Journal Logo

Musculoskeletal

Brain gamma-aminobutyric acid, but not glutamine and glutamate levels are lower in older adults with chronic musculoskeletal pain: considerations by sex and brain location

Cruz-Almeida, Yenisela,b,c,d,*; Forbes, Meganc,e; Cohen, Ronald C.c,e; Woods, Adam J.c,e; Fillingim, Roger B.a,b; Riley, Joseph L. IIIa,b; Porges, Eric S.c,e

Author Information
doi: 10.1097/PR9.0000000000000952
  • Open

1. Introduction

Chronic pain in older individuals is a growing public health problem that negatively affects physical and cognitive function, ultimately decreasing quality of life and overall well-being.38 Furthermore, effective pain treatments are lacking for this vulnerable population and the identification of potential neurobiological mechanisms underlying these pain conditions may potentially reveal effective therapeutic targets. Of particular relevance to the field of pain and aging neurobiology is the ability to measure the brain's major inhibitory and excitatory neurochemicals: gamma-aminobutyric acid (GABA) and the combined resonance of glutamate and glutamine (Glx)51 using the Mescher-Garwood point-resolved spectroscopy (MEGA-PRESS) sequence32,33 in proton magnetic resonance spectroscopy (1H-MRS).11,32,35 To date, this method has been used in the study of a wide range of domains, including nonnociceptive sensory processing,40 depression,16 Alzheimer disease,26 aging,14,39 and epilepsy.41

Chronic pain states also have altered brain neurochemical levels using 1H-MRS. Previous studies have shown negative associations between GABA and chronic pain.1,13,19,22,23,30,49 Similarly, acute pain stimulation leads to decreases in cerebral GABA and increases in Glx levels,6 although this may depend on the experimental pain modality used.28 However, most of the studies to date have included both younger and older individuals, raising the possibility that in older individuals, brain aging itself, and not chronic pain, may result in altered brain biochemistry. Previous studies indicate an age-related decline in cortical GABA levels,14,17 with lower frontal GABA levels being associated with worse cognitive performance.39 To the best of our knowledge, this is the first report of the relationship between GABA and Glx in prefrontal (ie, a region affected by aging processes and involved in the cognitive modulation of the pain experience) and sensorimotor (ie, a brain region involved in somatosensory and pain processing) brain regions. Based on the previous literature, we hypothesized that community-dwelling older adults with chronic musculoskeletal pain would exhibit lower GABA and higher Glx levels compared with younger and older controls. In addition, we hypothesized that GABA and Glx levels would be significantly associated with self-reported and experimental pain.

2. Methods

2.1. Participants

Community-dwelling younger (18–24 years old) and older (60–93 years old) individuals who spoke and understood written English were recruited and screened for a larger cross-sectional NIH-funded study at the University of Florida focused on the neurobiology of age-related differences in pain modulation (Neuromodulatory Examination of Pain and Mobility Across the Lifespan). Potential participants were first screened over the phone and then again in person from the fall of 2015 to fall of 2019. We excluded individuals reporting any of the following conditions either over the phone screening or during the in-person baseline session: (1) Alzheimer, Parkinson, or other diagnosis of a nervous system disorder; (2) serious psychiatric conditions (eg, schizophrenia, major depression, or bipolar disorder); (3) uncontrolled hypertension (blood pressure >150/95 mm Hg), heart failure, or history of acute myocardial infarction; (4) systemic rheumatic disorders (ie, rheumatoid arthritis, systemic lupus erythematosus, or fibromyalgia); (5) chronic opioid use; (6) magnetic resonance imaging (MRI) contraindications; (7) excessive anxiety regarding protocol procedures; (8) hospitalization within the preceding year for psychiatric illness; (9) HIV or AIDS; and (10) cognitive impairment (Modified Mini-Mental State Examination (3MS) score ≤ 77.48 Younger adults were further excluded if they reported any chronic pain condition currently or in the past. All procedures were reviewed and approved by the University of Florida's Institutional Review Board, and all participants provided verbal and written informed consent. We have previously reported on various aspects of the NEPAL study participants and its detailed methodology.5,8,9,31 To address the aims of this study, we included assessments of self-reported pain, depressive symptoms, a quantitative sensory testing battery, and MRS neuroimaging obtained from 3 separate laboratory visits.

2.1.1. Self-reported pain

Older participants were categorized to the pain group if they reported pain on most days during the past 3 months that interfered with their life consistent with the Task Force for the Classification of Chronic Pain consensus for the 11th version of the International Classification of Diseases of the World Health Organization.50 Pain categorization occurred after study completion and confirmed by a medical and pain history review. Participants were also asked about their average pain intensity in the past 3 months using an 11-point numerical rating scale (0 = no pain to 10 = the most intense pain imaginable), frequency of pain during the past week, pain duration, number of anatomical pain sites, as well as pharmacological and nonpharmacological treatments used in the past 3 months as part of a standardized pain history interview.

2.1.2. Depressive symptoms

The Center for Epidemiologic Studies Depression Scale questionnaire was used to measure the frequency of depressive symptoms during the past week on a 4-point Likert scale.42

2.1.3. Quantitative sensory testing

Quantitative sensory testing (QST) was used to assess experimental pain sensitivity, similar to the methodology previously reported by our group in older individuals.9 Standardized testing was performed at the left thenar eminence and the first metatarsal head on all participants. Thermal pain thresholds were obtained with the TSA-II NeuroSensory Analyzer and accompanying software (Medoc Ltd, Ramat Yishai, Israel) using the method of limits.

2.1.3.1. Thermal pain thresholds

Subjects were instructed to indicate as soon as the sensation changed from “just being cold to being painfully cold” or from “just being hot to being painfully hot.” Each trial began at 32°C and was either decreased (for cold pain = cold pain threshold) or increased (for heat pain = heat pain threshold) at a rate of 1.5°C/s until pain threshold was reached or the cutoff value was reached (0°C for cold pain and 50°C for heat pain). Each trial was separated by at least 20 seconds. The mean across 3 trials at each test site was calculated as the cold and heat pain thresholds.

2.1.3.2. Pressure pain thresholds

A handheld digital pressure algometer (AlgoMed; Medoc) was applied to the quadriceps and trapezius muscles at a constant rate of 30 kPa/s up to a maximum pressure level of 1000 kPa. Participants indicated when the pressure first became painful by pressing a button. The order of testing site was counterbalanced. Each trial was repeated 3 times at each site and averaged.

2.2. Z‐transformation of quantitative sensory testing data

Older adult's QST data were standardized by using the following expression18: Z-score = (Xsingle older subject − Meanyounger controls)/SDyounger controls.

This procedure results in a QST profile with zero mean, 1SD. For clarity of data presentation, we adjusted the algebraic sign of Z‐score values so that it reflects the participant's sensitivity. Z‐values above “0” indicates higher pain sensitivity, whereas Z‐scores below “0” indicate lower pain sensitivity.

2.2.1. Neuroimaging session

Magnetic resonance imaging data were collected at the McKnight Brain Institute on the Advanced Magnetic Resonance Imaging and Spectroscopy facility's Philips 3-Tesla scanner using a 32-channel radiofrequency coil. Participants were asked about their current pain intensity on the MRI table before starting the scanning procedures using a visual analogue scale (0–100). A T1-weighted anatomical image (magnetization-prepared rapid gradient-echo; repetition time/echo time = 7.2 ms/3.7 ms, 1-mm3 isotropic voxels) was acquired for MRS voxel placement and segmentation. GABA-edited MRS data were acquired using the MEGA-PRESS sequence.32 The prefrontal voxel was placed medial on the axial plane, aligned with the genu of the corpus callosum, inclusive of the pregenual ACC and medial prefrontal cortex (ie, mainly Brodmann areas 10 [ie, the anterior-most portion of the prefrontal cortex] and 32). When time allowed, we also collected data from the sensorimotor (S/M1) regions with the voxel centered over the hand knob area,55 parallel to the anterior–posterior axis. All voxels were 3 × 3 × 3 cm3 and representative voxel locations are shown in Figure 1. As part of the quality control procedures, voxel locations were verified after data collection to identify potential placement errors. Editing was performed with 14-ms sinc-Gaussian pulses applied at 1.9 ppm in the on experiment and 7.46 ppm in the off experiment. This editing scheme co-edits approximately 50% macromolecules at 3 ppm, which are coupled to spins at 1.7 ppm also inverted by editing pulses. Therefore, all GABA values reported refer to GABA+ macromolecules. Acquisition details were repetition time per echo time of 2 seconds/68 ms, 320 transients with on-off scans alternating every 2 transients, a 16-step phase cycle (with steps repeated for on and off), 2048 data points acquired at a spectral width of 2 kHz, and variable pulse power and optimized relaxation delays water suppression.20 Quantitative analysis was performed using the Gannet program (version 2.0).12 All time domain data were frequency-corrected and phase-corrected using spectral registration,36 filtered with a 3-Hz exponential line broadening and zero filled by a factor of 16. Metabolite levels were fitted in the difference spectrum using simultaneous GABA + glutamate and glutamine (Glx) signal fitting and both were acceptable34 (Fig. 2). The GABA peak was fit using a 3-Gaussian function and a nonlinear baseline with the Glx signal at 3.75 ppm fit using nonlinear least squares fitting. The metabolite peaks were quantified relative to water (fit with a Gaussian–Lorentzian model) in institutional units. The large interindividual variability in healthy brain aging processes (Table 1), even across regions (Table 2), necessitates the consideration of voxel tissue composition to ensure that metabolite changes are not unduly affected by the composition of the measurement voxel.21 In this study, 2 distinct approaches were implemented: (1) the most commonly used approach, the cerebrospinal fluid (CSF) correction that normalizes the voxel by the non-CSF voxel fraction (1-fCSF), with the implicit assumption that the metabolites are not contained in CSF or, if present, do not contribute to the MRS signal, and (2) a more recently proposed tissue correction, the α correction that accounts for the different concentrations of GABA and Glx in the gray and white matter in addition to accounting for differential tissue relaxations and signal constants to better address the underlying dependency of metabolites on the tissue composition of the voxel.21 These corrections involved the generation of a binary mask of the MRS voxel created with the same imaging matrix as the T1-weighted anatomical image, using Gannet tool GannetCoRegister,21 integrated voxel-to-image co-registration. Segmentation of the anatomical image was performed using GannetSegment21 and Segment in SPM12,2 which was used to obtain average coordinates of the center of ROIs in [X,Y,Z] format (prefrontal: [0.98, 47.96, 1.89] and S1/M1: [31.06, −19.53, 46.1]).

Figure 1.
Figure 1.:
Representative voxel placement from a single subject is superimposed in white, in both prefrontal (a) and sensorimotor (b) regions.
Figure 2.
Figure 2.:
Representative spectrum for data quality purposes obtained from the Gannet software. The edited spectrum is shown in blue and overlaid in red is the model of best fit using a simple Gaussian model. Below the plot, the residual between these two is shown in black.
Table 1 - Tissue composition in the voxels across participants.
Younger controls Older controls Older chronic pain P
Prefrontal voxel n = 24 n = 12 n = 29
 Gray matter fraction 0.586 ± 0.03 0.489 ± 0.04 0.505 ± 0.04 0.000
 White matter fraction 0.288 ± 0.04 0.280 ± 0.04 0.287 ± 0.05 0.851
 Cerebrospinal fluid fraction 0.125 ± 0.03 0.231 ± 0.07 0.208 ± 0.05 0.000
Sensorimotor voxel n = 12 n = 5 n = 14
 Gray matter fraction 0.293 ± 0.05 0.216 ± 0.02 0.263 ± 0.03 0.003
 White matter fraction 0.640 ± 0.06 0.644 ± 0.06 0.604 ± 0.05 0.234
 Cerebrospinal fluid fraction 0.067 ± 0.02 0.140 ± 0.04 0.133 ± 0.05 0.000
Bold entries represent probability values less than 0.05.

Table 2 - Paired correlations between metabolite levels at the prefrontal and the sensorimotor voxels in a subset of participants (n = 31).
Paired correlation Probability
CSF-corrected GABA 0.026 0.894
Tissue-corrected GABA −0.058 0.769
CSF-corrected Glx 0.380 0.046
Tissue-corrected Glx 0.368 0.054
GABA, gamma-aminobutyric acid; Glx, glutamate and glutamine.

2.3. Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics version 27. Descriptive statistics were examined across groups using t-tests for continuous variables and the χ2 test for nominal variables. Factorial analyses of covariance were used to determine differences in metabolite levels between groups with sex as a secondary between-subject factor while accounting for race. Partial correlation analyses controlling for age, sex, and race were used to assess associations between brain metabolite levels with clinical and experimental variables. A probability less than 0.05 was set a priori as statistically significant with a false discovery rate correction using Benjamini–Hochberg probabilities. We also evaluated significance using 95% bootstrapped confidence intervals based on 5000 bootstrap samples in all procedures. Partial eta squared was also reported to assess the magnitude of the differences where small, medium, and large effect sizes are represented by 0.01, 0.06, and 0.14, respectively.7

3. Results

A total of 186 younger and older individuals were screened with 84 participants being eligible for neuroimaging and undergoing the MRS MEGA-PRESS sequence acquisition. A number of MEGA-PRESS sequences failed preprocessing quality control measures in the prefrontal (n = 17) and sensorimotor (n = 23) regions. Of these 67 participants, 2 older adults (1 reporting chronic pain and 1 not reporting pain) were excluded as their GABA levels were significantly greater than 3 SDs from the sample mean, whereas 1 younger adult reported a chronic pain condition (ie, migraine headaches). Thus, this study includes 64 participants with high-quality MRS data in a prefrontal voxel. In a subset (n = 31), data from a second voxel in the sensorimotor cortex were also obtained. Participants were between the ages of 19 and 94 years and were mostly non-Hispanic White (80.3%) and female (60.7%). Most of our older adults (65.9%) reported chronic musculoskeletal pain during the past 3 months with 2 pain problems on average that were most commonly reported in the legs or knees (85.2%) and back (55.6%) regions. Approximately 31% of participants (n = 9) reported their pain to be constant or continuous while 48% of participants (n = 14) reported their pain to be intermittent and 20% reported their pain had no predictable pattern (n = 6). Summary demographic and clinical characteristics are presented in Table 3.

Table 3 - Differences in demographic and clinical characteristics between the groups.
Younger controls (n = 23) Older controls (n = 12) Older chronic pain (n = 29) P
Age, mean ± SD 21.7 ± 1.8 73.7 ± 6.7 72.0 ± 7.1 0.001*
Sex, no. 0.458
Male 11 5 9
Female 12 7 20
Race, no. 0.020
 Caucasian 13 12 26
 African American 1 0 2
 Asian 1 0 0
 Pacific Islander 1 0 1
 Hispanic 7 0 0
Education, mean ± SD 14.9 ± 1.5 16.4 ± 2.9 15.4 ± 2.6 0.187*
CES-D, mean ± SD 7.6 ± 6.1 7.4 ± 3.3 7.9 ± 5.6 0.969*
No. of anatomical pain sites, mean ± SD 4.1 ± 2.3
Current pain at MRI (0–100), mean ± SD 0.9 ± 4.1 0.9 ± 2.3 13.6 ± 16.0 0.001*
Ave. pain in the past 3 months (0–10), mean ± SD 0 ± 0 0 ± 0 4.8 ± 2.1 0.001*
Pain days in past week (0–7), mean ± SD 5.4 ± 2.2
Pain duration, mean ± SD 10.5 ± 2.1
Narcotic (PRN), no. (%) 5 (17.2)
NSAID/Acetaminophen, no. (%) 16 (55.2)
Nonpharmacological interventions, no. (%) 18 (62.1)
*Analysis of variance tests. Bold entries represent probability values less than 0.05.
χ2 tests.
CES-D, Center for Epidemiologic Studies Depression Scale; MRI, magnetic resonance imaging; NSAID, nonsteroidal anti-inflammatory drugs; PRN, as needed.

3.1. Age differences in gamma-aminobutyric acid and glutamate and glutamine metabolite levels

3.1.1. Prefrontal CSF-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 64)

There was a nonsignificant main effect of Age (F(1,59) = 1.3, P = 0.253, effect size = 0.022) and Age by Sex interaction (F(1,59) = 2.1, P = 0.155, effect size = 0.034) in prefrontal GABA. Similarly, there was a nonsignificant main effect of Age (F(1,59) = 1.1, P = 0.304, effect size = 0.109) and Age by Sex interaction (F(1,59) = 0.3, P = 0.609, effect size = 0.005) in prefrontal Glx.

3.1.2. Prefrontal tissue-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 64)

There was a nonsignificant main effect of Age (F(1,59) = 0.8, P = 0.369, effect size = 0.014) and Age by Sex interaction (F(1,59) = 2.1, P = 0.150, effect size = 0.035) in prefrontal GABA. Similarly, there was a nonsignificant main effect of Age (F(1,59) = 0.6, P = 0.431, effect size = 0.011) and Age by Sex interaction (F(1,59) = 0.3, P = 0.599, effect size = 0.005) in prefrontal Glx.

3.1.3. Sensorimotor CSF-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 31)

There was a significant main effect of Age (F(1,26) = 13.4, P = 0.001, effect size = 0.340) where older adults (n = 19) had lower sensorimotor GABA (M = 2.49, SEM = 0.05) compared with younger controls (n = 12, M = 2.78, SEM = 0.06, bootstrapped P = 0.002). There was no Age by Sex interaction in sensorimotor GABA (F(1,26) = 0.4, P = 0.540, effect size = 0.015). There was a significant main effect of Age group (F(1,26) = 5.4, P = 0.028, effect size = 0.172) where older adults (n = 19) had lower sensorimotor Glx (M = 3.9, SEM = 0.1) compared with younger controls (n = 12, M = 4.4, SEM = 0.2, bootstrapped P = 0.024, Fig. 3). There was no Age by Sex interaction in sensorimotor Glx (F(1,26) = 0.3, P = 0.561, effect size = 0.013).

Figure 3.
Figure 3.:
Differences in sensorimotor GABA and Glx metabolite levels between older and younger participants. *Statistically significant post hoc comparisons at a probability less than 0.05. GABA, gamma-aminobutyric acid

3.1.4. Sensorimotor tissue-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 31)

There was a significant main effect of Age (F(1,26) = 6.4, P = 0.018, effect size = 0.197) where older adults (n = 19) had lower sensorimotor GABA (M = 2.70, SEM = 0.05) compared with younger controls (n = 12, M = 2.93, SEM = 0.07, bootstrapped P = 0.037, Fig. 3). There was no Age by Sex interaction in sensorimotor GABA (F(1,26) = 0.5, P = 0.470, effect size = 0.023). There was a nonsignificant main effect of Age (F(1,26) = 3.1, P = 0.088, effect size = 0.108) and Age by Sex interaction in sensorimotor Glx (F(1,26) = 0.2, P = 0.649, effect size = 0.008).

3.2. Age and pain differences in gamma-aminobutyric acid and glutamate and glutamine metabolite levels

3.2.1. Prefrontal CSF-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 64)

There was a significant main effect of Age-Pain (ie, comparing the 3 groups: younger controls, older controls, and older adults reporting pain) (F(2,57) = 3.2, P = 0.045, effect size = 0.103). Older individuals with chronic pain had significantly lower prefrontal GABA (M = 1.9, SEM = 0.2) compared with older controls (M = 2.6, SEM = 0.2, bootstrapped P = 0.027, Fig. 4). There was a trend for Age-Pain by Sex interaction (F(2,57) = 3.1, P = 0.050, effect size = 0.100, Figure 4) in prefrontal GABA. There was a nonsignificant main effect of Age-Pain group (F(2,57) = 0.5, P = 0.583, effect size = 0.019) or Age-Pain by Sex interaction (F(2,57) = 0.2, P = 0.807, effect size = 0.008) in prefrontal Glx.

Figure 4.
Figure 4.:
Adjusted differences in CSF-corrected and tissue-corrected prefrontal GABA metabolite levels between the groups and stratified by sex. *Statistically significant post hoc comparisons at a probability less than or equal to 0.05. GABA, gamma-aminobutyric acid

3.2.2. Prefrontal tissue-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 64)

There was a significant main effect of Age-Pain (F(2,57) = 3.2, P = 0.049, effect size = 0.101). Older individuals with chronic pain had significantly lower prefrontal GABA (M = 2.2, SEM = 0.2) compared with older controls (M = 3.0, SEM = 0.3, bootstrapped P = 0.021, Fig. 4). There was a significant Age-Pain by Sex interaction (F(2,57) = 3.2, P = 0.039, effect size = 0.108) in prefrontal GABA where older males with chronic pain had significantly lower prefrontal GABA (M = 1.8, SEM = 0.3) compared with older (M = 3.3, SEM = 0.4, bootstrapped P = 0.022) and younger male controls (M = 3.0, SEM = 0.3, bootstrapped P = 0.028, Fig. 4). There was a nonsignificant main effect of Age-Pain (F(2,57) = 0.3, P = 0.711, effect size = 0.012) or Age-Pain by Sex interaction (F(2,57) = 0.3, P = 0.768, effect size = 0.010) in prefrontal Glx.

3.2.3. Sensorimotor CSF-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 31)

There was a significant main effect of Age-Pain (F(2,24) = 16.8, P < 0.001, effect size = 0.583) where younger controls had significantly higher sensorimotor GABA (M = 2.8, SEM = 0.1) compared to older controls (M = 2.3, SEM = 0.1, bootstrapped P = 0.001) and older participants with chronic pain (M = 2.6, SEM = 0.1, bootstrapped P = 0.020, Fig. 5). There was not a significant Age-Pain by Sex interaction (F(2,24) = 2.3, P = 0.120, effect size = 0.162) in sensorimotor GABA. There was a nonsignificant main effect of Age-Pain (F(2, 24) = 2.6, P = 0.098, effect size = 0.176) or Age-Pain by Sex interaction (F(2,24) = 0.2, P = 0.794, effect size = 0.019) in sensorimotor Glx.

Figure 5.
Figure 5.:
Adjusted differences in CSF-corrected and tissue-corrected sensorimotor GABA metabolite levels between the groups. * Statistically significant post hoc comparisons at a probability less than 0.05. There was no statistically significant Pain by Sex interactions (p's > 0.05). GABA, gamma-aminobutyric acid

3.2.4. Sensorimotor tissue-corrected gamma-aminobutyric acid and glutamate and glutamine (n = 31)

There was a significant main effect of Age-Pain (F(2,24) = 7.4, P = 0.004, effect size = 0.414) where older controls had significantly lower sensorimotor GABA (M = 2.4, SEM = 0.1) compared with younger controls (M = 2.9, SEM = 0.1, bootstrapped P = 0.008, Fig. 5). There was not a significant Age-Pain by Sex interaction (F(2,24) = 0.6, P = 0.576, effect size = 0.051) in sensorimotor GABA. There was a nonsignificant main effect of Age-Pain (F(2,24) = 1.5, P = 0.254, effect size = 0.122) or Age-Pain by Sex interaction (F(2,24) = 0.5, P = 0.596, effect size = 0.048) in sensorimotor Glx.

3.3. Associations between prefrontal gamma-aminobutyric acid and glutamate and glutamine levels with clinical and experimental pain in older adults

Experimental pain thresholds by study group and test site are presented in Figure 6. Post hoc comparisons revealed older individuals with chronic pain had significantly lower pressure pain thresholds in the quadriceps (M = 426.1, SEM = 43.9) compared with young controls (M = 666.5, SEM = 63.3, bootstrapped P = 0.006, Fig. 6). Prefrontal GABA metabolite levels were inversely correlated with self-reported pain variables surviving both multiple comparison corrections as well as bootstrapping procedures controlling for race, sex, and age of the participants. Similarly, heat and cold pain thresholds were inversely associated with prefrontal GABA levels and these associations were supported by bootstrapping procedures (Fig. 7). However, cold pain thresholds did not survive multiple comparison corrections. Prefrontal GABA levels were not associated with pressure pain thresholds. Prefrontal Glx levels were not significantly associated with any of the variables of interest. Findings are summarized in Table 4. Owing to the small number of participants who had data in the sensorimotor voxel and self-reported and experimental pain measures, we did not examine associations between these variables.

Figure 6.
Figure 6.:
Experimental pain thresholds by study group and test site where only pressure pain thresholds in the quadriceps survived post hoc corrections (P < 0.05). *Statistically significant post hoc comparison at a probability less than 0.05.
Figure 7.
Figure 7.:
Adjusted associations between tissue-corrected prefrontal GABA levels with clinical and experimental pain variables (QST variables as z-scores) controlling for age, sex, and race. GABA, gamma-aminobutyric acid; QST, quantitative sensory testing.
Table 4 - Partial correlation coefficients assessing associations between prefrontal gamma-aminobutyric acid and glutamate and glutamine metabolite levels with clinical and experimental pain measures in older participants controlling for race, age, and sex.
r, P Benjamini–Hochberg
P
Bootstrapped confidence intervals*
Pain at MRI
 GABA r = -0.407, P = 0.011 0.045 −0.679, -0.044
 Glx r = −0.041, P = 0.814 0.938 −0.315, 0.324
Pain frequency
 GABA r = -0.392, P = 0.015 0.045 −0.642, -0.078
 Glx r = 0.020, P = 0.908 0.938 −0.312, 0.462
Pain duration
 GABA r = -0.415, P = 0.010 0.045 −0.722, -0.030
 Glx r = 0.021, P = 0.905 0.938 −0.255, 0.528
No.of pain sites
 GABA r = -0.490, P = 0.002 0.028 −0.711, -0.170
 Glx r = 0.019, P = 0.910 0.938 −0.157, 0.296
Heat pain thresholds
 GABA r = -0.451, P = 0.016 0.045 0.710, 0.137
 Glx r = −0.121, P = 0.549 0.938 −0.476, 0.258
Cold pain thresholds
 GABA r = -0.408, P = 0.043 0.100 0.127, 0.714
 Glx r = −0.017, P = 0.938 0.938 −0.332, 0.420
Pressure pain thresholds
 GABA r = 0.088, P = 0.632 0.938 −0.382, 0.473
 Glx r = −0.093, P = 0.625 0.938 −0.431, 0.343
*Results are based on 5000 bootstrap samples. Bold entries represent probability values less than 0.05.
GABA, gamma-aminobutyric acid; Glx, glutamate and glutamine; MRI, magnetic resonance imaging.

4. Discussion

This study aimed to determine self-reported and experimental pain-related associations in brain levels of GABA and Glx, reflective of inhibitory and excitatory tone in community-dwelling older adults. Several key findings emerged. First, older adults with chronic pain had significantly lower prefrontal GABA levels that were driven, in part, by older males with chronic pain having lower levels of prefrontal GABA compared with younger and older male controls. Second, prefrontal, but not sensorimotor, GABA levels were associated with greater self-reported pain, greater pain frequency and duration, greater number of pain sites, and greater thermal pain sensitivity. Finally, prefrontal Glx levels were not associated with any self-reported or experimental pain characteristics in our older sample. Our findings suggest a key imbalance among inhibitory systems that may contribute to the increased prevalence of chronic pain in older individuals.

Our investigation furthers our understanding of the GABAergic and glutamatergic systems in the context of pain and aging brain processes. To the best of our knowledge, this is the first study to simultaneously examine clinical and experimental pain and their associations with GABA and Glx levels in a community-dwelling sample. Glutamate and GABA systems are key signal transducers in pain-processing pathways and are involved in various pain-gating mechanisms.4,15,23,27,29,52,53 Lower GABA levels in our older adults with chronic musculoskeletal pain are consistent with the existing literature using 1H-MRS where reduced cerebral GABA levels have been reported in fibromyalgia13 and chronic pelvic pain.19 Specifically, it seems that our differences were driven, in part, by the older male participants with chronic pain. To the best of our knowledge, such pain by sex differences have not been previously reported in humans, and multiple mechanisms may account for these findings. For example, lower GABAAR α3 subunit expression has been reported in older vs younger males that could account for GABAergic system dysfunction.37 Indeed, work in mice expressing genetically engineered GABAA receptor α subunits has shown that α2 and α3 GABAA receptors are the most relevant GABAA receptor subtypes for spinal analgesia.27,44 Future studies are needed to replicate the sex by pain interactions in the GABAergic system in older adults.

Moreover, lower GABA levels were associated with greater self-reported pain, greater pain frequency and duration, greater number of reported pain sites, and greater thermal pain sensitivity. These findings are consistent with previous research where GABA levels correlate with clinical pain severity45 and experimental pain sensitivity.49 In healthy younger individuals, acute experimental pain stimulation has been found to change GABA levels in various brain regions.6,28 Specifically, administration of painful stimulation momentarily decreased GABA levels in the anterior cingulate and occipital cortex,6 and tonic heat pain stimulation increased GABA levels in the prefrontal cortex. Although the above studies link cerebral GABA levels with chronic and acute pain processing, the directionality of the findings is less consistent. However, our findings are in sharp contrast to findings in persons with migraine1 and previous studies reporting associations between Glx concentrations and pain characteristics.46,52 This is likely due to the different brain regions investigated, different MRS acquisition and analyses parameters, and the age of the included participants. Most of the studies mentioned above have only included healthy younger individuals or a mixture of younger and older adults (except45 in knee osteoarthritis), which could account for these discrepancies because there are significant age-related changes in brain metabolite levels.3,14,17,39 In addition, we used a large prefrontal ROI covering several anatomically distinct brain areas, which are involved not only in descending inhibition but also in cognitive modulation through inhibitory GABAergic neurons. Furthermore, previous basic research showed hyperactivity in the amygdala leads to deactivation of prefrontal regions24 through GABAergic mechanisms emphasizing the need for studies in animal models.25 Future larger studies in humans must also differentiate between pain conditions in older vs younger populations and include appropriate age-control groups to examine metabolite levels across various regions.10

Our study is limited by its small sample size, particularly with respect to older controls. This is not surprising given the high prevalence of chronic pain in older individuals, which further highlights the need to identify and study older adults without any chronic pain. Thus, future investigations in older adults should identify these true “control” groups. Second, the source of the Glx and GABA levels measured cannot be ascertained using 1H-MRS. Glx levels reflect the combination of glutamate (Glu) and glutamine (Gln) levels (ie, Glx = Glu + Gln). Thus, we cannot determine whether the Gln and Glu levels are in opposing directions or evaluate their levels separately. On the other hand, the GABA measured using 1H-MRS is believed to represent mostly GABA linked to the “tonic” or GABAergic inhibitory tone of a localized brain region.43,47 Nonetheless, the 1H-MRS signal arises from the gray and white matter and is an ensemble average of multiple different cell types, including astrocytes, glia, and neurons and has been reported across a number of domains to be associated with increased inhibitory capacity.54 Therefore, the metabolite signal might be at regions distant from the synapse. Animal studies are needed to determine the exact source and mechanism underlying the association of pain with these metabolites. Third, this is a cross-sectional study and no causal inferences can be made. It is not known whether younger and older individuals have predisposing metabolite differences or whether the experience of pain or aging leads to such metabolite imbalances. Fourth, our older sample may not be representative of the general aging population because it included mostly Caucasian community-dwelling older individuals, who were cognitively intact and generally healthy. Longitudinal and basic studies including larger, more diverse samples with more severe pain and disability are needed to answer these questions in the context of an aging nervous system. Fifth, it is possible that our findings are driven by differences in momentary pain ratings, and future studies that test individuals with chronic pain when pain-free are needed to determine these associations. Furthermore, future studies should be designed to test whether medications taken by the study participants could drive potential metabolite differences, although our study participants were not taking medications affecting the GABAergic systems. Finally, given the small sample collected in the sensorimotor cortex, these findings, including the null findings, should be considered preliminary and require replication.

Despite these limitations, our study is the first to focus exclusively on age and pain differences including younger and older control cohorts, which is important to try to elucidate independent aging and pain contributions to brain GABAergic and glutamatergic processes. The present work includes 2 key neurochemicals to the forefront of our thinking regarding the molecular processes involved in chronic pain in older individuals. Our study supports prefrontal cortex GABA as a potential marker of pain severity and processing in chronic musculoskeletal pain states independent of age-related decline in these neurochemical systems. These findings may support the idea that GABAergic disinhibition in pain-processing brain networks may contribute to pain development and chronification in older individuals.29 Finally, these data may also indicate that pharmacologic interventions that specifically target GABAergic neurotransmission may be an effective target in these older individuals with chronic musculoskeletal pain and lower brain GABA levels.

Disclosures

The authors have no conflicts of interest to declare.

Acknowledgements

The authors are grateful to our study volunteers for their participation and the NEPAL study team (Paige Lysne, Lorraine Hoyos, Darlin Ramirez, Brandon Apagueno, and Rachna Sannegowda). This work was supported by the National Institutes of Health (NIA K01AG048259, R01AG059809, and R01AG067757 to Y.C.A.; NIAAA K01AA025306 to E.P.; and NIA K01AG050707 to A.J.W.), the University of Florida Clinical Translational Sciences Institute (NCATSUL1TR001427), the Cognitive Aging and Memory Clinical Translational Program, McKnight Brain Foundation, and the University of Florida Claude D. Pepper Older Americans Independence Center (P30AG028740). This work applies tools developed under NIH R01EB016089, R01EB023693, and P41EB015909.

References

[1]. Aguila MER, Lagopoulos J, Leaver AM, Rebbeck T, Hübscher M, Brennan PC, Refshauge KM. Elevated levels of GABA+ in migraine detected using1H-MRS. NMR Biomed 2015;28:890–7.
[2]. Ashburner J, Friston KJ. Unified segmentation. Neuroimage 2005;26:839–51.
[3]. Bañuelos C, Sofia Beas B, McQuail JA, Gilbert RJ, Frazier CJ, Setlow B, Bizon JL. Prefrontal cortical GABAergic dysfunction contributes to age-related working memory impairment. J Neurosci 2014;34:3457–66.
[4]. Bleakman D, Alt A, Nisenbaum ES. Glutamate receptors and pain. Semin Cell Dev Biol 2006;17:592–604.
[5]. Cardoso J, Apagueno B, Lysne P, Hoyos L, Porges E, Riley JL, Fillingim RB, Woods AJ, Cohen R, Cruz-Almeida Y. Pain and the montreal cognitive assessment (MoCA) in aging. Pain Med 2021;22:1776–83.
[6]. Cleve M, Gussew A, Reichenbach JR. In vivo detection of acute pain-induced changes of GABA+ and Glx in the human brain by using functional1H MEGA-PRESS MR spectroscopy. Neuroimage 2015;105:67–75.
[7]. Cohen J. Statistical power analysis for the behavioral sciences, 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers 1988. p. 110.
[8]. Cruz-Almeida Y, Cole J. Pain, aging, and the brain: new pieces to a complex puzzle. PAIN 2020;161:461–3.
[9]. Cruz-Almeida Y, Fillingim RB, Riley JL, Woods AJ, Porges E, Cohen R, Cole J. Chronic pain is associated with a brain aging biomarker in community-dwelling older adults. PAIN 2019;160:1119–30.
[10]. Cruz-Almeida Y, Porges E. Additional considerations for studying brain metabolite levels across pain conditions using proton magnetic resonance spectroscopy. Neuroimage 2021;224:117392.
[11]. Edden RAE, Barker PB. Spatial effects in the detection of gamma-aminobutyric acid: improved sensitivity at high fields using inner volume saturation. Magn Reson Med 2007;58:1276–82.
[12]. Edden RAE, Puts NAJ, Harris AD, Barker PB, Evans CJ. Gannet: a batch-processing tool for the quantitative analysis of gamma-aminobutyric acid-edited MR spectroscopy spectra. J Magn Reson Imaging 2014;40:1445–52.
[13]. Foerster BR, Petrou M, Edden RAE, Sundgren PC, Schmidt-Wilcke T, Lowe SE, Harte SE, Clauw DJ, Harris RE. Reduced insular γ-aminobutyric acid in fibromyalgia. Arthritis Rheum 2012;64:579–83.
[14]. Gao F, Edden RAE, Li M, Puts NAJ, Wang G, Liu C, Zhao B, Wang H, Bai X, Zhao C, Wang X, Barker PB. Edited magnetic resonance spectroscopy detects an age-related decline in brain GABA levels. Neuroimage 2013;78:75–82.
[15]. Gazerani P, Wang K, Cairns BE, Svensson P, Arendt-Nielsen L. Effects of subcutaneous administration of glutamate on pain, sensitization and vasomotor responses in healthy men and women. PAIN 2006;124:338–48.
[16]. Godlewska BR, Near J, Cowen PJ. Neurochemistry of major depression: a study using magnetic resonance spectroscopy. Psychopharmacology (Berl) 2015;232:501–7.
[17]. Grachev ID, Vania Apkarian A. Aging alters regional multichemical profile of the human brain: an in vivo 1H-MRS study of young versus middle-aged subjects. J Neurochem 2001;76:582–93.
[18]. Gullickson T. Review of statistical methods in education and psychology. Contemp Psychol A J Rev 1996;41:1224.
[19]. Harper DE, Ichesco E, Schrepf A, Halvorson M, Puiu T, Clauw DJ, Harris RE, Harte SE. Relationships between brain metabolite levels, functional connectivity, and negative mood in urologic chronic pelvic pain syndrome patients compared to controls: a MAPP research network study. Neuroimage Clin 2017;17:570–8.
[20]. Harris AD, Puts NAJ, Barker PB, Edden RAE. Spectral-editing measurements of GABA in the human brain with and without macromolecule suppression. Magn Reson Med 2015;74:1523–9.
[21]. Harris AD, Puts NAJ, Edden RAE. Tissue correction for GABA-edited MRS: considerations of voxel composition, tissue segmentation, and tissue relaxations. J Magn Reson Imaging 2015;42:1431–40.
[22]. Harris RE, Clauw DJ. Imaging central neurochemical alterations in chronic pain with proton magnetic resonance spectroscopy. Neurosci Lett 2012;520:192–6.
[23]. Jasmin L, Rabkin SD, Granato A, Boudah A, Ohara PT. Analgesia and hyperalgesia from GABA-mediated modulation of the cerebral cortex. Nature 2003;424:316–20.
[24]. Ji G, Sun H, Fu Y, Li Z, Pais-Vieira M, Galhardo V, Neugebauer V. Cognitive impairment in pain through amygdala-driven prefrontal cortical deactivation. J Neurosci 2010;30:5451–64.
[25]. Ji G, Neugebauer V. Pain-related deactivation of medial prefrontal cortical neurons involves mGluR1 and GABA(A) receptors. J Neurophysiol 2011;106:2642–52.
[26]. Killiany RJ, Gomez-Isla T, Moss M, Kikinis R, Sandor T, Jolesz F, Tanzi R, Jones K, Hyman BT, Albert MS. Use of structural magnetic resonance imaging to predict who will get Alzheimers disease. Ann Neurol 2000;47:430–9.
[27]. Knabl J, Witschi R, Hösl K, Reinold H, Zeilhofer UB, Ahmadi S, Brockhaus J, Sergejeva M, Hess A, Brune K, Fritschy JM, Rudolph U, Möhler H, Zeilhofer HU. Reversal of pathological pain through specific spinal GABAAreceptor subtypes. Nature 2008;451:330–4.
[28]. Kupers R, Danielsen ER, Kehlet H, Christensen R, Thomsen C. Painful tonic heat stimulation induces GABA accumulation in the prefrontal cortex in man. PAIN 2009;142:89–93.
[29]. Lau BK, Vaughan CW. Descending modulation of pain: the GABA disinhibition hypothesis of analgesia. Curr Opin Neurobiol 2014;29:159–64.
[30]. Li Q, Chen C, Gong T. High-field MRS study of GABA+ in patients with migraine: response to levetiracetam treatment. Neuroreport 2018;29:1007–10.
[31]. Lysne P, Cohen R, Hoyos L, Fillingim RB, Riley JL, Cruz-Almeida Y. Age and pain differences in non-verbal fluency performance: associations with cortical thickness and subcortical volumes. Exp Gerontol 2019;126:110708.
[32]. Mescher M, Merkle H, Kirsch J, Garwood M, Gruetter R. Simultaneous in vivo spectral editing and water suppression. NMR Biomed 1998;11:266–72.
[33]. Mescher M, Tannus A, Johnson N, Garwood M. Solvent suppression using selective echo dephasing. J Magn Reson Ser A, 123, 1996, 226–9, ISSN 1064-1858.
[34]. Mikkelsen M, Barker PB, Bhattacharyya PK, Brix MK, Buur PF, Cecil KM, Chan KL, Chen DYT, Craven AR, Cuypers K, Dacko M, Duncan NW, Dydak U, Edmondson DA, Ende G, Ersland L, Gao F, Greenhouse I, Harris AD, He N, Heba S, Hoggard N, Hsu TW, Jansen JFA, Kangarlu A, Lange T, Lebel RM, Li Y, Lin CYE, Liou JK, Lirng JF, Liu F, Ma R, Maes C, Moreno-Ortega M, Murray SO, Noah S, Noeske R, Noseworthy MD, Oeltzschner G, Prisciandaro JJ, Puts NAJ, Roberts TPL, Sack M, Sailasuta N, Saleh MG, Schallmo MP, Simard N, Swinnen SP, Tegenthoff M, Truong P, Wang G, Wilkinson ID, Wittsack HJ, Xu H, Yan F, Zhang C, Zipunnikov V, Zöllner HJ, Edden RAE. Big GABA: edited MR spectroscopy at 24 research sites. Neuroimage 2017;159:32–45.
[35]. Mullins PG, McGonigle DJ, O'Gorman RL, Puts NAJ, Vidyasagar R, Evans CJ, Edden RAE, Brookes MJ, Garcia A, Foerster BR, Petrou M, Price D, Solanky BS, Violante IR, Williams S, Wilson M. Current practice in the use of MEGA-PRESS spectroscopy for the detection of GABA. Neuroimage 2014;86:43–52.
[36]. Near J, Edden R, Evans CJ, Paquin R, Harris A, Jezzard P. Frequency and phase drift correction of magnetic resonance spectroscopy data by spectral registration in the time domain. Magn Reson Med 2015;73:44–50.
[37]. Pandya M, Palpagama TH, Turner C, Waldvogel HJ, Faull RL, Kwakowsky A. Sex- and age-related changes in GABA signaling components in the human cortex. Biol Sex Differ 2019;10:5.
[38]. Patel Kv, Guralnik JM, Dansie EJ, Turk DC. Prevalence and impact of pain among older adults in the United States: findings from the 2011 national health and aging trends study. PAIN 2013;154:2649–57.
[39]. Porges EC, Woods AJ, Edden RAE, Puts NAJ, Harris AD, Chen H, Garcia AM, Seider TR, Lamb DG, Williamson JB, Cohen RA. Frontal gamma-aminobutyric acid concentrations are associated with cognitive performance in older adults. Biol Psychiatry Cogn Neurosci Neuroimaging 2017;2:38–44.
[40]. Puts NAJ, Wodka EL, Harris AD, Crocetti D, Tommerdahl M, Mostofsky SH, Edden RAE. Reduced GABA and altered somatosensory function in children with autism spectrum disorder. Autism Res 2017;10:608–19.
[41]. Puts N, Edden R. In vivo magnetic spectroscopy of GABA: a methodological review. Prog Nucl Magn Reson Spectrosc 2012;60:29–41.
[42]. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385–401.
[43]. Rae CD. A guide to the metabolic pathways and function of metabolites observed in human brain 1H magnetic resonance spectra. Neurochem Res 2014;39:1–36.
[44]. Ralvenius WT, Benke D, Acuña MA, Rudolph U, Zeilhofer HU. Analgesia and unwanted benzodiazepine effects in point-mutated mice expressing only one benzodiazepine-sensitive GABAA receptor subtype. Nat Commun 2015;6:6803.
[45]. Reckziegel D, Raschke F, Cottam WJ, Auer DP. Cingulate GABA levels inversely correlate with the intensity of ongoing chronic knee osteoarthritis pain. Mol Pain 2016;12:1744806916650690.
[46]. Sharma NK, McCarson K, van Dillen L, Lentz A, Khan T, Cirstea CM. Primary somatosensory cortex in chronic low back pain: a H-MRS study. J Pain Res 2011;4:143–50.
[47]. Stagg CJ, Bachtiar V, Johansen-Berg H. What are we measuring with GABA Magnetic Resonance Spectroscopy? Commun Integr Biol 2011;4:573–5.
[48]. Teng EL, Chui HC. The Modified Mini-Mental State (MMS) examination. J Clin Psychiatry 1987;48:314–8.
[49]. Thiaucourt M, Shabes P, Schloss N, Sack M, Baumgärtner U, Schmahl C, Ende G. Posterior insular GABA levels inversely correlate with the intensity of experimental mechanical pain in healthy subjects. Neuroscience 2018;387:116–22.
[50]. Treede RD, Rief W, Barke A, Aziz Q, Bennett MI, Benoliel R, Cohen M, Evers S, Finnerup NB, First MB, Giamberardino MA, Kaasa S, Korwisi B, Kosek E, Lavand'homme P, Nicholas M, Perrot S, Scholz J, Schug S, Smith BH, Svensson P, Vlaeyen JWS, Wang SJ. Chronic pain as a symptom or a disease: the IASP classification of chronic pain for the international classification of diseases (ICD-11). PAIN 2019;160:19–27.
[51]. van Veenendaal TM, Backes WH, van Bussel FCG, Edden RAE, Puts NAJ, Aldenkamp AP, Jansen JFA. Glutamate quantification by PRESS or MEGA-PRESS: validation, repeatability, and concordance. Magn Reson Imaging 2018;48:107–14.
[52]. Widerström-Noga E, Cruz-Almeida Y, Felix ER, Pattany PM. Somatosensory phenotype is associated with thalamic metabolites and pain intensity after spinal cord injury. PAIN 2015;156:166–74.
[53]. Widerström-Noga E, Pattany PM, Cruz-Almeida Y, Felix ER, Perez S, Cardenas DD, Martinez-Arizala A. Metabolite concentrations in the anterior cingulate cortex predict high neuropathic pain impact after spinal cord injury. PAIN 2013;154:204–12.
[54]. Yoon JH, Maddock RJ, Rokem A, Silver MA, Minzenberg MJ, Ragland JD, Carter CS. GABA concentration is reduced in visual cortex in schizophrenia and correlates with orientation-specific surround suppression. J Neurosci 2010;30:3777–81.
[55]. Yousry TA, Schmid UD, Alkadhi H, Schmidt D, Peraud A, Buettner A, Winkler P. Localization of the motor hand area to a knob on the precentral gyrus. A new landmark. Brain 1997;120:141–57.
Keywords:

GABA; brain; MRS; musculoskeletal pain; aging

Copyright © 2021 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The International Association for the Study of Pain.