Warf Benjamin C, Mugamba John, Kulkarni Abhaya V
Department of Neurosurgery, Children's Hospital Boston, Harvard Medical School, Boston, Massachusetts, USA.
J Neurosurg Pediatr. 2010 Feb;5(2):143-8. doi: 10.3171/2009.9.PEDS09196.
In Uganda, childhood hydrocephalus is common and difficult to treat. In some children, endoscopic third ventriculostomy (ETV) can be successful and avoid dependence on a shunt. This can be especially beneficial in Uganda, because of the high risk of infection and long-term failure associated with shunting. Therefore, the authors developed and validated a model to predict the chances of ETV success, taking into account the unique characteristics of a large sub-Saharan African population.
All children presenting with hydrocephalus at CURE Children's Hospital of Uganda (CCHU) between 2001 and 2007 were offered ETV as first-line treatment and were prospectively followed up. A multivariable logistic regression model was built using ETV success at 6 months as the outcome. The model was derived on 70% of the sample (training set) and validated on the remaining 30% (validation set).
Endoscopic third ventriculostomy was attempted in 1406 patients. Of these, 427 were lost to follow-up prior to 6 months. In the remaining 979 patients, the ETV was aborted in 281 due to poor anatomy/visibility and in 310 the ETV failed during the first 6 months. Therefore, a total of 388 of 979 (39.6% and [55.6% of completed ETVs]) procedures were successful at 6 months. The mean age at ETV was 12.6 months, and 57.8% of cases were postinfectious in origin. The authors' logistic regression model contained the following significant variables: patient age at ETV, cause of hydrocephalus, and whether choroid plexus cauterization was performed. In the training set (676 patients) and validation set (303 patients), the model was able to accurately predict the probability of successful ETV (Hosmer-Lemeshow p value > 0.60 and C statistic > 0.70). The authors developed the simplified CCHU ETV Success Score that can be used in the field to predict the probability of ETV success.
The authors' model will allow clinicians to accurately identify children with a good chance of successful outcome with ETV, taking into account the unique characteristics and circumstances of the Ugandan population.
在乌干达，儿童脑积水很常见且难以治疗。对于一些儿童，内镜下第三脑室造瘘术（ETV）可能成功并避免依赖分流器。这在乌干达可能特别有益，因为与分流相关的感染风险高且长期失败率高。因此，作者开发并验证了一个模型，以预测ETV成功的几率，同时考虑到撒哈拉以南非洲大量人群的独特特征。
2001年至2007年期间在乌干达CURE儿童医院（CCHU）就诊的所有脑积水患儿均接受ETV作为一线治疗，并进行前瞻性随访。以6个月时ETV成功作为结局构建多变量逻辑回归模型。该模型在70%的样本（训练集）上推导得出，并在其余30%（验证集）上进行验证。
1406例患者尝试了内镜下第三脑室造瘘术。其中，427例在6个月前失访。在其余979例患者中，281例因解剖结构不佳/视野不佳而中止ETV，310例在最初6个月内ETV失败。因此，979例手术中有388例（39.6%，占完成ETV手术的55.6%）在6个月时成功。ETV时的平均年龄为12.6个月，57.8%的病例起源于感染后。作者的逻辑回归模型包含以下显著变量：ETV时的患者年龄、脑积水病因以及是否进行了脉络丛烧灼术。在训练集（676例患者）和验证集（303例患者）中，该模型能够准确预测ETV成功的概率（Hosmer-Lemeshow p值>0.60，C统计量>0.70）。作者开发了简化的CCHU ETV成功评分，可在现场用于预测ETV成功的概率。
作者的模型将使临床医生能够准确识别ETV手术成功几率高的儿童，同时考虑到乌干达人群的独特特征和情况。
