Edited by: Lutz Jäncke, University of Zurich, Switzerland
Reviewed by: Dahua Yu, Inner Mongolia University of Science and Technology, China; Nabin Koirala, Haskins Laboratories, United States
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Aging affects pain experience and brain functioning. However, how aging leads to changes in pain perception and brain functional connectivity has not yet been completely understood. To investigate resting-state and pain perception changes in old and young participants, this study employed region of interest (ROI) to ROI resting-state functional connectivity (rsFC) analysis of imaging data by using regions implicated in sensory and affective dimensions of pain, descending pain modulation, and the default-mode networks (DMNs). Thirty-seven older (66.86 ± 4.04 years; 16 males) and 38 younger healthy participants (20.74 ± 4.15 years; 19 males) underwent 10 min’ eyes-closed resting-state scanning. We examined the relationship between rsFC parameters with pressure pain thresholds. Older participants showed higher pain thresholds than younger. Regarding rsFC, older adults displayed increased connectivity of pain-related sensory brain regions in comparison to younger participants: increased rsFC between bilateral primary somatosensory area (SI) and anterior cingulate cortex (ACC), and between SI(L) and secondary somatosensory area (SII)-(R) and dorsolateral prefrontal cortex (PFC). Moreover, decreased connectivity in the older compared to the younger group was found among descending pain modulatory regions: between the amygdala(R) and bilateral insula(R), thalamus(R), ACC, and amygdala(L); between the amygdala(L) and insula(R) and bilateral thalamus; between ACC and bilateral insula, and between periaqueductal gray (PAG) and bilateral thalamus. Regarding the DMN, the posterior parietal cortex and lateral parietal (LP; R) were more strongly connected in the older group than in the younger group. Correlational analyses also showed that SI(L)-SII(R) rsFC was positively associated with pressure pain thresholds in older participants. In conclusion, these findings suggest a compensatory mechanism for the sensory changes that typically accompanies aging. Furthermore, older participants showed reduced functional connectivity between key nodes of the descending pain inhibitory pathway.
Pain in older adults is poorly understood. Aging seems to be associated with increased pain thresholds and poor functioning of endogenous pain inhibition mechanisms (Lautenbacher,
Age-related changes in the spontaneous organization of the brain have been linked to the cognitive, perceptual and motor alterations that frequently accompany aging (Ferreira and Busatto,
Therefore, the present study aimed to analyze the impact of aging on pain processing (pain pressure thresholds) and associated rsFC among sensory, affective and descending modulatory pain processing structures. Moreover, considering the last studies suggesting that the default mode network (DMN) is also active during experimental pain tasks in both young (Kong et al.,
Participants were recruited from the University of the Balearic Islands (the older group was recruited from a senior program of the University or University employees). The sample was composed of 37 healthy older adults (16 men; 66.86 ± 4.04, the age range of 60–79 years) and 38 healthy young adults (19 men; 20.74 ± 4.15, the age range of 18–26 years; see
Sociodemographic and clinical data of younger and older groups.
Younger ( |
Older ( |
Statistic | |||
---|---|---|---|---|---|
Age (years) | 20.74 (2.34) | 66.84 (4.15) | |||
Sex (males) | 19 | 14 | 0.602 | ||
Educational level | |||||
<8 | 0 | 3 | |||
8–12 | 1 | 4 | |||
>12 | 37 | 25 | |||
Medication | Anxiolytic | 0 | 1 | ||
Antidepressant | 0 | 5 | |||
Anti-inflammatory | 1 | 4 | |||
Cholesterol | 0 | 22 | |||
Hypertension | 0 | 15 | |||
Hypoglycemic | 0 | 3 | |||
Others | 7 | 27 | |||
Finger pain threshold (N) | 57.80 (24.67) | 78.18 (23.55) | |||
Wrist pain threshold (N) | 44.76 (21.23) | 55.30 (18.88) | |||
Shoulder pain threshold (N) | 42.81 (17.80) | 56.15 (24.39) | |||
Finger (0–100 pain rating) | 31.04 (19.86) | 50.16 (24.11) | |||
Wrist (0–100 pain rating) | 34.84 (18.63) | 51.41 (23.03) | |||
Shoulder (0–100 pain rating) | 30.76 (17.76) | 48.75 (22.70) | |||
Blood pressure (mmHg) | Systolic | 120.03 (15.25) | 131.66 (15.07) | ||
Diastolic | 73.95 (9.43) | 77.22 (17.90) | 0.332 | ||
PHQ-9 | 3.49 (2.80) | 2.56 (2.71) | 0.176 | ||
GAD-7 | 3.45 (3.37) | 3.28 (3.30) | 0.836 | ||
PANAS | Positive | 32.50 (6.21) | 37.09 (6.28) | ||
Negative | 12.42 (2.62) | 13.34 (3.59) | 0.219 |
All participants were interviewed in a previous screening session to exclude those who presented any of the following criteria: any current psychiatric or neurological condition, acute or chronic pain, uncontrolled hypertension, history of drug abuse, cognitive impairment (operationalized as a score below 27 in the Mini-Mental State Examination; Lobo et al.,
Before the day of the main experiment, all participants underwent an interview to assess clinical characteristics through a health interview and self-report questionnaires. They completed the Spanish versions of the Patient Health Questionnaire (PHQ-9; Kroenke et al.,
First, to control confounding variables, blood pressure was measured in the right arm with a tensiometer (OMRON MX2, OMRON Healthcare, Hoofddorp, Netherlands) after the participant was seated and after they had rested for 5 min. Then, pressure pain thresholds were assessed always by the same experimenter and applying a previously used procedure (Martínez-Jauand et al.,
After the measurement of pressure pain thresholds, all participants underwent an MRI and fMRI scanner on a GE 3T scanner (General Electric Signa HDx, GE Healthcare, Milwaukee, WI, USA) at the Son Espases University Hospital. For each participant, 240 whole-brain echo-planar images were acquired over 10 min with the eyes closed [36 transversal slices per volume; 3 mm slice thickness; 90° flip angle; repetition time (TR): 2,500 ms; echo time (TE): 35 ms; 64 × 64 matrix dimensions; 240 mm field of view; 3.75 × 3.75 × 3 mm voxel size]. The structural imaging data consisted of T1-weighted images. Twenty-five participants were acquired with the following parameters: 292 slices per volume; repetition time (TR): 7.84 s; echo time (TE): 2.976 ms; matrix dimensions, 256 × 256; 256 mm field of view; 1 mm slice thickness; 12 flip angle. Fifty participants were acquired with the following parameters: 220 slices per volume; TR: 7.9 s; TE: 3 ms; matrix dimensions, 256 × 256; 256 mm field of view; 1 mm slice thickness; 12 flip angle. T1 imaging data was only used to perform intraindividual coregistration and nuisance pre-processing. Scanner noise was passively reduced by using in-ear hearing protection. Also, foam cushions were placed over the ears to restrict head motion and further to reduce the impact of scanner noise.
The connectivity analyses were performed with the CONN-fMRI Fc toolbox v18a (Whitfield-Gabrieli and Nieto-Castanon,
To examine changes in functional connectivity within DMN structures and within pain-network areas due to aging, two separate region of interest (ROI) to ROI analyses were performed. DMN main regions were provided by the CONN toolbox and were originally derived from ICA analyses based on the human connectome project (HCP) dataset of 497 subjects. These regions included medial prefrontal cortex (mPFC), posterior cingulate cortex (PCC)/precuneus and bilateral lateral parietal (LP). Second, based on previous fMRI studies on experimental pain in patients with chronic pain and healthy controls (Gracely et al.,
Center of Montreal Neurological Institute coordinates for each region of interest (ROI) within the default-mode network and within the pain network, extracted from an anatomical atlas and previous studies.
ROI | |||
---|---|---|---|
mPFC | 1 | 55 | −3a |
LP (R) | 47 | −67 | 29a |
LP (L) | −39 | −77 | 33a |
PCC/precuneus | 1 | −61 | 38a |
ACC | 1 | 18 | 24b |
INS (R) | 37 | 3 | 0b |
INS (L) | −36 | 1 | 0b |
AMY (R) | 23 | −4 | −18b |
AMY (L) | −23 | −5 | −18b |
THA (R) | 11 | −18 | 7b |
THA (L) | −10 | −19 | 7b |
dlPFC (R) | 38 | 34 | 30c |
dlPFC (L) | −37 | 34 | 30c |
SI (R) | 43 | −28 | 53c |
SI (L) | −41 | −29 | 53c |
SII (R) | 52 | −37 | 37c |
SII (L) | −51 | −37 | 37c |
PAG | 0 | −32 | −12d |
Furthermore, we wanted to assess if changes in connectivity values were related to changes in pain perception. For this purpose, subjective pain ratings and pressure pain threshold indexes were computed by averaging the three-body locations (finger, wrist, and shoulder). Pearson’s correlations were computed between rsFC showing significant differences between groups and the pain threshold indexes by using SPSS (IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY, USA: IBM Corp). Finally, given that PANAS positive scores were higher in older in comparison to younger participants (see below), these scores were also correlated to rsFC scores. The new
Group differences in sex and educational level were analyzed with Chi-Square Tests. Age, questionnaire scores, systolic and diastolic blood pressure, pressure pain thresholds and related subjective pain ratings were analyzed with Student
Statistical analyses of sociodemographic and questionnaires data (
Older participants showed higher pressure pain thresholds than younger participants on the finger (
Pressure pain threshold index (in Newtons) and subjective pain rating index (0–100) in the younger and older groups. Older participants showed increased indexes in comparison to younger participants. **
Also, given the group difference in systolic blood pressure, one-way ANCOVA of the pressure pain threshold index and the subjective pain rating index controlling for systolic blood pressure were conducted. The differences between groups were maintained after controlling for systolic blood pressure (see
Functional connectivity analyses of the pain-network (
Functional connectivity differences derived from pain-network ROI to ROI analyses in older compared to younger groups.
p-unc | p-FDR | ||
---|---|---|---|
Older > Younger | |||
SI (L)—ACC | 3.38 | 0.0006 | 0.0076 |
SI (L)—SII (R) | 3.29 | 0.0008 | 0.0100 |
ACC—SI (R) | 2.57 | 0.0061 | 0.0394 |
SI (L)—dlPFC (R) | 2.40 | 0.0095 | 0.0411 |
Older < Younger | |||
INS (R)—AMY (R) | −3.85 | 0.0001 | 0.0016 |
INS (L)—ACC | −3.43 | 0.0005 | 0.0065 |
AMY (R)—AMY (L) | −3.41 | 0.0005 | 0.0069 |
AMY (R)—ACC | −3.01 | 0.0018 | 0.0116 |
AMY (R)—INS (L) | −2.86 | 0.0028 | 0.0180 |
PAG—THA (R) | −2.77 | 0.0036 | 0.0239 |
AMY (R)—THA (R) | −2.76 | 0.0037 | 0.0239 |
INS (R)—AMY (L) | −2.72 | 0.0041 | 0.0259 |
INS (R)—ACC | −2.58 | 0.0060 | 0.0259 |
PAG—THA (L) | −2.54 | 0.0066 | 0.0428 |
AMY (L)—THA (L) | −2.45 | 0.0083 | 0.0360 |
AMY (L)—THA (R) | −2.22 | 0.0147 | 0.0478 |
Functional connectivity differences between pain-related regions of interest (ROIs) in the older group as compared to the younger group.
Pearson’s correlational analyses showed that functional connectivity between SI (L) and SII (R) was positively associated with the pressure pain threshold index in the older group (
Scatter plots showing the correlation between left primary somatosensory cortex (SI) and right SII functional connectivity with pressure pain thresholds (upper panel), and the correlation between right insula (INS) and left amygdala (AMY) functional connectivity with pain intensity ratings (lower panel) in the older and younger groups.
This study aimed to analyze the age-related changes in spontaneous brain activity to find possible explanations for the increased pain thresholds and lack of pain inhibition that characterizes pain processing in older adults. The analysis of rsFC imaging data and the possible relationship with pressure pain thresholds in older compared to young participants revealed the following main results. First, older participants showed an aberrant hyperconnectivity of the SI with SII and frontal brain areas (dlPFC and ACC) at rest. Second, the hyperconnectivity between somatosensory regions was related to increased pressure pain thresholds in the older group. Third, older participants showed decreased rsFC between brain regions that constitute the brain circuitry defined as the descending pain modulatory system (ACC, INS, AMY, THA, and PAG). Fourth, DMN functioning was rather well preserved in the older participants and differences in connectivity concerning the young group were found only between the LP cortex and PCC/precuneus. Although, connectivity between DMN regions was not related to pain perception changes. The implications of these findings are discussed below.
The main findings from studies on pain sensitivity that have been carried out in humans include an increased threshold and decreased tolerance with advancing age mechanisms (Lautenbacher,
Indeed, we found increased connectivity of bilateral SI with ACC, and of left SI with right dlPFC and right SII in older compared to younger participants. SI is a critical component of the nociceptive pathway and is known to encode body location, intensity, and quality of nociceptive stimuli (Bushnell et al.,
Moreover, in agreement with previous studies showing a strong relationship between rsFC of INS and AMY with pain perception (Boly et al.,
Finally, we found increased rsFC between LP regions and the PCC/precuneus in older participants in comparison to younger ones. Studies have revealed that increased DMN activity in healthy aging is associated with a higher level of background sensory processing during cognitive tasks (Grady et al.,
There is a limitation of our study that merit further consideration. Most of the older participants were taking medication. To control for this confounding variable, we replicated the rsFC analyses excluding those subjects who were taking medication that can influence the central nervous system and/or pain perception (see
This study offers new insights into the evolution of cortical networks in normal aging and its relevance to pain perception. The clinical relevance of resting-state networks is notable because the degree of connectivity in these networks predicts individual cognitive, emotional and sensory functions. In our study, older participants showed an abnormal hyperconnectivity of the primary somatosensory area (SI) with other somatosensory and frontal brain regions. This result, together with the positive correlation found between SI-SII functional connectivity and pressure pain thresholds could be interpreted as a compensatory mechanism for the slowed pain processing that seems to accompany aging. Furthermore, older participants showed reduced functional connectivity between key nodes of the descending pain inhibitory pathway. Thus, our results are in line with the suggestion that in aging the pain system is activated lightly later, but, over time, dysfunction of pain modulatory and evaluation processes would lead to increased pain perception (Lautenbacher,
The datasets generated for this study are available on request to the corresponding author.
This study was reviewed and approved by Ethics Committee of the Balearic Islands (ref.: IB 3429/17 PI). The participants provided their written informed consent to participate in this study.
All the authors have read and approved the article and the procedures used. AG-R, CS, FA, MM, and PM discussed the original design of the experiment. AG-R acquired the data and drafted the original manuscript. AG-R and JT analyzed the data. AG-R, JT, CS, FA, MM, and PM revised the manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at:
1