Alterations of Cerebral Metabolites Are Associated with Frequency of Vasoocclusive Crises and Pain Sensitivity in Patients with Sickle Cell Disease

Purpose

Sickle cell disease (SCD) is an inherited hemolytic disorder complicated by acute and chronic pain. Acute vasoocclusive crises (VOCs) associated with SCD are extremely painful episodes that are recurrent, unpredictable, and frequently require hospitalization and opioids for pain control. Changes in brain metabolites were identified in other chronic pain populations but haven’t been well studied in SCD. Our ongoing trial (ClinicalTrials.gov Identifier: NCT05045820) aims to evaluate the characteristics of brain metabolites in SCD and understand the relationship between the altered cerebral metabolite(s) and clinical pain in patients with SCD.

Materials and Methods

Metabolite measurements from a single voxel (size: 2×3×3 cm³) located within the posterior insula cortex (PIC) were acquired from 26 patients with SCD (31.9 ± 13.9 years, 14-73; 14 females) and 19 healthy controls (HCs) (37.9 ± 16.4 years, 17-62; 10 females) with PRESS (TR/TE = 3000/25 ms, Averages =128) and MEGA-PRESS (TR/TE = 3000/68 ms, Averages =256) sequences using a Siemens Prisma 3T scanner. Written consent was obtained from each participant before scanning. LCModel was used for PRESS analysis to obtain concentration values of glutamine + glutamate (Glx), total creatine (tCr), N-acetylaspartate (NAA), total choline (tCho), and myoinositol (MI). Gannet was used for processing MEGA-PRESS data to estimate GABA+. Ratios of each metabolite to tCr were also quantified. Tissue correction for cerebrospinal fluid was performed for each voxel using the tissue composition estimated from SPM12. The Cramer-Rao Lower Bounds of each metabolite analyzed with LCModel was less than 20%. The frequency of VOCs in the preceding 12 months was recorded for each participant. A moderate pressure pain stimuli (using a 0-100 scale with “0” representing “no pain” and “100” representing “most intense pain imaginable”, pain rating at 40-60 abbreviated as P40 is considered as moderate pain) was delivered via a pressure cuff to the gastrocnemius area of the non-dominant leg using a validated MR-compatible air compressor to induce a tonic pressure pain for 8 min right before MRS scan. Higher pressure intensity (mmHg) of P40 indicated higher pain tolerance and lower pain sensitivity. Correlation between the brain metabolites and pain-related outcomes was studied.

Results

Due to the large range of age in the SCD group, the brain metabolite profile was compared between 17 SCD patients and 17 ethnicity-, age- and gender-matched HCs. The concentration of Glx (p =0.012) and ratio of Glx to tCr (p =0.034) in PIC were significantly higher in patients with SCD compared to HCs (Figure 1). Within the SCD patient group, there was a significant positive correlation between frequency of VOCs in the preceding 12 months and Glx/tCr (r= 0.39, p=0.047, Spearman’s), a significant negative correlation between P40 and Glx (r= -0.46, p=0.019, Spearman’s). There were no group differences in any other metabolites.

Conclusion

We find increased Glx in the PIC in patients with SCD. Previous studies have also found increased Glx and/ or Glx/tCr in the PIC as well as other brain regions including the amygdala, the posterior cingulate, the ventral lateral prefrontal cortex and the anterior insula in fibromyolgia. The PIC is thought to be involved in discriminative activities of sensory pain, which suggests that Glx in PIC is related to pain processing in patients with SCD. Glx/tCr was correlated with frequency of VOCs in patients with SCD, indicating that the SCD patients with higher level of Glx/tCr had more VOCs in the preceding 12 months. In addition, the SCD patients with higher level of Glx had lower pressure intensity of P40, suggesting that elevated Glx in the PIC was related to higher pain sensitivity. Our preliminary data provide the first evidence of a link between alterations in excitatory neurotransmitter levels in the PIC in patients with SCD and indicates new information of potential brain markers for pain in SCD, which may guide treatment.

Clinical Relevance Statement:

Treatment of pain in SCD remains very challenging in part due to a lack of understanding of the underlying mechanisms and pathophysiology of pain. Brain excitatory neurotransmitters and their relationship with pain have been largely unexplored in SCD. These findings indicate that the relationship between clinical outcomes and excitatory brain neurotransmitters may be a potential objective outcome measure in SCD.