CLINICAL ARTICLEJ Neurosurg 130:130–135, 2019

IdIopathIc normal pressure hydrocephalus (iNPH) is recognized by the following symptoms: disturbances of balance and gait, urgency incontinence, and cog-

nitive decline.1 Demonstration of typical morphological findings on neuroimaging plays an important role in sup-porting the diagnosis.4,13 The standard treatment is implan-tation of a ventriculoperitoneal shunt, which results in im-provement in 50%–80% of the patients.8,15

In patients who do not improve postoperatively, a mal-functioning shunt should be ruled out before accepting treatment failure. Ideally, shunt function should be evalu-

ated before proceeding with neurosurgical revision. Com-mon tests of shunt function include the lumbar infusion test10 and radionuclide shuntography.16 Both tests are inva-sive and not available at all neurosurgical centers, there-fore unnecessary revisions may be performed. Thus, easily available radiological methods to evaluate shunt function are warranted.

In rapidly developing noncommunicating and post-hemorrhagic hydrocephalus, the size of the ventricles is often markedly reduced after shunting, which obviously indicates a functioning shunt. Unfortunately, this is not al-

ABBREVIATIONS AC-PC = anterior commissure–posterior commissure; CA = callosal angle; ICC = intraclass correlation coefficient; iNPH = idiopathic normal pressure hydrocephalus; IQR = interquartile range; MMSE = Mini-Mental State Examination; QRAPMASTER = quantification of relaxation times and proton density by multiecho acquisition of saturation-recovery with turbo spin-echo readout.SUBMITTED March 3, 2017. ACCEPTED August 1, 2017.INCLUDE WHEN CITING Published online February 2, 2018; DOI: 10.3171/2017.8.JNS17547.

Increase in callosal angle and decrease in ventricular volume after shunt surgery in patients with idiopathic normal pressure hydrocephalusJohan Virhammar, MD, PhD,1 Katarina Laurell, MD, PhD,2 Kristina Giuliana Cesarini, MD, PhD,3 and Elna-Marie Larsson, MD, PhD4

Department of Neuroscience, 1Neurology and 3Neurosurgery; and 4Department of Surgical Sciences, Radiology, Uppsala University, Uppsala; and 2Department of Pharmacology and Clinical Neuroscience, Unit of Neurology, Umeå University, Umeå, Sweden

OBJECTIVE Postoperative decrease in ventricle size is usually not detectable either by visual assessment or by mea-suring the Evans index in patients with idiopathic normal pressure hydrocephalus (iNPH). The aim of the present study was to investigate whether the angle between the lateral ventricles (the callosal angle [CA]) increases and ventricular volume decreases after shunt surgery in patients with iNPH.METHODS Magnetic resonance imaging of the brain was performed before and 3 months after shunt surgery in 18 patients with iNPH. The CA and Evans index were measured on T1-weighted 3D MR images, and ventricular volume contralateral to the shunt valve was measured with quantitative MRI.RESULTS The CA was larger postoperatively (mean 78°, 95% CI 69°–87°) than preoperatively (mean 67°, 95% CI 60°–73°; p < 0.001). The volume of the lateral ventricle contralateral to the shunt valve decreased from 73 ml (95% CI 66–80 ml) preoperatively to 63 ml (95% CI 54–72 ml) postoperatively (p < 0.001). The Evans index was 0.365 (95% CI 0.35–0.38) preoperatively and 0.358 (95% CI 0.34–0.38) postoperatively (p < 0.05). Postoperative change of CA showed a negative correlation with change of ventricular volume (r = -0.76, p < 0.01).CONCLUSIONS In this sample of patients with iNPH, the CA increased and ventricular volume decreased after shunt surgery. The relative difference was most pronounced for the CA, indicating that this accessible, noninvasive radiological marker should be evaluated further as an indirect method to determine shunt function in patients with iNPH.https://thejns.org/doi/abs/10.3171/2017.8.JNS17547KEY WORDS normal pressure hydrocephalus; callosal angle; Evans index; postoperative radiology; quantitative MRI; NPH; shunt dysfunction

J Neurosurg Volume 130 • January 2019130 ©AANS 2019, except where prohibited by US copyright law

Unauthenticated | Downloaded 05/30/22 05:46 PM UTC

J Neurosurg Volume 130 • January 2019 131

J. Virhammar et al.

ways so evident in iNPH, and the Evans index,2 illustrated in Fig. 1, is often unchanged postoperatively.11,12 Measure-ment of ventricular volume based on quantitative MRI is more objective than visual estimation but has not yet been implemented in routine care.

The callosal angle (CA) is a radiological marker of iNPH that can easily be measured on CT or MRI as the angle between the lateral ventricles measured in a coro-nal plane (Fig. 2).18 The CA can be used diagnostically be-cause it is reported to be smaller in iNPH than in Alzhei-mer’s disease and healthy controls.6 The CA may also have positive predictive value.19 There are no reports of whether there is a change in CA after shunt surgery in iNPH.

The aim of this study was to investigate whether the CA increases after shunt surgery in patients with iNPH and if this difference correlates with change in ventricle size and clinical symptoms.

MethodsPatient Population

Twenty-six patients under evaluation for iNPH were recruited prospectively. These patients underwent clinical

examination, MRI of the brain, and lumbar puncture with withdrawal of high-volume CSF. Seventeen patients were classified as probable iNPH, 6 as possible iNPH, and 3 as unlikely iNPH, according to the iNPH guidelines.13

The 3 patients classified as unlikely iNPH were ex-cluded from the study. Also excluded were 2 patients who were not offered shunt surgery, one due to severe comor-bidity and the other because of very mild symptoms. Of the remaining 21 patients who were offered shunt surgery, 2 chose to drop out of the study and 1 died before surgery. Therefore, 18 patients (median age 73.5 years, range 65–81 years) were included in the statistical analysis (Table 1). All patients underwent insertion of a shunt system, i.e., adjustable Medtronic Strata valves. The patients were also described in a previous study.17 The study was approved by the local ethics committee in Uppsala, Sweden.

Clinical Investigation and Time ParametersClinical examinations were performed by neurologists,

physiotherapists, and occupational therapists. Patients were assessed with the Mini-Mental State Examination (MMSE) cognitive test, an ordinal balance scale,5 and a urinary incontinence scale.5 Gait function was evaluated with the timed “up and go” test, and time and number of steps required to walk 10 meters at maximum pace. Tests of gait function were performed twice and the mean value was used. The sum of the number of steps as well as the seconds for both gait tests was combined to produce a quantitative gait variable. The difference of the 4 clinical tests before and after shunt surgery were calculated and correlated with changes in radiological features. MRI of the brain as well as clinical examinations were performed preoperatively and at follow-up. The median duration be-tween preoperative evaluation and surgery was 21 weeks (interquartile range [IQR] 17–35 weeks), and the median time between surgery and follow-up was 18 weeks (IQR 12–33 weeks). A follow-up evaluation that included only clinical investigation was performed 12 months after shunt insertion.

MRI and ProtocolMRI was performed using a Philips 3-T unit (Philips

Achieva, Philips Healthcare) with a 32-channel head coil.

FIG. 1. Axial T1-weighted 3D images of preoperative (A) and postoper-ative (B) MR scans in a 78-year-old man with iNPH. The Evans index is the ratio between the maximum width of the frontal horns (x) of the lateral ventricles and the maximum inner diameter of the cranium (y), in the present study measured in the same slice. Figure is available in color online only.

FIG. 2. Callosal angle measured on T1-weighted 3D images in a 73-year-old woman with iNPH. A: Sagittal image used to identify the AC-PC plane. B: Preoperative investigation illustrating that the CA is measured in the coronal plane through the posterior commissure perpendicular to the AC-PC plane. C: Postoperative MR scan with an artifact due to the shunt valve. Figure is avail-able in color online only.

Unauthenticated | Downloaded 05/30/22 05:46 PM UTC

J. Virhammar et al.

J Neurosurg Volume 130 • January 2019132

A 3D T1-weighted gradient echo (turbo field echo) se-quence was obtained with the following parameters: voxel size 1.0 × 1.0 × 1.0 mm3, TR 8.2 msec, and TE 3.7 msec.

As a quantitative MR method, synthetic MRI was used. The synthetic MRI data were acquired using the multislice, multiecho, and multisaturation delay sequence QRAPMASTER (quantification of relaxation times and proton density by multiecho acquisition of saturation-re-covery with turbo spin-echo readout).21 Because of techni-cal problems, synthetic MRI data were not acquired in 2 of the patients.

Evans Index and CA MeasurementsMeasurements of the Evans index and CA were per-

formed using 3D T1-weighted images with 1-mm slice thicknesses. All axial slices were reconstructed parallel to the plane through the inferior margin of the pituitary gland and the fastigium of the fourth ventricle to make them geometrically comparable. The Evans index was cal-culated as the ratio between the maximum diameter of the frontal horns of the lateral ventricles and the maximum inner diameter of the skull in the same axial slice (Fig. 1). The CA was measured as the angle between the lateral ventricles on a coronal MR image through the posterior commissure, perpendicular to the anterior commissure–posterior commissure (AC-PC) plane.6 Multiplanar re-construction was performed using the radiological picture archiving and communication system in each patient to obtain a correct coronal image (Fig. 2).

Volumetric ImagingMeasurements of the lateral ventricle volume was

performed using the QRAPMASTER sequence,21 which provides rapid simultaneous quantification of longitudi-nal relaxation time (T1), transverse relaxation time (T2), and proton density. SyMRI Brain Studio (SyntheticMR) is a software that uses combinations of T1, T2, and proton density to estimate voxel-wise partial volumes of gray mat-ter, white matter, and CSF, which allows volumetric esti-mation (in milliliters) of these tissues. To avoid artifacts from the metal parts in the shunt valve, only the ventricle contralateral to the shunt was segmented. The volume was calculated semiautomatically using SyMRI (Fig. 3).

Statistical AnalysisThe differences between preoperative and postopera-

tive imaging data and quantitative gait tests were tested using a paired-samples t-test. Differences in symptom scales measured with ordinal scales were tested with the Wilcoxon signed-rank test. Correlations were assessed with Spearman’s correlation. The level of significance was set to p < 0.05, and all analyses were performed using IBM SPSS (version 23.0, IBM Corp.).

ResultsIn the total sample of 18 patients, the CA was signifi-

cantly smaller preoperatively (mean 67°, 95% CI 60°–74°) than postoperatively (mean 78°, 95% CI 69°–87°; p < 0.001). Concordantly, the mean unilateral ventricular vol-ume decreased from 73 ml (95% CI 66–80 ml) preopera-tively to 63 ml (95% CI 54–72 ml) postoperatively (p < 0.001). For the Evans index, the preoperative mean result was 0.365 (95% CI 0.35–0.38) and the postoperative result was 0.358 (95% CI 0.34–0.38), with a mean difference of -0.0071 (range -0.040 to 0.014; p < 0.05; Fig. 4). Results from clinical tests before and after shunt surgery are listed in Table 1.

The postoperative change in CA correlated with the

TABLE 1. Patient characteristics and imaging data before and after shunt surgery

Variable Preop Postop p Value

Median age in yrs (range) 73.5 (65–81) — —No. of men/women (%) 8/10 (44/56) — —Median MMSE score (IQR) 25 (21–27) 25.5 (23.75–29) NS*Median urgency incontinence scale score (IQR) 3 (1.75–4) 3 (2–4) NS*Median balance scale score (IQR) 4 (3.5–4.5) 4 (4–5) NS*Mean score of quantitative gait tests (95% CI) 28 (18–37) 22 (16–27) <0.05†Mean CA (95% CI) 67° (60°–73°) 78° (69°–87°) <0.001†Mean Evans index (95% CI) 0.365 (0.35–0.38) 0.358 (0.34–0.38) <0.05†Mean lateral ventricle volume in ml (95% CI) 73 (66–80) 63 (54–72) <0.001†

NS = not significant.* Wilcoxon signed-rank test.† Paired-samples t-test.

FIG. 3. Preoperative (A) and postoperative (B) axial MR images show-ing segmentation of the one lateral ventricle contralateral to the shunt valve measured semiautomatically using SyMRI. Figure is available in color online only.

Unauthenticated | Downloaded 05/30/22 05:46 PM UTC

J Neurosurg Volume 130 • January 2019 133

J. Virhammar et al.

change in quantified ventricular volume (r = -0.76, p < 0.01). There were no correlations between the Evans index and CA or quantified ventricular volume preoperatively. Nor were there any correlations between the differences in the clinical symptoms postoperatively with the difference in any of the radiological parameters in the whole sample of patients.

In 3 of 18 patients, the CA and quantified ventricular volume did not change postoperatively (< 1°), and clinical improvement was absent or moderate. In the other 15 pa-tients, the mean CA increased by 14° (range 5°–31°), mean ventricular volume decreased by 12 ml (95% CI -16 to -7 ml), and the mean Evans index decreased by 0.0065 (95% CI -0.014 to 0.00065). In addition, in these 15 patients mean gait performance (quantitative gait variable) im-proved by 22% (95% CI 8%–36%) and median MMSE by 2 points (IQR 0–3 points). After evaluation at the 3-month follow-up, the 3 patients who had not improved had the opening pressure of their Strata valves lowered from 1.5 to 1.0, from 2.0 to 1.5, and from 2.0 to 1.0. One of the patients also underwent shunt revision. At the 12-month follow-up of these 3 patients, gait had improved by 23%, 15%, and 16%, and MMSE changed by -1, -1, and +3 points. In ad-dition, in 2 of the patients, CT scans performed 6 weeks after the intervention revealed an increase in the CA of more than 20° compared with preoperative investigations. The corresponding findings in the other 15 patients at the 12-month follow-up were an improvement in gait of 8% (95% CI -27% to 44%) and a change in median MMSE of 1 point (IQR 0–3 points).

DiscussionIn this sample of patients with iNPH, the CA increased

and ventricular volume decreased 3 months after shunt surgery. The relative change was most pronounced for the CA with a mean increase of 11°. A postoperative decrease was shown for the Evans index as well, but it was too small to be visually detectable or to compensate for errors in measurement.

The CA has been reported to be of both diagnostic and

prognostic value in patients with iNPH.6,18,19 The increase of the CA after shunt surgery in the present study was highly significant, and unlike the Evans index, this dif-ference should be visible (Fig. 2). Measurement of the CA is reported to be reliable, with high intrarater agreement (intraclass correlation coefficient [ICC] 0.97) and interra-ter agreement (ICC 0.95–0.98).6,18 It is important, however, to perform the measurements as described by Ishii et al.6 because the angle varies depending on the plane and is larger in the frontal region. The CA was first described using pneumoencephalography9 and can be measured on any MR image with a 3D sequence and also on a standard multidetector CT scan.

Quantitative MRI (SyMRI) detected a significant de-crease of ventricular volume after shunt surgery in the present study. Reduction of intracranial CSF volume was also found after high-volume lumbar punctures in a pre-vious study, supporting the notion that the method can be used for assessing dynamic changes.20 The reliability of volumetric measurements using quantitative MRI is good,20 but as yet they are not being widely used in the routine clinical setting; at present, it requires specific se-quences and software.3

The Evans index decreased significantly after shunt treatment in the present study. Even though this reduction was measurable, it can be difficult to visualize. The mean difference in this study was -0.0071, which corresponds to a reduction in frontal horn diameter of 0.9 mm in a typical patient with iNPH with a frontal horn diameter of 45 mm and an inner skull diameter of 130 mm. The Evans index is a good screening tool to detect enlarged ventricles and is easily accessible because it can be measured on any CT or MRI scan. On the other hand, the index reveals no information about the shape of the ventricles.

There was a significant correlation between enlarge-ment of the CA and decrease of ventricular volume, but neither of these measures correlated with differences in the Evans index. This indicates that the Evans index is not as sensitive to change after shunting as the other tested methods.

A well-known sign of iNPH is compressed high-con-

FIG. 4. Before-and-after graphs with lines connecting the results of each individual patient before and after shunt surgery. Red bars represent the mean with 95% CI error bars. A: Evans index. B: Callosal angle. C: Quantified unilateral ventricular volume. *p < 0.05, ***p < 0.001. Figure is available in color online only.

Unauthenticated | Downloaded 05/30/22 05:46 PM UTC

J. Virhammar et al.

J Neurosurg Volume 130 • January 2019134

vexity sulci. In clinical practice, it is common to assess the high-convexity sulci, which are often better seen (less compressed) after shunt surgery.7 However, the interrater reliability of this finding is lower than for the Evans index and the CA.19

We found no significant correlations between differ-ences in pre- and postoperative radiological measures and clinical improvement, which might be due to the small sample size and the clinical tests used in this study. How-ever, in 3 patients with the smallest change of the CA and the ventricular volume, the clinical improvement was ab-sent or moderate compared with the rest of the patients. After intervention, all 3 patients had improved even more than the rest of the patient sample, and in 2 of them the CA increased by more than 20° compared with the preop-erative investigation. These findings may indicate that an unchanged CA postoperatively is a sign of underdrainage of CSF, and should be confirmed in larger samples as no meaningful statistical analysis can be performed in only 3 patients.

The available tests of shunt function are invasive and not accessible at all centers. Other noninvasive methods have been suggested, such as measurement of otoacoustic emissions to determine shunt function.14 Measuring the CA postoperatively is easily assessable because a postop-erative CT scan is performed routinely after shunt surgery and is repeated to rule out complications if the patient has not improved.

The importance of the ventricle size after shunt surgery in iNPH has been questioned; it has been reported that the diameter of the lateral ventricles is unchanged after shunt surgery in a majority of patients with iNPH and that there is no correlation between clinical outcome and change in ventricle size.11,12 In the present study there was a clearly measurable change in both size and shape of the lateral ventricles after shunt surgery, but no correlations were found between changes in imaging findings and clinical outcome at a group level. The subgroup of patients with unchanged CA and ventricle size was too small to be ana-lyzed statistically. If a relationship between increased CA and positive clinical outcome exists, it remains undefined; and if measurements of ventricle size after shunting are important, they have yet to be elucidated.

Some limitations need to be considered. The artifact from the metal parts of the shunt valve was quite large using 3-T MRI. It is possible that this affected the mea-surements postoperatively. However, the central parts of the brain, where the CA and Evans index are measured, are less affected by the artifact. The measurements of ventricular volume with SyMRI were only performed on the contralateral ventricle to the shunt to avoid the arti-fact. No blinding was performed because it was obvious to the investigator which scan was from the postoperative investigation because of the presence of the metal artifact. However, the investigators were blinded to clinical data. Even if measurements of the Evans index were standard-ized, the vertical positioning of the slice that is used for the measurements could potentially differ slightly between in-vestigations. The MMSE, which was the only cognitive test used in the study, has limitations in the context of iNPH because it underestimates subcortical deficits. Also, a clinical scale that combines all symptoms of iNPH (such

as the Hellström iNPH scale5) is probably more sensitive to change than 4 separate tests as were used in this study. This might have influenced the lack of correlations be-tween change in clinical symptoms and imaging features.

ConclusionsIn this sample of patients with iNPH, the CA increased

and ventricular volume decreased after shunt surgery. The relative difference was largest for the CA, suggesting that there could be an association between drainage of CSF and a larger CA postoperatively. The CA is an easily ac-cessible, noninvasive radiological marker that may have the potential to become an indirect method to assess shunt function in patients with iNPH.

AcknowledgmentsJohan Virhammar was supported by the independent Swedish

foundation Erik, Karin och Gösta Selanders Stiftelse; Katarina Laurell received research grants from “Syskonen Persons dona-tionsfond” and “ALF” (i.e., the financial agreement between the Swedish government and Universities concerning research and education of doctors) at Region Jämtland Härjedalen.

References 1. Adams RD, Fisher CM, Hakim S, Ojemann RG, Sweet WH:

Symptomatic occult hydrocephalus with “normal” cerebro-spinal-fluid pressure. A treatable syndrome. N Engl J Med 273:117–126, 1965

2. Evans WA Jr: An encephalographic ratio for estimating ventricular enlargement and cerebral atrophy. Arch Neurol Psychiatry 47:931–937, 1942

3. Granberg T, Uppman M, Hashim F, Cananau C, Nordin LE, Shams S, et al: Clinical feasibility of synthetic MRI in mul-tiple sclerosis: a diagnostic and volumetric validation study. AJNR Am J Neuroradiol 37:1023–1029, 2016

4. Hashimoto M, Ishikawa M, Mori E, Kuwana N: Diagnosis of idiopathic normal pressure hydrocephalus is supported by MRI-based scheme: a prospective cohort study. Cerebrospi-nal Fluid Res 7:18, 2010

5. Hellström P, Klinge P, Tans J, Wikkelsø C: A new scale for assessment of severity and outcome in iNPH. Acta Neurol Scand 126:229–237, 2012

6. Ishii K, Kanda T, Harada A, Miyamoto N, Kawaguchi T, Shimada K, et al: Clinical impact of the callosal angle in the diagnosis of idiopathic normal pressure hydrocephalus. Eur Radiol 18:2678–2683, 2008

7. Kiefer M, Eymann R, Meier U: Five years experience with gravitational shunts in chronic hydrocephalus of adults. Acta Neurochir (Wien) 144:755–767, 2002

8. Klinge P, Hellström P, Tans J, Wikkelsø C: One-year out-come in the European multicentre study on iNPH. Acta Neu-rol Scand 126:145–153, 2012

9. LeMay M, New PF: Radiological diagnosis of occult normal-pressure hydrocephalus. Radiology 96:347–358, 1970

10. Malm J, Sundström N, Cesarini KG, Edsbagge M, Kristensen B, Leijon G, et al: Implementation of a new CSF dynamic device: a multicenter feasibility study in 562 patients. Acta Neurol Scand 125:199–205, 2012

11. Meier U, Paris S, Gräwe A, Stockheim D, Hajdukova A, Mutze S: Is there a correlation between operative results and change in ventricular volume after shunt placement? A study of 60 cases of idiopathic normal-pressure hydrocephalus. Neuroradiology 45:377–380, 2003

Unauthenticated | Downloaded 05/30/22 05:46 PM UTC

J Neurosurg Volume 130 • January 2019 135

J. Virhammar et al.

12. Mocco J, Tomey MI, Komotar RJ, Mack WJ, Frucht SJ, Goodman RR, et al: Ventriculoperitoneal shunting of idio-pathic normal pressure hydrocephalus increases midbrain size: a potential mechanism for gait improvement. Neurosur-gery 59:847–851, 2006

13. Relkin N, Marmarou A, Klinge P, Bergsneider M, Black PM: Diagnosing idiopathic normal-pressure hydrocephalus. Neu-rosurgery 57 (3 Suppl):S4–S16, ii–v, 2005

14. Sakka L, Chomicki A, Gabrillargues J, Khalil T, Chazal J, Avan P: Validation of a noninvasive test routinely used in otology for the diagnosis of cerebrospinal fluid shunt mal-function in patients with normal pressure hydrocephalus. J Neurosurg 124:342–349, 2016

15. Sundström N, Malm J, Laurell K, Lundin F, Kahlon B, Cesa-rini KG, et al: Incidence and outcome of surgery for adult hy-drocephalus patients in Sweden. Br J Neurosurg 31:21–27, 2017

16. Thompson EM, Wagner K, Kronfeld K, Selden NR: Using a 2-variable method in radionuclide shuntography to predict shunt patency. J Neurosurg 121:1504–1507, 2014

17. Virhammar J, Laurell K, Ahlgren A, Cesarini KG, Larsson EM: Idiopathic normal pressure hydrocephalus: cerebral perfusion measured with pCASL before and repeatedly after CSF removal. J Cereb Blood Flow Metab 34:1771–1778, 2014

18. Virhammar J, Laurell K, Cesarini KG, Larsson EM: The callosal angle measured on MRI as a predictor of outcome in idiopathic normal-pressure hydrocephalus. J Neurosurg 120:178–184, 2014

19. Virhammar J, Laurell K, Cesarini KG, Larsson EM: Pre-operative prognostic value of MRI findings in 108 patients

with idiopathic normal pressure hydrocephalus. AJNR Am J Neuroradiol 35:2311–2318, 2014

20. Virhammar J, Warntjes M, Laurell K, Larsson EM: Quan-titative MRI for rapid and user-independent monitoring of intracranial CSF volume in hydrocephalus. AJNR Am J Neuroradiol 37:797–801, 2016

21. Warntjes JB, Leinhard OD, West J, Lundberg P: Rapid mag-netic resonance quantification on the brain: Optimization for clinical usage. Magn Reson Med 60:320–329, 2008

DisclosuresThe authors report no conflict of interest concerning the materi-als or methods used in this study or the findings specified in this paper.

Author ContributionsConception and design: Virhammar. Acquisition of data: Virham-mar, Cesarini. Analysis and interpretation of data: Virhammar, Laurell, Larsson. Drafting the article: Virhammar. Critically revising the article: all authors. Reviewed submitted version of manuscript: Virhammar. Approved the final version of the manu-script on behalf of all authors: Virhammar. Statistical analysis: Virhammar. Study supervision: Larsson.

CorrespondenceJohan Virhammar: Uppsala University, Uppsala, Sweden. johan.virhammar@neuro.uu.se.

Unauthenticated | Downloaded 05/30/22 05:46 PM UTC