Ultrasmall Superparamagnetic Iron Oxide Imaging Identifies Tissue and Nerve Inflammation in Pain Conditions

Abstract

Objective: Correlation between radiologic structural abnormalities and clinical symptoms in low back pain patients is poor. There is an unmet clinical need to image inflammation in pain conditions to aid diagnosis and guide treatment. Ferumoxytol, an ultrasmall superparamagnetic iron oxide (USPIO) nanoparticle, is clinically used to treat iron deficiency anemia and showed promise in imaging tissue inflammation in human. We explored whether ferumoxytol can be used to identify tissue and nerve inflammation in pain conditions in animals and humans.

Methods: Complete Freud's adjuvant (CFA) or saline was injected into mice hind paws to establish an inflammatory pain model. Ferumoxytol (20 mg/kg) was injected intravenously. Magnetic resonance imaging (MRI) was performed prior to injection and 72 hours postinjection. The changes in the transverse relaxation time (T2) before and after ferumoxytol injection were compared between mice that received CFA vs saline injection. In the human study, we administered ferumoxytol (4 mg/kg) to a human subject with clinical symptoms of lumbar radiculopathy and compared the patient with a healthy subject.

Results: Mice that received CFA exhibited tissue inflammation and pain behaviors. The changes in T2 before and after ferumoxytol injection were significantly higher in mice that received CFA vs saline (20.8 ± 3.6 vs 2.2 ± 2.5, P = 0.005). In the human study, ferumoxytol-enhanced MRI identified the nerve root corresponding to the patient's symptoms, but the nerve root was not impinged by structural abnormalities, suggesting the potential superiority of this approach over conventional structural imaging techniques.

Conclusions: Ferumoxytol-enhanced MRI can identify tissue and nerve inflammation and may provide a promising diagnostic tool in assessing pain conditions in humans.

Figure 1

Complete Freud’s adjuvant (CFA) induces inflammatory pain. (A, B) Mice were given CFA (N = 6) or saline control (N = 6) injection in the hind paws. (A) Hind paw mechanical withdrawal threshold was determined at indicated time points postinjection. CFA injection group exhibited significant mechanical pain, as suggested by decrease of withdrawal threshold. *P < 0.05. (B) Hind paw thickness was examined at indicated time points postinjection. CFA injection group exhibited significant paw swelling. *P < 0.05. (C, D) Inflammatory cell infiltration into hind paw. At day 3 after CFA or saline injection, hind paw tissues from six mice of each group were processed to isolate infiltrating cells, followed by flow cytometry staining on these cells. (C) Cells were stained with F4/80 and CD11b antibodies. Dot plots were gated on live cells based on forward and side scatter parameters. Each plot represents six stainings. (D) Cell counts for each group of animals. Numbers of CD11b⁺ F4/80⁺ macrophages and CD11b⁺ F4/80⁻ monocytes were compared between CFA group and saline control group. CFA group had significantly higher numbers of macrophage and monocyte infiltration. CFA = complete Freud’s adjuvant.

Figure 2

Ferumoxytol-contrasted magnetic resonance imaging (MRI) in inflammatory pain. (A, B) Mice were given complete Freud’s adjuvant (CFA; N = 4) or saline control (N = 4) injection in the hind paws. (A) Noncontrast MRI was performed at day 3 post–CFA and saline injection. Immediately after the MRI, ferumoxytol was administered to these mice, followed by MRI at 48 hours after ferumoxytol administration. MRI images represent four mice from each group. (B) Changes in T2 before and after ferumoxytol were calculated. CFA group showed significantly higher changes in T₂ than saline group.

Figure 3

Ferumoxytol-contrasted magnetic resonance imaging (MRI) in clinical radicular pain secondary to intervertebral disc protrusion. (A) T2-weighted MR images of a human pain subject with clinical diagnosis of right L3 radiculitis before and after ferumoxytol administration. Upper panels were MR images. Middle panels were heat map–coded MR images; white arrow: right L3 nerve root; black arrow: left L3 nerve root. Lower panels were the right L3 nerve root. T2-weighted signal strength was reduced on the right side, but not on the left side, after ferumoxytol. (B) T2*-weighted MR images of the same subject. Images showing right L3 nerve root (white arrow) as it descends in the spinal canal. Upper pannels were MR images. Lower pannels were heat map–coded MR images. (C) MR image of the same subject at L4/5 level. Upper pannels were MR images. Lower pannels were heat map–coded MR images. Arrow indicates a small disc protrusion touching on the exiting right L4 nerve root and narrowing of the bilateral subarticular zone touching on the right descending L5 nerve roots at the L4/5 level.

Figure 4

Ferumoxytol-contrasted magnetic resonance imaging (MRI) in a healthy subject. A subject without any pain symptoms underwent MR imaging study using the same protocol as pain subject. Upper panels were MR images; lower panels were heat map–coded MR images. There were no significant disc protrusions identified by conventional MR images. Heat map–coded MR images did not reveal signal strength changes around nerve roots before and after ferumoxytol administration.