The main findings of this study are hippocampal FC showed the between-group difference and it was correlated with catastrophic thinking as well as pain intensity. In addition, the patients with PHN showed increased FC among emotional pain regions and the cerebellum and decreased FC in the basal ganglia.
In this study, the hippocampus, cerebellum, and PCC showed significant FC increases in the PHN group compared with those in the HC group. Generally, the hippocampus engages in memory and learning12, the cerebellum is involved in motor control18, and the PCC is a core region of the DMN19. These regions are also involved in progression and maintenance of chronic pain. The hippocampus is thought to be a key structure involved in the development of chronic migraine20. It is well known that the hippocampus is associated with pain-related emotional coping as well as learning of emotional memory including pain experience21. Moreover, when transition from acute pain to chronic pain occurs, the FC of the hippocampus increases22. A study on trigeminal neuralgia showed that the hippocampal volume was decreased in the patient group23. Such functional and morphological changes of the hippocampus are implicated in the development of chronic PHN. In addition to the hippocampus, the PCC is also thought to have a key role in the development of chronic pain. Increased FC with the insula, which is known to process affective elements of pain, is considered as a form of maladaptive neuroplasticity leading to the development of chronic pain24. Our results correspond to this concept.
Our study revealed that FC between the bilateral primary somatosensory cortices and the left hippocampus was decreased in the PHN group. In a previous study, patients with high frequency migraine showed significant decreases in FC between the hippocampus and other brain regions that are involved in pain processing20. Although our finding is consistent with the results of the previous study in terms of hippocampal FC decrease in chronic pain conditions, it is a novel finding that FC between the hippocampus and primary somatosensory cortices was decreased in chronic pain patients. This finding could explain why patients with PHN show hypoesthesia.
The basal ganglia are also thought to have a critical role for chronic pain25. The basal ganglia receive nociceptive inputs from the cingulate cortex, dorsolateral prefrontal cortex, hippocampus, and amygdala26. Previous studies have shown a relationship between activities of the basal ganglia and pain in patients with fibromyalgia27 and patients with complex regional pain syndrome28. In the present study, the patients with PHN showed significant decreases in FC between the nucleus accumbens and the putamen and FC between the amygdala and the putamen. These results suggest that the basal ganglia also have a key role in the development of PHN. Specifically, these changes may be related to aberrant emotional regulation in patients with PHN. In fact, patients with PHN have a high tendency to develop anxiety and depression29.
Correlation analysis revealed that the FC of the hippocampus is related to pain intensity and psychological status of PHN patients. In the present study, a higher tendency of catastrophic thinking about pain was correlated with increased connectivity between the mPFC and the right hippocampus. The mPFC, which encompasses the rostral ACC, is a region involved in transition from acute pain to chronic pain30, and this region connects with limbic structures such as the amygdala and the ventral striatum31. The mPFC is also known to be a part of the DMN. Whereas the DMN shows deactivation while individuals focus on the external environment, it is activated when individuals do not engage in behavioral/cognitive tasks32. The hippocampus has a close relationship with the transition from acute pain to chronic pain20. Generally, sensitivity to pain is correlated with FC between brain areas associated with the DMN, including the PCC, mPFC and hippocampus, and pain regions33. Our results are consistent with this fact. Meanwhile, it is a new finding that FC of the mPFC has a relationship with catastrophizing thinking about pain. FC between bilateral SI and left hippocampus negatively correlated with pain intensity in PHN patients. This finding provides a new evidence about pathophysiology of PHN and supports the notion that maladaptive plastic changes in the central nervous system play an important role in the development and maintenance of chronic pain. SI is known to process the sensory aspect of pain9. On the other hand, the hippocampus is known to be involved in emotion and emotional memory including pain experience. Our result suggests that pain in patients with PHN is exaggerated by emotional modulation, and that its magnitude of modulation is related to the strength of SI-Hippocampus connectivity. In fact, previous studies has reported that the amygdala that densely connects with the hippocampus was related to emotional modulation of pain34 and that the amount of SI FC to other brain regions was related to pain in chronic low back pain condition35.
In the present study, the strength of SI-Hippocampus FC negatively correlated with pain intensity. That is, the patients with stronger pain showed weaker SI-Hippocampus FC. This relationship is counterintuitive. However, when a correlation between FC strength and variables is discussed, it is necessary to consider actual FC strength. In this study, the strength of SI-Hippocampus FC in the patients with weaker pain was around zero. In contrast, its strength in the patients with stronger pain were negative values. Therefore, although meaning of negative FC remains controversial, our result can be considered that patients with stronger pain showed stronger SI-Hippocampus FC.
We must consider limitations in this study. As mentioned above, the rs-fMRI data for the HC group and that for the PHN group were acquired at different institutions. As a result, even though the scanning parameters were almost the same in the two institutions, there was a possibility that the between-group differences we observed in this study merely reflect inter-scanner differences. However, this possibility can be ruled out because there were correlations between several FCs showing significant between-group differences and symptoms of PHN, particularly pain intensity and catastrophic thinking. Second, there is a possibility that analgesics prescribed for the patients with PHN, such as pregabalin, affected their rs-FC, as shown in a previous study36. Indeed, some of our patients with PHN were prescribed pregabalin for treatment of PHN. Therefore, we could not exclude the possible effects of pregabalin for the brain networks.
To conclude, we identified alterations in FC with the hippocampus in the patients with PHN compared with FC in the HC group. Furthermore, FC with the hippocampus was correlated with individual pain intensity and tendency of catastrophic thinking about pain. Our results suggest that functional alterations of the hippocampus are related to not only pain perception but also the pain-related cognitive process, especially catastrophizing, in patients with PHN as in patients with other types of chronic pain.