Outcomes of Chiari Malformation III: A Review of Literature

ABSTRACT

Purpose:

Chiari malformation type III (CM III) is the rarest type compared to other types of CMs. CM III usually reported as sporadic case reports which reflect the rarity of this anomaly. We report two cases of operated CM III at our institute with a reasonable outcome and reviewed the literature to illustrate the variability of prognosis and related hydrocephalus.

Materials and Methods:

We operated two cases of CM III in our hospital followed by ventriculoperitoneal shunt (VPS) placement with an accepted neurological outcome at 10 and 6 months follow-up. We reviewed the literature for other cases of CM III with focusing on prognosis to illustrate the real image of reported prognosis and related hydrocephalus.

Results:

After follow-up for 10 and 6 months, respectively, both cases had mild developmental delays. In this review, we report 51 cases of CM III over the last 30 years since 1989, there was slight male predilection, hydrocephalus was evident in 27 cases which was almost managed with VPSs and was evident in seven deaths.

Conclusion:

CM III is a rare anomaly which usually carries a bad prognosis, but death is not ultimate, and there may be a minority who carry good prognosis. This bad prognosis pushes some parents to refuse surgery otherwise repair should be done. With good pre- and postoperative care, physical therapy, and follow-up, the outcome is reasonable.

INTRODUCTION

Hans Chiari reported three types of hindbrain dysplasia according to the degree of descent of cerebellum after dissection of children’s autopsies.[1] Chiari Malformation (CM) type III reported as low occipital and/or high cervical encephalocoele and herniated posterior fossa dysplastic contents.[2] CM type III is an extremely rare anomaly with much more poor neurological outcome, developmental delay, and mortality compared to other types to CMs.[3] We report surgical repair of two cases of CM III with a reasonable prognosis and review of the literature for cases of CM III with focusing on prognosis and related hydrocephalus.

METHODS

Review of literature

PubMed search was done using terms “Chiari III,” “Chiari 3,” “malformation,” and “encephalocele” in a single and combination manner. Only articles written in English are included and articles mentioning other types of Chiari malformation other than type III are excluded.

Case Reports

Four- and 6-month-old female infants presented to our neurosurgery department in Cairo university hospitals in December 2018 and April 2019 subsequently with occipital swelling after birth that was expanding with their growth. Both infants had non-consanguineous parents and the mothers of both cases did not receive antenatal care as they live in rural areas, so they were not diagnosed until birth. No relevant history from both mothers during pregnancy. By examination, both had a normal tone in four limbs as well as reflexes, lax Anterior fontanel, and soft swelling in the occipital region covered by normal intact skin [Figures 1A and 2A, respectively], no spells of apnea reported and the infants had a normal cry. Mild developmental delay in milestones was evident in both of them. Cranial computed tomography (CT) showed a small defect in the posterior fossa region. Magnetic resonance imaging (MRI) showed small posterior fossa, herniation of cerebellum surrounded by cerebrospinal fluid (CSF), no apparent hydrocephalus nor syringomyelia [Figures 1B and 2B, respectively], and herniation of part of brain stem down to the cervical canal detected in the second case [Figure 2B]. Magnetic resonance venography (MRV) confirmed the patency of venous sinuses in both of them. In both cases, we performed primary repair through dissection in anatomical layers till reaching the bony edge of posterior fossa then removal and excision of herniated cerebellum till the bony edges because it is mostly contused and nonfunctioning [Figure 3], then untethering of the cerebellum from the bony edge to avoid the growth and recurrence of herniation again because this potential space that the cerebellum allowed easily to grow and finally closure of the dura with fascial graft completely to minimize CSF leak. Both cases were extubated postoperatively and on the second and the fourth day, respectively, the infants developed symptoms suggestive of increased intracranial tension in the form of repeated vomiting with bulging Anterior fontanels, CT brain showed hydrocephalus, and a ventriculoperitoneal shunt was placed for both of them and both were discharged 1 week later in good status and no deterioration. Follow-up for both cases 10 months and 6 months, respectively, showed both of them caught the supposed milestones with mild delay. The literature included all cases of CM III; i.e., the classic occipitocervical defect with herniation of posterior fossa contents and the cervical or occipital that are included in the expanded classification by many authors.[4] We found 30 case reports and two case series[5,6] with 51 as a total number of reported cases over 30 years from 1989 till 2019 as shown in Table 1.

Figure 1.

(A) Occipital mass in first 4-month-old child. (B) MRI T2 sagittal view for the child showing herniation of cerebellum surrounded by CSF through low occipital defect

Figure 2.

(A) Occipital mass in second 6-month-old child. (B) MRI T2 sagittal view showing herniation of cerebellum surrounded by CSF through low occipital defect and downward displacement of brain stem

Figure 3.

Describe the cranial part of surgery. * indicates the cranial part, # indicates the caudal part toward the cervical spine, (1) points to normal cerebellum, (2) points to the bone of posterior fossa from inside, the red arrow points to the bony edges of posterior fossa after coagulation and removal of herniated cerebellum, which is the cut off point for our removal

Table 1.

The results of literature review for cases of CM III

Author and date	Location of encephalocoele	Age at diagnosis	Preoperative hydrocephalus	Prognosis
Dyste et al. 1989[7]	High C	__ (2 cases)	__	__
Castillo et al., 1992[6]	High C, O	Newborn	No	4 cases surgery; 2 died, 1 severe MR and 1 not mentioned. 5 cases lost follow-up 3 males and 6 females
	HC, O	Newborn	No
	HC, O, P	Newborn	No
	HC, O	Newborn	No
	HC, O, P	Newborn	Yes
	O, P	Newborn	No
	O, P	Newborn	No
	O	Newborn	Yes
	O	Newborn	No
Kannegieter et al., 1994[8]	OC	Newbornfemale	No	Not mentioned
Cama et al., 1995[9]	OC	Not mentioned	Yes (VPS) Both cases	Not mentioned
	OC
Aribal et al., 1996[10]	C	40 days female	Dilated ventricles	___
	C	2 months female	___	___
Kernan et al., 1996[11]	__	Newborn /11 years male	Yes (VPS) 1month later	10 years: walk with braces and speech appropriate for age at 11 years: untethering for hindbrain
Snyder et al., 1998[12]	C	2 days female	Yes (VPS)	Repair at 30 months old 1 year later: doing well with some neurological signs
Sirikci et al., 2001[13]	O	Newborn female	No	11 years old: Asymptomatic
Häberle et al., 2001[14]	C	Newborn male	No	Died 3days later
Caldarelli et al., 2002[15]	OC	Newborn female	Yes (VPS)	18 months old: severe retardation
Lee et al., 2002[16]	OC	Newborn male	Yes (VPS) After repair	10 months old: developmental delay
Cakirer, 2003[17]	OC	Newborn male	Yes (VPS)	quadriparesis and not need for repair
	OC	2 months male	Yes (VPS)	Not mentioned
Cho et al., 2005[18]	Low O	Newborn female	Yes (ETV)	3 months FU: improved hydrocephalic changes
Smith et al., 2007[19]	OC	10 days Not mentioned	Yes (VPS) After repair	3 months old: seizures, hypertonia,
Jaggi and Premsagar , 2007[20]	C	1 month female	No	5 months FU: Asymptomatic
Muzumdar et al., 2007[21]	C	1 month male	NO	4 months: doing well 8 months: died (chest infection)
Işik et al., 2009[5]	OC	35 days female	Yes(VPS)	8 years FU:	good developmental milestones for age
	Low O	63 days male	Yes (VPS)
	High C	3 months male	Yes (VPS)	4 years FU:
	High C	7 months male	Yes (VPS)	2 years FU:
	OC	23 days female	Yes (VPS) 1day later	1 year FU:
	OC	2 days male	Yes (VPS) 1 month later	10 years FU:	Severe mental and motor retardation even unable to walk.
	Low O	2 years male	Yes (VPS)	2 years FU:
	OP	9 days male	Yes (VPS)	Died
Garg et al., 2008[2]	OC	Newborn male	No	Parents refused surgery
Furtado et al., 2009[22]	OC	15 months female	No (3 years FU)	Poor mental development (3 years FU)
Zolal et al., 2010[23]	OC	6 months, female	Yes (VPS)	Died 4 months later pneumonia
ile Birlikteliği, 2011[24]	Low O	4 years male	No ( 2 years FU)	2 years FU: decrease in falls down and no improvement of abnormal movement.
Agrawal et al., 2011[25]	OC	6 months female	Yes (VPS)	Not mentioned
Garg et al. 2011[26]	OC	18 months female	Yes(VPS)	1 month FU: doing well
Ambekar et al., 2011[27]	OC	7 years male	Yes(VPS)	1 year FU: doing well no deficit.
Rani et al., 2013[28]	O	Stillbirth (24 weeks)	Dilated Ventricles	Stillbirth
Andica and Soetikno, 2013[29]	OC	New born male	No	__
Ramdurg et al., 2013[30]	LO	1 month male	No	On regular follow-up
Bulut et al., 2013[31]	OC	Newborn male	No	Died the first day after surgery
Jeong et al., 2014[32]	OC	3 months female	Yes (VPS) 1 month later	36 months old: improved but delayed gross motor, not improved cognitive and NG tube for nutrition
Young et al., 2015[4]	OC	Newborn male	No (12 months FU)	12 months FU: Bilateral esotropia, mild hypotonia in trunks, unable to sit alone
Tan et al., 2018[33]	OC	Newborn male	Yes (VPS) at 61 days old	4 months FU: overall developmental delay
Ganeriwal et al., 2019[34]	OC	3 months___	Yes (VPS) at 3 months FU	Not mentioned
Our cases, 2019	OC	4 and 6 months females	NO	3 months and 7 months FU: Both has a mild developmental delay

Most of the cases reported radiological diagnosis with MRI study, exact structures included in the encephalocoele, local examination of the encephalocoele, and the neurological status of the child. There was no privilege for a specific gender; there were 23 males, 21 females, 6 cases in which gender was not mentioned, and a single stillbirth case. Both of our cases were female infants. Hydrocephalus was evident in 27 cases (56%); in 19 cases, hydrocephalus was diagnosed preoperative, in eight cases it was postoperative, and hydrocephalus was not evident in 19 cases (40%) neither preoperative nor postoperative with follow up ranging from 12 months up to 2 years.[4,24] The status of ventricles was not mentioned in three cases[7,10] and dilated ventricles was a description in two cases.[10,28] In our two cases, VPS was done postoperative due to hydrocephalic changes. The prognosis of CM III is not so dismal as predicted or reported by some authors. There was a single stillbirth case.[28] Seven cases died, in which two of them died 8 months[21] and 4 months,[23] respectively, postoperatively due to chest infection. Thirty-three cases had various degrees of motor and mental developmental delays, seizures, weakness, and nasogastric feeding, and the parents refused surgery in two kids. Although partial improvement or good development is not the main outcome, it was reported in minor cases. Both our cases have mild developmental delays and are doing well till 10 and 6 months postoperatively.

DISCUSSION

CM type III was first reported by Hans Chiari as herniated cerebellar tissue through bony defect associated with bifid cervical spine.[19,35] Cakirer[17] reported that up to 70% of cases of CM III have incomplete fusion at the posterior arch of C1. The presence of occipital and/or cervical defect and neural tissue herniation are essential for diagnosis that may be associated with other anomalies.[4] Many authors have considered occipital and/or cervical encephalocoele with cerebellar tissue within the sac and displaced brain stem downward in the cervical canal as the criteria for CM III diagnosis.[1,10,26,36] Young et al.[4] cleared that cervical myelomeningocele or occipital encephalocoele are different entities and misdiagnosed as CM III while the classic picture is occipitocervical defect with herniated posterior fossa dysplastic contents inside the encephalocoele or CSF sac.

The pathogenesis behind CM III is not clear and may be related to improper neutralization during the extension of ventricles giving the chance for cerebellum and brain stem prolapse.[32,37] Other theory that could explain the resultant occipitocervical defect is the persistent leakage of CSF with subsequent failure of closure of neural tube and ossification centers.[16] Although the dark reported prognosis of CM had been changed from being incompatible with life[38] to be dismal even after repair,[12,13,14,15] other authors considered occipitocervical encephalocoele is not a must for bad prognosis. There are some factors that can predict the prognosis such as neurological signs at birth like difficulty in breathing and swallowing, hypotonia, and muscle weakness;[4] in our cases, there is only mild delay in developmental milestones.

MRI brain is the standard method for accurate anatomical diagnosis for CMs and for the detection of associated anomalies like kinking of the medulla, beaking of the quadrigeminal plate of the tectum, syrinx of the spinal cord, and shallow posterior fossa,[9,10] and some neurological symptoms like cranial nerves palsy, hypotonia, and respiratory compromise may be owed to those abnormalities.[14] MRV has additional preoperative importance for localization of venous sinuses and detection of their patency; double and triple division of superior sagittal sinus has been reported.[16,19] This information could prevent massive bleeding.

The main aim of treatment is the closure of the dura in a water-tight manner as possible and CSF diversion if hydrocephalus is present that could be done before or after the repair.[12,16,20] In case of a large bone defect, Furtado et al.[22] reported the use of methyl-methacrylate cranioplasty flap and occipital scalp rotational flap based on occipital artery to close the defect. Amputation of herniated neural tissue may be obligatory for proper closure of the defect.[4] We used the bony edges of the posterior fossa as a cut-off point for excision.

Reported pathological examination of these tissues notified that those tissues were not functional gliotic and fibrotic tissues and were evidence of heterotopias.[8,13,28] Zolal et al.[23] could identify the corticospinal tract with the herniated sac using Diffusion Tensor imaging which added information about the functionality of the tissues.

Factor that could affect the outcome of surgical repair: timing of repair, neurological status, size of encephalocele and its covering, closed or ruptured sac, and respiratory distress.[5,20,26] In cases of large encephalocele that were accompanied by severe respiratory distress due to compressed brain stem, it is possible to relieve the compression gradually through drainage of CSF from the sac through external drain along 10 days, this will avoid rapid decompression then surgical repair came later on.[15] Although surgical repair can improve the prognosis, postoperative complications like thrombosis of the sagittal sinus were reported.[10,17]

CSF shunting is not mandatory in all cases and CSF diversion first policy may be the convenient decision if posterior fossa is crowded with eventual obstructive hydrocephalus or if the encephalocoele is not covered with full skin.[2,17] Some authors reported that hydrocephalus is a constant finding in all cases,[39,40] whereas Işik et al.[5] reported that the incidence of hydrocephalus is 88% in CM III. Maybe the presence of wide defect will give the chance for CSF to circulate through foramen of Magendi and Luschka without obstruction and this could explain why hydrocephalus is not present in all cases.[26,29]

In our review of 51 cases, hydrocephalus was evident in 27 cases (56%); in 19 cases, hydrocephalus was diagnosed preoperative and ventriculoperitoneal shunt (VPS) was done in 18 cases and in one case endoscopic third ventriculostomy (ETV) was done,[18] the ETV was functioning with follow up 3 months later. In eight cases, the hydrocephalus was postoperative and VPS was done. In our two cases, VPS was done postoperative due to early hydrocephalic changes. The prognosis of CM III is not so dark as predicted or reported by some authors; there was a single stillbirth case,[28] only seven cases died, and in 33 cases the prognosis varied from good development of the children to severe mental and motor retardation and in between those there were a lot of reported neurological deficits like weakness, seizures, walking with support, abnormal movement, and NG tube feeding. Signs of brain stem compression could predict worst prognosis, developmental delay was evident in most cases but there were cases that still near normal or asymptomatic till 11 years.[11,24,27] So, if a bad prognosis or severe delay in milestones was evident from the start, this means the dismal prognosis as expected and those children that had few symptoms or asymptomatic could continue their lives as well. The parents could not accept the anomaly from the start and refuse surgery.

CONCLUSION

CM type III is a hindbrain malformation anomaly that does not carry dismal prognosis in all cases as mentioned in old articles. Based on our two cases and as mentioned in most published articles; there are variations in the outcome and not all cases of CM III will die and severe developmental delay is not a must. Proper operative planning and technique followed by postoperative care and physiotherapy could lead to a reasonable outcome and decreased mortality.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.