Spinal Lipomas

Jeffrey P. Blount, MD, and Scott Elton, MD, Division of Neurosurgery, University of Alabama at Birmingham, Children's Hospital of Alabama, Birmingham, Alabama

Neurosurg Focus. 2001;10(1) 

In This Article

Surgical Treatment

Lipomas of the Conus Medullaris

The role of surgery in cases of lipomyelomeningocele remains controversial.[68] For the patient in whom symptoms or neurological deficits are present there is little controversy. Because of an unclear natural history, a significant incidence of perioperative neurological insults, postoperative complications (in some series), and a high rate of late neurological deterioration, some authors have been reluctant to advocate a prophylactic untethering procedure in the asymptomatic child.[28]

The natural history of untreated lipomyelomeningocele is imperfectly understood. Ethical issues preclude a methodologically perfect investigation in which, in a prospective manner, surgical resection and untethering is offered to one group of patients and denied to another matched group. In the absence of prospective data some consider it proper to pursue a conservative approach to spinal lipomas. Howeve r, indirect evidence obtained from multiple centers around the world supports the idea that pain, orthopedic deformity, and/or progressive neurological decline will develop in an asymptomatic child who harbors an untreated lipomyelomeningocele.[6,8,10.20,64] Kanev and Bierbrauer [30] have shown that neurological deterioration demonstrates a logarithmic relationship with advancing age. Further evidence culled from large series suggests that loss of neurological function is unlikely to be regained in the majority of cases.[10,27,40,60] Pain is quite amenable to surgical treatment even in adults; however, bladder function, once impaired, is not likely to be recovered.[11,22,23,27,30,60,72] Wu, et al.,[72] have found that the principal determinant of long-term bladder outcome was the patient's preoperative bladder status. Once neurologically related bladder dysfunction has occurred, the likelihood that normal bladder control will return is less than 20%.[30,46] As such, we support the opinion held by the majority of pediatric neurosurgeons that surgical exploration of lipomyelomeningoceles is warranted before neurological deterioration occurs.[6,30] The goals of such an exploration would include exposure of the plane between lipoma and spinal cord so that complete separation between the cord and the subcutaneous fibrofatty mass can occur and be followed by watertight closure of the dura and wound.

Surgical Technique and Approach

Following induction of general anesthesia and placement of appropriate intravenous lines and a Foley catheter, monitoring electrodes are placed in the muscles of the lower extremities to facilitate intraoperative nerve monitoring. The patient is then placed in the prone position, and pressure points are carefully padded. Following a routine surgical preparation, a linear midline incision is made that extends above and below the fatty lump on the back. After inferior and lateral dissection the lumbodorsal fascia can be identified circumferentially around the fatty mass. By working with blunt dissection (we find a Kittner dissector useful) lateral to medial along the surface of the lumbodorsal fascia the defect in the lumbo dorsal fascia, becomes evident. Occasionally the distal portion of the mass will blend with the muscular fibers of the gluteus maximus. The mass can often be significantly reduced at this point; however, it is critical to retain enough mass to maintain control of the lipoma.

Attention is then turned to the level above the defect, and the paraspinous muscle is separated on the midline. Monopolar cautery is used to isolate the first spinous process. A subperiosteal dissection is performed and self-retaining retractors are placed. Next, a laminectomy is performed to expose normal dura. Dorsal lipomas project through the lumbodorsal fascia and attach directly to the back of the descended conus medullaris. The nerve roots characteristically exit laterally or ventrally. Because caudal lipomas extend off of the distal end of the cord and envelop the exiting nerve roots in the substance of the lipoma, they may extend rostrally up the central canal of the cord for many levels. It is crucial to establish surgery-related goals clearly prior to beginning the procedure. Experience indicates that gross-total resection is not necessary and in fact carries an unacceptable risk of causing neurological injury. The primary surgery-related objective in the excision of a spinal lipoma is to remove enough of the mass so that normal anatomical planes may be surgically reconstructed. Some authors have advocated use of the CO2 laser for this purpose because it selectively dissolves fatty tissue with a minimum of contact or retraction., [3,46] We have found the ultrasonic aspirator to be a very useful for the internal debulking of these lesions (Video Clips 1-11). Once the dorsal plane of the spinal cord is appreciated, internal debulking of the lipoma is accomplished using the ultrasonic aspirator. Enough fat is resected to facilitate reconstruction of the neural tube within the dural sleeve. We have found this technique more efficient than the laser, and we use the laser selectively only for the last remaining bits of residual lipoma before closure is initiated.

Fatty and fibrous connections between the conus medullaris and paraspinous soft tissues are disconnected, thereby freeing the cord from tethering. To accomplish these objectives it is critical to establish the plane between the lipoma and the dorsal surface of the spinal cord during the dissection. Once this plane is appreciated, circumferential dissection can usually free the cord of its fatty attachments. Particular care needs to be taken laterally in the cul-de-sac in caudal lipomas as this is where exiting nerve roots can become lost and inadvertently injured or divided if care is not taken by the surgeon. Exiting nerve roots in the substance of the fatty fibrous tissue are at constant risk. Small fibrotic rootlets may be divided. Intraoperative monitoring (rectal electromyography, urethral electromyography, somatosensory evoked potentials, and rectal or bladder pressure) has proven to be a very useful adjunct for some surgeons in the dissection of viable neural tissue in the midst of a mass of fibrofatty tissue.[55,66] The use of monitoring depends on the experience of each surgeon.

Closure must then be performed with great care. Ideally the dural closure will be capacious enough that no contact remains between the distal cord and the dura. Tight reapproximation of the dura so as to cause constriction must be avoided. Often caudal lipomas are difficult to reestablish into a proper tube. Some authors have advocated the utility of a synthetic patch to help make the dura more capacious.[77] Layer-by-layer closure is accomplished in a systematic and compulsive means so as to reduce the likelihood of postoperative CSF leaks. Some surgeons advocate a minimal removal of subcutaneous fat because they believe that an enlarged fat pad provides an effective barrier against a postoperative CSF.[27] We recommend prescribing a period of 5 days of flatbed rest in the immediate postoperative period to further help reduce the likelihood of postoperative CSF leaks.

Lipomas of the Terminal Filum

In any rational consideration for surgical intervention the risk or the natural history of the untreated disease is weighed against the risk of the proposed surgical intervention. Because the natural history of terminal filum lipomas is incompletely understood, it is impossible to make recommendations with certainty. Nonetheless, the collective weight of evidence from multiple centers reporting [39,55,56] series of filum lipomas supports the notion that an enlarged fatty filum in which the conus medullaris is displaced is highly associated with a tethered spinal cord. Thus, the natural history of the untreated condition for those patients with fat in an enlarged filum and an abnormally caudally displaced conus medullaris can be reasonably anticipated to include progressive pain and neurological deterioration. Greater controversy surrounds the treatment of cases in which the usual clinical findings of the tethered cord are demonstrated because neuroimaging reveals the conus medullaris to be at a normal position. In such studies [70,71] the authors also report very low rates of complications associated with the division of such fatty filums. As such, an aggressive surgical approach seems warranted for those patients with a Grade 1 lipoma of the filum. Optimal therapy for asymptomatic patients or those in whom a normal conus medullaris position is shown is less clear.[69,70] Because we have found a low rate of complications associated with the surgical division of an enlarged filum, we have recommended that patients with either progressive symptoms or an abnormal conus medullaris position in conjunction with an enlarged fat infiltrated filum undergo surgical division of the fatty filum (Table 1). We emphasize, however, during the informed consent discussion with families of patients harboring Grade 2 or 3 lesions that the natural history of filum lipomas is unknown, that improvement in symptoms cannot be assured, and that the risks of the procedure, although real, are very low.

Surgical Approach

In cases of lipoma of the terminal filum the surgical objective is to interrupt the abnormal connection between the hypertrophied and fat-infiltrated filum with the conus medullaris. Because this is most simply accomplished by sectioning the enlarged filum, the surgical undertaking is significantly less challenging than that of a lipomyelomeningocele.

Usually the enlarged filum can be readily discerned from the exiting nerve roots of the cauda equina beneath the operating microscope. It usually is larger, has longitudinal striations, appears different in color (typically described as somewhat more pink or gray than the generally ivory colored exiting nerve roots), and is often apparent by virtue of the mass of infiltrating fat. Nonetheless, intraoperative monitoring may be useful in providing greater assurance that the structure cut is not an exiting nerve root. Therefore, the use of intraoperative monitoring is left to the discretion of the individual neurosurgeon.

Once general anesthesia is induced the patient is placed prone on rolls or a frame. It is not typically necessary to insert a Foley catheter. After the lumbar region is prepared, antibiotic agents are provided, and the patient is draped, a midline incision is made in the lumbar region. The site of the incision is approximated in the lumbosacral region to be over a region where the filum is robust and between two laminae. The paraspinous musculature is removed via subperiosteal dissection to expose adjacent laminae and spinous processes. Most of the work can be performed between the laminae; however, it is often useful to remove a small portion of the inferior aspect of the superior lamina (to the level of the spinous process) and a similar small portion of the superior aspect of the inferior lamina (also to the level of the spinous process). The ligamentum flavum is removed to create a small corridor of access to the dura. The dura is then opened, and dural retaining sutures are placed. The operating microscope is brought in, and the cauda equina is inspected. The hypertrophied filum is typically readily evident. Intraoperative monitoring may be used for stimulation of the filum prior to its transection to provide the surgeon with additional assurance that an important neural structure is not being divided. The filum is coagulated with bipolar coagulation as there is typically a small vein within the filum. Often the filum will significantly retract in length. We prefer to remove a segment of the affected filum and submit it to pathological examination for confirmation that the transected structure was terminal filum. Watertight dural closure followed by meticulous layer-by-layer closure finalize the operative procedure. The overall operation time is typically less than 90 minutes and is associated with a complication rate of less than 5%.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.