1 was shunted on postoperative day 3 due to persistent symptoms and a consistently large volume of CSF drainage, and 2 had a non-patent ETV with 1 undergoing shunt placement and the other discharged because of the absence of symptoms. The sensitivity of ventriculography was only 13%; however, the specificity was 98% and the accuracy 88%. CONCLUSION: After reviewing this series, the continued use of a postoperative EVD appears appropriate as the risk is low and it provides ventricular access to control ICP, thereby, improving patients' safety and reducing the need for CSF shunting on an urgent/emergent basis should the ETV prove to be unsuccessful. From our limited series, the usefulness of a 1-day postoperative ETV ventriculogram is less clear and would need confirmation with additional studies.