Breech delivery,rupture of Batson’s plexus,T8 epidural hematoma,and paraplegia

Abbreviations

1 Introduction

Spinal cord injury in neonates following delivery is rare. When it occurs, it is acute and associated with risk factors such as breech extraction with head hyperextension (in utero star gazer) or with after coming head entrapment in the pelvis. When associated with cephalic presentation, the mechanism of spinal cord injury is usually from shoulder dystocia and head traction, or forceps-assisted rotation greater than 90 degrees resulting in torsion. The most common site of injury is the upper cervical cord, followed less frequently by the upper thoracic cord. To our knowledge, distal cord injuries involving the lower thoracic or lumbar cord have not been documented at birth. We present a case of a newborn with delayed onset paraplegia following frank breech delivery, found to have an epidural hematoma at T8 and a thin subdural hematoma from T7 to L2. Ultrasound imaging was consistent with spinal cord compression by a localized epidural hematoma. The epidural hematoma resulted in distal edema of the thoracolumbar and conus regions. This report highlights the potential susceptibility of the newborn’s distal spinal cord to birth injury, particularly an epidural hemorrhage that is subacute. Early detection of symptomatic epidural hemorrhage may allow for emergency laminectomy and removal of a compressing hematoma. Since there are no prior reports of spontaneous spinal epidural hematomas (SSEH) in newborns, we assume that the pathophysiology is similar to adult and pediatric cases, i.e. that increased abdominal/thoracic pressure causes impedance of blood flow in the inferior vena cava, engorgement and rupture of the vertebral venous plexus (Batson’s plexus), and ultimately formation of an epidural hematoma [1].

2 Case report

A 2060 grams 33 weeks and six day gestation male infant was delivered from a 17 year old G1 mother. The infant’s head circumference was 31 cm, chest circumference was 26.5 cm, and length was 44 cm. The delivery was complicated by frank breech presentation. Although betamethasone was initiated on admission, due to a precipitous labor, a full course was not completed. Maternal vaginal culture was positive for group B streptococcus and chlamydia. The remaining prenatal labs, including hepatitis B, HIV, RPR, and gonorrhea were all negative and rubella titer was immune. Maternal urine drug screen was positive for tetrahydrocannabinol. The pregnancy was complicated by a threatened abortion evaluated at six weeks gestation and by hypertension throughout the remainder of the pregnancy.

After admission the mother rapidly progressed to complete cervical dilation and consented to breech vaginal delivery with possible C-section in the event of fetal distress. The mother easily pushed the infant’s sacrum to +3 station. The infant’s left leg spontaneously delivered and the right leg was then flexed and easily delivered through the cervix and vaginal canal. The right shoulder was flexed and easily delivered; the infant was rotated and the left shoulder was reduced and easily delivered. The infant was delivered to the level of the scapula and the head was flexed and easily delivered. In short, the delivery was atraumatic and Apgar scores were 8 at 1 minute and 9 at 5 minutes. Initial physical examination at birth and on admission to the NICU was normal, including normal tone and movement of all extremities. The lower spine was noted to be intact. Due to prematurity and inadequately treated maternal group B streptococcus, the infant was admitted to the NICU to rule out sepsis and for intravenous fluids.

At 24 hours, the newborn was noted to be diffusely hypotonic. He exhibited no response in the lower extremities to painful stimuli and his lower limb reflexes were absent. The NICU team attributed the hypotonia to prematurity and anticipated improvement over the next 24 hours. However, at 48 hours, he continued to exhibit decreased movement of the lower extremities. He also had decreased urine output and a possible patulous anus. Therefore an atonic bladder was suspected. An AP and lateral radiograph of the lumbosacral spine showed normal alignment with no evidence of distraction, fracture, or subluxation.

Abdominal ultrasound at 48 hours documented a large urinary bladder. Ultrasound of the spinal cord (Fig. 1a and 1b) showed a T8 epidural hematoma immediately above and anterior to an edematous distal spinal cord that had loss of visualization of the central canal starting below the epidural hematoma and extending to the tip of the conus medullaris at L2. There was also a hypoechoic collection anterior to the cord running longitudinally along the spinal canal from T10 to T12 which was compatible with a subdural hematoma. Laboratory investigations revealed normal PT/PTT/INR, fibrinogen, and platelet count. A normal clinical bleeding time was observed following heel sticks. MRI of the spine (Figs. 2a, 3, 4) confirmed an epidural hematoma at T8 – note a focal lesion with no extension along the axis of the spinal cord. Fat is normally found in the spinal epidural space and makes extension of hemorrhage along the spinal cord minimal. MRI also confirmed distal spinal cord edema from T 9 to L2 and a subdural hematoma running longitudinally along the axis of the spinal cord from T7 to L 2. Note that the subdural collection of blood was better visualized on the MRI while the epidural and loss of central canal definition was better visualized on the ultrasound.

OPEN IN VIEWER

Fig.1

a Longitudinal ultrasound image of the spine shows a focal epidural hemorrhage at T8/T9 which contours the spinal cord. A subdural hemorrhage is seen starting at T12 and running inferiorly along the spinal canal. Notice the abrupt termination of the central canal with spinal cord swelling which starts at T9 and extends through the rest of the cord. b. Longitudinal ultrasound image of the spine shows a more detailed image of the focal epidural hemorrhage at T8/T9.

OPEN IN VIEWER

Fig.2

a T2 weighted image from an axial MRI of the spine shows dark epidural hemorrhage with spinal cord edema and a biconvex shape (lens shape) of the extra-axial fluid collection typical for epidural blood. b Illustration showing epidural space limited by periosteum and dura. Within the epidural space is the vertebral venous plexus and fat deposition – with rupture of the venous plexus, blood accumulates as a space occupying epidural hematoma.

OPEN IN VIEWER

Fig.3

T2 weighted image from a sagittal MRI of the spine shows dark epidural hemorrhage with spinal cord edema. Notice the focal nature of the extra-axial fluid collection typical for epidural blood.

OPEN IN VIEWER

Fig.4

T2 weighted image from a sagittal MRI of the spine shows dark subdural hemorrhage. Notice how the extra-axial fluid collection extends freely from superior to inferior typical for subdural blood.

Immediately following the MRI, the infant underwent a one level mid-lumbar laminectomy which drained both the epidural and subdural hematomas. The infant had an uneventful recovery. Discharge exam on day 12 showed some improvement in that infant had regained some hip flexion and slight flexion of knees, but no ankle movement. At one year of age, bladder function was normal, but there was no significant recovery of movement in hips, knees, or ankles.

Vaginal breech delivery poses increased risk to the fetus. One important risk is that of spinal cord injury [2]. Exploring this relationship requires an understanding of both the anatomic factors and the positional factors that predispose fetal spines - particularly premature fetal spines - to trauma at the time of delivery. Although intrapartum spinal cord injury was first fully outlined by Parrot in 1869, it was not until well into the 1900’s that the unique anatomic elements of the fetal spine had been more thoroughly defined. The prenatal and early post-natal spine is more prone to traumatic injury partly due to: high levels of elasticity in the para-spinal soft tissues, more horizontal facet orientation, underdeveloped spinal musculature, and disproportionate size of the fetal head in relation to the trunk [3]. This extreme elasticity in the spinal cord of newborns is thought to explain the mechanism behind most intrapartum spinal cord injuries. Any longitudinal (axial) force on the neonatal spine, during breech extraction in particular, risks subsequent injury and potential rupture of the cord [4, 5]. Neonatal cervical spinal cord is vulnerable during vaginal breech delivery, when the head is hyperextended (Star Gazer) and must flex suddenly to traverse the pelvis. If adequate flexion does not occur then the head will present with a larger circumference compared to the body, and any traction exerted on the body to try and free the after coming head risks injury – the site of which is generally in the low cervical or upper thoracic spinal cord. In addition, neonatal cervical spinal cord is vulnerable to torsion during cephalic presentation with forceps rotation of the head greater than 90 degrees such as what may happen when rotated from occiput posterior or transverse presentations [6]. The site of spinal cord injury in both breech traction and cephalic torsion is generally in the low cervical spinal cord and rarely involves the upper thoracic spinal cord. Therefore the traction or torsion injuries result in either tetraplegia or paraplegia. Thoracic spinal cord injury is less common than cervical spinal cord possibly due to the splinting reinforcement provided by the rib cage. In newborns, a delivery induced spinal cord lesion has not been reported in the distal thoracic spinal cord. To add to the complexity of this mechanism, most spinal cord injuries in newborns are undetectable by radiograph. Often there is no evidence of distraction, fracture, or subluxation, although the clinical deficits may be extreme. This perplexing syndrome was coined by Pang and Wilberger in 1982 and has since been referred to as spinal cord injury without radiographic abnormality (SCIWORA) [4]. Partly due to lack of radiographic evidence, the incidence and prevalence of spinal cord injury following breech delivery has proven difficult to determine. Estimates have ranged from 1 in 29,000 to 1 in 80,000 live births [5, 7]. However, today a newborn with acute paraplegia or quadriplegia indicating spinal cord injury would have an emergency neuroimaging (ultrasound and/or MRI) of the spine that would confirm the pathology. Such an infant before ultrasound and MRI could initially be classified classic SCIWORA but after the scans would be correctly labeled a spinal cord injury with neuroimaging abnormality. In other words, the classic SCIWORA has become more of historical interest then clinically relevant – and classic SCIWORA is extremely unlikely to occur in a newborn. An infant who has acute or subacute paraplegia occurring after birth would have MRI and/or ultrasound findings that would account for the paraplegia.

Our infant was different from prior cases of newborn paraplegia. Presentation was delayed for 24 hours, i.e. subacute. The spinal cord lesion was thoracolumbar. The lesion compressing the spinal cord was an epidural hematoma at T8. This is the first account of an acute spinal cord injury following vaginal breech delivery that involved the distal thoracic and lumbar spinal cord and was caused by an epidural hematoma. Unlike previously reported cases, our infant did not have a spinal cord lesion resulting from traction or torsion and did not demonstrate an acute clinical course immediately following a breech extraction. MRI revealed no ligamentous injury or cervical trauma. However, it did support the development of a T8 epidural hematoma and edematous spinal cord extending from T10 to L2. This case introduces an important new etiology of a newborn acute spinal cord injury occurring around birth. The infant’s symptoms were caused by a T8 epidural hematoma causing spinal cord compression and distal spinal cord edema.

No cases of SSEH have been reported to have occurred at birth. SSEH has been reported to have occurred in utero [7], in an infant as young as 4 months old [8], in children [8, 9], and in adults [1]. Cervicothoracic epidural is more common in children presumably because of the disproportionate weight ratio of the head to the body in the pediatric population and increased cervical spine mobility in the presence of undeveloped neck muscles in children. In adults, thoracolumbar epidural (T10– L2) is more common [1]. In adults the differential or associated conditions include: straining, coughing, lifting, breath-hold diving [10], vascular malformations, bleeding blood dyscrasias, leukemia, lymphoma, anticoagulants, antiplatelet drugs, hypertension, and post epidural anesthesia. In one study, it was estimated that 54 % of adults with SSEH had a straining associated event preceding the SSEH [11]. The most widely accepted hypothesis for the source of bleeding is the venous system, as spinal epidural veins are unprotected from changes in abdominal or thoracic pressure. Reviewing the anatomy of the vertebral venous plexus or Batson’s plexus (Fig. 2b) helps us to understand the most probable pathophysiology of SSEH.

The vertebral venous plexus is a highly anastomotic network of valve-less veins running along the entire length of the vertebral column from the foramen magnum to the sacral hiatus. Owing to its valve-less nature, the vertebral venous plexus is subject to distension in cases of increased intrathoracic or intraabdominal pressure. This subsequently causes a decrease in blood flow through the inferior vena cava. The vertebral venous plexus is outside of the dura and surrounds the spinal cord and drains into the inferior vena cava (Fig. 2b). The epidural space (lies between the dura and the periosteum of the vertebral bone) where blood could accumulate if the thin walled venous plexus were to leak. This epidural space is laden with fat deposits. Hence a SSEH does not spread along the axis of the cord but rather remains as a localized epidural hematoma that can compress the spinal cord.

Ultrasonography is the preferred initial imaging modality to evaluate spinal cord injury in the neonate. Incomplete ossification of the posterior vertebral arch allows easy visualization of the neonatal spinal cord by ultrasound [5]. This has proven useful in the initial detection of spinal cord injuries, particularly because it serves as a fast, inexpensive, and non-invasive, bedside investigation that allows for frequent re-evaluations. However, MRI has become an increasingly important component of neonatal imaging in cases where acute cord compression is a concern. This is due to its presumed superiority in delineating soft tissue structures. Blount et al. suggest that urgent MRI studies should be a more consistent component of our work-up in the care of post-breech hypotonic newborns [7]. However, in our case ultrasound gave better detail with respect to identifying which hemorrhage was epidural, i.e. the hematoma that was localized by fat deposition allowed it to compress the spinal cord at T8 anteriorly. The subdural hematoma dissipated in spinal fluid, easily spread along the axis of the cord, and was less likely to compress the spinal cord. Ultrasound also gave better detail with respect to loss of the central canal from cord edema just below the epidural hematoma. Decreasing the delay from time of infant symptoms to diagnostic imaging is critical in improving prognosis and recovery from a lesion that requires emergency laminectomy.

4 Conclusions

Our report is unique in that it describes a newborn with subacute onset of paraplegia following an atraumatic vaginal frank breech delivery. Our case had a level of cord damage that has not been reported in a newborn. In our infant the thoracolumbar level of cord damage was caused by a T8 epidural hematoma. SSEH has also not been reported in the newborn. Although injury at the level of the cervical thoracic spinal cord is more common secondary to traction or torsion of the spinal cord, our case demonstrates that the distal spinal cord injury in neonates is also susceptible to injury secondary to frank breech extraction but not through traction or torsion. The acute flexion of hips in a frank breech presentation and/or abdominal compression by the delivering attendant would result in increased abdominal/thoracic pressure. This results in decreased blood flow through the inferior vena cava resulting in engorgement and rupture of the valve-less and thin walled vertebral venous plexus (Batson’s plexus) resulting in an epidural hematoma. This focal epidural would accumulate at the point of rupture and not spread along the axis of the spinal cord since the epidural space limits axial spread due to fat deposits that occupy the epidural space. An epidural hematoma from a venous source would result in a slower onset cord compression. This delayed process allows for possible surgical decompression. Such an isolated distal spinal cord injury that presents subacutely with paraplegia should prompt an immediate ultrasound. If ultrasound demonstrates an epidural hematoma, then rapid decompression with or without MRI may prevent permanent neurological damage.

Disclosure statements

Financial disclosure: The authors have indicated they have no financial relationships relevant to this article to disclose.

Funding

No external funding.

Potential conflict of interest

The authors have indicated they have no potential conflicts of interest to disclose.