Dear Editor, We read with great interest the article by Clemmensen et al. concerning infection after ventriculoperitoneal (VP) shunt placement in newborns with myelomeningocele (MMC). The authors described a higher infection rate in newborns with late VP shunt placement compared to early shunt surgery [1]. Although approximately 90% of patients with MMC eventually develop symptomatic hydrocephalus needing shunt placement, only 15% demonstrates active hydrocephalus at the time of MMC repair [2]. Despite the high rate of shunt-dependent hydrocephalus, many neurosurgeons are reluctant to place a shunt in patients with asymptomatic hydrocephalus at birth and try to manage them conservatively, given the high rate of shunt infection in patients with MMC relative to other etiologies [4]. Considering the higher risk of shunt infection in children younger than 6 months, the late presentation of hydrocephalus provides time for the child's immune system to develop before hydrocephalus reaches a surgically significant threshold. The fact that the authors only studied the patients who needed a VP shunt within their first 2 weeks of life indicates the presence of active progressive hydrocephalus in all patients where early shunt placement prevented the deleterious effects of increasing CSF pressure on the wound healing in that group. Unfortunately, the need for shunt placement at the time of MMC closure in an asymptomatic patient may not always be obvious; however, factors including the size and level of MMC may help in predicting the imminent need for shunt surgery. In some patients, an asymptomatic hydrocephalus progresses rapidly after the closure of MMC, needing VP shunt placement before complete wound healing, leading to the serious complications of cerebrospinal fluid (CSF) leakage and infection. CSF leakage from the MMC wound, which is one well-recognized independent risk factor for infection [3], is a potential drawback of this strategy. The authors observed high rates of infection (41%) and CSF leakage (43%), which, as expected, had a strong positive correlation [1]. They used univariate analysis to compare the two groups. To find real, significant factors and considering the role of CSF leakage as a confounding variable, we propose multivariate analysis with a logistical regression. In spite of a higher rate of infection in late-shunted neonates in the Clemmensen et al. study, there was no statistically significant difference of shunt infection in patients shunted early or late after MMC closure in our study of MMC children of any age [4]. Colonization of the fresh MMC sac with microorganisms throughout the days F. Radmanesh : F. Nejat (*) Department of Neurosurgery, Children’s Hospital Medical Center, Tehran University of Medical Science, Tehran, Iran e-mail: nejat@sina.tums.ac.ir