Migraine or idiopathic intracranial hypertension: Magnetic resonance venography and magnetic resonance imaging findings

Abstract

Background

Idiopathic intracranial hypertension (IIH) is a disease characterised by increased cerebral pressure without a mass or hydrocephalus. We aimed to differentiate migraine and IIH patients based on imaging findings.

Results

Patients with IIH (n = 32), migraine patients (n = 34) and control subjects (n = 33) were evaluated. Routine magnetic resonance imaging, contrast-enhanced 3D magnetic resonance venography and/or T1-weighted 3D gradient-recalled echo were taken with a 1.5 T magnetic resonance scanner. Optic-nerve sheath distention, flattened posterior globe and the height of the pituitary gland were evaluated in the three groups. Transverse sinuses (TS) were evaluated with respect to score of attenuation/stenosis and distribution. Pearson chi-square, Fisher’s exact test and chi-square trend statistical analyses were used for comparisons between the groups. A p-value of <0.05 was considered statistically significant. Decreased pituitary gland height, optic-nerve sheath distention and flattened posterior globe were found to be statistically significant (p < 0.001) in IIH patients. Bilateral TS stenosis was also more common in IIH patients than in the control group and migraine group (p = 0.02).

Conclusion

Decreased pituitary gland height, optic-nerve sheath distention, flattened posterior globe, bilateral stenosis and discontinuity in TS are significant findings in differentiating IIH cases from healthy individuals and migraine patients. Bilateral TS stenosis may be the cause rather than the result of increased intracranial pressure. The increase in intracranial pressure, which is considered to be responsible for the pathophysiology of IIH, is not involved in the pathophysiology of migraine.

Introduction

Idiopathic intracranial hypertension (IIH) is a disease characterised by increased cerebral pressure without a lesion occupying the intracranial space, venous occlusion or meningeal inflammation.¹ The aetiology and pathogenesis of the disease are not known precisely, but toxins, drugs (steroids, oral contraceptives, lithium carbonate and tetracyclines), obesity, pregnancy, renal and collagen tissue diseases, endocrine and haematological diseases can be included in the aetiology.² IIH is 2–10 times more common in females than in males, with an onset age of 11–58 years (M = 30 years). Although the incidence in the healthy population is 1/100,000, when the ideal body weight is ≥20%, the rate increases to 19.3/100,000.³ The most common clinical finding is headache, which is seen >90% of the patients. Headache, tinnitus, diplopia or temporary visual impairment may occur.⁴ More than 90% of headaches are primary, which consist of migraine, tension and cluster-type headaches. Those with an underlying disease are referred as secondary headaches and can be seen with IIH.⁵ Intracranial pressure changes can cause headaches and are difficult to diagnose because of the presence of accompanying migraine in these patients. Clinical presentation occasionally overlaps with migraine and IIH patients.⁶ They may have similar clinical signs such as headaches and similar radiological findings such as transverse sinus (TS) stenosis. Both groups of patients are generally female. Treatment is often different. Migraine can be misdiagnosed as IIH when it is accompanied by an increase in cerebrospinal fluid (CSF) pressure.

The diagnostic value of cranial magnetic resonance imaging (MRI) findings for IIH has been shown in various studies.^7,8 There is no comparative study of migraine and IIH patients based on cranial MRI and magnetic resonance venography (MRV) findings. In this study, we aimed to investigate the diagnostic difference of headache-associated IIH cases, migraine patients and control group subjects based on MRI findings.

Methods

This retrospective study was performed between 2014 and 2016, and it was approved by the local ethical committee. A total of 99 patients (77 female) with a mean age of 40 years (range 17–62 years) were included in this study. IIH patients (n = 32), migraine patients (n = 34) and control patients (n = 33) constituted groups 1, 2 and 3, respectively. The control group consisted of patients without typical clinical findings for IHH or migraine, especially headache. The majority of control cases were those who were investigated for metastases from extracranial malignancy, resulting in normal contrast-enhanced brain MRI. Those patients whose clinical findings overlapped between IIH and migraine were not included in the study. The age and sex distribution of 99 cases included in the study are shown in Table 1. Study inclusion criteria and radiological studies are summarised in Table 2.

Table 1.

Age and sex distribution of 99 cases included in the study.

	Total	Sex		Age, Mean ± SD; min–max
		Male	Female
Group 1 (IIH)	32	3	29	44.33 ± 7.57; 39–53
Group 2 (migraine)	34	8	26	34.38 ± 15.48; 18–62
Group 3 (control)	33	11	22	42.64 ± 14.02; 17–60
Total	99	22	77	39.86 ± 14.05; 17–62

IIH: idiopathic intracranial hypertension.

Table 2.

Inclusion criteria according to the groups.

Study groups	Routine brain MRI	Contrast enhanced MRV and/ or contrast 3D GRE T1	Migraine diagnosis with International Headache Society criteria	Not having headache or migraine
Group 1 (IIH)	+	+	−	−
Group 2 (migraine)	+	+	+
Group 3 (control)	+	+	−	+

MRI: magnetic resonance imaging; MRV: magnetic resonance venography; GRE: gradient-recalled echo.

Routine MRI sequences as well as 3D contrast-enhanced MRV and/or 3D contrast-enhanced T1-weighted (T1W) gradient-recalled echo (GRE) and maximum intensity projection (MIP) reconstructions obtained using a 1.5 T MRI scanner were studied. MRI examinations were done with 1.5 T Signa Excite and 1.5 T Optima MR 360 devices (General Electric, Boston, MA). Routine brain MRI was undertaken using 8- and 16-channel head coils: SE T1W FLAIR axial (TR/TE 1643/37; FA: 110; slice thickness: 5.5 mm; gap: 1.6 mm), TSE T2W axial (TR/TE 3600/73), TSE T1W FLAIR sagittal (TR/TE 1571/65; FA: 160; cross-sectional thickness: 5.5 mm; gap: 1.6 mm), TSE T2W FLAIR axial (cross-section thickness: 5.6 mm; gap 1.6 mm) 110 (cross-section thickness 5.5 mm; gap 1.5 mm) and TSE T2W coronal (TR/TE 2005/110, FA: 90; slice thickness: 6 mm; gap: 1 mm).

Contrast-enhanced MRV Fluoro/Auto Triggered 3B Gd-Enhanced MRV (TR/TE 6.1/1.6; FA: 30; slice thickness: 1.4 mm; NEX: 1 mm) was applied. An axial 3D GRE T1W (BRAVO) sequence was then made (TR = 10.7 ms; TE = 4.3 ms; matrix = 256 × 256, cross-section = 176, FOV = 256 × 256 mm²; NEX = 1, slice thickness = 1 mm). The voxel size in this sequence was set to 1 mm×1 mm×1 mm and isotropic voxel. These images were reconstructed using the MIP method with the same image quality on each of the three planes. MRV raw images were also evaluated by reconstruction of three plaques using the MIP method.

Headache, papilloedema, visual impairment, optic disc protrusion, visual loss and CSF opening pressure were evaluated in group 1. All patients and controls were assessed for optic-nerve sheath distention, flattened posterior globe/sclera and decreased height of the pituitary gland. Optic-nerve sheath distention is diagnosed in the presence of measured CSF distance around the optic nerve.⁷ For the evaluation, stage is graded as 0 if there is no CSF space around the optic nerve, grade 1 if the measured CSF space around the optic nerve is >1 but <2 mm and grade 2 if the CSF space around the optic nerve is >2 mm (Figure 1).⁹

Figure 1.

A 50-year-old woman with idiopathic intracranial hypertension (IIH). (a) The pituitary gland height decreased on sagittal 3D T1-weighted (T1W) image. (b) T2W axial image shows bilateral optic-nerve sheath distention (grade 2). (c) There was bilateral transverse sinus (TS) lateral part stenosis on magnetic resonance venography (MRV) maximum intensity projection (MIP) imaging. Transverse sinus stenosis score 1 + 1 = 2. (d) Posterior globe flattening is seen on 3D T1W sagittal image.

TS stenosis was also graded in all three groups. Sinus hypoplasia was considered normal. Unilateral focal stenosis was graded as 1 and focal discontinuity as 2. TS were evaluated using a total stenosis score and distribution of stenosis as unilateral or bilateral (0 = normal, 1 = unilateral stenosis, 2 = bilateral stenosis or unilateral discontinuity, 3 = unilateral stenosis and discontinuity, 4 = bilateral discontinuity; Figure 2).^8,9 Positive and negative predictive values of contrast-enhanced 3D GRE T1W were 91% and 92%, respectively, when grading TS stenosis compared to contrast-enhanced MRI. Most of the Group 1 cases and some of the group 2 and group 3 cases were evaluated with contrast-enhanced MRV. Contrast-enhanced 3D GRE T1W sequences and MIP reconstructions were used in other cases.

Figure 2.

A 40-year-old woman with IIH. (a) MRV MIP image showed left lateral TS stenosis and absence of the right TS. TS stenosis score 1 + 2 = 3. (b) T2W axial image showed bilateral optic-nerve sheath distention (grade 2) and flattening of the right posterior globe. (c) 3D T1WW sagittal image with contrast showed decreased pituitary gland height.

Pearson chi-square and Fisher’s exact tests were used in group comparisons. A p-value of <0.05 was considered statistically significant. Statistical analysis of the data was performed using IBM SPSS Statistics for Windows v22 (IBM Corp., Armonk, NY).

Results

The clinical features of IIH (group 1) are shown in Table 3. MRI findings which were significant in distinguishing IIH cases from healthy individuals and migraine patients are represented by pituitary gland thinning, posterior globe flattening and optic-nerve sheath distention (Table 4). While presence of bilateral TS stenosis or discontinuity is of importance in this differentiation of IIH from migraine cases, unilateral stenosis or discontinuity is not (Table 4).

Table 3.

Clinical features of IIH cases (group 1).

Clinical features	n	%
Headache
Present	29	90.6
Absent	3	9.4
Visual impairment
Present	3	9.4
Absent	29	90.6
Blurred vision
Present	21	65.6
Absent	11	34.4
Papilloedema
Present	28	87.5
Absent	4	12.5
CSF opening pressure (mm H2O)
≤200	3	9.4
201–250	6	18.8
>250	23	71.8

CSF: cerebrospinal fluid.

Table 4.

Distribution of groups according to radiological features.

Radiological features	Group						Total		p a
	Group 1		Group 2		Group 3
	n	%	n	%	n	%	n	%	1-2-3	1-2	1-3	2-3
Hypophysis height
<3 mm	13	40.6	3	8.8	1	3.0	17	17.2	<0.001	0.003	<0.001	0.614
≥3 mm	19	59.4	31	91.2	32	97.0	82	82.8
Optic-nerve sheath distention
0	1	3.1	23	67.6	23	69.7	47	47.5	<0.001	<0.001	<0.001	1.000
1	23	71.9	11	32.4	10	30.3	44	44.4	0.028	0.040	0.024	0.827
2	8	25.0	0	0.0	0	0.0	8	8.1	–	–	–	–
Flattened posterior globe
Present	29	90.6	2	5.9	1	3.0	32	32.3	<0.001	<0.001	<0.001	1.000
Absent	3	9.4	32	94.1	32	97.0	67	67.7
Optical disc protrusion
Present	9	28.1	–	–	–	–	–	–	–	–	–	–
Absent	23	71.9	–	–	–	–	–	–

Statistically significant values are shown in bold.

^aPearson chi-square test, Fisher’s exact test, chi-square trend analysis.

TS stenosis, stenosis scores and stenosis characteristics (unilateral, bilateral, stenosis, absence, discontinuity) were statistically evaluated among the groups (Table 5). Accordingly, there was a significant difference between the IIH and migraine groups in bilateral stenosis (p = 0.02). There was no significant difference in unilateral stenosis between the migraine and control groups. The distribution of TS stenosis among the groups is also shown in Figure 3.

Table 5.

Distribution of transverse sinus stenosis.

Score		Group 1		Group 2		Group 3		Total		1-2-3	1-2	1-3	2-3
0	Normal	8	25.0	23	67.6	23	69.7	54	54.5	0.016	0.007	0.007	1.000
1	Unilateral stenosis	8	25.0	6	17.6	5	15.2	19	19.2	0.692	0.593	0.405	0.763
2	Bilateral stenosis	12	37.5	3	8.8	4	12.1	19	19.2	0.021	0.020	0.046	0.705
2	Unilateral discontinuity	2	6.3	2	5.9	1	3.0	5	5.1	0.819	1.000	0.564	0.564
3	Unilateral stenosis+unilateral absence	2	6.3	0	0.0	0	0.0	2	2.0	–	–	–	–

Statistically significant values are shown in bold.

Figure 3.

Graphical distribution of transverse sinus stenosis among the groups.

Discussion

In our study, cases of IIH and migraine, both of which are characterised by headaches and in some cases overlapping clinical presentations, were compared together with a control group in terms of cranial MRI and MRV findings. Accordingly, among the MRI findings observed in IIH patients, the decrease in pituitary gland height, optic-nerve sheath distention and flattened posterior globe are significant findings in differentiating IIH cases from control patients and migraine patients. While bilateral narrowing or discontinuity in the TS is significant in this distinction, unilateral stenosis or discontinuity is insignificant.

Farb et al. diagnosed sino-venous stenosis in 27/29 IIH cases, scoring between 0 and 4 for TS and sigmoid sinus on both sides by using contrast-enhanced MRV.⁷ Using this scoring system, they reported that they were able to detect IIH cases with 93% sensitivity and 93% specificity.

However, in this study, hypoplasic sinus and severe sinus stenosis were equally scored, and very common sinus hypoplasia was assumed to be as important as stenosis in the pathogenesis of IIH. Higgins et al. used phase contrast (PC) or time of flight (TOF) MRV for the evaluation of the TS. Focal stenosis was numbered as 1 and focal discontinuity as 2, as in our study. They compared the total score obtained from both sides in IIH cases and the control group. According to their study, there was focal discontinuity in TS bilaterally in 13/20 IIH cases. This finding was not seen in the control group (0/40).⁸

In our series, bilateral TS stenosis was seen in 12/32 IIH cases. Unilateral stenosis and unilateral absence were detected in 2/32 cases. Bilateral stenosis or discontinuity was present in 14/32 (43.75%) cases in our series and in 16/20 (80%) in the series of Higgins et al.⁸ TOF or PC MRV could be one of the causes of incompatibility between these studies. These techniques are sensitive to current artefacts and can cause false absences or stenoses due to flow, especially in the lateral parts of the TS. In this series, instead of sinus stenosis, sinus discontinuity was seen, suggesting that stenosis due to flow artefacts was exaggerated. In our series, more contrast 3D MRV and less 3D GRE T1W contrast series were used, and both methods were not significantly affected by flow artefacts. Indeed, while bilateral stenosis was 12/32 in our series, discontinuity was only present in three cases.

Rohr et al. reported that 15/17 (88%) had bilateral TS stenosis in 3D contrast-enhanced MRI in IIH patients, and 7/15 (47%) returned to normal after treatment.¹⁰ Saindane et al. reported TS stenosis in patients with visual impairment of up to 90% in IIH patients.¹¹

According to the Monro and Kellie hypothesis, if volume increases in one of the intracranial components (blood, CSF or brain tissue), there should be a reduction in the volume of the other components.¹² Otherwise, intracranial pressure increases. There is no pathology that may lead to an increase in intracranial pressure in IIH. For the diagnosis, the dural sinus thrombosis should be excluded. For this purpose, a contrast-enhanced 3D GRE T1W should be performed, which may be MRV or equivalent. Some authors have reported lateral TS stenosis on MRI in patients with suspected sinus thrombosis.^7,8,10 The cause of these stenosis is debatable. In a group of patients, it was shown that clinical findings improved after stenting/angioplasty to treat sinus stenosis, thus assuming that these stenoses are the cause of the disease.^7,13–15

On the other hand, sinus stenosis has resolved in many patients with continuous CSF drainage and other CSF drainage methods, such as cervical or lumbar puncture.^7,16,17 This finding suggests that sinus stenosis is the result rather than the cause of the increase in intracranial pressure. Alternatively, a positive feedback mechanism between venous stenosis and intracranial pressure is suggested.¹⁸ Accordingly, increased venous pressure (intracranial pressure) leads to stenosis of the elastic TS walls by external pressure effect; sinus stenosis increases the venous pressure because the blood is prevented from flowing to the extracranial compartment. Endovascular stenting in the venous sinuses may be a good way to overcome this vicious cycle. In a review article in 2010, it was reported that up to 60 patients were treated with a stent, more than 80% had headache improvement and 88% had a reduction in papilloedema.¹⁹ Although these results seem to be successful, it is unclear whether postoperative headache and papilloedema are regressive, placebo-controlled, intervention-dependent or disease-associated in nature. Furthermore, venous stenting is an invasive procedure. In the treatment of IIH, weight loss, causative agent removal or medical treatment results in improvement. Stenting seems to be a method that can be selected in cases of resistance to other treatments.^14,19 In our study, bilateral TS stenosis was found to be statistically significant when compared to healthy people and migraine patients in IIH cases. Another feature of the stenosis in our study was that it was symmetrical on both sides. However, unilateral sinus stenosis or discontinuity was similar in all three groups. This finding suggests that stenosis aetiology may be due to external compression and secondary stenosis on bilateral TS lateral parts due to increased intracranial pressure. If the aetiological primary sinus stenosis was in the anterior segment, the stenosis was mostly unilateral (due to past thrombosis, etc.), it is expected to be asymmetric even if it is bilateral.

In IIH cases, the incidence of headache was 90.6%, papilloedema 87.5%, blurred vision 65.6% and vision loss 9.4%. Headache has been reported in most cases (90–94%). Optic disc oedema due to increased intracranial pressure is one of the main findings in IIH as well. The more the severe papilloedema, the greater the visual impairment is.²⁰

Papilloedema does not occur in every IIH case. Sometimes, it may also be unilateral. Clinical diagnosis becomes difficult in such cases.²¹ Papilloedema was found in 87.5% of our IIH cases. Mallery et al. reported that with the presence of at least three of the four criteria on MRI, IIH can be diagnosed even in the absence of papilloedema or sixth nerve palsy. These criteria are the presence of empty sellae, flattening of the posterior globe, TS stenosis and the distension of the subarachnoid space, which can be seen with or without a tortuous optic nerve.²² Although our MRI findings are generally consistent with this study, we noticed that it is important to distinguish IIH cases from migraine cases with bilateral TS stenosis. Permanent vision loss is the most feared complication of IIH. However, visual disturbances are usually transient and occur in 68–85% of cases.²² In our study, the CSF opening pressure was found to be >250 mm H₂O in 71.8% (23/32) of cases. In the literature, the issue of the minimum CSF opening pressure required for the diagnosis of IIH is controversial. But, as generally accepted, 250 mm H₂O is in accordance with IIH. Values <200 mm H₂O occur in healthy individuals. Values between 200 and 250 mm H₂O are controversial. Clinical findings were consistent with the literature.

In IIH, the height of the pituitary gland was <3 mm in 19/32 (59.4%) cases. In the literature, although there are many different results regarding the height of the pituitary gland in IIH cases, it is accepted that chronically increased intracranial pressure leads to arachnoid herniation downward from the diaphragma sellae and thus to a decrease in the height of the pituitary (partial empty sellae) or that it causes empty sellae, especially in chronic cases.²³ Differences in height between different IIH series may depend on the chronicity of the event and whether the patient has been treated.

In our IIH series of 31/32 (97%) cases, there were varying degree of optic-nerve sheath distention. The optic-nerve sheath distention was grade 1 in 23 cases and grade 2 in eight cases. In one study, the optic-nerve sheath distention was found to be 67% sensitive and 82% specific for IIH. The same authors found 43% sensitivity and 100% specificity for the flattened posterior globe or sclera. In this study, it was concluded that the flattened posterior globe/sclera, if any, is the most important cross-sectional imaging finding suggestive of IIH; other findings were reported to be relatively insignificant for diagnosis.²⁴

Intracranial pressure changes (intracranial hypertension or hypotension) can cause headaches.²⁴ A clinical challenge in this group of patients is accompanying migraine headaches. The clinical presentation often overlaps with headaches due to pressure changes and migraine headaches. Another clinical challenge is that the pain does not recover immediately, even if appropriate treatment is given. Because intracranial hypertension and migraine patients are mostly women, it is difficult to make a differential diagnosis.²⁵

TS stenosis is reported as a common radiological finding (47.5%) in chronic migraine and in chronic tension-type headache patients. Although TS stenosis was not statistically significant in these patient groups, it was noteworthy that bilateral TS stenosis was 44.4% in patients, with a CSF opening pressure of >200 mmH₂O.²⁶ This finding correlates with our conclusion. We can conclude that bilateral TS stenosis may be the cause not the result of increased intracranial pressure. Kosmorsky reported up to 10% of cases of IIH did not have headaches and sometimes did not have papilloedema.²⁷ Knash reported that a 16-year-old woman presented with a headache and had increased pressure in repeated CSF pressure measurements.²⁵

We investigated the frequency of imaging findings observed in the IIH group, healthy individuals and migraine patients. There was no difference between the migraine group and the control group in terms of IIH findings. We found that these findings were significantly higher in the IIH group compared to the other two groups. The pathophysiology of migraine is complex and not yet fully understood. In its aetiology, vascular and neurovascular mechanisms, neurotransmitters, vasoactive substances and oligaemia may be considered. Considering that the cross-sectional imaging findings observed in IIH are due to increased intracranial pressure, it can be assumed that this mechanism will not be valid in migraine cases.

Limitations

Our study is retrospective. The data were collected from the hospital information management system by evaluating patient records and images. The time between the onset of the disease and the first MRI series was not standardised. In some cases, the initial imaging was performed before the onset of the attack. So, the imaging findings may be partially regressed. Another limitation is the use of contrasted-enhanced 3D MRV in some cases and contrast-enhanced 3D GRE T1W sequence in the evaluation of sinus stenosis. Although these two methods give similar results in terms of stenosis, contrast-enhanced MRV is superior. In all cases, 3D contrast-enhanced MRV may be expected to be more accurate. TS stenosis was evaluated as unilateral or bilateral focal stenosis or absence, and the sinus volumes were not calculated. In our study, it was not possible to evaluate this condition together with the clinic aspect because the control MRV was not taken after the normalisation of CSF pressure. As the body mass index of all patients could not be reached, the relationship between obesity and the clinical findings of the disease could not be evaluated.

Conclusion

MRI of IIH revealed a decrease in the height of the pituitary gland, optic-nerve sheath distention and flattened posterior globe/sclera. They are significant findings in differentiating IIH cases from healthy individuals and migraine patients. Bilateral TS stenosis is also an important finding for the differentiation of IIH cases from patients with migraine and healthy individuals. Unilateral TS stenosis and discontinuity are insignificant for differential diagnosis. The presence of bilateral TS stenosis on MRV in patients with headache may indicate IIH rather than migraine.

We can conclude that bilateral TS stenosis may be the cause not the result of increased intracranial pressure. The increase in intracranial pressure, which is considered to be responsible for the pathophysiology of IIH, is not involved in the pathophysiology of migraine.

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Melda Apaydin https://orcid.org/0000-0001-6841-4988