Vascular loops mimicking herniated lumbar discs: fair warning

doi:10.21203/rs.3.rs-3987610/v1

Objective: The authors report their experience with twenty-one consecutive patients who presented with symptoms and imaging characteristics of a herniated lumbar disc; of whom, at the time of surgery had a vascular loop instead.

Methods The procedure was performed on 14 women and seven men with a mean age of 39 years. Clinical complaints included lumbar aching with one limb overt radiculopathy in all patients; with additional sphincter dysfunction in two cases. Symptoms had developed within a mean time period of three months. In all patients, the disc was exposed through an L5-S1 (n=10); L4-L5 (n=5) and L3-L4 (n=6) open minimal laminotomy. In 16 patients, rather than a herniated disc they had a lumbar epidural varix, while anarterio-venous fistula was found in the remaining five cases. In all cases, the vascular disorder was resected and its subjacent disc was left intact. One patient had a postoperative blood transfusion. While the radiculopathy dysfunction improved in all patients, four patients reported lasting lumbar pain following surgery. The postoperative imaging confirmed the resolution of the vascular anomaly and an intact disc. The mean length of the follow-up period was 47 months.

Conclusions Either epidural varix or arterio-venous fistula in the lumbar area may mimic a herniated disc on imaging studies. With the usual approach they can be operated safely. Resection of the anomaly can be sufficient for alleviating radiculopathy symptoms.

Epidural Varix

Arterio-Venous Fistula

Epidural

Lumbar

Herniated disc

Microsurgical

Discectomy

Lumbar radicular pain is usually caused by compressive pathology such as lumbar disc herniation. Some of the differential diagnoses are tumor pathology, abscess, hematomas and synovial cyst [16,24]. But in a few cases, it can be produced by vascular loop of the region which can mimic the symptoms produced by lumbar disc herniation [7,9,15]. Lumbar epidural varices (LEV) result from the dilatation of the internal vertebral venous plexus, and in the intervertebral foramina causes spinal canal narrowing, compression and consequently irritation of the nerve roots [15,20,22,23].

In the case of spinal dural arteriovenous fistulas (SDAVF), there is an abnormal vascular dural communication between the dural branch of a segmental artery and a subdural radicular vein that drains the perimedullary venous system. This leads to venous hypertension and congestive myelopathy depending on their location; if left untreated these loops can lead to serious morbidity and cause irreversible disability [2,8,9,19].

In light of the above, this study aimed to present the experience of the authors and surgical treatment performed in a series of 21 consecutive patients with overt radiculopathy and imaging characteristics of a herniated disc. Either a lumbar LEV or SDVAF, of which in no case was hernia or extrusion of the vertebral disc concomitantly found which could justify the symptoms, were disclosed during surgery.

We performed a retrospective, analytical, multicentric, cohort study. We included all patients older than 16 years with a magnetic resonance imaging (MRI) diagnosis of a lumbar herniated disc of whom was found a vascular loop (LEV and SDAVF), and no evidence of disc herniation at the time of surgery. The study was carried out in three centres: The Virgen Arrixaca University Hospital, Murcia Spain, Quiron Hospital, Torrevieja Spain and Giovanni Addolorata Hospital, Rome Italy, in the period of time between 2015 to 2022. IRB/ethics committee approval was obtained accordingly in each centre. Each patient signed a consent form.

Patients with incomplete data in the electronic medical record, missing patient consent forms, or patients with a follow-up time of fewer than 12 months were excluded. We collect data on gender, age at the time of surgery, symptoms and signs and their time of evolution until reaching the consultation. Preoperative lumbar MRI was evaluated to document the side and extent of the lesion. Since 2019 we routinely indicate postoperative lumbar MRI during admission. Preoperative lower limb EMG was found in half of the patients.

The patients were classified into two groups. The first group was made up of patients with an intraoperative diagnosis of LEV, within whom the presence or absence of thrombosis was discriminated; the second group was made up of patients with SDAVF, in the latter group was also classified according to the presence of thrombi within the malformation.

All the patients were treated surgically with an open minimal laminotomy at the indicated lumbar space. No percutaneous or endoscopic approach was used. Postoperative complications (up to 30 days after surgery) such as lumbar or wound infections, or new associated neuro deficit were recorded. Associated hemorrhage or other finding was documented on postoperative MRI. Categorical variables were presented as absolute frequency and percentage; continuous variables are presented as median since the sample was considered not normally distributed due to the limited number of patients.

Twenty-one patients (14 women and seven men) ranging in age from 21 to 72 years (mean 39 years) underwent surgery for lumbar microdiscectomy, of whom at the time of the exposure of the disc space, they had an overt vascular anomaly and no intraoperative evidence of a herniated disc in the correspondent lumbar disc space.

Inclusion criteria included a lumbosacral MRI which revealed imaging characteristics of a herniated disc at the level of L5-S1 (n = 10) or L4-L5 (n = 5) and L3-L4 (n = 6) and the clinical presentation of radicular symptoms, lumbosacral pain, or perineal pain, association with symptoms of sphincter dysfunction. Three patients had a previous history of more than two years of previous lumbar disc surgery, and it was assumed as a recurrent disc herniation. The intensity and quality of pain had to be severe aching (n = 10) or heavy (n = 6) or both (n = 5), as assessed by the Short-form McGill Pain Questionnaire (SF-MPQ). Obesity defined by a body mass index (BMI) over 30 was recorded in 8 patients, and smoking was found in 7 cases.

All cases were analyzed retrospectively and the characteristics found in this cohort are presented in Table 1. A Lumbar Epidural Varix (LEV) was diagnosed in sixteen patients. The median age of this group was 45 years and four were male. Signs and symptoms at diagnosis were radicular symptoms (n = 14), and lumbosacral pain (n = 12), associated with symptoms of sphincter dysfunction (n = 2). Symptoms developed over a mean time of twelve months with notable worsening in the last three months.

Patients were scheduled for elective microdiscectomy. Intraoperative, a Lumbar Epidural Varix with a thrombosed appearance was found in four cases, characterized by hardened consistency and primarily venous bleeding. In the remaining 12, the LEV appeared non thrombosed while in these, the varix was found softer and with a continuous venous bleed. LEV cases had no issues with intraoperative bleeding control (Figs. 2, 4, 6, 7).

During disc exposure, other five patients presented SDAVF. The median age was 37 years and three were male. Signs and symptoms at diagnosis were radicular symptoms (n = 5), and lumbosacral pain (n = 3) over a mean time of six months. Excruciating pain was experienced in three patients, while other characteristics are presented in Table 1.

Two patients presented thrombosis within the SDAVF, characterized by a hardened consistency of the reddish web with predominantly bleeding of the arterial component. In these cases, better control of bleeding was achieved during the surgical procedure. The remaining three patients in this group did not present thrombosis characterized by a softer consistency of a bright reddish web (Figs. 1, 5, 7). Intraoperative bleeding, mainly arterial, was found to be difficult to control in most of these cases. One male patient presented with intraoperative serious bleeding and required a blood transfusion in the postoperative period.

In the two groups studied, a relatively intact disc was observed. There was no blunt evidence of disc herniation or extruded disc. A bulging disc was found on half of the cases. Certainly, mild to moderate spondylosis was found in half of the cases, but not in direct association with the involved nerve root. Early postoperative follow-up was characterized in all cases with immediate relief of radicular symptoms. No added nerve root deficit was disclosed. Other than a minor wound infection in two cases, no other complications were recorded.

The initial postoperative lumbar angio-CAT scan confirmed no evidence of vascular anomaly and an intact disc. Since 2019, we indicate lumbar MRI with contrast for the initial follow-up in these cases during hospitalization. The mean length of the follow-up period was 47 months (range = 33–96 mos), and while all resolved their radicular and sphincter symptoms, four referred to persistent lower back pain. Postoperative MRI confirmed the absence of overt lumbar disc pathology.

In this retrospective, observational study, we report a surgical cohort of spinal vascular anomalies, in patients initially diagnosed with clinical and imaging characteristcs of a herniated lumbar disc. All cases included either a LEV or SDAF with or without inherent thrombosis at the time of the surgical resection.

Vascular anomalies of the spine and spinal cord comprise a rare but an important condition, which includes spinal cord ischemia of various etiologies and a spectrum of vascular malformations. With regard to the spinal dura, which represents a topographical marker for localization of spinal vascular malformations, these lesions can be found as Spinal Extradural AVMs or Paraspinal Arteriovenous Shunts. Both are very rare lesions characterized by a direct fistulous connection between a segmental artery with the majority occurring at the thoracic level. They may drain through a radicular vein into the perimedullary venous system causing significant venous ectasias with cord compression and venous congestion of the spinal cord. In addition, the arterial steal phenomenon may occur in cases with a high-volume arteriovenous shunt [9,19]. The Spinal Dural Arteriovenous Fistula or SDAVFs, also known as type I, are the most common spinal vascular malformation, with some estimates ranging as high as 80% of the total [17,21]. In contrast to spinal cord AVMs, SDAVFs are believed to be acquired lesions, possibly resulting from thrombosis of the extradural venous plexus. Evidence for their acquired nature includes their virtual absence in young patients and predominance in later life; the lack of association of SDAVF with other vascular anomalies; and their predilection for a location below the midthoracic spine, where increased venous pressure exists in a standing position [10,12]. Our findings support these conclusions, with all our cases located in the lumbar spine. It is important to note, that all our SDAVF lumbar cases presented with nerve root compression but no vascular stealing phenomena nor hemorrhage, except in one patient (Fig. 3). In these, brisk arterial bleeding was found intraoperatively, while doing the exposure of the disc. Dissection of the disc space revealed a vascular loop in the epidural space rather than a herniated disc. We assumed that the importance of the initial arterial bleeding, during exposure, was associated with the presence of an intravascular thrombus. The absence of thrombi made the hemorrhage brisker and rather profuse. Clinically, three patients presented with excruciating radicular pain, possibly in relation to subtle bleeding or thrombus formation.

On the other hand, our LEV cases, such as those previously reported [15,22,23], were characterized by profuse venous bleeding rather than oozing. Whether these lesions are solely varix or the late result of the natural evolution of SDAVF is a matter of controversy, and we can not give conclusive results. In four cases we found a thrombus within the venous dilation but, contrarily to SDAVF, this finding did not affect the intensity of venous bleeding. Most of LEV patients were middle aged. We can infer that the existence of thrombi, mainly in the SDAVF cases interferes with their bleeding condition at the time of surgery. As in one of our cases, the bleeding was so profuse that a blood transfusion was required in the early postoperative period. We found a slight trend of thrombosis in obese and smoker patients, but our sample is small to report significant results.

Radicular pain is a frequent reason for consulting a neurosurgeon and the pathology responsible for this symptom in 90% of cases is lumbar disc herniation [1]. Only on very rare occasions, this symptom corresponds to a vascular disorder. Spinal vascular conditions are rare, and, the fact that the symptoms it produces are nonspecific, there is usually a late diagnosis. Anatomical location is a key factor. In the lumbar area, as the clinical presentation and the characteristics of the imaging findings are similar, both in vascular disorders and in vertebral disc disease, an exhaustive analysis of the presurgical MRI is mandatory. Since the similarity between the pathologies could lead to an erroneous diagnosis [5,11,14], patients are frequently subjected to a variety of misdirected treatments, including surgery for stenosis [13]. Donghai et al. mention that misdiagnoses frequently occur in patients with SDAVF. In his series of patients, 83% of the cases suspected degenerative disc disease, myelitis, prostatic hyperplasia, intramedullary tumor, or other diseases. Age over 50 years, intramedullary length greater than 5 vertical segments, dilated perimedullary vessels, and subcervical injury were the four independent predictors that were significantly associated with SDAVF. Yet most of those cases were cervical, thoracic or both [4].

Our lumbar cases presented predominantly with non-hemorrhagic LEV. These lesions have an incidence of 0.067-1.2% [6]. Dilation of the venous system that passes through the intervertebral foramen causes root compression. It is important to note that although LEV does not present specific symptoms, it can be suspected in those people who have conditions that generate increased abdominal pressure, such as obesity, pregnancy or tumor mass. The latter can generate venous congestion and the ensuing dilation of the veins within the spinal canal [3]. As reported [18], the diagnosis of LEV is made at surgery, and coagulation and microsurgical decompression of the nerve root is adequate for the improvement of radiculopathy such as in our cases. Although spinal vascular disorders are rare, and overall the most common is the arteriovenous fistula, to our understanding whether to include LEV presenting with clinical radiculopathy is a matter of debate.

In this cohort the initial neurosurgical approach was performed for microdiscectomy surgery. Because the initial hypothetical diagnosis was disc pathology, once encountered a vascular loop, we proceed to coagulate and resect it, which were sufficient for the nerve decompression also with resolution or at least improvement of symptoms in nearly all cases. Given the limited number of patients, the rarity of this disorder, the retrospective and observational nature of this study, we are not allow to give blunt conclusions about this condition. Still, this is a relatively large series about this topic in the lumbar spine. With a standard surgical approach, the proper elimination of the varix leaving the disc intact is possible adding little or no morbidity.

We have presented a series of cases in which disc herniation was first thought of, but it was found intraoperatively a vascular loop. Either an epidural varix or arterio-venous fistula may mimic a herniated disc in imaging studies in the lumbar area. They can be operated safely with a proper approach. Resection of the anomaly can be sufficient for alleviating radiculopathy symptoms.

IRB/ethics committee approval was obtained accordingly in each centre: The Virgen Arrixaca University Hospital, Murcia Spain, Quiron Hospital, Torrevieja Spain and Giovanni Addolorata Hospital, Rome Italy, in the period of time between 2015 to 2022.

Funding “not applicable”

Availability of data and materials: upon request strictly for educational purposes

Equal Contribution Designation: All authors have contributed equally to this paper. All authors read and approved the final manuscript.

Details of Previous Publication: None.

Conflict of interest: None.

Informed consent: Each patient signed a consent form.

Author Contribution

I, Marcelo Galarza, certify that this manuscript is a unique submission and is not being considered for publication, in part or in full, with any other source in any medium. Equal Contribution Designation: All authors have contributed equally to this paper. All authors read and approved the final manuscript.I personally checked all patient information. Details of Previous Publication: None.Conflict of Interest: None.Disclosure of Funding: None. Informed consent: Each patient signed a consent form. Ethical approval: IRB/ethics committee approval was obtained accordingly in each centre.

Aljawadi A, Sethi G, Islam A, Elmajee M, Pillai A (2020) Sciatica Presentations and Predictors of Poor Outcomes Following Surgical Decompression of Herniated Lumbar Discs: A Review Article. Cureus 12(11), e11605. https://doi.org/10.7759/cureus.11605
Ando T, Yoshimura T, Kawakami O (2021) [Early Diagnosis of Spinal Dural Arteriovenous Fistula]. Brain Nerve 73(6):685–696. Japanese. 10.11477/mf.1416201819. PMID: 34127565
Bursali A, Akyoldas G, Guvenal AB, Yaman O (2016) Lumbar Epidural Varix Mimicking Disc Herniation. J Korean Neurosurg Soc 59(4):410–413. https://doi.org/10.3340/jkns.2016.59.4.410
Cao W, Huang L, Ge L, Lu G, Zhang X, Dong Q (2014) A simple score (AVFS) to identify spinal dural arteriovenous fistula before spinal digital subtraction angiography. J stroke Cerebrovasc Dis 23(8):1995–2000. https://doi.org/10.1016/j.jstrokecerebrovasdis.2014.01.021
Donghai W, Ning Y, Peng Z, Shuo X, Xueen L, Peng Z, Bin H, Xingang L (2013) The diagnosis of spinal dural arteriovenous fistulas. Spine 38(9):E546–E553. https://doi.org/10.1097/BRS.0b013e31828a38c4
Endres S (2011) Epidural varicosis as a possible cause of radicular pain: a case report. J Med Case Rep 5:537. https://doi.org/10.1186/1752-1947-5-537
Im IK, Son ES, Kim DH (2018) Lumbar Epidural Varix Causing Radicular Pain: A Case Report and Differential Diagnosis of Lumbar Cystic Lesions. PM & R: the journal of injury. function rehabilitation 10(11):1283–1287. https://doi.org/10.1016/j.pmrj.2018.04.002
Jellema K, Tijssen CC, van Gijn J (2006) Spinal dural arteriovenous fistulas: a congestive myelopathy that initially mimics a peripheral nerve disorder. Brain 129(12):3150–3164. https://doi.org/10.1093/brain/awl220
Jeng Y, Chen DY, Hsu HL, Huang YL, Chen CJ, Tseng YC (2015) Spinal Dural Arteriovenous Fistula: Imaging Features and Its Mimics. Korean J Radiol 16(5):1119–1131. https://doi.org/10.3348/kjr.2015.16.5.1119
Koch C, Gottschalk S, Giese A (2004) Dural arteriovenous fistula of the lumbar spine presenting with subarachnoid hemorrhage. Case report and review of the literature. J neurosurgery: spine 100:385–391. https://doi.org/10.3171/spi.2004.100.4.0385
Maimon S, Luckman Y, Strauss I (2016) Advances and technical standards in neurosurgery. 43:111–137. https://doi.org/10.1007/978-3-319-21359-0_5. Spinal Dural Arteriovenous Fistula: A Review
Marshman LA, David KM, Chawda SJ (2007) Lumbar extradural arteriovenous malformation: case report and literature review. spine J 7(3):374–379. https://doi.org/10.1016/j.spinee.2006.03.013
Morris JM (2012) Imaging of dural arteriovenous fistula. Radiol Clin North Am 50(4):823–839. https://doi.org/10.1016/j.rcl.2012.04.011
Muralidharan R, Saladino A, Lanzino G, Atkinson JL, Rabinstein AA (2011) The clinical and radiological presentation of spinal dural arteriovenous fistula. Spine 36(25):E1641–E1647. https://doi.org/10.1097/BRS.0b013e31821352dd
Pennekamp PH, Gemünd M, Kraft CN, von Engelhardt LV, Lüring C, Schmitz A (2007) Epidurale Varikose als seltene Ursache eines akuten radikulären Lumbalsyndroms mit kompletter Fussheber- und Fusssenkerparese–Kasuistik und Literaturübersicht [Epidural varicosis as a rare cause of acute radiculopathy with complete foot paresis–case report and literature review]. Z Orthop Ihre Grenzgeb 145(1):55–60. https://doi.org/10.1055/s-2007-960503
Pekindil G, Yalniz E (1997) Symptomatic lumbar foraminal epidural varix. Case report and review of the literature. Br J Neurosurg 11(2):159–160. https://doi.org/10.1080/02688699746546
Petrin Z, Marino RJ, Oleson CV, Simon JI, McCormick ZL (2020) Paralysis After Lumbar Interlaminar Epidural Steroid Injection in the Absence of Hematoma: A Case of Congestive Myelopathy Due to Spinal Dural Arteriovenous Fistula and a Review of the Literature. Am J Phys Med Rehabil 99(9):e107–e110. https://doi.org/10.1097/PHM.0000000000001325
Raghavendra V, Papanaik A, Kumar A, Ajith K (2014) Spinal Epidural Varices, a great Mimic of Intervertebral Disc Prolapse - A Case Series. J Orthop case Rep 4(4):3–5. https://doi.org/10.13107/jocr.2250-0685.212
Sato K, Terbrugge KG, Krings T (2012) Asymptomatic spinal dural arteriovenous fistulas: pathomechanical considerations. J Neurosurgery: Spine 16(5):441–446. https://doi.org/10.3171/2012.2.SPINE11500
Slin’ko EI, Al-Qashqish II (2006) Surgical treatment of lumbar epidural varices. J Neurosurgery: Spine 5(5):414–423. https://doi.org/10.3171/spi.2006.5.5.414
Sundbye FBG, Karabegovic S, Birkeland P (2018) [The spinal dural arteriovenous fistula is an underdiagnosed vascular malformation] [Article in Danish]. Ugeskr Laeger 16(29):V10170723
Tofuku K, Koga H, Yone K, Komiya S (2007) Spontaneous regression of symptomatic lumbar epidural varix: a case report. Spine 32(4):E147–E149. https://doi.org/10.1097/01.brs.0000255811.64922.47
Wong CH, Thng PL, Thoo FL, Low CO (2003) Symptomatic spinal epidural varices presenting with nerve impingement: report of two cases and review of the literature. Spine 28(17):E347–E350. https://doi.org/10.1097/01.BRS.0000090500.10184.7A
Zimmerman GA, Weingarten K, Lavyne MH (1994) Symptomatic lumbar epidural varices. Report of two cases. J Neurosurg 80(5):914–918. https://doi.org/10.3171/jns.1994.80.5.0914

TABLE 1

Vascular loops mimicking lumbar herniated discs: Patient characteristics.

Spinal dural arteriovenous fistulas (SDAVF); Lumbar epidural varices (LEV); * In months.
# HDS: Herniated disc surgery

Age & Sex	Vascular anomaly	Location	Thrombosis intraoperative finding	Clinical outcome	Followup *	Other features #
21, M	LEV	L5-S1	NO	resolution	47	smoker; obesity
27, F	SDAF	L4-L5	NO	resolution	45	obesity
41, F	LEV	L3-L4	NO	resolution	44
47, F	LEV	L4-L5	YES	improvement	30	obesity
53, M	LEV	L5-S1	NO	resolution	34	smoker
56, F	LEV	L3-L4	NO	resolution	46
31, M	SDAF	L5-S1	NO	resolution	57	previous HDS; smoker
57, F	LEV	L3-L4	NO	resolution	64
49, M	LEV	L4-L5	YES	improvement	87	obesity
40, M	LEV	L3-L4	NO	improvement	96	previous HDS; obesity
45, F	SDAF	L5-S1	YES	resolution	90
51, F	LEV	L5-S1	NO	resolution	44	smoker; obesity
46, F	LEV	L4-L5	YES	resolution	47	obesity
52, F	LEV	L5-S1	NO	resolution	77
49, F	LEV	L3-L4	NO	resolution	65	smoker
42, M	SDAF	L4-L5	YES	improvement	53	previous HDS; obesity
50, F	LEV	L5-S1	NO	improvement	33
32, M	SDAF	L3-L4	NO	improvement	77	blood transfusion smoker
50, F	LEV	L5-S1	NO	improvement	75

55, F	LEV	L5-S1	YES	improvement	42	smoker
72, F	LEV	L5-S1	NO	resolution	41

No competing interests reported.

Vascular loops mimicking herniated lumbar discs: fair warning

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

Results

Discussion

Conclusion

Declarations

Author Contribution

References

Tables

Additional Declarations

Status:

Version 1