Three CSF Drainage Strategies Show Similar Clot Clearance in Brain Haemorrhage Study

Patients with bleeding inside the brain’s ventricles complicated by hydrocephalus often need urgent cerebrospinal fluid (CSF) diversion to survive, yet the best way to drain the fluid has never been settled. The study, published in the Egyptian Journal of Neurosurgery, also recorded a strikingly high catheter-related infection rate in the standard external ventricular drainage group, though the difference fell short of statistical significance.

Intraventricular haemorrhage (IVH) occurs either as a primary event, when vessels lining or near the ventricles bleed, or secondary to extension of an intracerebral haemorrhage, aneurysmal subarachnoid haemorrhage, or ruptured arteriovenous malformation. Acute obstructive hydrocephalus develops in up to half of affected patients, and mortality in severe cases has historically exceeded 50 percent. Timely CSF diversion is central to management, but the optimal technique remains debated. External ventricular drainage (EVD), the most widely used first-line method, allows continuous fluid egress and intracranial pressure monitoring but carries reported catheter-related infection rates of 8 to 22 percent and a substantial rate of shunt dependency after removal.

Ommaya reservoir-based strategies have been proposed as alternatives that may reduce the time an externalised, open system stays in contact with the environment. One option is intermittent needle aspiration, or tapping, of a reservoir implanted under the scalp; the other is connecting the reservoir to a closed free-drainage system. Both permit graded CSF removal and, where needed, intraventricular fibrinolytic instillation. Comparative clinical data across the three approaches have been sparse, and no consensus exists on which best balances drainage efficiency, clot clearance, infection risk, and downstream shunt dependency.

The research team at the Department of Neurosurgery, SKIMS Soura, enrolled 60 consecutive adult patients with radiologically confirmed IVH and acute hydrocephalus requiring CSF diversion. Allocation was not randomised: patients were assigned according to institutional protocol and treating-surgeon preference to Group A, standard lateral ventricular EVD; Group B, an Ommaya reservoir with repetitive percutaneous CSF taps; or Group C, an Ommaya reservoir connected to free continuous drainage into a closed external bag via a 22G butterfly cannula secured with sutures and a semi-occlusive dressing inspected daily. Twenty patients were allocated to each group. The study followed the STROBE reporting guideline and received institutional ethics approval with written informed consent from patients or their legal representatives.

The groups were well matched at baseline for age, admission Glasgow Coma Scale score, haemorrhage volume, blood pressure, and radiological severity scores including the Graeb, modified Graeb, and LeRoux scores and the Evans and bicaudate indices. One imbalance did emerge: the distribution of IVH type differed significantly, with 100 percent secondary IVH in Group A, 70 percent secondary and 30 percent primary in Group B, and 95 percent secondary and 5 percent primary in Group C, a difference the authors attribute to real-world case-mix variation and treating-team decision-making rather than designed allocation. Overall, 40 of the 60 patients were male and the mean age was 56.9 years.

The most consistent finding was the difference in drainage volume. Mean daily CSF output was 184.0 millilitres per day in the EVD group and 171.2 millilitres per day in the free-drainage reservoir group, compared with only 95.6 millilitres per day with intermittent tapping, a difference that was highly significant. Total CSF drained followed the same pattern, at roughly 1,897 millilitres, 1,648 millilitres, and 923 millilitres respectively. Yet the duration of catheter drainage was nearly identical across groups, at about 9.7 to 10.3 days, and radiological clot clearance did not differ. Modified Graeb scores declined progressively from admission to postoperative day 7 in all three groups, and the clot clearance rate at day 7 was 60 percent in the EVD group versus 56 percent in both reservoir groups, a non-significant difference. Time to complete clot resolution was also similar, at roughly 12 days.

Catheter-related infection or ventriculitis, defined using 2017 Infectious Diseases Society of America criteria for healthcare-associated ventriculitis, occurred in 9 of 20 patients (45 percent) in the EVD group, compared with 4 of 20 (20 percent) in each reservoir group. This difference did not reach statistical significance. Postoperative day 7 CSF culture positivity followed a similar numerical pattern, and one patient in the tapping group met clinical and cytochemical criteria for ventriculitis despite a negative culture. Notably, two of the sixteen culture-positive patients grew Candida species rather than bacteria, one in the EVD group and one in the free-drainage group. Both developed progressive fungal ventriculitis despite catheter removal and systemic antifungal therapy and subsequently died. The authors report this observation at the level of individual cases, emphasising the importance of sending fungal cultures in IVH patients with persistently culture-positive CSF who fail to improve on antibacterial therapy.

Permanent ventriculoperitoneal shunt dependency occurred in 45 percent of the EVD group, 45 percent of the tapping group, and 25 percent of the free-drainage group, again without statistical significance. Ninety-day modified Rankin Scale and Glasgow Outcome Scale scores, mortality of roughly 35 to 40 percent, ICU and hospital stay, and discharge neurological status were all statistically similar across the three strategies. When 90-day outcomes were dichotomised into favourable and unfavourable, 55 percent of the EVD group, 50 percent of the tapping group, and 40 percent of the free-drainage group achieved a favourable outcome, a non-significant difference. Rates of catheter blockage, dislodgement, tract haemorrhage, CSF leak, overdrainage, and revision were low in all groups.

The authors are careful to frame these results as hypothesis-generating rather than confirmatory. No formal sample-size calculation was performed, the 20 patients per group were enrolled pragmatically, and p-values were not adjusted for multiple comparisons. With only 20 patients per arm, the study is substantially underpowered to detect or rule out clinically meaningful differences in secondary outcomes such as infection or shunt dependency, and the authors explicitly state that the numerical trends should not be read as evidence of superiority of reservoir-based drainage over EVD. The 45 percent infection rate in the EVD group is also high relative to previously reported ranges of 8 to 22 percent, and the authors caution that centre- or period-specific factors in catheter insertion and maintenance practice may have contributed. The non-randomised design, the imbalance in IVH type, the single-centre setting, the absence of outcomes beyond 90 days, and the lack of systematic intraventricular fibrinolytic therapy further limit generalisability.

What the study does establish is that intermittent Ommaya tapping, which removed roughly half the daily and total CSF volume of the continuous strategies, did not translate into worse radiological clearance, longer catheter dwell time, or worse functional outcomes. The authors suggest that the total volume of fluid removed per day may matter less than achieving adequate, sustained ventricular decompression, and that closed, reservoir-based systems can achieve clearance comparable to an open EVD circuit. They also note that intermittent tapping may reduce nursing workload and avoid continuous connection to an external collection system, and could theoretically lower overdrainage risk, though this advantage was not clearly demonstrated. The findings sit alongside trials of intraventricular fibrinolysis such as CLEAR III, which showed that enhancing clot clearance through an EVD can improve functional outcome, underscoring that the drainage conduit is only one component of a broader management strategy. The team calls for an adequately powered randomised controlled trial to determine whether Ommaya reservoir-based drainage can match the efficacy of standard EVD while reducing catheter-related morbidity in IVH-associated hydrocephalus.

Subject of Research: Comparative outcomes of external ventricular drainage and Ommaya reservoir-based cerebrospinal fluid diversion strategies in intraventricular haemorrhage with hydrocephalus

Article Title: Comparative outcomes of three cerebrospinal fluid drainage strategies in primary and secondary intraventricular haemorrhage with hydrocephalus: a prospective cohort study

Article References: Khan, M. A., Hela, A. H., Arif, S. H., & Chhiber, S. S. (2026). Comparative outcomes of three cerebrospinal fluid drainage strategies in primary and secondary intraventricular haemorrhage with hydrocephalus: a prospective cohort study. Egyptian Journal of Neurosurgery, 41(1), Article 166. https://doi.org/10.1186/s41984-026-00675-y

Image Credits: AI Generated

DOI: 10.1186/s41984-026-00675-y

Keywords: intraventricular haemorrhage, hydrocephalus, cerebrospinal fluid drainage, external ventricular drain, Ommaya reservoir, ventriculitis, clot clearance, shunt dependency, neurosurgery, prospective cohort study, modified Graeb score, Candida ventriculitis