Decadelong low basal ganglia NAA/tCr from elevated tCr supports ATP depletion from mitochondrial dysfunction and neuroinflammation in Gulf War illness

Reduced N-acetylaspartate (NAA)/total creatine (tCr) ratio found with long echo-time proton magnetic resonance imaging (¹H-MRS) of deep brain structures in a Seabees Battalion in 1997-1998 was soon replicated by two studies but not in a later one using intermediate echo time. We undertook this study in 2008-2009 to repeat the previous long echo-time ¹H-MRS study with ¹H-MRS at both long (TE = 270 ms) and short (TE = 30 ms) echo time and higher field strength (3T) to test whether the abnormality of NAA/tCr affecting this Battalion had normalized or been obscured by effects of the T₂ decay curve. Under investigator blinding, 39 Seabees in the three GWI variant groups and 16 Seabees controls prospectively underwent ¹H-MRS at both short (TE = 30 ms) and long (TE = 270 ms) echo time to measure metabolites and at five TE values between 60 ms and 270 ms to measure transverse relaxation time (T₂) in the basal ganglia. A mixed-effects linear model adjusting for age tested group differences. Findings supported the observations of the prior studies, demonstrating that veterans with GWI no longer had reduced NAA/tCr at long echo time but had significantly lower basal ganglia NAA/tCr than controls at short echo time (left: 1.22 ± 0.02 vs. 1.38 ± 0.03, P < 0.0001; right: 1.12 ± 0.02 vs. 1.18 ± 0.03, P = 0.059). The group differences were mainly due to higher [tCr] (left: 14.1%, P = 0.0001; right: 9.1%, P = 0.009) rather than lower [NAA] in the ill groups. Longer echo time substantially reduced the sensitivity of ¹H-MRS. Chronic metabolite abnormalities persisted in GWI for 10 more years but remained detectable only at short echo time. That reduced NAA/tCr is due primarily to increased [tCr] rather than decreased [NAA] supports recent studies implicating mitochondrial dysfunction with ATP depletion and neuroinflammation as causative factors and therapeutic targets in GWI.