Absence of the interthalamic adhesion (ITA) as a neuroanatomical association or risk factor for neuropsychiatric disorders: A systemic review and meta-analysis

Abstract

Background:

This study aimed to provide an up-to-date account of the frequency of “the absence of interthalamic adhesion (AITA) as a risk factor or association” in healthy subjects and neuropsychiatric patients. Owing to the increased interest in the contribution of ITA to neurological function in previous literature, a meta-analysis of its frequency and sex dependency is required.

Aim:

This study aimed to study whether the AITA is associated with neuropsychiatric disorders.

Settings and Design:

This study is a meta-analysis and systemic review.

Methods and Material:

Literature searches were conducted in PubMed, Web of Science, and Google Scholar using the keywords “interthalamic adhesion,” “massa intermedia,” “adhesio interthalamica,” and “adhesion” along with the Boolean operators (OR, AND, and NOT). Three reviewers independently assessed the abstracts and full texts for validation based on the inclusion criteria. The meta-analysis was performed using Microsoft Excel 2019 for descriptive studies and RevMan 5.2 for comparative studies.

Results:

The incidence of absent ITA was 15.3% in healthy subjects and 28.76% in neuropsychiatric subjects. The relative probability of AITA was 2.30 [95% confidence interval (CI), 1.96–2.70] in neuropsychiatric illness. Healthy men were 1.91 times more likely, and men with neuropsychiatric disorders were 1.82 times more likely to have absent ITA than women.

Conclusions and Relevance:

In this study, a consistent association of AITA with psychiatric disorders was observed, rendering the condition to be treated as an associated risk factor affecting the function of the habenula nuclear complex via the stria medullaris thalami. A cohort or longitudinal study is needed to compare the incidence of psychiatric disorders in individuals with or without ITA and to calculate the attributed risk.

INTRODUCTION

The interthalamic adhesion (ITA) connects the medial surface of the thalamus to the contralateral thalamus near the roof of the third ventricle. It is flat and grayish, typically extending beyond the interventricular foramen. Its average diameter is about 1 cm and consists of midline-crossing axons and neurons [Supplementary Figure 1 ^{(232.6KB, tif)} ].[1]

The ITA was first described by Giovanni Battista Morgagni in 1719 using the name “transversa lamina cinereal,” which translates as “transverse ash-gray lamina.”[2] Alexandre Jamain, a French anatomist, was the first to allude to the ITA as a “mass”[3]; the “mass” was given the name “massa intermedia,” which translates as “middle mass,” to emphasize its closeness to both thalami.[4] An extensive study of the ITA by Marchi and Tenchini in the 1880s, cited by Malobabić et al. (1987), revealed that it involved commissural fibers but not neuronal cell bodies. Malobabić et al.[5] refuted the second assertion. Although the “adhesio interthalamica” has been shown to include commissural fibers, Terminologia Anatomica continues to use this term.[6] Because of their similar connotations, the terms “interthalamic adhesion,” “massa intermedia,” and “interthalamic mass” are now sometimes used interchangeably (Basel and Jena versions of Nomina Anatomica).[7]

The ITA is a midline structure observed during the 13th–14^th gestational weeks.[8] It is well developed in lower mammals and has multiple nuclei.[5] In several autopsy studies, it was absent in 15–25%, more frequently in males.[5,9,10] Because of its unknown function, the ITA has long been considered a residual component.[11] However, recent studies have implicated it in cognitive processes.[12,13,14] The absence of ITA (AITA) is a neuroanatomical abnormality arising early in development caused by malnutrition or maternal viral infection during pregnancy.[15] Other midline morphological changes, for example, agenesis of the corpus callosum, cerebellar vermis, and cavity septum pellucidum, are also observed in schizophrenia.[12]

The incidence rate of AITA has been reported in previous publications, which ranged from 0% to 51%. Research into its frequency began over two centuries ago using a variety of techniques, including direct examination of the postmortem or dead brain, pneumoencephalography, ventriculo-encephalography, and magnetic resonance imaging (MRI).[9,16,17,18] There is contradictory information about sexual dimorphism. While there is emerging evidence that the ITA is more frequent among young people and women (age and gender dependency), other research has not substantiated this.[8,10]

Owing to the increased interest in the contribution of ITA to neurological function, a meta-analysis of its frequency and sex dependency is required. This study aimed to provide an up-to-date account of the frequency of “AITA as risk factor or association” in healthy subjects and neuropsychiatric patients. The odds ratio (OR), or risk ratio, was calculated to assess its relationship with neuropsychiatric disorders.

The following outcomes are evaluated:

1. Incidence of AITA in healthy subjects and neuropsychiatric patients.

2. The distribution of AITA in subgroups based on sex, ethnicity, and research method.

3. OR of AITA in healthy subjects and neuropsychiatric patients.

METHODS

Inclusion criteria

1. Population: cadaveric samples, healthy subjects, and patients with neuropsychiatric illness.

2. Exposure: neurodevelopmental or neurodegeneration.

3. Comparator: healthy or control subjects.

4. Outcome: frequency of AITA.

Exclusion criteria

1. Case report or case series.

2. Review article.

3. Data from textbooks.

4. Animal studies.

Search strategy

Using the keywords “interthalamic adhesion,” “massa intermedia,” “adhesio interthalamica,” and “adhesion” along with the Boolean operators (OR, AND, and NOT), search strings were constructed for three major databases: PubMed, Web of Science, and Google Scholar. The searches were conducted between June and September 2021, with no language or time period filter. The citations, titles, and abstracts obtained from the searches were transferred to a reference manager (Zotero).

Selection of studies and risk of bias

Additional citations were checked in the references of the selected publications. Three reviewers independently assessed the abstracts and full texts for validation based on the inclusion criteria. Discrepancies in selecting relevant articles were resolved by consensus. The Anatomical Quality Assessment (AQUA) tool was used for descriptive studies.[19] The New Castle Ottawa Score was used to assess the quality of the comparative studies.[20]

Data extraction and statistical analysis

Two authors extracted the data, and the relevant data from all the included studies are presented in Tables 1 and 2. The meta-analysis was performed using Microsoft Excel 2019 for descriptive studies and RevMan 5.2 for comparative studies. For each study, the incidence rate, the OR, and their 95% confidence intervals were calculated. The fixed-effects or random-effects model was used to weight the studies by inverse variance or Peto's method. Cochrane Q and Higgins I² statistics were calculated to determine heterogeneity between and within studies. The final effect size is calculated based on heterogeneity. The fixed-effects model was applied only when the heterogeneity was less than 50%; otherwise, the random-effects model was used. Egger's regression test and a funnel plot were implemented to assess publication bias. Subgroup analysis and meta-regression were used to assess heterogeneity. Subgroup analyses were used to determine the influences of race, sex, study modality, and study participants’ health status. The effect of mean subject age on the occurrence of AITA was examined using a random-effects meta-regression model.

Table 1.

Characteristics of descriptive studies are included along with the estimated risk of bias

Author (year)	AITA	Sample size	Incidence	Lower CI	Upper CI	Weight	Status	Ethnicity	Mode of investigation	Risk of bias
Allen (1991)	27	100	0.27	0.1922	0.3651	3.7374	Healthy	North American	Cadaveric	Low
Belkovsky (2019)	12	57	0.2105	0.1236	0.3352	3.2324	Healthy	South American	Cadaveric	Moderate
Borghei (2020)	42	402	0.1045	0.0781	0.1384	4.0136	Healthy	North American	MRI	Low
Borghei (2021a)	179	1410	0.127	0.1106	0.1454	4.278	Healthy	North American	MRI	Low
Borghei (2021b)	11	90	0.1222	0.069	0.2074	3.2483	Healthy	North American	MRI	Low
Damle (2017)	11	233	0.0472	0.0263	0.0832	3.3153	Healthy	North American	MRI	Low
Davie & Baldwin (1967)	44	111	0.3964	0.3099	0.49	3.8822	Neuropsychiatric	European	Pneumoencephalogram	Low
Davie & Baldwin (1967)	4	20	0.2	0.0771	0.4278	2.1405	Healthy	European	Cadaveric	High
Lansdell (1972)	28	74	0.3784	0.2756	0.4933	3.6672	Neuropsychiatric	North American	Pneumoencephalogram	High
Macedo (1889)*	43	215	0.2	0.1518	0.2588	3.9834	Healthy	European	Cadaveric	Unclear
Malobabic (1987)	11	50	0.22	0.1262	0.3551	3.1471	Healthy	European	Cadaveric	Moderate
Miró-Padilla (2022)	31	240	0.1292	0.0923	0.1778	3.8892	Healthy	European	MRI	Low
Morel (1947)	166	823	0.2017	0.1757	0.2305	4.262	Neuropsychiatric	European	Cadaveric	Unclear
Morel & Weissfeilar (1931)*	32	175	0.1829	0.1323	0.2472	3.8754	Healthy	European	Cadaveric	Unclear
Park (1993)	17	146	0.1164	0.0736	0.1793	3.576	Healthy	Asian	Cadaveric	Low
Parra (2021)	16	31	0.5161	0.3453	0.6833	3.0546	Healthy	South American	Cadaveric	Moderate
Patra (2022)	7	50	0.14	0.0682	0.2657	2.8126	Healthy	Asian	Cadaveric	Moderate
Patra (2022)	6	50	0.12	0.0549	0.2424	2.6788	healthy	Asian	MRI	Moderate
Pavlovic (2020)	8	41	0.1951	0.1007	0.3442	2.8793	Healthy	European	Cadaveric	Low
Rabl (1958)	292	921	0.317	0.2878	0.3478	4.2974	Neuropsychiatric	European	Cadaveric	Moderate
Samra & Cooper (1968)	5	32	0.1562	0.0666	0.3247	2.4412	Healthy	European	Cadaveric	Moderate
Samra & Cooper (1968)	146	860	0.1698	0.1461	0.1964	4.2523	Healthy	European	Ventriculographic	Unclear
Sen (2005)	14	161	0.087	0.0522	0.1415	3.4658	Healthy	Asian	MRI	Low
Tenchini (1882)*	20	100	0.2	0.1328	0.2898	3.6162	Healthy	European	Cadaveric	Unclear
Tsuneda (1935)	19	100	0.19	0.1246	0.2788	3.5916	Neuropsychiatric	Asian	Cadaveric	Moderate
Tsutsumi (2021)	20	205	0.0976	0.0638	0.1464	3.6883	Healthy	Asian	MRI	Low
Viller (1887)*	5	21	0.2381	0.1027	0.4603	2.3307	Healthy	European	Cadaveric	Unclear
Wenzel (1812)*	10	66	0.1515	0.0835	0.2592	3.1373	Healthy	European	Cadaveric	Unclear
Yasaka (2018)	15	153	0.098	0.06	0.1563	3.5058	Healthy	Asian	MRI	Low
Random-effects model	1241	6937	0.1759	0.1457	0.2108	100

*Data Collected from Samra and Cooper (1968); AITA: Absent interthalamic adhesion. CI: Confidence interval, MRI: Magnetic resonance imaging

Table 2.

Characteristics of the comparative studies included, with risk of bias assessment

Author (year)	Population/ location	Methods of investigation	Neuropsychiatric illness	Control	Selection	Comparability	Exposure/outcome	Total score
Agrawal (2008)	European/ Italy	MRI 1.5 T	n=71, M/F=46/25, RH/LH=59/12, AITA=6 (schizophrenia)	n=75, M/F=39/36, RH/LH=61/14, AITA=2	****	**	***	9
Ceyhan (2008)	Asia/ Turkey	MRI 1.5 T	n=35 (schizophrenia)+21 (bipolar disorder), M/F 18/17+6/15 Mean age 37.6±11.2 years schizophrenia patients and 32±10.3 years AITA=20% in schizophrenia patients and 14.3% in bipolar patients	n=89, M/F 39/50 Mean age 36.6±1.28 years. AITA=13.5% of the healthy controls and females lower incidence (6%) compared with males (23.1%)	****	**	***	9
Crippa (2006)	South America/ Brazil	MRI 1.5 T	n=38, M/F=26/12, RH/LH 36/2 Mean age 29.8±10 yrs AITA=18.42% (all male) (schizophrenia)	n=38, M/F 26/12, RH/LH 26/12 Mean age 29.7±9.7 yrs AITA=10.53% (three male, one female)	****	**	***	9
Erbagci (2002)	Asia/ Turkey	MRI 1.5 T	n=26, M/F=11/15, mean age 34.69±19.65 years AITA=34.61% (4 male, 5 female) (schizophrenia)	n=29, M/F=11/18, mean age 28.59±7.52 years). AITA=13.79% (1 male, 3 female).	****	**	***	9
Ettinger (2007)	North America/USA	MRI 1.5 T	n=69, M/F=49/20, R/L 55/14 AITA=5 (schizophrenia)	n=54, M/F 34/20, R/L45/9 AITA=8	***	**	**	7
Filipovic (2013)	European/Serbia	Postmortem analysis 479 autopsied (64.7% male) (M=57.44±15.37).	n=110 AITA=27/110 (schizophrenia)	n=369 AITA=44/369	***	**	**	7
Haghir (2013)	Asian/ Iran	MRI 1.5 T	n=29, M/F=19/10, mean age 36.4±12.7 years AITA=8 patients (27.58%, including 7 males and 1 female) (schizophrenia)	n=29, M/F=19/10, RH/LH 23/6, mean age 36.5±12.6 years AITA=4 (13.79%, including 3 males and 1 female)	****	**	***	9
Landin-Romero (2015)	Oceanian/ Australia	MRI 1.5 T	n=639 AITA=115/260 (personality disorder) 14/23 (psychosis) 95/219 (schizophrenia)	n=223 M/F=99/124 AITA=53/223	***	**	**	7
Meisenzahl (2000)	European/ Germany	MRI 1.5 T	n=30, mean age=29.4±7.97) AITA=7/30 (schizophrenia)	n=30, mean age=29.2±8.0 AITA=4/30	***	**	**	7
Meisenzahl (2002)	European/ Germany	MRI 1.5 T	n=50 males, mean age 30.0±8.41 years AITA=11/50 (schizophrenia)	n=50 males, mean age=30.2±8.8 years AITA=8/50	***	**	***	8
Nopulas (2001)	North American/ USA	MRI 1.5 T	n=114, M/F=56/58, mean age M/F 28.6±8.83/31.2±11.9 years AITA total 31.58% M/F 30.36%/32.76% (schizophrenia)	n=112 M/F=53/59, mean age=M/F 27.69±7.87/29.4±12.2 yrs AITA total 22.32% M/F 32.08%/13.56%	****	**	***	9
Shimizu (2008)	Asian/ Japan	MRI 1.5 T	n=64, M/F=30/34, R/L=64/0 Mean age 36.3±11 yrs, AITA=4.7% (schizophrenia)	n=51, M/F=22/29, RH/LH=51/0 Mean age=36.1±8.4 yrs, AITA=5.8%	****	**	***	9
Snyder (1998 m)	European/ Germany	MRI 1.0 T	n=82 M/F=54/28, mean age 23.4±6.2 years, AITA=28/82 (schizophrenia)	n=52 M/F=30/22, mean age 27.7±6.5 years, AITA=7/52	****	**	***	9
Snyder (1998p)	European/ Germany	Postmortem Study	n=41, M/F=18/23 AITA M=14/18 AITA F=10/23 (schizophrenia)	n=53, M/F=29/24 AITA M=14/29 AITA F=13/24	****	**	**	8
Takahashi (2008b)	Asian/ Japan	MRI 1.5 T	n=62, M/F=32/30 Mean age=25.8±4.9 AITA=17/72 (schizophrenia) 11/47 (personality disorder)	n=63, M/F=35/38 Mean age 24.4±5.4 AITA=7/81	***	**	***	8
Takahashi (2008a)	Oceanian/ Australia	MRI 1.5 T	n=89, M/F 76/13, RH/MH/LH 74/5/6 Mean age 34.9±9.6 years, AITA=19/89 (schizophrenia) 10/162 (psychosis)	n=87 M/F 55/32 RH/MH/LH 80/2/5 Mean age 26.9±10.1 years AITA=2/87	****	**	**	8
Takahashi (2009)	Oceanian/ Australia	MRI 1.5 T	Group I n=29 M/F=7/22 Mean age 32.5±8.3 yrs, AITA=4/29 (psychosis) Group II, n=27 M/F=9/18 Mean age 35.1±10.0 yrs AITA=0/27 (psychosis)	n=33 M/F 12/21 Mean age 34.0±9.9 years AITA=1/33	****	**	***	9
Takahashi (2010)	Oceanian/ Australia	MRI 1.5 T	n=26, M/F=8/18, mean age 38.4±10.9 years, AITA=4/26 (personality disorder)	n=24, M/F 7/17, mean age 38.7±11.1 years AITA=1/24	****	**	***	9
Trzesniak (2012)	South American/ Brazil	MRI 1.5 T	n=122, M/F=66/56, RH/LH=111/1, mean age 28.6±8.4 years, AITA=13/62 (schizophrenia), 5/46 (personality disorder), 4/14 (psychosis)	n=94, M/F=53/42, RH/LH=91/3, mean age 30.2±8.4 years AITA=11/94	****	**	**	8
Whitehead & Najim (2020)	North American/ USA	MRI 3.0 T	n=103 patients M/F=50/55, mean age 4.1±5.5 yrs AITA=41/103 (schizophrenia)	n=105 M/F=51/52, mean age 11.7+5.3 yrs AITA=13/105	****	**	***	9

n=Sample size, M/F=Male/female, RH/MH/LH=Handedness (right, ambidextrous, left), AITA=Absent interthalamic adhesion; T: Tesla; m: MRI; P: Postmortem

RESULTS

Characteristics of included studies

We collected 368 citations from the three databases and by manual searching. After duplicate citations were removed, 170 remained. Based on our criteria, abstract and full-text screening excluded 103 and 15 citations. During data extraction, three more citations were excluded owing to insufficient data. Finally, 45 citations or studies were included in this meta-analysis [Figure 1].[5,9,10,12,13,14,16,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54] These studies were classified according to study design. Among the 45, 25 were descriptive studies calculating the incidence of ITA, and the other 20 were comparative studies, either case–control or cohort, in which the OR of AITA was calculated. Davie and Baldwin (1967), Patra et al. (2022), Samra and Cooper (1968), and Snyder et al. (1998)[9,16,24,54] assessed the incidence of ITA by MRI and cadaver or postmortem dissection, so data were separated based on methods. Borghei (2021a, b) and Takahashi (2008) were different publications published in the same year. The characteristics of the descriptive and comparative studies are presented in Tables 1 and 2, respectively.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta Analyses (PRISMA) flow diagram of the search strategy

1. Descriptive meta-analysis

   1. The incidence of AITA in healthy subjects was 15.30% [95% confidence interval (CI), 0.13–0.18]. The incidences of AITA in males and females were 19.64% and 11.89%, respectively. The absence in men was significantly higher. Details of the incidences of AITA in different subgroups are given in Table 3.

   2. The incidence of AITA in neuropsychiatric subjects: Overall, this was 28.76% [95% CI, 0.21–0.37], almost twice that in healthy subjects. The incidences of AITA in males and females were 33.68% and 23.24%, respectively [Table 3]. Again, the incidence was higher in men.

   3. Effects of age: No significant correlation between mean age and the incidence of AITA was found in either descriptive or comparative studies by meta-regression. Thus, age as a covariate does not affect AITA incidence.

   4. Publication bias: The funnel plots were symmetrical, and the studies were almost evenly distributed on either side of the line of effect. Egger's linear test, Begg–Majumdar rank, correlation test, and Rosenthal's fail-safe number were insignificant, indicating no publication bias.

2. Comparative meta-analysis

   1. OR of AITA in comparative studies: The incidence of AITA in males was 1.91 times higher than in women [95% CI, 1.56–2.31] in healthy subjects and 1.87 times [95% CI, 1.63-2.15] with minimal heterogeneity (17%) higher in neuropsychiatric subjects [Figure 2]. Similarly, there was a higher incidence of AITA in neuropsychiatric patients, OR = 2.31 [95% CI, 1.98–2.70], and in men, OR = 1.87 [95% CI, 1.63-2.15] [Figure 3]. The risk ratio of AITA in neuropsychiatric to healthy subjects, computed from the incidence rates, was 1.86 [95% CI, 1.66–2.09].

   2. Subgroup analysis based on neuropsychiatric disorder and methods: The relative probability of AITA in schizophrenia was 1.73 [95% CI, 1.12–2.679] with nil heterogeneity in postmortem studies and 2.34 [95% CI, 1.89–2.90] in MRI studies with zero or minimal heterogeneity (0% and 37%, respectively) [Figure 3]. The frequencies of AITA in psychoses and personality disorders were also significantly higher than in healthy controls, and their ORs were 3.65 [95% CI 1.89–7.03] and 2.30 [95% CI, 1.66–3.18] without heterogeneity, respectively [Figure 3].

   3. Publication bias: The funnel plot was almost symmetrical; the studies were evenly distributed on either side of the effect line in all subgroups. Therefore, the likelihood of publication bias was negligible [Figure 4].

Table 3.

Result of a meta-analysis of descriptive studies examining the incidence of absent interthalamic adhesion (AITA)

Subgroup analysis	No of studies	Sample size	Incidence	Lower CI	Upper CI	Higgin i2	Tau-square
Total	29	6937	0.1759	0.1457	0.2108	90.8861	0.3039
Cadaveric dissection	17	2948	0.2118	0.1759	0.2528	78.4947	0.1464
MRI	9	2944	0.1051	0.0882	0.1248	46.4943	0.0354
Pneumoencephalogram	2	185	0.3892	0.3217	0.4613	0.0000	0.0000
Ventriculographic	1	860	0.1698	0.1461	0.1964	NA	NA
Healthy	24	4908	0.1530	0.1297	0.1795	76.0993	0.1455
Male	16	1920	0.1964	0.1716	0.2239	34.2745	0.0321
Female	16	1875	0.1189	0.0923	0.1519	58.3150	0.1475
Asian	6	765	0.1041	0.0843	0.1280	0.0000	0.0000
European	11	1820	0.1744	0.1575	0.1926	0.0000	0.0000
North American	5	2235	0.1211	0.0799	0.1792	87.1621	0.2251
South American#	2	88	0.3458	0.1197	0.6729	87.7864	0.8435
Neuropsychiatric illness	5	2029	0.2876	0.2138	0.3747	91.2003	0.1716
Male	5	977	0.3368	0.2745	0.4053	71.0712	0.0711
Female	5	1052	0.2324	0.1521	0.3382	86.6871	0.2645
Asian	1	100	0.1900	0.1246	0.2788	NA	NA
European	3	1855	0.2943	0.2011	0.4085	94.7387	0.1832
North American	1	74	0.3784	0.2756	0.4933	NA	NA

^#Over-estimated owing to low sample size

Figure 2.

Forest plot computing odds ratios of absent interthalamic adhesion (AITA) in both sexes

Figure 3.

Forest plot computing odds ratios of absent interthalamic adhesion (AITA) in neuropsychiatric disorders along with subgroup analysis. M: MRI data, p: postmortem data

Figure 4.

Funnel plot showing negligible publication bias. (a) Healthy and neuropsychiatric studies (symmetrical). (b) Subgroups of neuropsychiatric studies (almost symmetrical)

DISCUSSION

Summary of findings

While the ITA is a small structure in the human brain, and its importance cannot necessarily be attributed solely to its specific function, its presence or absence could indicate differences in genetics, development, or functioning. Imaging technology has advanced dramatically over the last few decades, enabling in vivo studies of deep brain regions and their anatomical relationships. For the first time, Kochanski and coworkers discovered crossed medullary stria medullaris thalami fibers in the ITA and established the structure as a midline commissure.[55] However, it is still unclear why certain people have no ITA and whether this absence is relevant to normal or pathological neurobehavioral processes.

The incidence of AITA was 15.3% in healthy subjects and 28.76% in neuropsychiatric subjects and was almost twice as common in males as in females. The AITA was slightly more common in European populations than in Asian or North and South American populations. However, the confidence intervals overlapped, except for the Asian population. This applied to both normal and neuropsychiatric individuals. The slight difference could be attributed to sampling error. The risk ratio of AITA in comparative studies for neuropsychiatric patients was 1.86 [95% CI, 1.66–2.09]. When we calculated it from incidences computed in descriptive studies (28.76%/15.3%), it was 1.88 [95% CI, 1.62–2.06]. The two estimates are almost similar. Therefore, the AITA could be considered a risk factor or association with neuropsychiatric disorders.

Trzesniak et al. (2011)[12] performed a meta-analysis of eleven studies on the incidence of ITA in schizophrenia, schizophreniform, and/or schizoaffective disorder (SSD). They included comparative data from 822 patients and 718 healthy subjects with and without an ITA. SSD patients had a significantly higher incidence of AITA than controls, almost twice the frequency [OR = 1.98; 95% CI, 1.33–2.94].

The current findings are consistent with previous meta-analyses: AITA is twice as common in men as in women. Several studies on sex differences in the occurrence of AITA[5,9,10,27] support this conclusion. Its functional implications remain unknown. The study shows that men are more likely to develop schizophrenia. Most research indicates that males contract the disease at a younger age than females.[56]

Neuroanatomy of ITA

The ITA's comparatively small size is not an indicator of its importance; even small structures in the brain have essential functions. It is a part of the highly conserved “dorsal diencephalic conduction system” that gathers fibers from septal nuclei (center of motivation and pleasure), anterior cingulate (decision-making center), lateral hypothalamus (center of arousal and pain), and basal ganglia via the stria medullaris thalami [Figure 5].[57,58] The ITA pushes the fibers of the stria medullaris thalami more dorsally. Kochanski et al. (2018)[55] examined the course of the stria medullaris thalami using diffusion tensor imaging. They showed that ITA has crossing fibers of the stria medullaris thalami connecting it to the habenular nuclei, the anterior part of the cingulate and insular areas, and the dorsal orbitofrontal cortex, but the pathway is unidirectional. The habenular outflow via the retroflex fasciculus modulates the release of serotonin, dopamine, and norepinephrine. The unidirectional connection of the frontal limbic region to the core complex of the habenula through the stria medullaris thalami is larger and more developed in humans than in other species. Thus, the habenular complex's direct relationship with neurotransmitter release could potentially have a role in neuropsychiatric conditions such as schizophrenia, bipolar disorder, and depression.[57] The ITA is related to the ventral striatum and is said to be involved in the emotions of loneliness.[14]

Figure 5.

A schematic representation of the dorsal diencephalic pathway possibly connected to the ITA via the stria medullaris thalami. AC: anterior commissure, SMT: stria medullaris thalami, ITA: interthalamic adhesion, Habenula: habenular nucleus, pineal: pineal gland

ITA: Neurodevelopmental or neurodegenerative?

The various brain anomalies observed in psychiatric disorders (e.g., schizophrenia and bipolar disorder) include small temporal lobe, ventriculomegaly, abnormal cavum septum pellucidum, and AITA. Two general hypotheses have been proposed to account for these abnormalities in neuropsychiatric disorders, particularly schizophrenia: neurodevelopmental and neurodegenerative.[59] Schizophrenia has been associated with hereditary B-cell lymphoma 2 gene deficiency, which promotes pathological apoptosis. Some autopsy examinations have revealed neuropathological changes such as the absence of gliosis and marked abnormalities in cytoarchitecture, indicating neurodegeneration. However, “pathological neuronal apoptosis” could not explain all those phenomena; apoptosis does not produce gliosis, though necrosis does. A few authors have proposed that changes in body weight, alcohol intake, steroid or anti-psychotic drug use, or hormonal status can also affect brain morphology, as seen in structural neuroimaging. Animal models can provide critical information for establishing hypotheses about neurodevelopment. For example, animal models for schizophrenia have shown early injury to specific brain regions after prenatal exposure to viruses such as influenza and Borna virus and prenatal hypoxic or ischemic insults.[60] A meta-analysis by Cannon et al. (2002)[61] demonstrated the role of prenatal disorders or stress and obstetric complications in causing neuropsychiatric disorders. Takahashi et al.[37] documented a smaller ITA with advancing age in schizophrenia and bipolar disorder patients and suggested neurodegenerative changes. In the present study, we found no association between the incidence of ITA and mean age. ITA could therefore be a neurodevelopmental disorder.

AITA in neuropsychiatric illness: A neuroanatomical association or risk factor

Neuropsychiatric disorders are multifactorial, involving genetic predisposition, environmental factors, traumatic head injury, infection, and side effects of medication or substance abuse. Although AITA is not necessarily a direct cause of neuropsychiatric disorders, it could increase disease susceptibility in the presence of the abovementioned factors. AITA can be considered to represent the disturbed malformation of neural networks, which includes the thalamic and related subcortical regions during the period of neurodevelopment.[62] The ITA is connected with midline thalamic nuclei and is involved in regulating dopamine release in the mesolimbic pathway. This may explain the variety of schizophrenia symptoms.[63]

Therefore, AITA might be responsible for increased vulnerability to developing a psychotic disorder.

The higher incidence of AITA in neuropsychiatric patients (almost double that in normal subjects) is a very consistent association, suggesting that it is an unmodifiable risk factor like genetic predisposition. We found no study comparing the incidence of neuropsychiatric disorders in patients with and without ITA. Owing to such data's absence, we still hesitate to deem it a neuroanatomical risk factor.

Limitations

The limitations of this study must be considered in future studies to assess risk stratification. A cohort or longitudinal study is needed to compare the incidence of psychiatric disorders in individuals with or without ITA and to calculate the attributed risk. Despite this limitation, the results of this study are important because the calculated effect size has minimal heterogeneity and is highly consistent across populations, regardless of age and sex.

CONCLUSIONS

Given its solid and consistent association with psychiatric disorders, AITA can be treated as an associated risk factor affecting the function of the habenula nuclear complex via the stria medullaris thalami. It could contribute to increased vulnerability to psychiatric disorders caused by genetics, environmental factors, traumatic head injury, infection, drug side effects, or drug addiction. Contrary to Macedo's claim, no neurological injury was recognized in association with AITA. The current revelations could change this view.

Abbreviations

ITA: interthalamic adhesion; AITA: absent interthalamic adhesion; OR: odds ratio; MRI: magnetic resonance imaging.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Supplementary Figure 1

Sagittal section of brain showing ITA (1), thalamus (2), habenula (3), anterior commissure (4), stria medullaris thalami (5), pineal gland (6), and septum pellucidum (7)