Open Access
CC BY 4.0 · Indian Journal of Neurosurgery
DOI: 10.1055/s-0045-1811656
Original Article

Magnetic Resonance Imaging–Based Morphometric Analysis of the Tentorial Notch in a North Indian Population: A Descriptive Anatomical Study

Authors

  • Ashwin K. Panda

    1   Department of Neurosurgery, All India Institute of Medical Sciences Rishikesh, Rishikesh, Uttarakhand, India
  • Saravanan Sadhashivam

    1   Department of Neurosurgery, All India Institute of Medical Sciences Rishikesh, Rishikesh, Uttarakhand, India
  • Rajnish K. Arora

    1   Department of Neurosurgery, All India Institute of Medical Sciences Rishikesh, Rishikesh, Uttarakhand, India
  • Radhey Shyam Mittal

    1   Department of Neurosurgery, All India Institute of Medical Sciences Rishikesh, Rishikesh, Uttarakhand, India
  • Sudhir Saxena

    2   Department of Radiodiagnosis and Imaging, All India Institute of Medical Sciences Rishikesh, Rishikesh, Uttarakhand, India
  • Khanak Nandolia

    2   Department of Radiodiagnosis and Imaging, All India Institute of Medical Sciences Rishikesh, Rishikesh, Uttarakhand, India
 

Abstract

Background

Anatomical variabilities in the morphometric characteristics of the tentorial notch have been demonstrated through cadaveric studies on humans. Nevertheless, it is important to note that the measurements may be subject to variability due to postmortem effects. The aim of the study was to assess the morphometry of the tentorial notch in living individuals using magnetic resonance imaging (MRI).

Methods and Materials

A retrospective cross-sectional study was conducted and 1,043 cases were analyzed for the institutional archive from January 2021 till December 2022. Variables measured were maximum notch width (MNW), anterior notch length, notch length (NL), interpedunculoclival distance, and apicotectal distance using RadiAnt DiCom viewer. A quartile distribution technique was applied to all measurements and patients were classified into different types of tentorial notch.

Results

We studied 1,043 patients between the ages of 2 months and 84 years with normal MRI scans. Quartile analysis was applied to all measurements. Based on the quartile group, MNWs were classified into wide, narrow, and midrange, and NLs were classified into long, short, and midrange. According to criteria established by Adler and Milhorat, classification of tentorial notch was done and we found the following: short and wide (4.2%), short and medium (12%), short and narrow (9.1%), medium and wide (12.4%), classical (25.3%), medium and narrow (12.4%), long and wide (8%), long and medium (12.6%), and long and narrow (4%).

Conclusion

The dimension of tentorial notch significantly varies among different age groups. NL and MNW are significantly larger in those above 60 years of age. Typical type of notch was the most common variety in our sample and short wide and long narrow were least common types.


Introduction

Transtentorial herniation refers to the displacement of brain tissue through the tentorial notch from one intracranial compartment to another. It can present in different forms, including central, uncal, and upward herniation. These conditions are critical medical emergencies that require prompt surgical or medical intervention to potentially reverse the damage and improve outcomes.[1] The tentorial notch is frequently visualized using contemporary neuroimaging techniques (magnetic resonance imaging [MRI] and computed tomography scanning), but aside from a few classic studies conducted in the 1960s[2] [3] that examined its involvement in specific types of brain herniation and emphasized its significance in trauma-related brain shifts such as those occurring in concussive or inertial injuries, limited emphasis has been placed on its anatomical variations and regional structure.[2]

The tentorial aperture is an anatomically complex region exhibiting considerable variation in its dimensions. Even though the free borders of the tentorium cerebelli define this area, its three-dimensional morphology, lack of veins along its edge, and infrequent calcification have made it anatomically elusive.[3] For many years, neurosurgeons and neuroanatomists struggled to visualize and quantify this structure. Studies have shown that broader and more elongated tentorial notches tend to provide greater exposure of the cerebellum, whereas smaller ones reveal less. Logically, smaller openings reveal less brain parenchyma. Variation in exposure linked to the measurements of the tentorial notch may affect the risk of experiencing herniation syndromes, whether descending or ascending; however, further clinical studies are needed to substantiate this association.[4] [5] [6]

With modern neuroimaging advancement, these structures become more easily visible and quantifiable to further enhance our knowledge about these structures. A thorough understanding and classification of the tentorial notch may provide insights into the mechanisms underlying various herniation syndromes. However, existing literature offers limited comprehensive data on morphometric characteristics of the tentorial notch.

In this study, using currently available MRI methods, we examined typical anatomical differences in the dimensions of the tentorial notch through various measurements, which will further improve our understanding on traumatic brain injury and herniation syndromes. Morphometric measurements facilitate precise preoperative planning by defining the lesion's location and surgical pathway, helping to minimize disruption to normal tissue and reduce postoperative complications and mortality.


Materials and Methods

MRI brain scans performed from January 1, 2020 to December 31, 2022 were retrieved from the institution's imaging archive. Patients who underwent brain MRI without evidence of organic brain lesions, or those undergoing brain screening in the context of spinal pathology—whether as inpatients or outpatients—were included in the study. Only MRI scans that were reported as “normal” by a senior consultant radiologist were included in the study. To ensure the exclusion of any intracranial pathology, all scans were independently reviewed by two investigators. Any scans showing signs of space-occupying lesions, mass effect, hydrocephalus, prior neurosurgical intervention, or other structural abnormalities were excluded. This two-step review process ensured that only anatomically normal scans were included in the final analysis. After recruiting the patients meeting inclusion criteria who underwent MRI at our imaging center, three-dimensional (3D) T1/T2/FLAIR (fluid-attenuated inversion recovery) images of all patients were retrieved from institutional archive and analyzed using Medixant (RadiAnt DICOM Viewer [software], Version 2021.1, Jun 27, 2021; available from: https://www.radiantviewer.com).

Study variables analyzed were ([Fig. 1]):

Zoom
Fig. 1 MRI brain showing measurement of study variables. (A) Anterior notch width: the width of the tentorial notch in the axial plane through the posterior aspect of the dorsum sellae. (B) Maximum notch width: maximum width of the notch in the coronal plane. (C) Notch length: distance from superoposterior margin of the dorsum sellae to the apex of the tentorial notch in the mid-sagittal plane using T2-weighted sagittal MRI sequences. (D) Apicotectal distance: distance was measured in the median plane from the tectum to a perpendicular line extending from the apex of the tentorial notch to the surface of the cerebellum. (E) Interpedunculoclival distance: distance was measured from the interpeduncular fossa to the superoposterior border of the dorsum sellae in T2 sagittal sequence. MRI, magnetic resonance imaging.
  • Anterior notch width (ANW): the notch width was measured on T2-weighted axial images at the level of the posterior margin of the dorsum sellae.

  • Maximum notch width (MNW): maximum width of the notch in the T2 coronal plane.

  • Notch length (NL): distance from superoposterior margin of the dorsum sellae to the apex of the tentorial notch in the mid-sagittal plane using T2-weighted sagittal MRI sequences.

  • Apicotectal (AT) distance: the distance was assessed in the mid-sagittal plane, extending from the tectum to a vertical line drawn from the tip of the tentorial notch to the cerebellar surface.

  • Interpedunculoclival (IC) distance: the distance was measured from the interpeduncular fossa to the superoposterior border of the dorsum sellae in T2 sagittal sequence.

The tentorial notch was classified based on quartile distributions of parameters including NL, MNW, IC distance, and AT distance, as described by Adler and Milhorat.[2] According to the MNW quartiles, the notch was categorized as narrow, midrange, or wide, while the NL quartiles were used to classify it as long, midrange, or short. By combining the NL and MNW grouping, eight unique notch types were delineated as per Adler and Milhorat's methodology.

Statistical Analysis

Variables measured on a continuous scale, like age, were described using the mean and standard deviation (SD). In contrast, categorical data were reported as frequencies with corresponding percentages. Age distribution was examined visually via histogram and formally tested with the Shapiro–Wilk test, in which the null hypothesis assumes a normal distribution.

As age and morphometric parameters exhibited nonparametric distribution, the differences between male and female groups were evaluated using the Mann–Whitney U-test. Group comparisons for categorical data were conducted using the Chi-square method.

To determine the strength and direction of associations, either Pearson's or Spearman's correlation coefficient (r) was used for variables such as ANW, MNW, NL, AT, IC, age, and sex. A p-value less than 0.05 was considered statistically meaningful.

All statistical procedures were conducted using IBM SPSS Statistics version 25 (SPSS Inc., Chicago, Illinois, United States). After organizing and defining the variables, appropriate statistical tests were applied accordingly. Logistic regression was performed with age and sex as predictors, and morphometric parameters (ANW, MNW, NL, AT, and IC) as outcome variables. The nomogram was generated using RStudio software (RStudio Team, Boston, Massachusetts, 2020). Available from: https://www.rstudio.com/.



Results

This study analyzed 1,043 MRI scans retrieved from the institution's imaging archive. The mean age was 35.97 (SD ± 21.49) years. There were 602 (57.7%) males and 441 (42.3%) females. The age ranged from 2 months to 84 years ([Table 2]).

Table 1

Comparison measurements of tentorial notch between sexes

Measurements

Total cases (N = 1,043)

Mean ± SD

Females (N = 441)

Mean ± SD

Males (N = 602)

Mean ± SD

p-Value

ANW

21.53 ± 3.95

21.48 ± 3.79

21.56 ± 4.07

0.7469

MNW

31.16 ± 4.40

31.01 ± 3.92

31.26 ± 4.71

0.3642

NL

50.75 ± 5.61

50.93 ± 5.53

50.61 ± 5.67

0.3631

AT

16.92 ± 4.01

16.93 ± 3.64

16.91 ± 4.27

0.9367

IC

17.30 ± 3.04

17.24 ± 2.99

17.34 ± 3.08

0.6001

Abbreviations: ANW, anterior notch length; AT, apicotectal distance; IC, interpedunculoclival distance; MNW, maximum notch width; NL, notch length; SD, standard deviation.


Out of all the cases, 5.3% of individuals belonged to 0 to 2 years, while 9.3% of individuals belonged to 3 to 10 years, and 8.8% of individuals belonged to 11 to 17 years. Most of the cases (60.0%) belonged to the age group of 18 to 60 years. Lastly, 16.6% of the cases were in the age group of over 60 years.

The mean ANW, MNW, NL, AT, and IC were 21.53 mm (±3.95), 31.16 mm (±4.4), 50.75 mm (±5.61), 16.92 mm (±4.01), and 17.30 mm (±3.04), respectively ([Table 1]). The mean values of all the measurements vary significantly between the age groups ([Table 2]). All these measurements show positive correlation with age ([Fig. 2A–E]), but did not show any association with sex. A statistically significant and strong positive correlation was evident between AT and NL (r = 0.66, p < 0.001; [Fig. 3F]). Moderate positive correlations were observed between NL and MNW (r = 0.3, p < 0.001), between AT and MNW (r = 0.15, p < 0.001), and IC and MNW (r = 0.31, p < 0.001). Similarly, we found positive correlation between IC and NL (r = 0.43, p < 0.001) and between MNW and ANW (r = 0.5, p < 0.001).

Table 2

Summary table for association between age and other parameters

Parameter

Age

p-Value

0–2 years

(n = 55)

3–10 years

(n = 97)

11–17 years

(n = 92)

18–60 years

(n = 626)

 > 60 years

(n = 173)

Gender

<0.001[a]

 Male, 602 (57.7%)

41 (74.5%)

65 (67.0%)

54 (58.7%)

319 (51.0%)

123 (71.1%)

 Female, 441 (42.3%)

14 (25.5%)

32 (33.0%)

38 (41.3%)

307 (49.0%)

50 (28.9%)

ANW

17.69 ± 3.29

19.66 ± 3.19

21.08 ± 3.87

1.93 ± 3.80

22.59 ± 4.07

<0.001[b]

MNW

24.79 ± 4.06

28.69 ± 4.01

31.40 ± 3.26

31.70 ± 3.92

32.45 ± 4.72

<0.001[b]

NL

42.82 ± 5.61

47.54 ± 4.95

51.25 ± 6.27

51.73 ± 5.19

51.23 ± 4.33

<0.001[c]

AT

12.90 ± 3.33

15.51 ± 3.91

16.41 ± 3.40

17.44 ± 4.02

17.36 ± 3.64

<0.001[c]

IC

14.78 ± 3.31

16.25 ± 3.12

17.11 ± 3.44

17.65 ± 2.81

17.52 ± 3.01

<0.001[c]

Abbreviations: ANW, anterior notch length; AT, apicotectal distance; IC, interpedunculoclival distance; MNW, maximum notch width; NL, notch length.


a Chi-squared test.


b One-way ANOVA.


c Kruskal–Wallis test.


Zoom
Fig. 2 Scatter plot showing association between age and various study variables.
Zoom
Fig. 3 Scatter plot showing correlation between study variables.

These findings indicated a notable but weak positive association between age and ANW (r = 0.27, p < 0.001), MNW (r = 0.29, p < 0.001), and AT (r = 0.21, p < 0.001). However, no statistically meaningful correlation was found between AT and ANW (r = 0.03, p = 0.276).

The measurement thresholds applied to categorize the tentorial notch, derived from MNW and NL quartiles, were used to determine distribution patterns, as summarized in [Table 3]. The most frequently occurring type in this dataset was the typical configuration, aligning with midrange MNW and NL values. Conversely, the mixed configuration was the least commonly identified in this analysis.

Table 3

Type of tentorial notch

Type of notch

Dimensions

Range in mm

Percentage of total cases

Wide

NL (midrange)

MNW (wide)

47.90–54.60

33.91–46.20

12.4%

Narrow

NL (midrange)

MNW (narrow)

47.90–54.60

11.3–28.29

12.4%

Long

NL (long)

MNW (midrange)

54.61–66.2

28.30–33.91

12.6%

Short

NL (short)

MNW (midrange)

21.1–47.89

28.30–33.91

12%

Typical

NL (midrange)

MNW (midrange)

47.90–54.60

28.30–33.91

25.3%

Large

NL (long)

MNW (wide)

54.61–66.2

33.91–46.20

8%

Small

NL (short)

MNW (narrow)

21.1–47.89

11.3–28.29

9.1%

Mixed

NL (long)

MNW (narrow)

54.61–66.2

11.3–28.29

4%

NL (short)

MNW (wide)

21.1–47.89

33.91–46.20

4.2%

Abbreviations: MNW, maximum notch width; NL, notch length.


Regression Analysis

The regression analyses conducted for various dependent variables, including ANW, AT, IC, MNW, and NL, provide insights into the relationships between these variables and their predictors. In each analysis, both univariable and multivariable regression coefficients were examined to understand the individual and combined effects of predictor variables on the dependent variable.

For ANW, AT, IC, MNW, and NL, age consistently showed a significant positive association with the dependent variable in both univariable and multivariable analyses. This suggests that as age increases, the dependent variables tend to increase as well. Gender, on the other hand, showed mixed results across the different dependent variables. In some cases, such as ANW and IC, gender did not show a significant association with the dependent variable in either univariable or multivariable analyses. However, for AT, MNW, and NL, gender did not show a significant association in univariable analysis but showed a significant association in multivariable analysis, albeit with differing coefficients.

The models' goodness of fit was assessed using the F-statistic, AIC (Akaike information criterion), and R-squared values. Generally, the F-statistic was significant for all models, indicating that the models were a good fit for the data. AIC values were utilized to compare the models, where lower values suggested a superior fit. R-squared values represented the proportion of variance explained by the model, with higher values indicating greater explanatory strength.

In summary, the regression analyses contributed important observations regarding the connection between predictor and outcome variables, aiding in identifying the elements that impact each dependent measure. Nonetheless, further studies are warranted to investigate other influencing variables and reduce possible confounding factors that may influence these relationships.



Discussion

The tentorium is a unique structure that has evolved over time in higher mammals, including humans and primates.[7] It separates the cerebral and cerebellar hemispheres and surrounds the midbrain structures. The tentorial opening, known as the tentorial notch or hiatus, has undergone evolutionary enlargement, being more prominent in higher mammals. Research into the evolutionary origin and development of the tentorium indicates that it appeared later in evolution as paired dural extensions flanking the midbrain within the cerebro-cerebellar fissure. As species evolved, the straight sinus formed where the falx cerebri joins the tentorium—became progressively longer, contributing to the development of the tentorial hiatus. Several authors in the early decades of the 20th century explained the indications, symptoms, and neuroanatomical alterations that arise when there is parenchymal herniation through the tentorial opening.[8] [9] [10] [11] The presence of herniation of the uncus at the level of notch was identified, and medical terminology like “temporal pressure cone,” “tentorial pressure cone,” and “transtentorial herniation of the brain-stem” were used to describe this abnormal condition.[12] It is well recognized that a larger and broader tentorial incisura results in greater exposure of cerebellar tissue compared to a smaller, narrower one.[2] It stands to reason that smaller apertures expose less brain parenchyma. Variation in exposure associated with the size of the tentorial notch has significant implications for the risk of both descending and ascending herniation syndromes.

In the course of autopsy-based examination of the human brain, the falx cerebri and tentorium are typically separated to aid in the movement and delivery of the brain. However, this process can inevitably distort the brain's normal anatomy, leading to increased alterations that already exist in a deceased brain specimen. Studies investigating morphometry of tentorial notch are limited in both Indian and Western populations, with most studies utilizing cadaveric samples. Additionally, live in vivo studies are rare, and previous studies have generally had small sample sizes. Thus, future studies should aim to examine the effect of postmortem examination on brain anatomy and to explore this topic in larger, more diverse populations. Our study incorporates large sample size and found variability in size of tentorial notch.

To prevent the anatomical alterations that occur during postmortem analysis, utilizing widely accessible MRI techniques to study the brain in its natural, living state offers a practical alternative. This method enables precise in situ measurement and clear identification of anatomical landmarks, all without posing any risk to the patient.

Our study explored the morphological characteristics of the tentorial notch through 1,043 patients with normal MRI brain scan. In this study population of 1,043, 58% of the cases were males, and 42% were females. The mean values of all measurements of tentorial notch varied significantly between age groups. The NL and MNW happened to be significantly maximum beyond 60 years. Regression analysis also suggested that the age is the only predictor of various tentorial dimensions. Although age predicted tentorial notch width and length, gender did not play any significant role in determining the tentorial notch dimension. Also, there was no significant gender-based difference in either the total length or the maximal width of the tentorial notch.

Various measurements of the tentorial notch are also correlated with one another. Both IC and AT showed a positive correlation with NL, suggesting that these values determine the length of the tentorial notch. Similarly, MNW showed positive correlation with ANW and NL. Both NL and ANW values can influence the maximal width of the tentorial notch.

In 1958, Sunderland classified the tentorial notch as either wide or narrow, whereas Corsellis suggested that variations in its shape and size might influence the manifestation of herniation patterns.[13] [14] While they illustrated anatomical differences in the tentorial notch and its relationship with the brainstem, no clear system of classification was established, and no hypotheses were put forth to explain the varied clinical outcomes that result from concussive and inertial brain injuries or patterns of transtentorial herniation. Adler and Milhorat later developed a classification system consisting of eight types through morphometric analysis.[2] By organizing these groups into a matrix, the tentorial notch can be categorized into nine distinct types according to the NL and MNW variables ([Table 3]). Among these, the typical notch was the most common type, accounting for 25.3% of total cases. The second most common type was the long and medium notch, representing 12.6% of total cases, followed by the medium wide and medium narrow notches with 12.4% each. The least common types were the short and wide and the long and narrow notches, accounting for 4.2 and 4% of total cases, respectively.

The comparison of various dimensions of the tentorial incisura between our study and other studies for the same age groups is shown in [Table 4]. The various dimensions of the tentorial incisura in our study are almost comparable to those of other studies with the exception that the NL is slightly on the lower side in our study. This can be due to the fact that our result is based on MRI, while most other studies, on cadaver.

Table 4

Table summarizing findings from different studies

This study

Arrambide et al (2022)[15]

Das et al (2021)[16]

Staquet et al (2020)[17]

Adler and Milhorat (2002)[2]

Sunderland (1958)[13]

Ono et al (1984)[18]

Country

India

Mexico

United States

France

United States

Australia

United States

Sample

626 MRI

60 MRI

40 cadavers

40 CT

100 cadavers

30 cadavers

25 cadavers

Age

18–60 years

41.5 ± 12.1

20 to 65

42.5

ANW

21.93 ± 3.80

26.92 ± 2.14

25.5 ± 3.5

26.6 ± 2.7

27.06 ± 3.5

MNW

31.70 ± 3.92

31.2 ± 12.1

29.77 ± 2.26

31.0 ± 2.5

29.6 ± 3.0

30.16 ± 3.21

29.6 (26–35)

NL

51.73 ± 5.19

55.55 ± 4.82

57.98 ± 4.52

55.0 ± 5.3

57.7 ± 5.6

54.6 ± 6.93

52 (46–67)

AT

17.44 ± 4.02

19.45 ± 3.81

25.81 ± 8.04

16.8 ± 5.4

19.8 (13–27)

IC

17.65 ± 2.81

21.03 ± 3.17

21.21 ± 3.72

20.4 ± 3.2

12.1 (7.8–15.6)

Abbreviations: ANW, anterior notch length; AT, apicotectal distance; CT, computed tomography; IC, interpedunculoclival distance; MNW, maximum notch width; MRI, magnetic resonance imaging; NL, notch length.


While our study included a broad age range—from infancy to advanced age—we primarily employed linear correlation models to assess age-related morphometric changes. Although nonlinear relationships were explored, they did not significantly enhance model performance in our dataset. Additionally, pediatric scans were carefully evaluated using standardized anatomical landmarks, and inter-observer reliability was maintained. Nonetheless, we recognize that ongoing neurodevelopmental changes in infants and structural brain alterations in the elderly may introduce subtle morphological variations that warrant further investigation in age-stratified cohorts.

As a preliminary exploratory observation, we reviewed MRI scans of 11 patients (not included in the primary cohort) who presented with supratentorial mass lesions and radiological features suggestive of transtentorial herniation. All patients demonstrated a “typical” tentorial notch morphology based on our classification. Interestingly, most did not exhibit overt clinical signs of herniation; only two patients presented with oculomotor palsy and two exhibited features consistent with Kernohan's notch phenomenon. While these observations may suggest that notch morphology alone does not determine clinical manifestations, they are anecdotal and lack sufficient methodological rigor for statistical inference. We recognize that herniation symptomatology is multifactorial and likely influenced by the rate and direction of herniation, associated edema, and other dynamic factors. These preliminary findings underscore the need for dedicated studies comparing notch dimensions between patients with and without clinical herniation to better understand potential anatomical contributions.


Conclusion

We analyzed various measurements of the tentorial notch in a large number of patients who lacked any evidence of intracranial pathology using 3D MRI. The dimension of the tentorial incisura significantly varies between different age groups. NL and MNW are significantly larger in those above 60 years of age. The notch dimensions do not vary significantly between males and females. Age is the only factor that determines the incisura length and width. The tentorial notch can be classified into nine types based on NL and its maximal width. The classical type is the most common variety of tentorial notch accounting for 25.3% of all study participants. Short-wide and long-narrow tentorial notch configurations represent the least prevalent morphotypes identified in the North Indian population. The tentorial notch may be classified into types based on normative morphometric data; however, clinical relevance of these types remains to be established through future pathological correlation studies.



Conflict of Interest

None declared.


Address for correspondence

Rajnish K Arora, MCh
Department of Neurosurgery, All India Institute of Medical Sciences Rishikesh
Rishikesh, Uttarakhand, 249203
India   

Publication History

Article published online:
23 September 2025

© 2025. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (https://creativecommons.org/licenses/by/4.0/)

Thieme Medical and Scientific Publishers Pvt. Ltd.
A-12, 2nd Floor, Sector 2, Noida-201301 UP, India


Zoom
Fig. 1 MRI brain showing measurement of study variables. (A) Anterior notch width: the width of the tentorial notch in the axial plane through the posterior aspect of the dorsum sellae. (B) Maximum notch width: maximum width of the notch in the coronal plane. (C) Notch length: distance from superoposterior margin of the dorsum sellae to the apex of the tentorial notch in the mid-sagittal plane using T2-weighted sagittal MRI sequences. (D) Apicotectal distance: distance was measured in the median plane from the tectum to a perpendicular line extending from the apex of the tentorial notch to the surface of the cerebellum. (E) Interpedunculoclival distance: distance was measured from the interpeduncular fossa to the superoposterior border of the dorsum sellae in T2 sagittal sequence. MRI, magnetic resonance imaging.
Zoom
Fig. 2 Scatter plot showing association between age and various study variables.
Zoom
Fig. 3 Scatter plot showing correlation between study variables.