Abstract：
OBJECTIVE: In patients requiring prophylactic cranial irradiation (PCI) or whole-brain radiotherapy (WBRT) for brain metastases (BMs), hippocampal avoidance (HA) has been shown to preserve neurocognitive function and quality of life. Here, we aim to estimate the incidence of hippocampal and perihippocampal BMs and the subsequent risk of local undertreatment in patients undergoing hippocampal sparing radiotherapy.
METHODS: MEDLINE, Embase, and Scopus were searched with the terms \"Hippocampus\", \"Brain Neoplasms\", and related terms. Trials reporting on the incidence of hippocampal and/or perihippocampal BMs or hippocampal failure rate after PCI or WBRT were included.
RESULTS: Forty records were included, encompassing a total of 5,374 patients with over 32,570 BMs. Most trials employed a 5 mm margin to define the HA zone. In trials reporting on BM incidence, 4.4 % (range 0 - 27 %) and 9.2 % (3 - 41 %) of patients had hippocampal and perihippocampal BMs, respectively. The most common risk factor for hippocampal BMs was the total number of BMs. The reported failure rate within the HA zone after HA-PCI or HA-WBRT was 4.5 % (0 - 13 %), salvageable with radiosurgery in most cases. SCLC histology was not associated with a higher risk of hippocampal failure (OR = 2.49; p = 0.23). In trials comparing with a conventional (non-HA) PCI or WBRT group, HA did not increase the hippocampal failure rate (OR = 1.90; p = 0.17).
CONCLUSIONS: The overall incidence of hippocampal and perihippocampal BMs is considerably low, with a subsequent low risk of local undertreatment following HA-PCI or HA-WBRT. In patients without involvement, the hippocampus should be spared to preserve neurocognitive function and quality of life.
METHODS: MEDLINE, Embase, and Scopus were searched with the terms \"Hippocampus\", \"Brain Neoplasms\", and related terms. Trials reporting on the incidence of hippocampal and/or perihippocampal BMs or hippocampal failure rate after PCI or WBRT were included.
RESULTS: Forty records were included, encompassing a total of 5,374 patients with over 32,570 BMs. Most trials employed a 5 mm margin to define the HA zone. In trials reporting on BM incidence, 4.4 % (range 0 - 27 %) and 9.2 % (3 - 41 %) of patients had hippocampal and perihippocampal BMs, respectively. The most common risk factor for hippocampal BMs was the total number of BMs. The reported failure rate within the HA zone after HA-PCI or HA-WBRT was 4.5 % (0 - 13 %), salvageable with radiosurgery in most cases. SCLC histology was not associated with a higher risk of hippocampal failure (OR = 2.49; p = 0.23). In trials comparing with a conventional (non-HA) PCI or WBRT group, HA did not increase the hippocampal failure rate (OR = 1.90; p = 0.17).
CONCLUSIONS: The overall incidence of hippocampal and perihippocampal BMs is considerably low, with a subsequent low risk of local undertreatment following HA-PCI or HA-WBRT. In patients without involvement, the hippocampus should be spared to preserve neurocognitive function and quality of life.
摘要：
目的：对脑转移瘤（BMs）需要预防性头颅放疗（PCI）或全脑放疗（WBRT）的患者，海马回避（HA）已被证明可以保护神经认知功能和生活质量。这里,我们的目的是评估海马和海马周围BMs的发生率以及随后在接受海马保留放疗的患者中局部治疗不足的风险.
方法：MEDLINE，Embase,和Scopus用“海马”“脑肿瘤”,和相关术语。包括关于PCI或WBRT后海马和/或海马周围BMs发生率或海马失败率的试验报告。
结果：包括40条记录，涵盖总共5,374名患者，超过32,570名BMs。大多数试验采用5毫米的边缘来定义HA区。在报告BM发病率的试验中，4.4%(范围0-27%)和9.2%(3-41%)的患者有海马和海马周围的BMs，分别。海马BMs最常见的危险因素是BMs的总数。HA-PCI或HA-WBRT后HA区内报告的失败率为4.5％（0-13％），在大多数情况下，放射外科是可以挽救的。SCLC组织学与海马衰竭的高风险无关（OR=2.49；p=0.23）。在与常规（非HA）PCI或WBRT组比较的试验中，HA并不增加海马失败率（OR=1.90；p=0.17）。
结论：海马和海马周围BMs的总体发生率相当低，随后HA-PCI或HA-WBRT后局部治疗不足的风险较低。在没有参与的患者中，海马体应幸免，以保持神经认知功能和生活质量。
方法：MEDLINE，Embase,和Scopus用“海马”“脑肿瘤”,和相关术语。包括关于PCI或WBRT后海马和/或海马周围BMs发生率或海马失败率的试验报告。
结果：包括40条记录，涵盖总共5,374名患者，超过32,570名BMs。大多数试验采用5毫米的边缘来定义HA区。在报告BM发病率的试验中，4.4%(范围0-27%)和9.2%(3-41%)的患者有海马和海马周围的BMs，分别。海马BMs最常见的危险因素是BMs的总数。HA-PCI或HA-WBRT后HA区内报告的失败率为4.5％（0-13％），在大多数情况下，放射外科是可以挽救的。SCLC组织学与海马衰竭的高风险无关（OR=2.49；p=0.23）。在与常规（非HA）PCI或WBRT组比较的试验中，HA并不增加海马失败率（OR=1.90；p=0.17）。
结论：海马和海马周围BMs的总体发生率相当低，随后HA-PCI或HA-WBRT后局部治疗不足的风险较低。在没有参与的患者中，海马体应幸免，以保持神经认知功能和生活质量。
