Purpose: To evaluate whether the BOLD signal from skeletal muscle can be modulated by exercise and ingestion of vasoactive substances.
Materials and Methods:The right calf muscles of healthy adult volunteers were imaged using a GE 1.5-Tesla scanner and a gradient-echo sequence with spiral readout. Timevarying changes in the BOLD signal were induced through cyclic phases of normoxia (90 seconds of 20.8% O 2 ) and hyperoxia (45 seconds of 100% O 2 at 22 L/minute). Superimposed on this paradigm were pre-and post-exercise regimes, with and without ingestion of caffeine (100 mg) or antihistamine (4 mg chlorpheniramine). The numbers of voxels within slow-twitch (soleus) and fast-twitch (gastrocnemius) muscles that significantly responded to the paradigms were scored and compared using the AFNI software (NIMH).Results: Cycling-inspired O 2 produced a corresponding BOLD modulation that increased in magnitude with exercise. Chlorpheniramine significantly (P Ͻ 0.01) prevented the overall increase in exercise-induced soleus muscle BOLD signal, while caffeine accentuated the increase (P Ͻ 0.05) in the gastrocnemius relative to control (no vasomodulator) conditions.
Conclusion:BOLD signal changes with exercise can be modulated by standard doses of chlorpheniramine (antihistamine) and caffeine. We suggest that chlorpheniramine may act detrimentally on slow-twitch muscle contractility, while caffeine appears to improve fast-twitch muscle function. THE BLOOD OXYGEN LEVEL-DEPENDENT (BOLD) signal in MRI, which was first described by Ogawa et al (1) in 1990, is attributed to a change in the blood ratio of oxyhemoglobin (oxyHb) to deoxyhemoglobin (deoxyHb). This signal modulation is the basis for functional MRI (fMRI) studies of brain activation, in which increased local metabolism disproportionately drives increased blood flow and volume, resulting in elevated oxyHb/deoxyHb and BOLD signal enhancement. Recently the BOLD signal has been applied to the study of other tissues, including muscle (2) and tumors (3), and BOLD signal changes have been observed with inspiration of 100% O 2 . In addition, the potential for MR to characterize myocardial perfusion reserve, as indexed by the BOLD response to maximal pharmacological vasodilation, has been documented (4,5).Even though the blood volume of the brain is fairly low (on the order of 4%), BOLD signal changes with activation are still visible. Skeletal muscle at rest has a comparable blood volume; however, activation (through exercise) results in comparatively larger increases in blood flow and volume. In a previous report (2) we demonstrated that cycling inspired gas between hyperoxia (100% O 2 ) and air (20.8% O 2 ) resulted in a time-varying BOLD signal in the lower human leg that was analyzable with routine "block design" fMRI-based methods and software. Using this method we showed clear differences between the soleus (slow-twitch oxidative muscle) and gastrocnemius (fast-twitch anaerobic/glycolytic muscle), which were hypothesized to be due largely to the comparatively greater mic...