Men with pedophilia who committed child sexual offenses on average had a 47 mm3 smaller hypothalamus per side

Hypothalamic volume in pedophilia with or without child sexual offense. Melanie Storch, Maria Kanthack, Till Amelung, Klaus M. Beier, Tillmann H. C. Krueger, Christopher Sinke, Henrik Walter, Martin Walter, Boris Schiffer, Stephanie Schindler & Peter Schoenknecht. European Archives of Psychiatry and Clinical Neuroscience, Nov 12 2022. https://link.springer.com/article/10.1007/s00406-022-01501-w

Abstract: The hypothalamus regulates sexual behavior and is simultaneously associated with aggression and violence. Consequently, this brain region is relevant in research of pedophilia and child sexual offenses (CSO). The distinction between these two phenomena is of great importance and was the object of consideration of this study. We analyzed exclusively men, including 73 pedophilic offenders who committed CSO, an equal number of people with pedophilia but without such offenses, and 133 non-pedophilic, non-offending subjects who formed the control group. All data were collected in a multicenter in vivo study and analyzed using a semi-automated segmentation algorithm for 3-Tesla magnetic resonance images. Men with pedophilia who committed CSO on average had a 47 mm3 smaller hypothalamus per side than people without committed CSO. This effect was driven by both the group of non-offending people with pedophilia and the control group. By contrast, the exploratory comparison of pedophilic persons without CSO with the control group showed no significant difference. The present study demonstrates a deviant hypothalamic structure as a neurobiological correlate of CSO in pedophiles, but not in people with pedophilia who have not committed CSO. Thus, it strengthens the argument to distinguish between sexual offending and paraphilic sexual preferences.

Discussion

Child sexual offenders show hypothalamic volume reduction

In agreement with our first hypothesis of a reduction of hypothalamus volume in CSO, we observed a hypothalamic reduction in persons with pedophilia who committed CSO. The effect was evident in both hypothalamic hemispheres in absolute volumes and after correction for the two confounders ICV and age separately and simultaneously after outlier exclusion. Following contrasts confirmed our hypotheses that the effect was driven by the control as well as the P-CSO group. In our study CSO is related to hypothalamic volume reduction in pedophilic men and this effect is driven by the control group as well as the pedophilic non-offender group.

Significance was initially lost after simultaneous correction for ICV and age in the global univariate group comparisons of hypothalamus volumes and reemerged after excluding the statistical outliers. This suggests suppression effects due to multicollinearity and a sensitivity of the parametric models to non-normal outliers. The statistical outliers could not be explained by variations in clinical characteristics, measurement, or sampling. To account for the known sensitivity of parametric tests to outliers, we reported results before and after excluding statistical outliers for the purpose of transparency and reliability.

Since one contrast of volume reduction was not significant in the right hypothalamus, we may expect that the effect is more prominent on the left side. This would be consistent with the findings of a volume reduction of the right amygdala [24] and a functional connectivity between the right amygdala and the left hypothalamus in pedophilic offenders [65]. Blinding the rates to the hemispheres minimized the likelihood that the algorithm was applied differently.

Pedophilic participants both with and without histories of committed CSO consumed material depicting child sexual abuse, indicating the alteration in hypothalamic structure appears to be associated with implementation of CSO at the behavioral level. Our findings corroborate studies highlighting the hypothalamus and its subsequent cascades and regulatory mechanisms in violence [13,14,15,16,17]. Furthermore, our results are consistent with previous studies that focused on sexual violence against children and showed an activity reduction in the hypothalamus in pedophilic offenders [27] and acquired pedophilia and CSO after hypothalamic damages [18,19,20]. More precisely our findings possibly confirm our initial assumption that a reduced hypothalamic volume may indicate a reduced HPA axis activity. The deficit of glucocorticoids, induced by the hypofunction of this axis, may be related to aggression and CSO through epigenetic changes in the prefrontal cortex [31, 32]. In our sample Kruger et al. [37] found no cortisol reduction in CSO, but it is not unlikely that this may be due to methodological limitations, such as measuring at different times, despite cortisol levels fluctuating throughout the day.

Voxel-based morphometry studies are less sensitive for small structures, such as the hypothalamus [66]. For this reason, this analyses of this area [23,24,25,26] may not have yielded results. Additionally, except for Schiffer et al. [25], the field strengths of the MRIs were lower in the mentioned studies and the sample sizes were smaller.

No significant differences in non-offending people with pedophilia

In contrast to the results regarding CSO, we found no significant difference in exploratory comparisons of left or right hypothalamic volumes between non-offending pedophilic men and the control group. This was in line with our expectations. Differences remained non-significant even with correction for ICV and/or age. This suggests an unchanged hypothalamic macrostructure in non-offending people with pedophilia. Interestingly, there was no gradually progressive increase in hypothalamic volume between the groups (e.g. P + CSO < P-CSO < controls), not even descriptively. Thus, the P-CSO group does not appear to be an intermediate stage between pedophilic offenders and the control group. Contrary to our conclusions previous studies attributed structural changes of the hypothalamus or other brain regions to pedophilia [22, 23, 26]. It can be speculated that their results were influenced by sexual offenders.

General assessments of the observed hypothalamus volumes

The bilateral hypothalamus volumes measured in vivo with a total mean of 1543 mm³ (SD = 146.6 mm³) are slightly higher than the volumes previously measured with the same method (1427 mm³ to 1478 mm³) [52]. However, this was to be expected, as our sample consisted exclusively of men and a sexual dimorphism of the hypothalamus, postulated to be larger in men, has been shown before [67,68,69,70,71]. Furthermore, the correlations of hypothalamic volume with age and ICV are consistent with previous reports [66, 67, 72, 73]. The P + CSO group also had the lowest ICV. This finding has to be questioned in future studies by exploring several distinct brain regions involved in control of behaviour and sexual functioning.

Strengths and limitations

Probably one of the greatest strengths of this study is the distinction between offenders and non-offenders and thus also the fundamental distinction between offenders and people with pedophilia. A disadvantage is that the classification regarding the offender status was necessarily dependent on the self-reporting of the participants, which risks a probability of false statements in, theoretically, both directions. Attempts have been made to mitigate this by ensuring anonymity. Balancing too many or too few exclusion criteria is difficult as excluding individuals with specific diagnoses or medications increases homogeneity at the cost of generalizability. The distribution of lifetime mental and personality disorder diagnoses was significantly different between all groups and the hypothalamic volume may be affected by these. Violent crimes other than CSO, which were not exclusion criteria, may have influenced the results. Even though the sample size is large compared to previous studies in the subject area, it is not sufficient to calculate an equivalence test [64] between the hypothalamus volumes of pedophilic non-offenders and the control group. Since the study is designed cross-sectionally, no conclusions about cause and effect can be drawn. Unbalanced sample distribution among different locations and thus scanner models may have an effect on the GM-TPMs. The evaluated 3-Tesla MR images provided a strong basis for measurements of the brain structures, but a higher field strength would reduce the partial volume effect. Manual segmentation at submillimeter resolution is the most accurate method for this, which is unfortunately hardly feasible for higher case numbers such as in our study due to high time expenditure.

Our large-scale multicenter sample consisted of only men; therefore we can only draw conclusions about males. However, men are most relevant for the research on sexual violence against children, as they are the major group of offenders [74]. To answer whether the results are not only valid for people with pedophilia who committed CSO, but also for CSO in general, exploring non-pedophilic CSO subjects is needed. The structural analysis was based on group comparisons and cannot serve as a basis for a diagnostic criterion or to draw reliable conclusions about individuals.

To our knowledge, no other study has accurately investigated the structure of the hypothalamus in pedophilia with or without CSO. In addition to theoretical considerations about understanding of the neurobiological underpinnings of CSO in pedophilia, the study may add to additional impact in the future, as trait markers of risk factors for committing CSO are needed to stimulate early in the clinical course specific sexual therapeutic treatment which covers more than general psychotherapeutic intervention. The present results need to be replicated in further studies and assessed in relation to, for example, endocrinological and behavioral functions before practical implication can be raised. Implementing an equivalence test would be an important challenge for future studies to discuss the hypothesis of similar brain structures in pedophilic non-offenders and subjects of a control group. However, the required sample sizes for the populations of interest are difficult to realize. Further studies are warranted using functional brain imaging to investigate emotional processing according to the development of pedophilia or CSO over the lifespan. Another interesting question to examine is whether the hypothalamic volume reduction can also be found in a (large) non-pedophilic CSO group to clarify whether our results apply explicitly to the P + CSO group or are valid for CSO exclusively.

The topics of pedophilia and CSO are undeniably emotionally charged. Research such as ours not only provides a better understanding of neural mechanisms underlying pedophilia and CSO, but also contributes to education and public discussion about these matters, rather than reinforcing threats to child welfare with silence and stigmatization.