NeuroImage: Clinical (Jan 2021)

Quantitative susceptibility mapping in the thalamus and basal ganglia of systemic lupus erythematosus patients with neuropsychiatric complaints

  • Marjolein Bulk,
  • Thijs van Harten,
  • Boyd Kenkhuis,
  • Francesca Inglese,
  • Ingrid Hegeman,
  • Sjoerd van Duinen,
  • Ece Ercan,
  • César Magro-Checa,
  • Jelle Goeman,
  • Christian Mawrin,
  • Mark van Buchem,
  • Gerda Steup-Beekman,
  • Tom Huizinga,
  • Louise van der Weerd,
  • Itamar Ronen

Journal volume & issue
Vol. 30
p. 102637

Abstract

Read online

Systemic lupus erythematosus (SLE) is an auto-immune disease characterized by multi-organ involvement. Although uncommon, central nervous system involvement in SLE, termed neuropsychiatric SLE (NPSLE), is not an exception. Current knowledge on underlying pathogenic mechanisms is incomplete, however, neuroinflammation is thought to play a critical role. Evidence from neurodegenerative diseases and multiple sclerosis suggests that neuroinflammation is correlated with brain iron accumulation, making quantitative susceptibility mapping (QSM) a potential hallmark for neuroinflammation in vivo. This study assessed susceptibility values of the thalamus and basal ganglia in (NP)SLE patients and further investigated the in vivo findings with histological analyses of postmortem brain tissue derived from SLE patients. We used a 3T MRI scanner to acquire single-echo T2*-weighted images of 44 SLE patients and 20 age-matched healthy controls. Of the 44 patients with SLE, all had neuropsychiatric complaints, of which 29 were classified as non-NPSLE and 15 as NPSLE (seven as inflammatory NPSLE and eight as ischemic NPSLE). Mean susceptibility values of the thalamus, caudate nucleus, putamen, and globus pallidus were calculated. Formalin-fixed paraffin-embedded post-mortem brain tissue including the putamen and globus pallidus of three additional SLE patients was obtained and stained for iron, microglia and astrocytes. Susceptibility values of SLE patients and age-matched controls showed that iron levels in the thalamus and basal ganglia were not changed due to the disease. No subgroup of SLE showed higher susceptibility values. No correlation was found with disease activity or damage due to SLE. Histological examination of the post-mortem brain showed no increased iron accumulation. Our results suggest that neuroinflammation in NPSLE does not necessarily go hand in hand with iron accumulation, and that the inflammatory pathomechanism in SLE may differ from the one observed in neurodegenerative diseases and in multiple sclerosis.

Keywords