Displaced Cortical Vein Sign on CT: A Reliable Predictor to Distinguish Low Attenuation Subdural Collections from Benign Enlargement of Subarachnoid Spaces

doi:10.21203/rs.3.rs-4979219/v1

Download PDF

Research Article

Displaced Cortical Vein Sign on CT: A Reliable Predictor to Distinguish Low Attenuation Subdural Collections from Benign Enlargement of Subarachnoid Spaces

https://doi.org/10.21203/rs.3.rs-4979219/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Differentiating benign enlargement of subarachnoid spaces (BESS) from low attenuation subdural collections (SDC) on CT imaging can be challenging. The distinction between SDC and BESS is crucial in children, as SDC may raise the concern for non-accidental trauma (NAT).

Objectives

To evaluate the utilization of the displaced cortical vein sign on CT as a predictor of a pathological subdural collection confirmed by MRI and assesses the reproducibility of this finding between radiologists with different levels of clinical experience.

Methods and Materials:

A total of 104 imaging exams were independently reviewed by a junior and senior level radiologist (52 head CTs and 52 corresponding brain MRI exams). These exams included 43 MRI confirmed cases of SDC and 9 MRI confirmed cases of BESS from patients aged 0–2 years. The presence of pathological collections, sulci flattening, the displaced cortical vein sign, and the presence of thrombosed bridging veins were recorded for both reviewers along with attenuation and signal characteristics of the collections. Confirmed diagnosis of NAT was determined by chart review. The sensitivity, specificity, and inter-observer reliability were analyzed between reviewers.

Results

The average patient age was 6.4 months with a median age of 4 months. The sensitivity of the displaced cortical vein sign on CT for the presence of SDC confirmed by MRI was 79.0% and 69.6% for the two reviewers respectively. The specificity of this finding was 100% for both reviewers, with no false negative cases. Interobserver reliability was the highest among all features analyzed for the depiction of the displaced cortical vein sign on CT (κ = 0.63, 95% CI 0.45–0.82) and MRI (κ = 0.96, 95% CI 0.87-1.00). All cases where at least 1 reviewer noted the displaced cortical vein sign on CT and were later confirmed to be traumatic SDC on MRI were concluded to have high concern for NAT upon further chart review.

Conclusions

The displaced cortical vein sign on head CT is a specific and reproducible finding associated with subdural collections. Our data suggests that this sign is a more reliable and readily identifiable finding of pathological subdural collections compared to other traditional imaging findings, such as flattening of the cerebral sulci or bridging vein thrombosis. The displaced cortical vein sign when present on head CT should raise suspicion for NAT.

subdural collection

cortical vein

Computed Tomography

Magnetic Resonance Imaging

abusive head trauma

child abuse

non-accidental trauma

benign enlargement of subarachnoid spaces

infants

children

pediatric

Non-accidental trauma (NAT) is a critical problem that requires prompt and careful attention from all providers and individuals involved in the affected infant’s care. Mortality as a result of NAT occurs mainly from head trauma – a leading cause of pediatric death and disability in the United States [1–3]. Meta-analyses investigating outcomes in infants who have suffered abusive head trauma report mortality rates up to 20% among those aged 2 years and younger [4–6]. Infants who survive such insults may be left with debilitating neurological sequelae that significantly impact their quality of life ranging from tetraplegia and epilepsy, to blindness and memory/attention difficulties [4–7]. Missed diagnoses of NAT can also be catastrophic as the child has a high likelihood of suffering repeat insult that can cause further harm and death [8–12]. Therefore, accurate identification of children who may have sustained NAT is crucial to be able to connect victims to appropriate care and protective services.

Radiologists hold a pivotal role in raising suspicion for potential NAT [4, 13]. In addition to retinal hemorrhages and associated skeletal fracture patterns, a key finding that can indicate abusive head trauma is the presence of unexplained subdural collections (SDC) [4, 7, 14]. SDC result from the rupture or tearing of bridging veins due to a combination of abrupt acceleration, deceleration, shearing, and rotational forces [4, 15–17]. Although SDC may be present in accidental head trauma, these pathologic collections warrant further imaging studies along with evaluation by the child abuse team if no reliable history of trauma exists or if associated with other suspicious injuries. Even with the increased utilization of fast MRI trauma protocols by emergency departments, CT remains the most utilized method for initial evaluation of pediatric head trauma.¹⁸ SDC in the chronic stage appears as low attenuation collections on head CT and can mimic benign enlargement of the subarachnoid spaces (BESS), a self-limiting condition caused by CSF accumulation into the subarachnoid spaces without long-term neurologic or developmental effects (Fig. 1) [16, 19]. Distinguishing low attenuation SDC from BESS can be challenging, given the similarities in their appearance on CT imaging (Fig. 1) [15, 16, 20, 21].

The previously described displaced cortical vein sign may play a key role in this important distinction between BESS and pathological SDC on CT [22]. Cortical veins are located in the subarachnoid space and merge to form the bridging veins. The bridging veins cross the arachnoid and dura matter to drain into the superior sagittal sinus [23, 24]. As a result of a shearing force injury to the bridging veins, blood may collects in the subdural and subarachnoid spaces. When blood collects in the subdural space, the cortical veins that are located in the subarachnoid space may appear displaced by the SDC towards the brain parenchyma and away from the inner table of the skull [15, 16, 20, 25].Other findings that have been mentioned in existing literature to help identify pathological SDC include flattening of the cerebral sulci and bridging vein thromboses, which may be visible in addition to displaced cortical veins [7, 16, 17, 24, 26].

Our hypothesis is that the presence of the displaced cortical vein sign on CT is a reliable finding for the diagnosis of subdural collections. This study evaluates the utilization of the displaced cortical vein sign on head CT as a predictor of a pathological subdural collection confirmed by brain MRI and assesses the reproducibility of this finding between radiologists with different levels of clinical experience.

This project was reviewed and approved by the authors’ Institutional Review Board (IRB) prior to the commencement of study activities. No protected health information or patient identifiers were analyzed as part of the study efforts. All patient imaging studies were de-identified prior to study participants’ review.

A retrospective search of keywords from January 2018 to December 2023 was conducted in our institution’s Sectra PACS™ imaging system for patients between 0–2 years of age. The searched keywords included “subdural”, “subarachnoid”, “benign enlargement”, and “extra-axial” to identify cases of benign enlargement of the subarachnoid spaces or subdural collections. Only cases that had both CT and a corresponding MRI performed within a 2-week period to confirm a SDC or BESS were included. Cases with hyperdense SDC were excluded as these collections are not difficult to differentiate from BESS and are the same attenuation as the cortical vein which limits their visualization. A total of 104 imaging exams were included for final analysis – 52 head CTs and 52 corresponding MRIs.

After conducting a thorough literature review, a subset of key imaging findings relevant to the distinction between SDC and BESS were identified. These included: the presence of pathological SDC, the signal attenuation of the collections, the presence of sulci flattening, the visualization of the displaced cortical vein sign, and the presence of thrombosed bridging veins. Study limitations such as motion artifact and lack of multiplanar reconstruction were also documented. Additional relevant information recorded for each case included patient age at the time of the imaging exam, study indication, additional injuries on skeletal survey, and final child abuse team determination regarding the suspicion for NAT.

All imaging exams were independently and blindly reviewed by a junior radiologist (radiology fellow) and a senior-level radiologist (with 15 years of neuroimaging experience). These exams included 43 MRI confirmed cases of SDC and 9 MRI confirmed cases of BESS. In an effort to minimize recall bias, CT scans were reviewed first and on different dates from their corresponding MRI. Sensitivity, specificity, and interobserver reliability were analyzed between reviewers for imaging variables assessed on CT and MRI imaging. Inter-observer reliability was measured using the kappa statistic.

The average patient age of our study population was 6.4 months with a median age of 4 months. The most common study indication for performing initial head CT among our cohort was fall/head trauma. Other commonly listed indications for the imaging studies reviewed included worsening seizures, fullness of the anterior fontanelle, skull fracture, and concern for NAT.

Approximately 23.3% of cases confirmed to be SDC on MRI were found to have calvarial fractures associated with their presentation. Out of the SDC cases analyzed, 41.9% of cases had flattening of the sulci on head CT and 41.9% of cases had thrombosed bridging veins on MRI that were visualized by both reviewers. The sensitivity of the displaced cortical vein sign on head CT in identifying the presence of SDC confirmed by MRI was 79.0% and 69.6% for the senior and junior reviewers respectively. The specificity of this finding was 100% for both reviewers, with no false negative cases.

As shown in Table 1, the interobserver reliability was the highest for the depiction of the displaced cortical vein sign on CT (κ = 0.63, 95% CI 0.45–0.82) and MRI (κ = 0.96, 95% CI 0.87-1.00). There was low interobserver reliability for the visualization of sulci flattening on head CT (κ = 0.32, 95% CI 0.07–0.57) and MRI (κ = 0.56, 95% CI 0.33–0.79) (Table 1). Interobserver reliability was also low for identifying thrombosed bridging veins (κ = 0.31, 95% CI 0.10–0.53) on MRI (Table 1).

Table 1

Inter-Observer Reliability for Evaluated Imaging Findings
Imaging Finding	Modality Evaluated	Inter-Observer Reliability (κ)	95% Confidence Interval
Displaced Cortical Vein	CT	0.63	0.45–0.82
Displaced Cortical Vein	MRI	0.96	0.87-1.00
Flattening of Sulci	CT	0.32	0.07–0.56
Flattening of Sulci	MRI	0.56	0.33–0.79
Thrombosed Bridging Vein	MRI	0.31	0.10–0.53

Subsequent skeletal survey studies were performed in 91.7% of cases in which the displaced cortical vein sign was noted on initial head CT. The cases in which the displaced cortical vein sign was noted yet skeletal survey studies were not performed was due to low suspicion for child abuse from providers caring for the infant. Out of the cases where skeletal surveys were performed, 52.6% of infants had other associated injuries identified on imaging. The most common injuries found on skeletal survey in patients with confirmed SDC in this study were fractures of the calvarium and ribs, followed by the tibia, fibula, femur, and upper extremities including the clavicle, radius, ulna, humerus, and acromion process (Table 2). All cases where at least 1 reviewer noted the displaced cortical vein sign on initial head CT and were later confirmed to be SDC on MRI were concluded to have high concern for NAT upon review of documentation in the patient’s electronic medical record at the conclusion of the investigation.

Table 2

Additional Skeletal Injuries Associated with SDC Identified in Infants
– Skeletal Survey – Common Injuries Noted
• Calvarial Fracture (23.3%) • Rib Fracture(s) (23.3%) • Tibia Fracture (9.3%) • Fibula Fracture (7.0%) • Femur Fracture (4.7%) • Clavicle Fracture (4.7%) • Radius/Ulna/Humerus/Acromion Fracture (2.3%)

The displaced cortical vein sign was first described in 1992 by McCluney et al. as a finding associated with SDC on CT [22]. However, this sign has not been widely utilized and disseminated in the radiology community since its first description. Additionally, the false concept that the subdural compartment has no vessels adds to the difficulty and confusion while evaluating for chronic SDC [27]. It is true that the subdural space is a virtual space in the absence of pathological collections and has a paucity of vessels. However, it is important to remember that the bridging veins (which are the result of merging cortical veins), do in fact cross the subdural space to reach the superior sagittal sinus, and thus traverse the subdural space with or without a collection. The bridging veins, which are fewer in number, are located along the parasagittal vertex and are not displaced by a subdural collection (Figs. 2 and 3) [15,16,28.].

Identification of unexplained acute or chronic subdural collections in infants is imperative to ensure adequate protection for potential NAT victims and to connect them with appropriate care [4–6]. Missed cases of NAT can have significant consequences from repeat insult to mortality [4, 8–12]. Therefore, accurate assessment and rigorous attention must be given when reviewing imaging studies for characteristic findings suggestive of possible child abuse. This underscores the importance of the radiologist’s role in performing meticulous imaging interpretation to raise the concern for abuse and differentiating significant findings from conditions that may simulate NAT [4, 13].

In cases of both accidental and non-accidental head trauma, the shearing of bridging veins can cause the formation of subdural collections which can have serious or sometimes fatal consequences for infants [4, 7, 15–17]. Chronic SDC raises the concern for a prior insult and may be an additional imaging finding seen in infants who have suffered NAT. These chronic collections appear with low attenuation on CT and are similar to findings seen in BESS, making this differentiation extremely difficult [16, 19]. As both BESS and SDC present as prominent extra-axial spaces on CT, certain features like flattening of the sulci and bridging vein thrombosis have been described in the existing literature to help differentiate between the two conditions [7, 16, 17, 24, 26]. Our study closely investigated the often forgotten displaced cortical vein sign in association with subdural collections. These vessels, which are located in the subarachnoid space, are displaced away from the inner table of the skull by a pathological subdural collection and run parallel to the cortex [22]. We found that the displaced cortical vein sign is a reliable and reproducible finding for diagnosing pathological subdural collections on CT, with a sensitivity varying from 69.6–79% and specificity of 100%. Although there was a 10-year difference in clinical experience between the junior and senior radiologists readers, the interobserver reliability was the highest for the depiction of the displaced cortical vein sign on both CT (κ = 0.63, 95% CI 0.45–0.82) and MRI (κ = 0.96, 95% CI 0.87-1.00) when compared to other traditionally described findings like flattening of the cerebral sulci and bridging vein thrombosis [7, 16, 17, 24, 26]. This highlights the reproducibility of this important finding that can be a key differentiator between BESS and pathological SDC. To our knowledge, this is the first study characterizing the sensitivity and specificity of the displaced cortical vein sign and whether this finding may be recognized by radiologists of varying levels of expertise in practice.

Another important on further chart review for our patient cohort was that all cases where at least 1 reviewer had noted the displaced cortical vein sign on initial head CT were found to have high concern for NAT requiring further imaging and coordination of care. Considering the most common indication for initial head CT performed for affected patients in our study was fall/head trauma, this emphasizes the importance of being aware of the displaced cortical vein sign and its association with pathological chronic subdural collections so that this finding can be incorporated regularly into search patterns for relevant imaging studies. Other common indications for initial head CT in cases of confirmed pathological SDC were worsening seizures, skull fracture, and fullness of the anterior fontanelle. Macrocephaly in young children has also been discussed as an indication to consider traumatic SDC in the differential diagnosis of BESS [15, 20, 21].

It should be noted that skeletal survey studies were performed in over 90% of cases in which the displaced cortical vein sign was visualized, and in all cases where this finding was present on initial head CT in combination with clinical history that was suspicious for NAT. There is a vast amount of literature describing the many characteristic skeletal injuries that serve as radiological indicators for potential child abuse, which may guide healthcare providers to perform more thorough assessments to uncover evidence towards NAT [4, 14, 16, 24]. Among the skeletal surveys completed in our analyzed patient cohort, 52.6% of children had other associated injuries with the two most common being fractures of the calvarium and ribs. Having a sound understanding of the relevant anatomy in the formation of SDC and ability to recognize the cortical vein sign as a predictor of subdural extra-axial collection location may improve our ability to avoid missing cases of NAT in children who present with symptoms suggestive of isolated intracranial pathology and undergo head CT. The presence of this finding may even increase a radiologist’s confidence in discussing the possibility of NAT with other members of the healthcare team.

There are some limitations to this study. Both the junior and senior level radiologists who reviewed the imaging studies had specialized training and experience specific to the field of Pediatric Neuroradiology, and may have been more readily aware of the signs relevant to abusive head trauma on imaging. We also had a limited number of studies to review based on those available through our health system imaging records. Some studies that were included for analysis were of variable quality which may have affected the reviewer’s ability to accurately evaluate some of the collected imaging data. Identifying cases of BESS to include for review was challenging as this finding may be considered a normal variant and is not typically reported unless clinically relevant such as in cases of macrocephaly. Future directions to expand upon this work include conducting a similar study with radiology residents, as well as including a broader range of CT and MRI studies to review across various imaging systems. This would allow us to better study the identification of key imaging findings despite variation in level of training, prior experience, and imaging protocols or techniques. Including more cases of BESS to review alongside SDC would also improve the strength of the study. Additionally, some work has also been done in the use of other modalities like ultrasound in characterizing pericerebral fluid collections which could be an interesting area for further research in distinguishing BESS from SDC in young infants with open fontanelles [21, 28, 29].

Differentiating a pathological subdural collection from the benign self-limited enlargement of subarachnoid spaces is imperative to help identify children who may have sustained nonaccidental trauma, especially in cases where children may not have overt signs of potential child abuse. When present, the displaced cortical vein sign is a reliable and reproducible finding to confidently classify pathological low-attenuation subdural collections on CT. MRI serves as a helpful tool to further elucidate the presence or absence of a subdural collection in equivocal CT cases.

Author Contribution

D.B - Gathered data and wrote the manuscript, primary authorS.S and C.G - reviewed all the images, gathered data, guided manuscript writing, revised manuscriptW.P - Guided manuscript writing, revised manuscriptT.B - Statistical analysis, revised manuscript

Pinto PS, Poretti A, Meoded A, Tekes A, Huisman TA (2012). The unique features of traumatic brain injury in children. Review of the characteristics of the pediatric skull and brain, mechanisms of trauma, patterns of injury, complications and their imaging findings--part 1. J Neuroimaging 22(2):1–17. https://doi.org/10.1111/j.1552-6569.2011.00688.x
Dias MS (2004). Traumatic brain and spinal cord injury. Pediatr Clin North Ame 51(2):271–303. https://doi.org/10.1016/S0031-3955(03)00211-6
Keenan HT, Bratton SL (2006). Epidemiology and outcomes of pediatric traumatic brain injury. Dev Neurosci 28(4-5):256–263. https://doi.org/10.1159/000094152
Adamsbaum C, Méjean N, Merzoug V, Rey-Salmon C (2010). How to explore and report children with suspected non-accidental trauma. Pediatr Radiol 40(6): 932–938. https://doi.org/10.1007/s00247-010-1591-0
Wittschieber D, Karger B, Pfeiffer H, Hahnemann ML (2019). Understanding Subdural Collections in Pediatric Abusive Head Trauma. AJNR. Am J of Neuroradiol 40(3): 388–395. https://doi.org/10.3174/ajnr.A5855
Minns RA, Jones PA, Barlow KM (2002). Outcome and prognosis of nonaccidental head injury in infants. Clinics in Developmental Medicine (162):364-414.
Choudhary AK, Bradford R, Dias MS, Thamburaj K, Boal DKB (2015). Venous injury in abusive head trauma. Pediatr Radiol 45(12):1803-1813. https://doi:10.1007/s00247-015-3399-4
Pierce MC, Bertocci GE, Berger R, Vogeley E (2002). Injury biomechanics for aiding in the diagnosis of abusive head trauma. Neurosurgery Clinics of North America 13(2):155-168. https://doi:10.1016/S1042-3680(01)00006-7
Alexander R, Crabbe L, Sato Y, Smith W, Bennett T (1990). Serial abuse in children who are shaken. Am J Dis Child. 44(1):58-60. https://doi:10.1001/archpedi.1990.02150250068033
Ewing-Cobbs L, Prasad M, Kramer L, et al (2000). Acute neuroradiologic findings in young children with inflicted or noninflicted traumatic brain injury. Childs Nerv Syst 16(1):25-34. https://doi:10.1007/s003810050006
Ewing-Cobbs L, Kramer L, Prasad M, et al (1998). Neuroimaging, physical, and developmental findings after inflicted and noninflicted traumatic brain injury in young children. Pediatrics 102(2 Pt 1):300-307. https://doi:10.1542/peds.102.2.300
Jenny C, Hymel KP, Ritzen A, Reinert SE, Hay TC (1999). Analysis of missed cases of abusive head trauma. JAMA 281(7):621-626. doi:10.1001/jama.281.7.621
Kleinman PK (2008). “The roentgen manifestations of unrecognized skeletal trauma in infants”--a commentary. AJR Am J Roentgenol 190(3):559-560. https://doi:10.2214/AJR.07.3217
Fernando S, Obaldo RE, Walsh IR, Lowe LH. (2008). Neuroimaging of nonaccidental head trauma: pitfalls and controversies. Pediatr Radiol 38(8):827-838. https://doi:10.1007/s00247-007-0729-1
Orrù E, Calloni SF, Tekes A, Huisman TAGM, Soares BP (2018). The Child With Macrocephaly: Differential Diagnosis and Neuroimaging Findings. American Journal of Roentgenology 210(4):848-859. https://doi:10.2214/AJR.17.18693
Raissaki M, Adamsbaum C, Argyropoulou MI, et al (2023). Benign enlargement of the subarachnoid spaces and subdural collections—when to evaluate for abuse. Pediatr Radiol 53(4):752-767. https://doi:10.1007/s00247-023-05611-y
Zahl SM, Andersson J, Wester K, Wikström J (2023). Neuroradiological findings in children with subdural hematoma and suspected abusive head trauma. Annals of the Child Neurology Society (1):44-52. https://doi:10.1002/cns3.3
Kessler BA, Goh JL, Pajer HB, et al (2021). Rapid-sequence MRI for evaluation of pediatric traumatic brain injury: a systematic review. J Neurosurg Pediatr 28(3):278-286. https://doi:10.3171/2021.2.PEDS20852
Greiner MV, Richards TJ, Care MM, Leach JL (2013). Prevalence of Subdural Collections in Children with Macrocrania. AJNR Am J Neuroradiol 34(12):2373-2378. https://doi:10.3174/ajnr.A3588
Wilms G, Vanderschueren G, Demaerel PH, et al (1993). CT and MR in infants with pericerebral collections and macrocephaly: benign enlargement of the subarachnoid spaces versus subdural collections. AJNR Am J Neuroradiol 14(4):855-860.
Caré MM. (2021). Macrocephaly and subdural collections. Pediatr Radiol 51(6):891-897. https://doi:10.1007/s00247-020-04848-1
McCluney KW, Yeakley JW, Fenstermacher MJ, Baird SH, Bonmati CM (1992). Subdural hygroma versus atrophy on MR brain scans: “the cortical vein sign.” AJNR Am J Neuroradiol 13(5):1335-1339.
Uddin MA, Haq TU, Rafique MZ (2006). Cerebral venous system anatomy. J Pak Med Assoc 56(11):516-519.
Adamsbaum C, Rambaud C (2012). Abusive head trauma: don’t overlook bridging vein thrombosis. Pediatr Radiol 42(11):1298-1300. https://doi:10.1007/s00247-012-2434-y
Wittschieber D, Karger B, Niederstadt T, Pfeiffer H, Hahnemann ML (2015). Subdural Hygromas in Abusive Head Trauma: Pathogenesis, Diagnosis, and Forensic Implications. AJNR Am J Neuroradiol 36(3):432-439. https://doi:10.3174/ajnr.A3989
Zimmerman RA, Bilaniuk LT, Farina L (2007). Non-accidental brain trauma in infants: diffusion imaging, contributions to understanding the injury process. Journal of Neuroradiology 34(2):109-114. https://doi:10.1016/j.neurad.2007.01.124
Chen CY, Chou TY, Zimmerman RA, Lee CC, Chen FH, Faro SH (1996). Pericerebral fluid collection: differentiation of enlarged subarachnoid spaces from subdural collections with color Doppler US. Radiology 201(2):389-392. https://doi:10.1148/radiology.201.2.8888229
Veyrac C, Couture A, Baud C (1990). Pericerebral fluid collections and ultrasound. Pediatr Radiol 20(4):236-240. doi:10.1007/BF02019655
De Vries LS, Smet M, Ceulemans B, et al (1990). The role of high resolution ultrasound and MRI in the investigation of infants with macrocephaly. Neuropediatrics 21(2):72-75. https://doi:10.1055/s-2008-1071464

No competing interests reported.

Download PDF

Editorial decision: Revision requested
11 Oct, 2024
Reviews received at journal
15 Sep, 2024
Reviews received at journal
15 Sep, 2024
Reviews received at journal
03 Sep, 2024
Reviewers agreed at journal
03 Sep, 2024
Reviewers agreed at journal
03 Sep, 2024
Reviewers agreed at journal
30 Aug, 2024
Reviewers invited by journal
28 Aug, 2024
Editor assigned by journal
27 Aug, 2024
Submission checks completed at journal
27 Aug, 2024
First submitted to journal
26 Aug, 2024

You are reading this latest preprint version

Displaced Cortical Vein Sign on CT: A Reliable Predictor to Distinguish Low Attenuation Subdural Collections from Benign Enlargement of Subarachnoid Spaces

Status:

Version 1

Abstract

Background

Objectives

Methods and Materials:

Results

Conclusions

Figures

Introduction

Materials and Methods

Results

Discussion

Conclusion

Declarations

Author Contribution

References

Additional Declarations

Status:

Version 1