In the current study, we analyzed DTI microstructural parameters of 21 WM tracts in adults with ADHD and the healthy controls to test our hypothesis regarding microstructural disruptions in ADHD in our regions of interest. These analyses revealed areas of decreased FA and increased diffusivity indices. Mentioned findings were consistent with earlier studies on adults with ADHD. Further exploration of macrostructural variables and the NQA of the same tracts provided additional insight; our study revealed macrostructural expansions in IFOF and decrements in CC (body and forceps minor) in ADHD subjects. Moreover, a novel finding of highly significant NQA increments was discovered in all of the tracts examined, except left fornix.
To investigate the microstructural properties of WM tracts, we examined four commonly used DTI measures (FA, MD, RD, AD). In our analysis, in the ADHD group, the left Uncinate Fasciculus demonstrated increased MD, RD, and AD; CC Forceps Major demonstrated increased MD and decreased FA and GFA; Right Fornix showed increased MD and RD; Left fornix and right Uncinate Fasciculus demonstrated increased MD and AD; Right Cingulum Rarolfactory demonstrated increased RD and decreased FA and GFA; right Cingulum Frontal Parietal demonstrated increased RD, and SLF demonstrated reduced FA and GFA. As expected, the alterations in FA and GFA followed similar trends.
Our analysis detected decreased FA values in several tracts in ADHD subjects. The lower FA in the CC (Forceps Major) is consistent with a previous meta-analysis, which found the CC to demonstrate the most significant FA decrement among all the regions in ADHD patients (Aoki et al., 2018). Earlier studies on FA alterations in ADHD subjects have revealed inconsistent findings. In several studies, widespread FA reductions were detected in ADHD subjects (Aoki et al., 2018; Van Ewijk et al., 2014). Other studies, however, showed no significant FA alteration in any of the explored tracts (de Zeeuw et al., 2012; Lawrence et al., 2013). The heterogeneity in FA results can be explained by the difference in age groups within the existing literature. However, more studies analyzing multiple variables are required for each group to validate these findings.
Furthermore, we found several areas with elevated values of MD in ADHD patients. However, in contrast to previous studies reporting higher MD levels in SLF and CC Forceps Minor, our results revealed unchanged values of MD in these regions (Konrad et al., 2010; Lawrence et al., 2013; Pavuluri et al., 2009). We also found regions with higher AD values in ADHD subjects, which is in line with prior studies (Lawrence et al., 2013; Silk et al., 2009; Tamm et al., 2012). AD values have been linked to axonal pruning (Bockhorst et al., 2008), and altered AD values may be associated with disrupted axonal architecture and pruning. We also detected elevated AD in Left Fornix and left and right Uncinate Fasciculus. Similar to the AD and MD values, regional patterns of increment in RD values in ADHD subjects were observed in our analysis, which is in line with previous studies (Lawrence et al., 2013; Nagel et al., 2011).
Additionally, to further investigate the macrostructural qualities of the WM tracts included in our analysis, we evaluated tract volume, mean length, tract span, diameter, and the number of tracts. Several studies have discovered global and tract-based significant WM volumetric differences in ADHD patients compared to normal controls (Cao et al., 2010; Pironti et al., 2014; Seidman et al., 2006; Tremblay et al., 2020). Our findings in these areas were consistent with previous research. As discussed earlier, alterations in the IFOF, CC body, and CC Forceps minor were observed. It was previously discovered that the presence and severity scores of ADHD significantly correlate with WM volume expansions in certain parts of the right IFOF tract (Pironti et al., 2014). In addition to volume expansion, increments in tract diameter and the number of tracts were observed in ADHD subjects in our study. According to several studies, size alterations in the IFOF have a potential role in the pathophysiology of the disease (Braskie et al., 2012; Tremblay et al., 2020). Another part of the brain that has frequently undergone volumetric investigations in ADHD patients is the CC (Cao et al., 2010; Dramsdahl et al., 2012). Prior studies have reported decrements in the CC size among ADHD children regardless of their treatment (Schnoebelen et al., 2010). In our study, similar reductions were observed in the adult ADHD patients, specifically in the Body and the Forceps Minor of the CC. However, these findings are in contrast with another study, in which the macrostructure of CC, unlike its microstructure, was not disrupted in the ADHD group (Dramsdahl et al., 2012).
A novel finding in our research was the widespread increments in NQA values in almost every analyzed tract. NQA is another microstructural measure for anisotropy based on oriental distribution function (ODF) (F.-C. Yeh et al., 2010) and is normalized by calculating the quota of the maximum QA (F.-C. Yeh et al., 2013; F. C. Yeh et al., 2011). NQA reduces the measurement variability, providing a higher resolution for the microstructure (F. C. Yeh et al., 2010). To our knowledge, NQA has not been formerly investigated in ADHD studies; thus, further research in different settings is required to interpret these findings as it was not our focus area.
To date, no detailed pathophysiological explanations underlying the WM patterns observed in ADHD patients have been discovered. Moreover, there are heterogeneous interpretations regarding the alterations in DTI metrics. For instance, increments in MD have been attributed to either decreased cellular density or myelin breakdown in earlier research (Alexander et al., 2007; Tievsky et al., 1999). Additionally, reduced neurofilaments or microtubules in the myelinated axons have been proposed to be potential etiologies for the increased AD (Kinoshita et al., 1999). Furthermore, reduced FA could result from disrupted myelination, reduced size, and the altered number of axons passing through a tract (Dramsdahl et al., 2012; Wu et al., 2017). The abovementioned alterations in MD, AD, and FA were observed in our study.
Certain tracts have been an area of focus in tractographic studies, suggesting that they contribute to the pathophysiology of the disease. For instance, in previous studies, IFOF, with smaller tract volume, diameter, and the number of tracts in ADHD probands, has been connected to compromised inhibitory control and executive performance (Pironti et al., 2014; Takeuchi et al., 2012). Moreover, some genetic alterations in the nerve growth factor pathways have been linked to disrupted myelination and WM developmental characteristics in the IFOF (Braskie et al., 2012). The disruptions in the IFOF have been related to altered connectivity of the default mode network (DMN) (Tremblay et al., 2020). It has been argued that defective communication between the ventral and dorsal attentional network (VAN and DAN, respectively), and DMN, contributes to impaired inhibitory control in ADHD patients (Bhaijiwala et al., 2014; Sonuga-Barke & Castellanos, 2007; Sudre et al., 2017).
CC is another region of interest with significant findings in our study and contributes to the pathophysiology according to the existing literature. Our analysis revealed decreased FA and GFA and increased MD and RD in the CC Forceps Major. In CC Forceps Minor and the body of CC, significant size reductions were observed. Similar findings, specifically in the isthmus of the CC, were linked to the number, density, and diameter of neuronal and non-neuronal components, along with the myelin sheath integrity and thickness (Aboitiz et al., 1992). Previous studies on children with ADHD have revealed significant alterations in microstructural and macrostructural metrics of the CC (Cao et al., 2010). However, in a subsequent study conducted exclusively on adults with ADHD, although specific microstructural alterations were correlated with ADHD in adulthood, no macrostructural changes were observed (Dramsdahl et al., 2012). These findings suggest that delayed maturation is a mechanism responsible for ADHD pathophysiology. Most of these abnormalities are compensated during maturation in adolescence, making the macrostructure in CC comparable to the average population; however, the maturation is not sufficient to completely correct the underlying microstructural defects (Dramsdahl et al., 2012). In contrast, the presence of both macrostructural and microstructural abnormalities in our study invalidates the abovementioned theory.