Skip to main content
Missouri Medicine logoLink to Missouri Medicine
. 2019 Jul-Aug;116(4):287–288.

Groundhog Day: My Constant Headache Due to Spontaneous Cerebrospinal Fluid Leak

Alisha Wright 1,, Cindy Schmidt 1
PMCID: PMC6699807  PMID: 31527975

The headache started halfway through my (AW) shift. Busy and thirsty, I diagnosed dehydration, chugged a bottle of water and three ibuprofen. By the end of my shift, it had worsened substantially. Every day, I woke up with the same headache. All day, every day, no matter what time of day, the same headache. At times it would intensify, particularly if I stayed upright for an extended period of time. Regardless, it was there, constantly, even when I would awaken during the night. It reminded me of Bill Murray in the movie Groundhog Day, who relives the same day over and over.1

I simply tried to ignore it by lying down, closing my eyes or hiding in my room by myself. I avoided most social situations altogether as my headache persisted twenty-four hours a day. I treated myself for sinusitis, then for viral meningitis. When I couldn’t take it anymore, I became a patient in my own Emergency Department. CT, MRI, and lumbar puncture (LP) revealed only two abnormal findings – pachymeningeal enhancement on MRI and a low opening pressure on LP. A few days later, the pain was too intense and distracting to work, and my neurologist recommended admission. Imaging didn’t inform, and medication didn’t help. Three days passed without progress on the pain or diagnosis, so my neurologist suggested we pursue my cerebrospinal fluid (CSF) leak theory by doing a diagnostic/therapeutic lumbar blood patch. Within minutes I felt some relief – the first since the onset of my headache. Twelve hours later, however, all my symptoms returned. A second lumbar blood patch brought the same, temporary relief.

I was transferred to KU Medical Center in Kansas City, Ks., where they found the focus of my leak – near C2. Three more blood patches directed at C2 didn’t help, so I decided to find a facility with more expertise. After hours and hours of research, I found only three centers in the U.S. that seemed to have what I was looking for, years of experience and a large volume of patients with spontaneous CSF leaks. The three that met my criteria were Cedars-Sinai Medical Center, Duke University Medical Center, and the Mayo Clinic in Rochester, Minnesota. I arranged my travel to Mayo Clinic with hopes of a cure for my headache.

Imaging had indicated I had a “fast” leak, so the Mayo Clinic performed a dynamic CT myelogram and found the leak at T3. A sixth unsuccessful blood patch there led me to my cure. I subsequently underwent removal of the spinous process of T2 with complete laminectomies at T3/T4 to allow enough mobilization of my spinal cord to open the dura, find the leak on the anterior side and repair it. Even with tremendous back pain, I arose from bed the next day and realized my headache of five months duration was gone!

My cure came with quite the cost. I still have back pain and am hopeful that I may regain my old fitness regimen – pushups, pullups, lifting weights, etc. I had been in pretty good physical condition prior to the onset of the headache and worked out routinely. After the back surgery, I have had a lot to overcome with pain and loss of muscle. Initially, the fatigue was a huge hurdle but I am thankful to say that has already improved substantially. After eight months of medical leave, though, I’ve returned to my full professional capacity in the ED as physician and Program Director.

Spontaneous intracranial hypotension (SIH) from a dural leak classically presents with an orthostatic headache.24 This is not always the case, however. Many features of SIH headache have been described, including pain that improves in the upright position. The headache may be absent upon awakening with immediate onset after sitting or may progressively worsen throughout the day. It may be described as annoying or completely incapacitating. The postural component may decrease with time.5 There may be brainstem signs of tinnitus, diplopia, and vertigo. Pulsatile tinnitus is a strong sign of CSF leak. Historically thought to be due to CSF leakage from lumbar puncture, today it is recognized as leakage in the spine, and quite rarely cranially.2 It typically occurs from minor trauma (e.g., sports injury, sudden twist or sneeze, sexual intercourse, military combat or training experience) tearing the dural nerve sheath or possibly from an underlying connective tissue disorder. Furthermore, some have occurred in association with osseous abnormalities, congenital defects in the dura itself or with diverticuli.3,69

Spontaneous dural leaks are difficult to diagnose. Neurologic exam is usually normal.6 Head CT will typically be normal, and brain MRI may be normal in up to 20% of patients.3 Recommended imaging is use of brain MRI with gadolinium and spinal MRI without gadolinium.8 Look for SEEPS: subdural fluid, enhanced pachymeninges, engorged venous structures, pituitary enlargement, and “sagging” of the brain.3 Sagging of the brain may mimic a Chiari I malformation identified by cerebellar tonsillar descent below the foramen magnum, descent of the Sylvian aqueduct below the incisural lines, obliteration of the prepontine or perichiasmatic cisterns, crowding of the posterior fossa or flattening of the optic chiasm and/or anterior pons. Enlargement of the pituitary may be evident on MRI T2, as well as engorged cerebral venous sinuses and ventricular collapse.10 Pooling of contrast material may be apparent in the epidural space on spinal MRI when associated with a boney lesion and a fast leak. This may be extensive, impeding localization. Slow leaks usually lack this finding.5

Treatment of a dural leak varies from institution to institution although conservative measures predominate initially. Davidson et al. proposed an algorithm for diagnosis and treatment of suspected spontaneous intracranial hypotension.11 When suspected, admit the patient and obtain hourly neuro checks. The patient should lie flat while awaiting a lumbar epidural blood patch (EBP). A repeat CT should be obtained after 24 hours.11 For patients who improve clinically within the first 24 hours and also show stable or improved imaging, have the patient sit up gradually over the next 24 hours.11 If tolerated and asymptomatic, patients may be discharged and follow-up outpatient with CT.11 If upright positioning is not tolerated and patients remain symptomatic, conduct a repeat EBP, maintain patient lying flat for 24 hours and repeat CT.11 For patients who do not improve clinically within the first 24 hours, anesthesia, interventional neuroradiology, and neurosurgery consults are appropriate.11 If the patient clinically declines after the first EBP, Trendelenberg positioning, repeat EBP, and MRI of the spine should be completed to attempt localization of the leak.11 If this approach does not identify a CSF leak, repeat EBP, keep patient flat for 24 hours and repeat CT.11 If this approach does identify a possible CSF leak, CT/MR myelogram ± DS myelogram should be performed.11 If the leak is identified, conduct a targeted EBP.11 If patient improves, gradually sit up over 24 hours.11 If patient does not improve with the targeted EBP, surgical repair of the leak is recommended.11

graphic file with name ms116_p0287f1.jpg

MRI of brain sagging with low cerebrospinal fluid pressure headache

Brain MRI of a patient with low CSF pressure headache (spontaneous intracranial hypotension). Noncontrast T1-weighted sagittal images were obtained before (A) and after (B) treatment. There is crowding and pointing of the cerebellar tonsils at the foramen magnum mimicking a Chiari 1 malformation on the first study (arrow, image A). In addition, there is sagging of the brain resulting in inferior displacement of the floor of the third ventricle (arrowhead, image A) as well as loss of CSF space in the suprasellar cistern. These findings have resolved on MRI six months later after treatment (B), which included draining of associated subdural collections.

MRI: magnetic resonance imaging; CSF: cerebrospinal fluid.

Graphic 77679 Version 3.0

Spontaneous CSF leaks occur in 5 in 100,000 people.3 The typical patient would be like me – female, 40 years old, healthy, thin, low blood pressure.3,9 Although this is not common, it is a treatable cause of headaches warranting inclusion in the differential.

Footnotes

Alisha Wright, DO, (left), practices Emergency Medicine in the Freeman Health System in Joplin, Mo. Cindy Schmidt, PhD, is Director of Scholarly Activity and Faculty Development and Assistant Professor, College of Osteopathic Medicine, Kansas City University of Medicine and Biosciences

Contact: drwright2b@hotmail.com

References

  • 1.Ramis H, Ramis H, Rubin D. Groundhog Day. United States: Columbia Pictures; 1993. [Google Scholar]
  • 2.Mokri B. Spontaneous intracranial hypotension. Continuum Lifelong Learning in Neurology. 2015;21(4):1086–1108. doi: 10.1212/CON.0000000000000193. [DOI] [PubMed] [Google Scholar]
  • 3.Schievink WI. Spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension. Journal of the American Medical Association. 2006;295(19):2286–2296. doi: 10.1001/jama.295.19.2286. [DOI] [PubMed] [Google Scholar]
  • 4.Sinnaeve L, Vanopdenbosch L, Paemeleire K. Association of cerebral venous thrombosis and intracranial hypotension: Review of 3 cases. Journal of Stroke and Cerebrovascular Diseases. 2017;26(8):e165–e169. doi: 10.1016/j.jstrokecerebrovasdis.2017.05.015. [DOI] [PubMed] [Google Scholar]
  • 5.Friedman DI. Headaches due to low and high intracranial pressure. CONTINUUM: Lifelong Learning in Neurology. 2018;24(4):1066–1091. doi: 10.1212/CON.0000000000000623. [DOI] [PubMed] [Google Scholar]
  • 6.Greif S, Mandel S, Langer DJ, Ortiz RA. Spontaneous intracranial hypotension. Practical Neurology. 2014 May;:24–27. [Google Scholar]
  • 7.Nosik WA. Intracranial hypotension secondary to lumbar nerve sleeve tear. Journal of the American Medical Association. 1955;157:1110–1111. doi: 10.1001/jama.1955.02950300038008. [DOI] [PubMed] [Google Scholar]
  • 8.Schievink WI, Ebersold MJ, Atkinson JLD. Rollercoaster headache due to spinal cerebrospinal fluid leak. Lancet. 1996:347–1409. doi: 10.1016/s0140-6736(96)91048-x. [DOI] [PubMed] [Google Scholar]
  • 9.Schievink WI, Meyer FB, Atkinson JLD, Mokri B. Spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension. Journal of Neurosurgery. 1996;84(4):598–605. doi: 10.3171/jns.1996.84.4.0598. [DOI] [PubMed] [Google Scholar]
  • 10.Mokri B. Spontaneous low pressure, low CSF volume headaches: spontaneous CSF leaks. Headache: The Journal of Head and Face Pain. 2013;53(7):1034–1053. doi: 10.1111/head.12149. [DOI] [PubMed] [Google Scholar]
  • 11.Davidson B, Nassiri F, Mansouri A, Badhiwala JH, Witiw CD, Shamji MJ, Peng PW, Farb RI, Bernstein M. Spontaneous intracranial hypotension: A review and introduction of an algorithm for management. World Neurosurgery. 2017;101:343–349. doi: 10.1016/j.wneu.2017.01.123. [DOI] [PubMed] [Google Scholar]

Articles from Missouri Medicine are provided here courtesy of Missouri State Medical Association

RESOURCES