LAUSR

To the Editor: We thank Dr Barami for his insightful letter.1 We have read with interest the referenced literature regarding the role of the cerebral venous system in governing intracranial pressure and the theory of cerebral venous overdrainage (CVO).2,3 The concept of cerebrospinal fluid (CSF) and cerebral venous compartment coupling and its regulation by the Starling resistor was of particular interest to us as we did not include it in our publication.4 This model offers yet another explanation for the development of low-pressure hydrocephalus (LPH) following lumbar puncture (LP).5,6 Dr Birani states that CSF diversion causes CSF hypotension in the subarachnoid space, leading to the loss of the Starling effect on draining cortical veins. This loss of venous outflow regulation, magnified by the upright position, allows for unregulated siphoning of cortical venous blood as the lateral lacunae no longer function as resistors to venous drainage. The ventriculomegaly and periventricular edema that defines LPH is due to venous hypotension in the superior sagittal sinus, deep cerebral veins, and periventricular veins. This hypotension, in turn, creates a pressure gradient between the ventricle and surrounding parenchyma that leads to expansion of the ventricles and centrifugal flow of CSF flow from ventricle to parenchyma. Dysfunction of the Starling resistor leading to LPH is supported by Rekate et al7 as well. He described the role of the cortical subarachnoid space in various models of hydrocephalus and did include a description of a resistor between the cortical subarachnoid space and the superior sagittal sinus; however, it was not named a “Starling resistor.” It is clear that restoring the normal hierarchy of pressures within the intracranial compartment in LPH—as is dictated by the CVO theory—is of utmost importance. This can be achieved by increasing venous pressures (eg, enforced recumbency, cervical neck wrapping), or by reversing the ventricle pressure gradient (eg, aggressive CSF diversion, endoscopic third ventriculostomy in appropriate patients). While derangements leading to LPH following LP are easier to conceptualize (and for which CVO does seem to offer a satisfying explanation), there are questions that cannot be answered by this model. If CSF diversion by itself causes CSF hypotension in the subarachnoid space leading to the loss of the Starling effect on draining cortical veins, then why do we not see LPH develop spontaneously in patients with chronic CSF diversion by way of ventricular or subdural shunts? Or for that matter, in any person without a shunt that undergoes an LP? While CVO may certainly be valid, there are likely other factors involved. If this were simply a by-product of overdrainage, it seems the phenomenon would not be as rare as current data suggest. Why do only a very small fraction of children with an existing shunt who undergo an LP develop LPH? What about those children in our series who developed LPH spontaneously or semispontaneously, such as after a shunt revision, intracranial infection, or hemorrhage? The CVO theory is but one of many theories that have been postulated, but are as of yet unproven. It is clear that there is still much to learn in regard to the pathogenesis of this rare and fascinating phenomenon, and it is possible that there may not be 1 unifying mechanism. However, while the underlying pathophysiology of LPH is not known, treatment options are well established as detailed in our manuscript.4 In summary, cerebral venous outflow is an important concept when it comes to understanding the pathophysiology and treatment modalities of LPH. As venous blood comprises one of the changeable compartments of the cranial vault per theMonroeKellie doctrine, it is certainly implicated in perturbations leading to LPH and understanding its dynamics satisfactorily explains why current treatment modalities are effective in treating this condition.