The utility of middle ear fluid biomarkers in spontaneous cerebrospinal fluid otorrhoea

Introduction

The development of cerebrospinal fluid (CSF) leakage is a complex process with various aetiologies, including trauma, neoplasia, surgery, and congenital malformations (1,2). Spontaneous CSF otorrhoea, a distinct subset of cases without an apparent cause, has garnered recognition but remains poorly understood (3-7). Factors such as increased intracranial pressure, obesity, and obstructive sleep apnoea almost certainly play an aetiological role (4-16). The pathomechanism likely originates from chronically impaired CSF absorption within structures such as arachnoid granulations, the choroid plexus, and the glymphatic system (5,17-19). Consequently, increased pulsatile hydrostatic forces are exerted at anatomically vulnerable sites along the skull base, such as the tegmen or sites of aberrant arachnoid granulations (3,5,6,18-21). Over time, these can erode and slowly enlarge bony defects, ultimately leading to the formation of an osteodural defect and subsequent CSF leakage (5,6,8,10).

Complications from CSF leaks carry significant risks, including life-threatening intracranial infections, seizure disorders, septic emboli, and neurological deficits (22-24). The risk of bacterial meningitis varies depending on the cause of the CSF leak, and is yet to be quantified for spontaneous leaks (11,13,25). Spontaneous CSF leaks are unlikely to resolve on their own and have the highest recurrence rates among all causes (4,6,18,26). Currently, accurate predictors of meningitis in unrepaired spontaneous leaks are lacking, and much of our understanding comes from small observational studies (11,27). Prompt diagnosis of spontaneous CSF leakage is crucial, as surgery is indicated to help prevent severe complications. Further research is needed to identify prognostic factors, particularly for those who may be unfit or unwilling to undergo surgical intervention.

Relying solely on clinical history and examination can make it challenging to achieve a definitive and timely diagnosis of a CSF leak. Symptoms such as unilateral hearing loss, aural fullness, and headaches often overlap with more common conditions (5,8,11). Radiological studies—such as computed tomography (CT), magnetic resonance imaging (MRI), radionuclide scans or cisternography—can aid in visualising skull base defects and confirming CSF leaks but may expose patients to certain risks (23,28). An accurate and timely diagnosis of CSF otorrhoea depends on the clinician’s heightened vigilance and expertise, with biochemical testing for CSF-specific proteins being essential.

The current gold standard for diagnosing CSF otorrhoea is laboratory analysis of middle ear fluid. Beta-2 transferrin was the first biochemical marker widely adopted for this purpose due to its high sensitivity and specificity (23,29). The most commonly used detection method, immunofixation electrophoresis, is time-consuming, costly, and not universally available, often resulting in delayed result availability (23,30). Interpretation can also be challenging due to variability in fluid composition, laboratory protocols, and operator expertise (31). More recently, beta trace protein (BTP) has gained prominence as a diagnostic marker due to its lower cost and faster assay turnaround time (1,2,30). Unlike beta-2 transferrin, which provides a binary result indicating the presence or absence of CSF, BTP can be quantitatively measured, offering the potential for stratification or prognostic use.

The primary role of BTP analysis remains the confirmation or exclusion of CSF in a given sample (2). In native CSF compartments, such as in fluid obtained from lumbar punctures, additional biomarkers—like protein, glucose, and bilirubin—can provide additional diagnostic insights, indicating infection or a haemorrhage. However, to our knowledge, analysis of these biomarkers has yet to be conducted on middle ear fluid. Additionally, validated normative values for these biomarkers in middle ear fluid do not exist, unlike the well-established reference ranges for lumbar CSF. These biomarkers could potentially offer more information on the severity of a CSF leak, especially given the limited understanding of the natural history of this condition.

Aim

The primary aim of our study is to describe the range of values of middle ear fluid biomarkers—specifically bilirubin, proteins, and glucose—in patients diagnosed with spontaneous CSF otorrhoea. We aim to evaluate their correlation with clinical features and determine whether these biomarkers can provide actionable insights into diagnostic accuracy and guide patient management in clinical settings.

Methods

Study design and setting

A retrospective case review was performed for individuals with this pathology. Given the limited prevalence of spontaneous CSF otorrhoea and lack of well-defined exposure variables and appropriate control groups, alternative designs were less feasible. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Institutional ethics approval was obtained from the Committee of Quality Improvement of Sir Charles Gairdner Hospital under protocol (No. HoD 50556 SCGOPHCG). Patient confidentiality was maintained by anonymising all data, and individual consent was waived for this retrospective analysis. This study is reported according to the STROBE reporting guidelines (available at https://www.theajo.com/article/view/10.21037/ajo-25-10/rc).

Patient selection and data collection

Adult patients diagnosed with CSF otorrhoea and treated by Sir Charles Gairdner Hospital from 2018 to 2023 were included in the study. This cohort includes patients referred from outpatient clinics across both private and public health services. The number of cases identified during the study period determined the sample size. Patients with otorrhoea secondary to other pathologies, such as iatrogenic causes, trauma or cholesteatoma, were excluded from the study. All included subjects had a valid BTP result.

Two independent clinicians conducted a comprehensive review of the patients’ medical records, encompassing demographic information, symptomatology, audiometric assessments and radiological findings from CT and MRI. Surgical interventions were also examined, including intraoperative findings, surgical techniques employed, and clinical outcomes related to CSF leak repair. The data collection period concluded in December 2023.

Diagnostic methodology

A transtympanic aspiration of middle ear fluid was performed using a fine-bore spinal needle under either general anaesthesia or local anaesthesia. In the case of local anaesthesia, 5% lignocaine ointment was applied topically to an intact tympanic membrane under direct vision using an operating microscope and then removed by suction clearance. All fluid samples were analysed by a single validated laboratory (PathWest). The BTP assay was performed using the Atellica NEPH 630 nephelometer (Siemens Healthineers, Forchheim, Germany), requiring a minimum of 50 µL of non-viscous fluid. According to this laboratory’s reference range, BTP values ≥1.3 mg/L are indicative of CSF. Patients meeting this criterion (i.e., BTP above ≥1.3 mg/L in middle ear fluid) were included in the study.

Additional biochemical analyses included measurement of bilirubin, glucose, and total protein levels using the Architect analyser (Abbott Diagnostics, Lake Forest, IL, USA), employing the same assays consistent with routine blood biochemistry. These additional biomarker results were not used to inform clinical management, and were only accessible to the investigators at the conclusion of the study period.

Our missing data analysis procedures assumed data were missing at random (MAR), with no clear associations between missingness and observed patient characteristics. Multiple imputation techniques were applied to estimate missing biomarker levels for glucose, bilirubin and total proteins. Rubin’s rules were used to pool the results and account for the uncertainty inherent in imputation. A three-month post-treatment observation window was selected based on prior studies indicating that initial leak repairs typically stabilise within this period. While secondary CSF leaks and complications can develop later, our choice of a three-month window was based on the clinical approach to immediate post-surgical recovery.

Statistical analysis

Descriptive statistics were used to analyse quantitative variables, such as biomarker levels, alongside frequency counts for categorical variables, including sex, ethnicity, and the laterality of ear involvement. Comparative analysis was conducted to assess the impact of different treatments or patient demographics on outcomes, using t-tests and analysis of variance (ANOVA) for quantitative data and Chi-squared tests for categorical variables.

In our correlation analysis, we explored the relationships between biomarker levels (bilirubin, glucose, BTP, and other proteins) and outcome severity. To assess the relationship between biomarker levels and clinical outcomes, we developed a novel severity scoring system, as no validated tool exists for spontaneous CSF otorrhoea. This de novo composite score was created in consultation with the otology team at Sir Charles Gairdner Hospital to reflect clinically relevant symptoms and complications encountered in CSF leak management. Each parameter was weighted to reflect its impact on morbidity, with greater emphasis on outcomes associated with long-term sequelae. Raw scores and associated clinical data for each patient are tabulated to support transparency and reproducibility (Table 1).

The scoring system includes the following clinical parameters:

• Headache: frequency and intensity as described by the patient (1 point).
• Meningism: specific symptoms indicative of meningeal irritation, such as neck stiffness, photophobia, and phonophobia (2 points).
• Nausea/vomiting: incidence and severity of nausea and vomiting episodes (1 point).
• Length of hospitalisation: total duration of hospital stay required for treatment and recovery (1 point per day over 10 days of standard hospitalisation).
• Neurological or surgical complications: presence and severity of complications arising from CSF otorrhoea, including infections, neurological deficits, or other adverse outcomes (3 points).
• Need for re-exploration: requirement for additional surgical interventions due to recurrent symptoms or complications within the first three months post-treatment (3 points).

A Pearson’s correlation coefficient matrix was used to assess the linear associations between biomarker levels and the composite outcome severity score. Further regression analyses were performed to model the relationship between independent variables and outcomes: linear regression was applied to continuous outcomes, while logistic regression was used for binary outcomes. All statistical analyses were conducted using R version 4.2.1 and Tableau version 2024.3.

Results

A total of 24 patients met the inclusion criteria and were diagnosed with spontaneous CSF otorrhoea. All patients underwent surgery, during which skull base dehiscences were visualised and repaired. The age range of patients at diagnosis was from 30 to 81 years. The cohort included 13 male patients (54%) and 11 female patients (46%). Eleven patients had recorded glycosylated haemoglobin (HbA1c) results during the study period, with values ranging from 5.1% to 5.9%. Of these, three patients had a known diagnosis of diabetes mellitus and were receiving medical management. One patient developed meningitis with a temporal lobe abscess following the diagnosis of CSF otorrhoea. Two patients died of unrelated pathologies during the study period.

BTP levels were measured in all 24 patients. The mean BTP concentration was 9.14 mg/L, with a standard deviation of 5.70 mg/L, and BTP values ranged from 0.20 to 25.50 mg/L. The 25th percentile was 6.05 mg/L, the median was 9.65 mg/L, and the 75th percentile was 11.18 mg/L, indicating substantial variability in BTP concentrations among patients. One patient with a surgically and radiologically confirmed CSF leak demonstrated a BTP level below the established diagnostic threshold of 1.3 mg/L. Despite the confirmed diagnosis, the BTP assay result fell into a non-diagnostic range, highlighting a potential false negative.

Regarding the other biomarkers, the mean bilirubin level was 18.9 µmol/L (range, 10–32 µmol/L), and the mean glucose level was 2.8 mmol/L (range, 1.40–4.30 mmol/L). The mean total protein level was 126.29 g/L, with a standard deviation of 17.07 g/L, ranging from 96 to 147 g/L. Table 2 summarises the descriptive statistics of bilirubin, glucose and BTP levels in the CSF, highlighting the variability of these biomarkers in patients with CSF otorrhoea. Bilirubin levels clustered around the mean but displayed a skew toward higher values, while glucose levels were tightly distributed within a narrower range, with some outliers at the higher end (Figure 1).

Figure 1 Histograms illustrating the distribution of bilirubin and glucose levels among the patient cohort.

In our comparative analysis, a t-test comparing bilirubin levels between male and female patients yielded a P value of 0.633. Due to data limitations, a conclusive analysis of glucose differences between males and females could not be conducted. A Chi-squared test examining laterality versus surgical approach (transmastoid in 10 cases, middle cranial fossa repair in 8, and combined approach in 6) yielded a Chi-squared statistic of 34.00 and a P value of 0.281.

Correlation analysis revealed a slight negative correlation between bilirubin levels and outcome severity (r=−0.208), while glucose levels showed a strong positive correlation with outcome severity (r=0.923). The correlation between total protein levels and outcome severity was moderate (r=0.581). The visualisation matrix (Figure 2) underscores the strong positive correlation between glucose levels and outcome severity. This heatmap helps quickly identify and understand the strengths and directions of the relationships between the variables, supporting the findings from the correlation analysis.

Figure 2 Heatmap visualising the correlation matrix of biomarkers (bilirubin, glucose, proteins, and outcome severity). This provides a colour-coded representation of the correlation coefficients, with warmer colours (towards red) indicating positive correlations and cooler colours (towards blue) indicating negative correlations.

In regression analysis, model coefficients for bilirubin, glucose, and other proteins were −3.12, 1.59, and 2.17, respectively. This suggests that, holding other variables constant, a one-unit increase in glucose and other proteins is associated with an increase in the outcome severity, while a one-unit increase in bilirubin is associated with a decrease in outcome severity. The mean squared error (MSE), a measure of the average squared difference between the observed actual outcomes and the outcomes predicted by the model, was 3.55, indicating a good fit of the model to the data.

Discussion

BTP

BTP is a low molecular weight glycoprotein primarily synthesised by the epithelial cells of the choroid plexus (22,32). It is the most abundant protein found in CSF after albumin (33). In serum, BTP levels range from 0.5 to 0.6 mg/L, originating from the resorption of CSF (1,2,34). CSF BTP concentrations are 33 to 35 times higher than in serum and about 50 times higher than in nasal secretions, establishing BTP as a reliable marker for CSF (1,2,9,34,35). BTP is found at even higher concentrations in the perilymph, likely attributable to the communication between the cochlea and subarachnoid space via the cochlear aqueduct (35,36).

However, the absence of an international cutoff and standardised analysis method for BTP levels complicates its diagnostic use. Some laboratories advocate for concurrent serum analysis and suggest a positive CSF detection if BTP levels in fluid range from 0.7 to 1.29 mg/L and the BTP fluid/serum ratio is ≥2 (23). Furthermore, it is almost exclusively renally excreted, and patients with chronic kidney disease are known to exhibit elevated serum BTP values (22). For the rare patient with renal failure and possible CSF leakage, this could result in a falsely positive detection of BTP in other body fluids (22,34). This was not an exclusion criterion in our study design, thus warranting careful interpretation.

A study by Bachmann-Harildstad et al. found elevated BTP in over half of paediatric patients with otitis media with effusion, with a median BTP value of 2.4 mg/L (35). Though no imaging had confirmed CSF leaks, chronic effusions or inflammatory mediators may increase permeability, allowing BTP to infiltrate the tympanic cavity. In one notable case, the BTP level was below the commonly accepted diagnostic cutoff of 1.3 mg/L, yet a CSF fistula was definitively confirmed on CT imaging and subsequently during surgical exploration and repair. This apparent false negative does raise important questions regarding the reliability of BTP testing in certain scenarios. This BTP level is on average lower than we would expect for unaltered CSF, which would about approximately 13 to 20 mg/L (37).

Possible mechanisms contributing to a falsely low BTP result include sample dilution from coexistent middle ear effusion, protein degradation due to local inflammation, or patient-specific metabolic and renal differences affecting BTP expression and clearance. While BTP is a useful biomarker due to its rapid turnaround and accessibility, this case underscores its limitations and the importance of interpreting results in conjunction with clinical, radiological, and operative findings.

Tumani et al. demonstrated that CSF BTP is significantly decreased in cases of active bacterial meningitis, whereas in other diseases, such as spinal canal stenosis, it was elevated (37). This was also observed in an earlier study by Meco et al. (33). These findings support the theory that, as CSF makes its way from the intracranial compartment to the middle ear, it may be subject to a process of filtration, dilution, or metabolism, hence lowering the BTP levels. No patients in our study had active meningitis at the time of fluid sampling, and often the increased viscosity and turbidity of secretions during active infection preclude adequate testing. One patient had a past history of meningitis, but their biochemical results were not found to be outlying. Future studies may clarify these complex dynamics and determine whether there is a demonstrable relationship between BTP levels and the risk of complications.

Glucose

In normal physiology, CSF glucose levels range from 2.2 to 4.4 mmol/L, or 60–70% of plasma levels, except in conditions like hypoglycorrhachia from meningitis. Comparatively, serum and perilymph glucose levels vary from 3.9 to 6.9 mmol/L. Secretions such as tears and mucus will typically exhibit glucose levels of less than 0.3 mmol/L. Tan et al. demonstrated that CSF glucose levels can be influenced by serum glucose levels in the few hours preceding sampling and are vulnerable to misinterpretation (38). Therefore, in patients with abnormal glucose metabolism, such as those with diabetes mellitus, the sample/serum glucose ratio may be a more reliable measure.

In this small cohort, only three patients had a known diagnosis of diabetes mellitus. HbA1c measurements were available for 11 patients during the study period, which is expected, as routine testing is typically not performed in individuals without a history of diabetes. Notably, no recorded HbA1c value exceeded 6%, including in those with diabetes, indicating satisfactory glycaemic control across the cohort. As a result, it is not possible to assess the impact of poorly controlled diabetes on middle ear glucose levels. However, these findings suggest that diabetes mellitus is unlikely to be an independent contributor to elevated middle ear glucose levels or to influence outcome severity in this population.

The mean middle ear fluid glucose level in our cohort was 2.78 mmol/L with a standard deviation of 0.99 mmol/L, marginally lower than expected for CSF. This discrepancy suggests a disruption in normal CSF production or circulation, possibly due to the pathophysiological processes underlying CSF otorrhoea (38). Literature supports that CSF glucose levels can be altered in various conditions, including infections or inflammation within the central nervous system (CNS). Low CSF glucose is linked to diseases such as bacterial and fungal meningitis, tuberculosis, and carcinomatous meningitis.

In our cohort, a higher glucose level strongly correlated with more severe outcomes. While elevated glucose levels in CSF are not typically a direct causal factor, they can serve as an indirect indicator of underlying pathology. Elevated glucose may reflect the severity of cellular dysfunction, or impaired CSF circulation or clearance, all of which can contribute to poorer outcomes. While HbA1c levels could potentially influence middle ear glucose levels, the correlation between glucose and outcome severity observed in our study was not significantly affected by diabetes status.

Bilirubin

Bilirubin, a byproduct of haemoglobin breakdown, serves as a marker for haemorrhagic complications in spontaneous CSF leaks. It is produced in vivo exclusively through the enzymatic conversion of haemoglobin and, under normal conditions, its levels in CSF are negligible. Thus, any significant increase is suggestive of the presence of blood. Increased CSF bilirubin, or xanthochromia, may appear within 12 hours of subarachnoid haemorrhage and can persist for up to three weeks (39). Tympanocentesis alone is unlikely to raise bilirubin levels, as it is a largely bloodless procedure when performed correctly. While bilirubin is often detected in lumbar puncture samples due to blood contamination, this risk is reduced in tympanocentesis, which is performed under direct visualisation with minimal tissue disruption. The use of topical local anaesthesia for venepuncture does not affect clinical chemistry or haematology measurements, including bilirubin levels (40).

Our cohort showed elevated bilirubin, averaging 18.9 mg/dL with a range of 10–32 mg/dL, a novel finding in middle ear fluid. The presence of bilirubin in middle ear fluid has not, to our knowledge, been previously documented. This suggests a possible link between middle ear effusions and blood extravasation, warranting further investigation into its diagnostic relevance in otologic conditions. The greater standard deviation and range of bilirubin levels within our cohort suggest it may be more readily influenced than glucose or other proteins by disease mechanisms.

Our study demonstrated a slight negative correlation between bilirubin levels and outcome severity. Higher bilirubin levels may be weakly associated with lower severity of outcomes, although the relationship is not strong. Although no literature currently links CSF with middle ear inflammation, the presence of bilirubin in CSF could still reflect broader pathophysiological processes associated with CSF otorrhoea. Further research is needed to investigate the diagnostic potential of elevated bilirubin in CSF otorrhoea and to better understand its underlying mechanisms.

Clinical implications

No differences were observed in biomarker levels between the sexes, indicating that bilirubin, glucose, and BTP levels remain consistent across male and female patients with CSF otorrhoea. The standard deviation for glucose levels (0.99 mmol/L) is smaller than that for bilirubin (6.67 µmol/L), suggesting that bilirubin levels may be more influenced by underlying pathophysiological mechanisms. Conversely, the smaller standard deviation and tighter clustering of glucose levels may support its utility as a stable benchmark for differentiating normal from abnormal levels. Understanding these distinctions is essential for clinicians, as it shapes the diagnostic and predictive value of each biomarker in managing CSF otorrhoea.

The variability in bilirubin levels may require a more individualised interpretation, while the consistent glucose levels could serve as a reliable indicator of patient status. The inverse relationship between bilirubin and glucose indicates that an increase in bilirubin levels is associated with a decrease in glucose levels, and vice versa. The range of values underscores the heterogeneity of CSF otorrhoea presentations and emphasises the need for biomarker profiling to improve diagnostic accuracy and inform treatment strategies.

This analysis aligns with recent findings across various neurological conditions. Yang et al. reported a significant association between CSF biomarker levels and both diagnosis and injury severity in acute traumatic spinal cord injury, though without a direct link to short-term outcomes (41). Similarly, studies on acute ischemic stroke demonstrated positive correlations between CSF biomarkers, infarct volume, and severity scores (42-45). Bachmann-Harildstad et al., in a systematic review, highlighted that altered CSF concentrations of proteins related to cytoskeletal damage, inflammation, apoptosis, and oxidative stress were predictive of poorer neurological outcomes in patients with acute brain injuries. Although their primary focus was on brain injuries, these insights may be extrapolated to CSF leak cases, suggesting that certain biomarkers could reflect ongoing pathophysiological processes in these patients (2).

In addition, Lafer and colleagues identified an association between elevated BTP levels and the acute-phase reaction, indicated by a rise in C-reactive protein (CRP) levels (32). These findings collectively underscore the complexity of CSF otorrhoea and the need for individualised assessments informed by a nuanced understanding of CSF biomarker dynamics. This study does not suggest that biomarkers should influence the immediate urgency of treatment for CSF leaks. However, biomarkers like glucose and bilirubin may provide prognostic information to guide long-term management decisions, including the likelihood of complications or recurrence.

Surgical approaches

Our results indicate no statistically significant correlation between the ear involved and the selected surgical approach, suggesting that surgical decisions are influenced by factors beyond the laterality of fluid presence. When choosing a surgical approach, factors such as defect location and size, hearing status, patient body habitus, comorbidities, and the surgeon’s experience and preference all play a role.

Strengths

Our analysis of biomarkers in patients with CSF otorrhoea reveals a complex interplay between clinical observations and normative ranges. Although understanding biochemical parameters in CSF otorrhoea is critical, there is limited literature examining these biomarkers in relation to clinical severity. To the best of our knowledge, this paper represents the first study of biochemical markers beyond BTP in CSF otorrhoea. The consistency in sample collection, processing, and reporting—ensured by using a single clinician and laboratory—minimises variability and enhances the reliability of findings. Conducting the study retrospectively allowed for statistical analyses and the identification of correlations between biomarker levels and clinical outcomes.

Our research, utilising Pearson correlation coefficients, uncovered a strong association between glucose levels in CSF otorrhoea and outcome severity (P<0.05). The statistically significant correlations support findings from previous studies by Stukas et al. and Guasp et al., adding to existing evidence on the prognostic value of molecular markers in CNS pathologies (46,47). Specifically, higher glucose levels may indicate greater disease severity, positioning glucose as a potential biomarker for predicting patient outcomes.

Limitations

The limitations of this study include its small sample size, single-centre design, and retrospective approach, which may reduce the generalisability of the findings. While this ensures consistency in diagnosis and management, it may limit broader applicability to groups with different demographic, geographic, or healthcare system variations. The relatively small number of cases, due to the rarity of the condition, may also restrict extrapolation to larger or more diverse populations. Since this study captures biomarker levels at a single time point rather than longitudinally, it may not account for fluctuations in biomarker concentrations over time or across different stages of disease progression. A longitudinal design would strengthen conclusions about biomarker dynamics.

Although care was taken to remove debris via direct microscopy and not to introduce contaminants with the use of a fine-bore needle, lignocaine itself does primarily bind to protein, so theoretically, this can influence total protein count in the sample fluid. Additionally, the lack of concurrent blood sampling prevents direct comparisons of biomarker concentrations in serum and middle ear fluid. In clinical practice, blood samples are often obtained alongside lumbar punctures to assess glucose and bilirubin levels; however, this protocol was not applied here. Although three patients had a known diagnosis of diabetes mellitus, the impact of diabetes on middle ear glucose levels was not directly studied in this cohort.

The outcome severity scoring system used in this study is a novel tool and has not undergone external validation. This introduces a potential limitation regarding the internal validity, reproducibility, and generalisability of severity-based conclusions. While the score was developed in collaboration with experienced otologists and based on routinely observed clinical outcomes, it remains exploratory in nature. Future studies should aim to validate or refine this scoring metric in larger, prospective, and multicentre cohorts to enhance its utility. Until such validation occurs, findings involving severity correlations should be interpreted cautiously and primarily as hypothesis-generating.

Furthermore, the retrospective design and subjective scoring for outcome severity may introduce selection and interpretation biases. Surgical outcomes and BTP results, readily available to clinicians, may have influenced case management. A lack of long-term follow-up data prevents conclusions about recurrence, which has been reported to occur years after surgery (14). Future research should investigate diabetes status and meningitis history as potential confounders, as well as biochemical profiles in non-CSF cases, as limited data exist on protein, bilirubin, and glucose levels in middle ear fluid from infection or trauma. Prospective, multicentre studies are needed to validate these results and further explore biomarkers’ role in diagnosing and managing CSF otorrhoea.

Conclusions

The biochemical composition of CSF serves as a dynamic reflection of the delicate homeostatic balance between the CNS and the peripheral. This retrospective analysis provides insights into the pathophysiology and diagnosis of spontaneous CSF otorrhoea, particularly regarding the presence of bilirubin in middle ear fluid and the utility of glucose level as a possible biomarker to predict patient outcomes. This paper contributes to the expanding body of knowledge surrounding spontaneous CSF otorrhoea and suggests pathways for refining diagnostic criteria, enhancing prognostication, and optimising clinical management. Our findings, in line with current literature, underscore the importance of validated biomarker levels in CSF otorrhoea diagnosis and treatment. Future research, including well-powered, prospective studies, is essential to build upon these insights and advance care for patients facing this unique clinical challenge.

Acknowledgments

The authors would like to thank PathWest Laboratory Medicine of Western Australia, specifically medical scientists John Joseph, Samuel Abbs and Monika Zacharek for their expertise.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://www.theajo.com/article/view/10.21037/ajo-25-10/rc

Data Sharing Statement: Available at https://www.theajo.com/article/view/10.21037/ajo-25-10/dss

Peer Review File: Available at https://www.theajo.com/article/view/10.21037/ajo-25-10/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://www.theajo.com/article/view/10.21037/ajo-25-10/coif). J.K. serves as an unpaid editorial board member of Australian Journal of Otolaryngology from January 2019 to December 2027. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study has received institutional ethics approval from the Sir Charles Gairdner Hospital under protocol number HoD 50556 SCGOPHCG. Patient confidentiality was maintained by anonymising data, and individual consent was waived for this retrospective analysis.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. de Almeida SM, Castoldi JR, Riechi SC. Comparison between cerebrospinal fluid biomarkers for differential diagnosis of acute meningitis. Diagnosis (Berl) 2023;10:298-308. [Crossref] [PubMed]
2. Bachmann-Harildstad G. Diagnostic values of beta-2 transferrin and beta-trace protein as markers for cerebrospinal fluid fistula. Rhinology 2008;46:82-5.
3. Boo SH, Goh YB, Han CS. Repair of spontaneous cerebrospinal fluid otorrhea from defect of middle cranial fossa. Korean J Audiol 2013;17:148-51. [Crossref] [PubMed]
4. Lobo BC, Baumanis MM, Nelson RF. Surgical repair of spontaneous cerebrospinal fluid (CSF) leaks: A systematic review. Laryngoscope Investig Otolaryngol 2017;2:215-24. [Crossref] [PubMed]
5. Pan DW, Vanstrum E, Doherty JK. Idiopathic Intracranial Hypertension: Implications for the Otolaryngologist. Otolaryngol Clin North Am 2022;55:579-94. [Crossref] [PubMed]
6. Schlosser RJ, Bolger WE. Nasal cerebrospinal fluid leaks: critical review and surgical considerations. Laryngoscope 2004;114:255-65. [Crossref] [PubMed]
7. Son HJ, Karkas A, Buchanan P, et al. Spontaneous cerebrospinal fluid effusion of the temporal bone: repair, audiological outcomes, and obesity. Laryngoscope 2014;124:1204-8. [Crossref] [PubMed]
8. Allen KP, Perez CL, Kutz JW, et al. Elevated intracranial pressure in patients with spontaneous cerebrospinal fluid otorrhea. Laryngoscope 2014;124:251-4. [Crossref] [PubMed]
9. Almela MT, Navarro-Zaragoza J, Laorden ML, et al. Cut-off value for β-trace protein (β-TP) as a rapid diagnostic of cerebrospinal fluid (CSF) leak detection. Laryngoscope Investig Otolaryngol 2023;8:1233-9. [Crossref] [PubMed]
10. Hendriks T, Thompson A, Boeddinghaus R, et al. Radiological findings in spontaneous cerebrospinal fluid leaks of the temporal bone. J Laryngol Otol 2021;135:403-9. [Crossref] [PubMed]
11. Quimby AE, De Ravin E, Eliades SJ, et al. Meningitis Risk and Role of Prophylactic Antibiotics in Spontaneous Lateral Skull Base CSF Leaks. Ann Otol Rhinol Laryngol 2023;132:1600-9. [Crossref] [PubMed]
12. Spinos D, Geropoulos G, Vavoulis G, et al. The Association Between Obesity and Spontaneous Temporal Bone CSF Leak Outcomes: A Systematic Review and Meta-Analysis. Laryngoscope 2024;134:2012-8. [Crossref] [PubMed]
13. Thomeer HG, Schreurs C, van Doormaal TP, et al. Management and Outcomes of Spontaneous Cerebrospinal Fluid Otorrhoea. Front Surg 2020;7:21. [Crossref] [PubMed]
14. Kari E, Mattox DE. Transtemporal management of temporal bone encephaloceles and CSF leaks: review of 56 consecutive patients. Acta Otolaryngol 2011;131:391-4. [Crossref] [PubMed]
15. LeVay AJ, Kveton JF. Relationship between obesity, obstructive sleep apnea, and spontaneous cerebrospinal fluid otorrhea. Laryngoscope 2008;118:275-8. [Crossref] [PubMed]
16. Rabbani C, Saltagi MZ, Ye MJ, et al. Association of Obstructive Sleep Apnea With Calvarial and Skull Base Thinning. JAMA Otolaryngol Head Neck Surg 2018;144:513-8. [Crossref] [PubMed]
17. Iliff JJ, Goldman SA, Nedergaard M. Implications of the discovery of brain lymphatic pathways. Lancet Neurol 2015;14:977-9. [Crossref] [PubMed]
18. Kenning TJ, Willcox TO, Artz GJ, et al. Surgical management of temporal meningoencephaloceles, cerebrospinal fluid leaks, and intracranial hypertension: treatment paradigm and outcomes. Neurosurg Focus 2012;32:E6. [Crossref] [PubMed]
19. Schlosser RJ, Woodworth BA, Wilensky EM, et al. Spontaneous cerebrospinal fluid leaks: a variant of benign intracranial hypertension. Ann Otol Rhinol Laryngol 2006;115:495-500. [Crossref] [PubMed]
20. Cheng E, Grande D, Leonetti J. Management of spontaneous temporal bone cerebrospinal fluid leak: A 30-year experience. Am J Otolaryngol 2019;40:97-100. [Crossref] [PubMed]
21. Gacek RR. Arachnoid granulation cerebrospinal fluid otorrhea. Ann Otol Rhinol Laryngol 1990;99:854-62. [Crossref] [PubMed]
22. Baranyi A, Amouzadeh-Ghadikolai O, Lewinski DV, et al. Beta-trace Protein as a new non-invasive immunological Marker for Quinolinic Acid-induced impaired Blood-Brain Barrier Integrity. Sci Rep 2017;7:43642. [Crossref] [PubMed]
23. Bernasconi L, Pötzl T, Steuer C, et al. Retrospective validation of a β-trace protein interpretation algorithm for the diagnosis of cerebrospinal fluid leakage. Clin Chem Lab Med 2017;55:554-60. [Crossref] [PubMed]
24. Durand ML, Calderwood SB, Weber DJ, et al. Acute bacterial meningitis in adults. A review of 493 episodes. N Engl J Med 1993;328:21-8. [Crossref] [PubMed]
25. Rao N, Redleaf M. Spontaneous middle cranial fossa cerebrospinal fluid otorrhea in adults. Laryngoscope 2016;126:464-8. [Crossref] [PubMed]
26. Savva A, Taylor MJ, Beatty CW. Management of cerebrospinal fluid leaks involving the temporal bone: report on 92 patients. Laryngoscope 2003;113:50-6. [Crossref] [PubMed]
27. Hendriks T, Bala A, Kuthubutheen J. Spontaneous cerebrospinal fluid leaks of the temporal bone - clinical features and management outcomes. Auris Nasus Larynx 2022;49:26-33. [Crossref] [PubMed]
28. Hendriks T, Bala A, Rodrigues S, et al. Spontaneous cerebrospinal fluid (CSF) otorrhoea in Western Australia—an emerging entity? Aust J Otolaryngol 2019;2:4.
29. Prosser JD, Vender JR, Solares CA. Traumatic cerebrospinal fluid leaks. Otolaryngol Clin North Am 2011;44:857-73. vii. [Crossref] [PubMed]
30. Schnabel C, Di Martino E, Gilsbach JM, et al. Comparison of beta2-transferrin and beta-trace protein for detection of cerebrospinal fluid in nasal and ear fluids. Clin Chem 2004;50:661-3. [Crossref] [PubMed]
31. Eide JG, Mason W, Ray A, et al. Systematic review of errors on beta-2 transferrin gel electrophoresis testing of rhinorrhea and otorrhea. Int Forum Allergy Rhinol 2024;14:1016-25. [Crossref] [PubMed]
32. Lafer I, Michaelis S, Schneider C, et al. Beta-trace protein concentrations at the blood-cerebrospinal fluid barrier - acute phase affects protein status. EXCLI J 2021;20:1446-52. [Crossref] [PubMed]
33. Meco C, Oberascher G, Arrer E, et al. Beta-trace protein test: new guidelines for the reliable diagnosis of cerebrospinal fluid fistula. Otolaryngol Head Neck Surg 2003;129:508-17. [Crossref] [PubMed]
34. Reiber H, Walther K, Althaus H. Beta-trace protein as sensitive marker for CSF rhinorhea and CSF otorhea. Acta Neurol Scand 2003;108:359-62. [Crossref] [PubMed]
35. Bachmann-Harildstad G, Müller R, Michel O. Beta-trace protein in pediatric otitis media with effusion. Int J Pediatr Otorhinolaryngol 2014;78:659-62. [Crossref] [PubMed]
36. Michel O, Bamborschke S, Nekic M, et al. Beta-trace protein (prostaglandin D synthase)--a stable and reliable protein in perilymph. Ger Med Sci 2005;3:Doc04.
37. Tumani H, Reiber H, Nau R, et al. Beta-trace protein concentration in cerebrospinal fluid is decreased in patients with bacterial meningitis. Neurosci Lett 1998;242:5-8. [Crossref] [PubMed]
38. Tan QC, Xing XW, Zhang JT, et al. Correlation between blood glucose and cerebrospinal fluid glucose levels in patients with differences in glucose metabolism. Front Neurol 2023;14:1103026. [Crossref] [PubMed]
39. Marcolini E, Hine J. Approach to the Diagnosis and Management of Subarachnoid Hemorrhage. West J Emerg Med 2019;20:203-11. [Crossref] [PubMed]
40. Amdisen A, Glud V. No influence from topical application of EMLA cream before blood sampling on routine clinical chemistry and haematology measurements. Eur J Clin Pharmacol 1991;41:619-20. [Crossref] [PubMed]
41. Yang Z, Apiliogullari S, Fu Y, et al. Association between Cerebrospinal Fluid and Serum Biomarker Levels and Diagnosis, Injury Severity, and Short-Term Outcomes in Patients with Acute Traumatic Spinal Cord Injury. Diagnostics (Basel) 2023;13:1814. [Crossref] [PubMed]
42. Li H, Qiu S, Li X, et al. Autophagy biomarkers in CSF correlates with infarct size, clinical severity and neurological outcome in AIS patients. J Transl Med 2015;13:359. [Crossref] [PubMed]
43. Xiong X, Zhang L, Li Y, et al. Calcium Channel Subunit α2δ-1 as a Potential Biomarker Reflecting Illness Severity and Neuroinflammation in Patients with Acute Ischemic Stroke. J Stroke Cerebrovasc Dis 2021;30:105874. [Crossref] [PubMed]
44. Naik A, Adeleye O, Koester SW, et al. Cerebrospinal Fluid Biomarkers for Diagnosis and the Prognostication of Acute Ischemic Stroke: A Systematic Review. Int J Mol Sci 2023;24:10902. [Crossref] [PubMed]
45. Hjalmarsson C, Bjerke M, Andersson B, et al. Neuronal and glia-related biomarkers in cerebrospinal fluid of patients with acute ischemic stroke. J Cent Nerv Syst Dis 2014;6:51-8. [Crossref] [PubMed]
46. Stukas S, Cooper J, Gill J, et al. Association of CSF and Serum Neurofilament Light and Glial Fibrillary Acidic Protein, Injury Severity, and Outcome in Spinal Cord Injury. Neurology 2023;100:e1221-33. [Crossref] [PubMed]
47. Guasp M, Muñoz-Sánchez G, Martínez-Hernández E, et al. CSF Biomarkers in COVID-19 Associated Encephalopathy and Encephalitis Predict Long-Term Outcome. Front Immunol 2022;13:866153. [Crossref] [PubMed]