Use of a Prophylactic Retrograde-Flushing Device in High-Risk Pediatric Patients with Ventriculoperitoneal Shunts: A Technical Note

Introduction: Ventriculoperitoneal shunt (VPS) malfunction rates are as high as 40% in the first year with posthemorrhagic hydrocephalus (PHH) patients having the highest proximal occlusion risk. Debris, protein, and cellular ingrowth most commonly obstruct the proximal ventricular catheter and/or valve. Historically, no preventative methods have demonstrated efficacy. We present a technical note and case series describing the use of a retrograde proximal flushing device and prophylactic flushing protocol to maintain ventricular catheter patency and reduce proximal shunt occlusions. Methods: We present our 2.8–4-year follow-up data on the first 9 pediatric cases of ReFlow (Anuncia Inc, Scottsdale, AZ) device implantation combined with routine prophylactic flushing. Rationale for device implantation, patient selection, surgical procedure details, postoperative follow-up, and prophylactic flushing protocol are discussed as well as pre- and postimplantation ventricular catheter obstruction rates. We include a technical note on the device setup and prophylactic flushing protocol. Results: Patient average age was 5.6 years and all patients had PHH. Minimal follow-up was 2.8 years (range 2.8–4 years). Prophylactic flushing was initiated between 2 and 14 days after ReFlow implantation and has continued as of the last follow-up. In 7 patients, ReFlow implantation occurred during the revision of an existing shunt and in two, implantation was coincident with initial VPS placement. In the 2 years preceding ReFlow and prophylactic flushing, 14 proximal shunt failures occurred in the 7 patients with existing VPS. This was reduced to only one proximal shunt failure in all 9 patients during the full follow-up period after ReFlow and prophylactic flushing. Conclusion: Pediatric VPS placement carries high rates of proximal catheter occlusion, often leading to emergency surgery, morbidity, or even death. The ReFlow device along with routine prophylactic flushing may reduce proximal obstruction and need for revision surgery. Higher patient numbers and longer follow-up periods are necessary to further elucidate the safety and effect of such a device on longer term shunt failures and revision surgery.


Introduction
Pediatric hydrocephalus is a common condition that occurs congenitally or from underlying pathologies including neonatal hemorrhage, tumor, trauma, or infection [1]. Accounting for approximately 70,000 hospital admissions annually, patients with hydrocephalus often require cerebrospinal fluid (CSF) diversion, typically with placement of a ventriculoperitoneal shunt (VPS) [2]. Approximately 30,000 VPS procedures are performed annually in the USA. The cost of care for hydrocephalic children has risen over 30% accounting for up to $2 billion annually [2,3]. As implants, VPSs are not just prone to infection but also malfunction causing repeated neurosurgical procedures [4][5][6]. Malfunction rates are reported to be as high as 40% within 1 year and 85% by 15 years in children [7]. Repeated VPS surgery has also been directly correlated to higher infection rates and reduced intelligence quotients [8,9]. Malfunctions occur most commonly in the proximal portions of the shunt system (ventricular catheter) [8]. In contrast, distal malfunctions occur more commonly in the valve and even less commonly in the distal peritoneal tubing. Debris, protein, and cellular growth are the leading causes of proximal ventricular catheter and valve obstruction [6,10]. Patients with posthemorrhagic hydrocephalus (PHH) carry the highest proximal occlusion risk [10,11]. The number of patients born and surviving with high-grade germinal matrix hemorrhage is increasing with improved neonatal intensive care, leading to higher rates of infants with PHH [12]. Historically, there have been no preventative methods known to reduce the risk of proximal shunt failures [6,13]. Recently, a device designed to retrogradely flush the proximal portion of the shunt has been introduced [14]. In this technical report and retrospective case series with long-term follow-up, we demonstrate the utility and outcome of a retrograde ventricular catheter flushing device implanted immediately distal to the catheter in-line with a typical VPS system.

Clinical Data Collection
Institutional Review Board approval was obtained for a retrospective review of all patients who underwent the ReFlow (Anuncia Inc, Scottsdale, AZ) device implantation between 2018 and 2021. Inclusion criteria included patients under 18 years old with an implanted ReFlow device for hydrocephalus. Patients selected for implantation of the ReFlow device with prophylactic flushing presented with shunt malfunction demonstrable through both imaging and clinical evaluation or were first-time VP shunt implantation patients. Implanted patients demonstrated: a history frequent shunt failure or were deemed clinically high risk for proximal shunt obstruction (ventricular infection <3 months prior, history of intraventricular hemorrhage, high CSF protein counts >400 μg/dL, or pre-malfunction baseline chronically slit ventricles). The device was not placed in infants under 1 year old due to the size of the device. Data on VPS revision were collected on each patient prior to and after ReFlow implantation with prophylactic flushing protocol. All patients' families were instructed on the flushing method and frequency by their primary board-certified pediatric neurosurgeon. Compliance with flushing was assessed via self-report survey questions at clinic visits from each patient's guardian throughout their follow-up period and collected retrospectively. Data variables collected include age at implant, etiology of hydrocephalus, reason for surgery at time of implantation, previous VPS revisions, time from surgery to initiation of prophylactic flushing, surgical complications, and need for and cause of subsequent VPS revisions. Qualitative statistical analysis was performed.

Technical Note
The full ReFlow system received 510(k) clearance from FDA as well as the CE Mark in 2017 and includes both a flushing device and a unique proximal catheter with a relief membrane designed to open if the ventricular catheter is completely obstructed and the flushing dome is depressed. The individual parts that make up an implanted ReFlow system in-line with a shunt valve are the ventricular catheter, valveless flushing device, connection tubing that attaches to the shunt valve, and distal peritoneal tubing ( Fig. 1). While the shunt is functioning properly, fluid passes freely through the ventricular catheter and ReFlow flushing chamber into the implanted valve of choice then down the distal catheter. The flushing dome of the ReFlow houses~0.41 mL of the patient's own CSF at all times. Pressing the flushing dome causes complete and immediate obstruction of anterograde flow into the valve, while delivering 0.41 mL of CSF into the ventricular catheter in a retrograde fashion. Upon release of the ReFlow dome (either quickly or slowly), there is no siphoning effect, as the same amount of CSF that was in the proximal system is allowed back into the system. This process may open ventricular catheter holes which are obstructed or maintain patency of open ventricular catheter inlets if done prophylactically. Releasing the flushing dome allows for a non-siphon flow of CSF to return into the flushing dome through the shunt's baseline flow rate. Within the proximal catheter, there is also a relief membrane just above the intraventricular catheter holes that function as a backup proximal opening. During normal function, the membrane remains intact. When an obstruction occurs which cannot be dislodged through a single flush of the ReFlow dome, the ReFlow reservoir creates resistance, rupturing the membrane and creating a new proximal opening to restore proximal function.
A prophylactic flushing protocol was developed following implantation of the ReFlow system. This protocol was established with the goals of consistent and frequent flushing, ease of use, and convenience and compliance for the patient's family. Although our protocol is novel and therefore does not yet have proven validity, discussion with several pediatric neurosurgical colleagues ensured that we are starting with a reasonable and logical protocol to ensure high compliance. Proximal flushing is ideally initiated within 48 h of surgery. An MRI is obtained prior to initiating flushing to ensure adequate intraventricular catheter placement. Flushing may be initiated within 2 weeks of surgery if the patient does not tolerate due to site tenderness from surgery. Furthermore, time to flushing is also dependent on a risk stratification of patients (Table 1). High-risk stratification was determined by finding of pre-shunt malfunction slit-like ventricles based on frontal-occipital horn ratio <0.2. [15], hemorrhage or infection within the last 3 months, or need for cautery for removal of prior catheter. The latter was deemed Prophylactic Flushing Device in High-Risk Pediatric VP Shunt Patients as a risk factor due to likely scar formation surrounding the indwelling catheter as well as increased risk of intraventricular blood products and debris following removal. Patients with larger ventricles at baseline or prior to malfunction, or without infection or hemorrhage within the past 3 months were deemed to be at lower risk with a later initiation of flushing.
The prophylactic flushing protocol is described in detail in Table 2. The patient's guardian is educated on the location of the flushing device's reservoir by first demonstrating on a manikin and then on the child's head. The pediatric neurosurgeon performs the first flushing while the caregiver is present and ensures the caregiver can demonstrate understanding and reproduce the proper flushing sequence. This includes finding, depressing the flushing dome fully, and allowing full recoil before flushing again. For high-risk patients, the protocol encourages families to initiate flushing within 2 days. Flushing the ReFlow can be uncomfortable for some patients, especially those with postoperative tenderness at the implantation site, therefore low-risk patients were allowed 2 weeks of recovery before initiation of flushing. Patients undergo 3 separate sessions/day (5 flushes per session) for the first 6 months postimplantation, then deescalate to 2 sessions/day between 6 and 12 months following surgery, then to 1 session/day. The family is instructed on signs and symptoms of proximal occlusion, any increased resistance to pressing the reservoir, or inability to locate the reservoir. Postoperative instructions for possible shunt malfunctions are unchanged from patients in whom the ReFlow was not implanted. Standard of care guidelines for emergency situations are carried out in all patients.

Results
Nine patients were identified to have an implanted proximal flushing device and underwent prophylactic flushing. All patients had PHH as the etiology of hydrocephalus.
The average age was 5.6 years old at the time of implantation of the ReFlow device. Six of the 9 patients presented with symptomatic proximal VPS occlusions, one following infection and explanted shunt, and two at the time of their first shunt placement. Prophylactic flushing of the device was performed immediately (within 48 h) in 6 patients (high-risk group), while 3 patients (low-risk group) had delayed initiation of flushing (2 weeks following surgery). Overall compliance with prophylactic flushing observed was 100% in 7 patients and 50% in 1 patient at 6 months; 100% in 6 patients and 50% in 2 patients at 1 year; and 100% in 7 patients and 50% in 1 patient at 2 years. During the 2 years prior to implantation of ReFlow and prophylactic flushing, 7 patients with prior shunts accounted for 14 total proximal VPS revisions, or an average of 2.0 revisions/patient (Table 3).
After 2.8-4 years of prophylactic flushing, one shunt revision was required for acute proximal malfunction 10 months post ReFlow implantation with prophylactic flushing. The other 8 patients continued to have no clinical or radiographic signs of proximal malfunction at follow-up. Two non-proximal shunt revisions were also performed: one for distal peritoneal malfunction and another for infection following wound breakdown. In summary, during the follow-up period, four complications were noted (shunt infection, nonoperative subdural hematoma, proximal occlusion requiring revision, and distal shunt malfunction). One elective shunt revision was performed during this followup period for ventricular catheter migration over time with asymptomatic ventricular enlargement on routine imaging. Vinzani/Alshareef/Eskandari

Discussion
Hydrocephalus treatment continues to be one of the costliest pediatric implant-related conditions at >$2 billion annually [3]. VPS malfunction rates are as high as 85% by 15 years with proximal occlusions being the most common cause [8]. Debris, protein, and cellular ingrowth most commonly obstruct the proximal ventricular catheter and/or valve [16]. Historically, no preventative methods demonstrated efficacy. The use of a proximal flushing device in-line with a traditional VPS is a novel idea that allows for disruption of proximal debris. This maneuver is thought to reduce settling of debris and disrupt the ingrowth of living cells and choroid plexus, all of which have been demonstrated to ultimately lead to proximal shunt failure. The use of flushing protocols has aided in creating a strict regimen with high compliance, which minimizes the need for clinician involvement and allows at-home delivery of  preventative retrograde fluid flushes. In allowing families to perform this task, we find that an implantable retrograde flushing device is a feasible preventative tool. In our study, we noted a reduction of shunt revisions from 2.0 revisions/ patient in the previous 2 years to only 1 proximal shunt failure between all 9 patients in the up to 4 years following device implantation with prophylactic flushing. The singular proximal malfunction requiring revision was in the only patient without 100% compliance in flushing at 6 months follow-up (recorded at 50%). This patient who is now 2 years post-revision has 100% compliance without any more proximal malfunctions. This is a limited study describing the technique for device attachment and protocol implementation. However, the findings in 9 patients show the potential for such a device to be utilized, with high compliance, in patients who are at risk for multiple future shunt failures.

Conclusion
VP-shunting for pediatric hydrocephalus has a risk for failure requiring revision surgery. The use of a proximal flushing device in-line with the traditional VPS may provide a benefit to reducing proximal shunt failures. Our long-term follow-up data suggest that prophylactic flushing of the ReFlow Ventricular System could be a safe and reliable means of shunt occlusion prevention, with patient families being educated in routine, prophylactic flushing. Combination of implantation of a retrograde flushing device and compliant prophylactic flushing reduced VP shunt revision rates in this high-risk cohort. We believe that larger, multiinstitutional evaluations with longer follow-up periods are necessary to further elucidate the safety and effect of such a protocol on lifetime shunt failure and revision surgery rates.

Statement of Ethics
This study protocol was reviewed and considered exempt from Human Research Subject Regulations and the requirement of written informed consent by the IRB-II -Medical University of South Carolina in accordance with 45 CFR 46.101(b). Written informed consent was not required according to the Medical University of South Carolina's Institutional Review Board.

Conflict of Interest Statement
Ramin Eskandari is a consultant for Anuncia Inc.

Funding Sources
No funding sources were used for this work.

Author Contributions
R.E. and M.A. designed research; R.E., M.A., and M.V. performed chart reviews for the collection of data, wrote the paper, and analyzed and interpreted data. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement
All data upon which the stated conclusions rely are included in Table 3. Further inquiries can be directed to the corresponding author.