Secondary Logo

Management of Anesthesia and Delivery in Women With Chiari I Malformations

Waters, Janet F. R., MD, MBA; O'Neal, M. Angela, MD; Pilato, Madison, MD; Waters, Samuel, BS; Larkin, Jacob C., MD; Waters, Jonathan H., MD

doi: 10.1097/AOG.0000000000002943
Contents: Medical Complications of Pregnancy: Original Research
Journal Club

OBJECTIVE: To estimate whether vaginal delivery or neuraxial anesthesia poses a risk of neurologic deterioration in women with uncorrected Chiari I malformation.

METHODS: To assemble this case series, electronic record databases were used to identify women with Chiari I malformation who delivered on two busy tertiary care obstetric services over a 5-year period from January 2010 through December 2015. Women who had undergone surgical decompression were not included in the study. The size of the Chiari malformation, neurologic symptoms before delivery, mode of delivery, anesthetic method used, and neurologic complications were recorded.

RESULTS: Ninety-five deliveries in 63 patients were identified. The size of the Chiari malformation was 9.3±4.3 mm (mean±SD). In 58 pregnancies, women reported no headaches; in 36 they did. There was no association between the size of the Chiari malformation and the incidence of headache. Forty-four neonates were delivered by cesarean delivery and 51 were delivered vaginally. No neurologic deterioration occurred in either group. Neuraxial anesthesia was administered before 62 deliveries. No neurologic complications occurred. None of the women who delivered vaginally or received neuraxial anesthesia had signs of increased intracranial pressure. The upper limit of the 95% CI for the risk of neurologic complications from our study of 95 deliveries was 3.1%.

CONCLUSION: This case series support that in patients with Chiari I malformation who have no signs of increased intracranial pressure, the mode of delivery should be based on obstetric rather than neurologic considerations. The absence of complications in patients who received epidural or spinal anesthesia suggests that these procedures should be made available to women with Chiari I malformation.

In this case series, neuraxial anesthesia and vaginal delivery were not associated with any neurologic deterioration in patients with Chiari I malformation.

Department of Neurology, Division of Women's Neurology, the Department of Neurology, and the Departments of Anesthesiology and Bioengineering, McGowan Institute for Regenerative Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; the Department of Neurology, Brigham & Women's Hospital, Harvard Medical School, Boston, Massachusetts; and the Department of Bioengineering, University of Pittsburgh, and the Department of Obstetrics, Gynecology and Reproductive Sciences, Division of Maternal-Fetal Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania.

Corresponding author: Janet F. R. Waters, MD, MBA, 3471 Fifth Avenue, Suite 810, Pittsburgh, PA 15213; email:

Financial Disclosure The authors did not report any potential conflicts of interest.

Each author has indicated that he or she has met the journal's requirements for authorship.

Received May 27, 2018

Received in revised form July 27, 2018

Accepted August 02, 2018

Types I–III Chiari1 malformations are a group of congenital posterior fossa abnormalities affecting the structural relationships among the bony cranial base, cerebellum, brainstem, and cervical cord. Chiari I is defined as a descent of the cerebellar tonsils below the foramen magnum by greater than 5 mm.2 There may be an associated cervical syringomyelia. It is relatively common and occurs in 0.6% of the population.3 Chiari II–III are rare malformations (Fig. 1). Although most patients with Chiari I malformations are asymptomatic, 30% may experience symptoms including headaches, which are exacerbated by coughing, Valsalva maneuver, or positional change. They may also experience neck pain, vertigo, and tinnitus. More severely affected patients can present with imbalance and limb weakness as a result of brainstem compression. Some patients may develop signs of increased intracranial pressure including confusion, dysconjugate gaze, and papilledema. These findings may be associated with hydrocephalus.4 During labor, contractions are associated with increases in cerebrospinal pressure, which can be exacerbated by pain.5,6 This has led to concern that women with Chiari I malformation who undergo vaginal delivery are at risk for neurologic deterioration during labor. Despite the prevalence of Chiari I malformation, there have been no reported cases of neurologic deterioration resulting from labor attributable to a Chiari I malformation.

Obstetric anesthesiologists have been concerned that neuraxial anesthesia in women with Chiari I malformations would cause neurologic deterioration from tonsillar herniation.7 Although there are scattered case reports of patients with Chiari I malformation who have undergone vaginal delivery, neuraxial anesthesia, or both without event, the obstetric and anesthetic management of women affected by Chiari I malformation have scarcely been addressed in the literature.8 We hypothesized that neither vaginal delivery nor neuraxial anesthesia would be associated with poor neurologic outcome in women with Chiari malformations and no signs of elevated intracranial pressure.

Back to Top | Article Outline


After institutional review board approval at both the University of Pittsburgh Medical Center and Brigham and Women's Hospital, electronic hospital databases were used to identify all women who delivered at Magee Women's Hospital in Pittsburgh and Brigham and Women's Hospital in Boston between January 2010 and December 2015, who also had a known diagnosis of Chiari I malformation based on neuroimaging, either a head computerized axial tomography or brain magnetic resonance imaging (MRI). Women who had undergone surgical decompression before delivery were excluded from the study because the safety of vaginal delivery and neuraxial anesthesia in this population has been well documented.8–10 A retrospective chart review of each case was conducted. Information on demographics, neurologic history, radiology reports, choice of mode of delivery, anesthetic method, and outcome were recorded. All reported neurologic symptoms were reviewed to determine whether they were attributable to the Chiari malformation. This included information about headaches during pregnancy, at delivery, and in the postpartum period. Information was also recorded on signs of increased intracranial pressure including confusion, dysconjugate gaze, papilledema, and hydrocephalus. Patients who had undergone neuraxial anesthesia were routinely seen 1 day postprocedure by an obstetric anesthesiologist to ascertain whether there were any complications, including pain, weakness, numbness, and headache.

Data were assessed for normality with the D'Agostino and Pearson normality test. The size of the Chiari malformation was compared between patients who had headaches before delivery and those who were asymptomatic with the Mann-Whitney test. Likewise, a Mann-Whitney test was used to compare the size of the Chiari malformation between women who delivered by cesarean delivery compared with a vaginal delivery. A P<.05 was considered statistically significant.

The upper limit of the 95% CI for patients having neurologic complications was calculated by the Rule of 3. The formula R=1−[(0.05)1/N] was used with R being an estimate of the upper limit of the 95% CI and N is the number of patients without an event in the study.11

Back to Top | Article Outline


Ninety-five deliveries in 63 women with Chiari I malformation were identified (Table 1). In 44 pregnancies delivery was by cesarean and in 51 delivery occurred vaginally. In the latter group, no women had signs suggestive of increased intracranial pressure. Six women underwent two sequential cesarean deliveries. Thirteen women had two or more vaginal deliveries. There were no women who underwent both cesarean delivery and vaginal delivery. There were no significant differences between the two institutions in terms of the mode of delivery.

Table 1

Table 1

The level of descent through the foramen magnum ranged between 5 and 22 mm with a mean of 9.3 mm. Two women had an associated cervical syrinx. Symptoms during first and subsequent pregnancies were the same for all women regardless of their gravidity. Most women had a single MRI to assess the size of their Chiari I malformation. In those patients who underwent more than one MRI, there was no difference in the size of the Chiari I malformation in subsequent imaging. For the 63 unique women, median Chiari size was 8 mm (range 5–19 mm) for those who were asymptomatic and 10 mm (range 5–22 mm) for those who had headaches. There was no difference in the median size between women who had headaches when compared with those who were asymptomatic (P=.80) (Fig. 2).

The median Chiari malformation size for women who underwent cesarean delivery was 8 mm (range of 5–22 mm), which was the same as the median value of 8 mm (range 5–19 mm) for women who delivered vaginally (P=.33) (Fig. 3). Ten cesarean deliveries were performed at the recommendation of their physicians as a result of the presence of the Chiari I malformations. Median Chiari malformation size in this group was 12 mm, with a range of 6–22 mm. This Chiari size was no different than in all of the other women who underwent cesarean delivery for other indications (P=.13). Three of the 10 women had associated hydrocephalus, one had papilledema, four had headache alone, and two were asymptomatic. There were multiple indications listed in the records for cesarean delivery (Table 2).

Table 2

Table 2

Of the 51 vaginal deliveries, 21 were associated with headaches before delivery and 30 were asymptomatic. By review of the record, there was no reported worsening headache or neurologic symptoms during delivery nor during the postpartum period. Ten of the patients were advised to undergo general anesthesia rather than neuraxial anesthesia as a result of the presence of their Chiari I malformation. One patient received a general anesthetic as a result of an inability to achieve adequate neuraxial block and one underwent general anesthesia as a result of obstetric considerations. General anesthesia was not used in any women undergoing vaginal delivery.

Neuraxial anesthesia was administered before 62 deliveries. None of the women who received this form of anesthesia had signs suggestive of increased intracranial pressure. Thirty-eight epidurals and 24 spinal anesthetics were administered. No patients received combined spinal–epidural anesthesia. No complications were reported for any of these anesthetics. Two women had severe headaches before administration of anesthesia and further evaluation led to a diagnosis of severe preeclampsia. All headaches resolved after delivery. One patient developed a postpartum headache as a result of postpartum eclampsia. None developed low-pressure headaches.

Back to Top | Article Outline


There are two important findings in this retrospective case series. The first is that in 62 deliveries in women with Chiari I malformation, epidural or spinal anesthesia was administered without worsening headaches or any symptoms of tonsillar herniation. The second is that 51 women with Chiari malformation were able to labor and deliver vaginally without neurologic detriment. These findings have significant ramifications regarding the obstetric management of women with Chiari I malformations. Each is discussed subsequently.

Obstetric anesthesiologists have been reluctant to offer neuraxial anesthesia to women with Chiari I malformation as a result of concerns for neurologic deterioration resulting from changes in pressure gradients associated with dural puncture. This fear originated from a case report of a woman with a headache after spinal anesthesia given before cesarean delivery.7 The headache was postural and initially improved after placement of an epidural blood patch. The headache returned 4 days later. A brain MRI was done, which revealed a Chiari I malformation. The authors hypothesized that “an increased gradient of CSF pressure develops between the brain and the spinal canal” in a patient with a Chiari I malformation, which increased the risk of a postdural puncture headache. In retrospect, it seems equally plausible that the patient in this case had descent of her cerebellar tonsils through the foramen magnum as a result of lowered cerebrospinal fluid pressure from her dural puncture causing a “pseudo-Chiari malformation.”

Chiari I malformations are quite common and can be found on neuroimaging studies in 0.6% of the population.3 If this malformation truly posed a risk for women receiving epidural or spinal anesthesia, neurologic complications would be expected to be widely reported. In contrast, the limited existing evidence supports the safety of neuraxial anesthesia in this group (Orth T, Gerkovich M, Babbar S, Porter B, Lu G. Maternal and pregnancy complications among women with Arnold Chiari malformation: a national database review [abstract]. Am J Obstet Gynecol 2015;212:S349.).12–16 This case series in conjunction with previous reports suggest that spinal and epidural anesthesia should not be withheld from women with Chiari I malformation.

The second issue addressed by this case series concerns the safety of vaginal delivery in women with Chiari I malformation. Pregnant women with Chiari I malformation have been considered high-risk patients by their obstetricians as a result of the concern for increasing intracranial pressure during active labor causing worsening neurologic deficits. As the use of brain imaging has become more common, Chiari I malformations are frequently discovered, often as an incidental finding. In some institutions, pregnant women with a radiologic diagnosis of Chiari I malformation are advised to undergo delivery by cesarean delivery with its increased risks.

In our series, 51 women with Chiari malformations up to 19 mm were able to safely deliver vaginally. Almost half of these patients had headaches. Although this finding supports the safety of vaginal delivery in women with Chiari malformation, there are complexities involved in the decision-making process. A major drawback of this retrospective study is the selection process that determined which patients were delivered vaginally and which by cesarean. Although the majority of the 44 cesarean deliveries were performed for obstetric considerations, 10 were performed at the recommendation of the neurologist or neurosurgeon owing to the presence of their Chiari malformation. Five women had either headaches or were asymptomatic, much like the women who underwent vaginal delivery. However, there were also three cesarean deliveries performed in the setting of hydrocephalus, one with papilledema and another with headache and sensory changes. It is unclear whether these women with more severe features of Chiari malformation associated with signs of increased intracranial pressure could have labored safely.

In consideration of the findings in this series of cases in conjunction with review of the available literature on the subject, the following guidelines for choosing mode of delivery in women with Chiari I malformation are proposed:

  • Choice of mode of delivery in women with asymptomatic Chiari I malformation should be based on obstetric considerations.
  • Women with Chiari I malformation who are experiencing headache as the only manifestation of their Chiari I malformation can be delivered based on obstetric considerations.
  • Women with Chiari I malformation with hydrocephalus and papilledema (with or without headache) may be considered to be high risk for both vaginal delivery and neuraxial anesthesia. Coordination of care among obstetricians, anesthesiologists, and neurologists and neurosurgeons is recommended to optimize patient outcome.
Back to Top | Article Outline


1. Chiari H. Über Veränderungen des Kleinhirns infolge von Hydrocephalie des Grosshirns. Deutsch Medizinische Wocherschrift 1891;17:1172–5.
2. Tubbs RS, Lyerly MJ, Loukas M, Shoja MM, Oakes WJ. The pediatric Chiari I malformation: a review. Childs Nerv Syst 2007;23:1239–50.
3. Hidalgo JA, Dulebohn SC. Arnold Chiari malformation. StatPearls. Treasure Island (FL): 2017. Available at: Retrieved November 13, 2017.
4. Ropper AH, Samuels MA, Klein JP. Developmental diseases of the nervous system. In Sydor AM, Davis KJ, editors. Adams and Victor's principles of neurology. 10th ed. New York (NY): McGraw Hill; 2014. p.1015–17.
5. Hopkins EL, Hendricks CH, Cibils LA. Cerebrospinal fluid pressure in labor. Am J Obstet Gynecol 1965;93:907–16.
6. Karmaniolou I, Petropoulos G, Theodoraki K. Management of idiopathic intracranial hypertension in parturients: anesthetic considerations. Can J Anesth 2011;58:650.
7. Hullander RM, Bogard TP, Leivers D, Moran D, Dewan DM. Chiari I malformation presenting as recurrent spinal headache. Anesth Analg 1992;75:1025–6.
8. Ghaly RF, Candido KD, Sauer R, Knezevic NN. Anesthetic management during cesarean section in a woman with residual Arnold-Chiari malformation type I, cervical kyphosis, and syringomyelia. Surg Neurol Int 2012;3:26.
9. Mueller DM, Oro’ J. Chiari I malformation with or without syringomyelia and pregnancy: case studies and review of the literature. Am J Perinatol 2005;22:67–70.
10. Sicuranza GB, Steinberg P, Figueroa R. Arnold-Chiari malformation in a pregnant woman. Obstet Gynecol 2003;102:1191–4.
11. Bernstein J, Hua B, Kahana M, Shaparin N, Yu S, Davila-Velazquez J. Neuraxial anesthesia in parturients with low platelet counts. Anesth Analg 2016;123:165–7.
12. Newhouse BJ, Kuczkowski KM. Uneventful epidural labor analgesia and vaginal delivery in a parturient with Arnold-Chiari malformation type I and sickle cell disease. Arch Gynecol Obstet 2007;275:311–3.
13. Semple DA, McClure JH. Arnold-chiari malformation in pregnancy. Anaesthesia 1996;51:580–2.
14. Chantigian RC, Koehn MA, Ramin KD, Warner MA. Chiari I malformation in parturients. J Clin Anesth 2002;14:201–5.
15. Garvey GP, Wasade VS, Murphy KE, Balki M. Anesthetic and obstetric management of syringomyelia during labor and delivery: a case series and systematic review. Anesth Analg 2017;125:913–24.
16. Parker JD, Broberg JC, Napolitano PG. Maternal Arnold-Chiari type I malformation in syringomyelia: a labor management dilemma. Am J Perinatol 2002;19:445–50.
© 2018 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.