CSF rhinorrhoea after endonasal intervention to the anterior skull base (CRANIAL): proposal for a prospective multicentre observational cohort study

Abstract Background The endonasal transsphenoidal approach (TSA) has emerged as the preferred approach in order to treat pituitary adenoma and related sellar pathologies. The recently adopted expanded endonasal approach (EEA) has improved access to the ventral skull base whilst retaining the principles of minimally invasive surgery. Despite the advantages these approaches offer, cerebrospinal fluid (CSF) rhinorrhoea remains a common complication. There is currently a lack of comparative evidence to guide the best choice of skull base reconstruction, resulting in considerable heterogeneity of current practice. This study aims to determine: (1) the scope of the methods of skull base repair; and (2) the corresponding rates of postoperative CSF rhinorrhoea in contemporary neurosurgical practice in the UK and Ireland. Methods We will adopt a multicentre, prospective, observational cohort design. All neurosurgical units in the UK and Ireland performing the relevant surgeries (TSA and EEA) will be eligible to participate. Eligible cases will be prospectively recruited over 6 months with 6 months of postoperative follow-up. Data points collected will include: demographics, tumour characteristics, operative data), and postoperative outcomes. Primary outcomes include skull base repair technique and CSF rhinorrhoea (biochemically confirmed and/or requiring intervention) rates. Pooled data will be analysed using descriptive statistics. All skull base repair methods used and CSF leak rates for TSA and EEA will be compared against rates listed in the literature. Ethics and dissemination Formal institutional ethical board review was not required owing to the nature of the study – this was confirmed with the Health Research Authority, UK. Conclusions The need for this multicentre, prospective, observational study is highlighted by the relative paucity of literature and the resultant lack of consensus on the topic. It is hoped that the results will give insight into contemporary practice in the UK and Ireland and will inform future studies.


Introduction
The endonasal transsphenoidal approach (TSA) has emerged as the preferred approach in order to resect pituitary adenoma and related sellar pathologies owing to its superior effectiveness and safety profile when compared to transcranial approaches. 1,2 This approach is defined by its purpose of accessing the sella turcica through the sphenoid bone. Whilst traditionally performed microscopically, recent technological advances have allowed the TSA to be performed with success endoscopically. 1, 3 Furthermore, building on these endoscopic techniques, the development of the expanded endonasal approach (EEA) has further improved access to the anterior skull base. 4 This approach refers to accessing an area beyond the sella alone, bounded by the frontal sinus, cribriform plate, medial orbital wall, cavernous sinus, posterior clinoid processes, and clivus. 5 The EEA is used for the surgical management of many pathologies including large pituitary adenomas, craniopharyngiomas, meningiomas, Rathke's pouch cysts, clival chordomas and chondrosarcomas. 5 Despite the advantages these approaches offer, cerebrospinal fluid (CSF) rhinorrhoea remains a common complication and may lead to significant morbidity including prolonged hospital stay, headaches, pneumocephalus, and meningitis. [6][7][8] CSF rhinorrhoea occurs with the disruption of the tissue between the subarachnoid space and sinonasal cavity, namely the meninges, skull base, sinonasal mucosa. 8 Arguably, therefore, the most important risk factor for the development of a CSF rhinorrhoea is the method of reconstruction of the skull base 5 (Figure 1). Other risk factors for postoperative CSF rhinorrhoea include elevated BMI, prior cranial radiotherapy, prior cranial surgery, tumour size, local tumour infiltration, high-flow intraoperative CSF leak and surgeon experience. 5,[9][10][11][12] Commonly cited skull base repair methods include the use of fat or fascia grafts, nasoseptal flaps and lumbar drains. 13 There is, however, a multitudinous array of techniques and combinations available, including direct dural closure (for example, with sutures), dural replacement (for example, Durepair TM or fascia lata), synthetic grafts (for example, Tachosil V R , Gelfoam V R ), buttresses (for example, Titanium mesh or Medpor V R ), tissue glues (for example, Eviceal V R or Adherus V R ) and nasal packing (for example, ballooned catheters or Nasopore V R ). 5,14 There is a suggestion that the use of nasoseptal flaps is particularly beneficial in the setting of large defects (>3cm) and/or high CSF flow. 15,16 Similarly, a recent randomised controlled trial concluded that perioperative lumbar drain use decreased CSF rhinorrhoea rates when combined with nasoseptal flap repair (in the context of dural defects >1cm 2 and high flow intra-op CSF leak). 17 Overall, however, there is a lack of comparative evidence to guide the ideal choice of skull base reconstruction. 13 This is the circumstance in both first and second attempts of leak repair, as well in both high and low CSF flow situations. 13 Thus, there is considerable heterogeneity in current practice and is based mostly on surgeon preference. 13 Similarly, there is marked variation in resultant CSF leak rates, estimated at up to 5% for TSA and up to 20% for EEA. 5,6,11 To this end, this study aims to determine: (1) the scope of the methods of skull base repair; and (2) the corresponding rates of postoperative CSF rhinorrhoea in contemporary neurosurgical practice in the UK and Ireland.

Design
We will adopt a multicentre, prospective, observational cohort study design. 18 All neurosurgical units (NSUs) in the UK and Ireland performing the relevant surgeries (TSA and EEA) will be eligible to participate. The study will be registered as a quality improvement project on a local level, with registration in accordance with the local audit department and Caldicott guardian approvals if needed.
The project will be run through the Neurology and Neurosurgery Interest Group (NANSIG; https://nansig.org/) and British Neurosurgical Trainee Collaborative (BNTRC; https:// www.bntrc.org.uk/) networks. Each participating centre will have an appointed consultant, trainee and junior doctor or student lead for the project. Consultant neurosurgeons will be contacted in advance and invited to join the project steering group by the central study team before local students and trainees are recruited. Local teams will be provided with supporting materials to facilitate the uniform set-up of the project, for example, project registration templates (Supplementary Appendix A) and explanatory figures/definitions (Supplementary Appendix B).

Eligible patients
Included cases will be patients of all ages undergoing TSA for sellar tumours and EEA for skull base tumours. Exclusion criteria include patients undergoing transcranial surgery and those with a history of preoperative CSF rhinorrhoea. Figure 1. This image illustrates an example method by which the anterior skull base may be repaired following transsphenoidal surgery.

Case recruitment
As the study aim is to capture contemporary practice over the study period, stopping criteria will be time-basedwith eligible cases being prospectively recruited over six months from the study launch date, and with six months of postoperative followup for each case. This time period was chosen for pragmatic reasons, allowing for trainees to support the study for its entire one-year duration. We estimated this would include sufficient patients for meaningful analysis.

Data collection
Data points collected will include: demographics, tumour characteristics, operative data, and postoperative outcomes ( Table  (1 -5)). Baseline, operative and postoperative data points will be collected within 30 days of admission whilst follow-up outcomes will be collected within six months of surgery.
Pseudo-anonymised data will be collected locally and submitted to a secure web-based central database hosted by Castor Electronic Data Capture (https://www.castoredc.com/). Local data sources will include patient case files, multidisciplinary team discussions, theatre lists/logbooks and local registries/databases. Data will be collected by a member of the clinical team caring for the patient or member of the approved audit team. Importantly, the primary outcomes of the study will be: (1) methods of intraoperative skull base reconstruction used, and (2) postoperative CSF rhinorrhoea biochemically confirmed and/or requiring intervention (CSF diversion and/or operative repair). These primary outcomes will be compared with rates reported from the literature ( Table 6). Secondary outcomes will be: (1) Intraoperative CSF leak; (2) operating time; (3) rates of other postoperative complications; and (4) length of hospital stay.

Data accuracy
All data points collected by medical students must be approved for accuracy by the local trainee or consultant lead before final submission into the Castor EDC system. Furthermore, specific data points must be discussed with the operating surgeon(s) before submission and this is highlighted by the Castor datasheet, for example, presence and grading of intra-op CSF leak, 20 max diameter of skull base defect and exact methods of skull base repair used). Illustrations and clear definitions will be presented to support the accurate recognition of the various skull base repair techniques and facilitate standardised discussion ( Supplementary Figures 1-3, Appendix B). The study procedure has been piloted in three NSUsthe National Hospital for Neurology and Neurosurgery (London), the John Radcliffe Hospital (Oxford) and Addenbrooke's Hospital (Cambridge). Our pilot experience was formative in refining the data collection proforma and illustrated the feasibility and acceptability of the project process. Of note, adaptations to operative notes by surgeons (to explicitly display CRANIAL data points), impacted data collection efficiency and accuracy and will be encouraged going forward.
Local student and trainee leads must meet with the supervising consultant at the half-way mark (three months of case recruitment) for review of data collected, progress update and to troubleshoot any problems encountered. Additionally, the local student and trainee leads must meet with the supervising consultant again at the end of the case recruitment period (six months). Lastly, a final review meeting will occur at the end of data collection (at 12 months). This is a final review and sign off of data collected and marks the end of the local study.
Finally, data validation will be performed in all centres to audit data accuracy before data analysis. This will involve an independent data validator (who is not part of the local CRANIAL team) who is from the centre in which the data was collected. This is to facilitate working within the agreements set out by local audit/service evaluation processes and Caldicott guardian approval. 10% of the centre's cases (selected randomly) will be reviewed, comparing the data submitted to raw data sources for accuracy. The target for data is accuracy is >95% with no case duplication. Conflicts between actual and submitted data will be resolved by discussion between the validator and local team, with oversight from a steering committee member. If data accuracy is <95%, the local team will then be asked to update all local data accordingly. A re-audit of 10% of the centre's cases (selected randomly) will then be repeated. If the requested updating of data is not performed or data accuracy remains <95%, data from the respective centre will be analysed separately or excluded.

Data analysis
Pre-processing steps will include re-categorising free text entries into existing similar data categories and grouping free text entries into new data categories.  Pre-procedure (before the patient was taken to theatre), Pre-procedure (in theatre, under the same GA but before skull base surgery begins), Immediately post-procedure (e.g. in theatre or under the same GA) as a prophylactic measure. (continued) With respect to primary study aims, the scope of the methods of skull base repair, will initially be described using descriptive statistics exploring the incidence density of individual repair methods and repair method combinations within TSA/EEA and CSF leak grade subgroups. Corresponding rates of postoperative CSF rhinorrhoea will be presented as incidence percentages per TSA/EEA subgroups and per repair method used. CSF rhinorrhoea rates for individual centres will not be presented separately. Multivariable logistic regression models will be used to assess the impact of baseline characteristics and skull base repair methods used on postoperative CSF rhinorrhoea. Odds ratios and 95% confidence intervals will be reported. Sub-group analysis will be performed where possible.
Descriptive statistics will be used to summarise baseline characteristics (demographic and operative data points) and surgical outcomes. This includes study secondary outcomes of rates of other postoperative complications (will be presented as incidence percentage for TSA and EEA subgroups), operating time (will be presented as median and interquartile ranges for TSA and EEA subgroups) and length of hospital stay (will be presented as median and interquartile ranges for TSA and EEA subgroups).

Ethics and dissemination
Formal institutional ethical board review was not required owing to the nature of the study (seeking to evaluate local services) and this was confirmed with the Health Research Authority, UK. 21 Pseudo-anonymised data will be collected locally and submitted to a secure web-based central database hosted by Castor Use of nasal pack following repair? Nasal packing refers to using a material to occupy a nasal space and provide structural support through its local pressure effects. They can also be coated with substances (e.g. bismuth) to augment particular qualities (e.g. haemostasis).   Electronic Data Capture (https://www.castoredc.com/). Only anonymised data will be published and disseminated.

Conclusions
The heterogeneity of literature and a lack of consensus on the incidence and management of CSF rhinorrhoea following endonasal skull base procedures supports the need for this multicentre, prospective, observational study. It is hoped that the results will give insight into contemporary practice in the UK and Ireland. Additionally, this study aims to inform future studies and facilitate the establishment of national benchmarks for clinical practice. Finally, we hope that the established CRANIAL network of medical students, trainees and consultants will become a platform for future qualitative and quantitative studies aiming to consolidate evidence-based practice on this topic.