We have identified several independent predictors for shunt-dependent hydrocephalus in a consecutive cohort of patients with SAH. Patients with age > 55 years, presence of intraventricular hemorrhage, a Fisher grade ≥ 3, a H&H ≥ 3, a history of acute hydrocephalus, a history of CNSI, a history of EVD placement and a history of postoperative LP had higher odds of developing shunt-dependent hydrocephalus. Many of these risk factors have been found in prior studies[5, 6, 13, 14]. Our 14.3% rate of shunt dependence is consistent with past reports[8, 14]. Our study showed that elderly patients had a higher incidence of shunt dependency than younger patients. The susceptibility of elderly patients to SDHC may be related to the mechanism; with age, the enlargement of the ventricle and subarachnoid space caused by brain atrophy allows the subarachnoid space to accommodate more blood, thus promoting meningeal inflammation and fibrosis. CSF absorption decreased and CSF circulation in the ventricular system slowed[15]. However, these factors, including angiographic vasospasm, ICH, GCS score, WFNS grade, BNI score, surgical clipping and posterior circulation aneurysm, were found to be significant in the univariate analysis, but not in the multivariate analysis. Several reports also documented the other risk factors for SDHC, including hypertension, smoking and alcohol abuse[12, 16]. In our study, these factors were more frequently present in the SDHC group, but the differences were not significant. Fisher grade and BNI score reflect the volume of hemorrhage in the ventricle and subarachnoid space. It was reported that a higher Fisher grade is positively correlated with SDHC[8]. Likewise, our study found this correlation. Treatment choice, surgical clipping or endovascular coiling, was determined according to multidisciplinary discussions involving both surgical and endovascular teams. Surgical clipping was the preferred modality of treatment in our study. Two-thirds of aSAH patients were treated using surgical clipping. Furthermore, our study findings can be consist with those from previous studies that have shown no significant differences in the incidences of SDHC between surgical clipping and endovascular treatment[6, 8]. However, A large registration study involving 1448 patients, which showed that coil embolization, when compared with clipping, is associated with significantly lower risks for development[7].
Acute hydrocephalus is a cause of early brain injury following subarachnoid hemorrhage and is often considered to be a form of noncommunicating or obstructive hydrocephalus due to the presence of intracisternal or intraventricular blood products limiting the outflow of CSF from the cranial vault[17]. The pathophysiological mechanism of chronic hydrocephalus after aSAH is not very clear. However, a generally accepted hypothesis suggests that it results from the impairment of CSF reabsorption and circulation caused by arachnoid granule adhesions and ventricular system obstruction[17–19]. There is growing evidence showing that SDHC after aSAH is a multifactorial process that involves the up-regulation of pro-inflammatory cytokines and fibroblasts, increased production of collagen, toxic effects of iron, and an increased secretion of serotonin, catecholamine, angiotensin and other hormones[20–23]. Kanat et al. suggest that both arachnoid granule inflammation and fibrosis lead to CSF reabsorption disorder rather than an increased CSF in secretion, which results in a chronic communicating hydrocephalus[24]. Tan Q, et al. believe that the dysfunction of CSF reabsorption caused by fibrosis and arachnoid granules adhesions is caused by a local inflammatory reaction after the subarachnoid hemorrhage[25]. CNSI and Requirement for EVD were independent risk factors for shunt dependency in our series, and this association has also been found in other series[10, 22]. CNSI is a risk factor that could, in principle, be prevented with comprehensive management and therefore should be the focus of prevention efforts to minimize the patient’s risk for shunting. CSF diversion is generally considered as the main management approach for symptomatic acute hydrocephalus and patients with altered level of consciousness after aSAH. Drainage is required to aid CSF flow dynamics, thought to be affected by blockage caused by blood product obstructing arachnoid granulations and the ventricular and cisternal drainage pathways[22].
With that in consideration, we performed regular LPs to release hemorrhagic CSF in some patients after surgery, avoiding a large number of blood cells in the ventricular circulatory system metabolic decomposition, and then promote early recovery of the physiological state of CSF. But in fact our research shows that the incidence of SDHC in patients who underwent an LP was significantly higher than that in patients who did not undergo an LP. The possible mechanism are as follows: (1) long-term and regular LPs make the body dependent on it, and the absorption and compensation function of CSF decreases gradually; (2) excessive release of CSF reduces the pressure gradient on both sides of the arachnoid granules and slows down CSF absorption; (3) excessive CSF drainage causes the subarachnoid to collapse, which leads to the poor drainage of inflammatory factors, adhesion of subarachnoid space, and poor CSF circulation; (4) excessive drainage of CSF reflexively leads to the increased secretion of it; and (5) iatrogenic injury caused by multiple LPs leads to local coagulation, inflammation, and thickening of the dura mater or arachnoid membrane. In conclusion, the occurrence of SDHC after aSAH may be closely related to the CSF drainage volume and time during an LP; however, this finding requires further study.
Our study has several limitations: (1) all cases were from a single center, indicating that our findings were influenced by hospital and physician bias during the patient selection and treatment process; (2) selection bias exists, as patients were not randomly assigned to the LP or No LP group in our retrospective cohort study. Although propensity score matching of baseline characteristics minimized differences between LP and No LP cohorts, other unmeasured, latent variables and their interactions may have contributed to selection bias; and (3) to evaluate LP more precisely as a risk factor for the development of chronic hydrocephalus after SAH, additional analysis for the volume and time of LP is needed.