Abstract:1. We recently showed that fatigue of the inspiratory muscles via voluntary efforts caused a time-dependent increase in limb muscle sympathetic nerve activity (MSNA) (St Croix et al. 2000). We now asked whether limb muscle vasoconstriction and reduction in limb blood flow also accompany inspiratory muscle fatigue. 2. In six healthy human subjects at rest, we measured leg blood flow (.Q(L)) in the femoral artery with Doppler ultrasound techniques and calculated limb vascular resistance (LVR) while subjects performed two types of fatiguing inspiratory work to the point of task failure (3-10 min). Subjects inspired primarily with their diaphragm through a resistor, generating (i) 60 % maximal inspiratory mouth pressure (P(M)) and a prolonged duty cycle (T(I)/T(TOT) = 0.7); and (ii) 60 % maximal P(M) and a T(I)/T(TOT) of 0.4. The first type of exercise caused prolonged ischaemia of the diaphragm during each inspiration. The second type fatigued the diaphragm with briefer periods of ischaemia using a shorter duty cycle and a higher frequency of contraction. End-tidal P(CO2) was maintained by increasing the inspired CO(2) fraction (F(I,CO2)) as needed. Both trials caused a 25-40 % reduction in diaphragm force production in response to bilateral phrenic nerve stimulation. 3. .Q(L) and LVR were unchanged during the first minute of the fatigue trials in most subjects; however, .Q(L) subsequently decreased (-30 %) and LVR increased (50-60 %) relative to control in a time-dependent manner. This effect was present by 2 min in all subjects. During recovery, the observed changes dissipated quickly (< 30 s). Mean arterial pressure (MAP; +4-13 mmHg) and heart rate (+16-20 beats min(-1)) increased during fatiguing diaphragm contractions. 4. When central inspiratory motor output was increased for 2 min without diaphragm fatigue by increasing either inspiratory force output (95 % of maximal inspiratory pressure (MIP)) or inspiratory flow rate (5 x eupnoea), .Q(L), MAP and LVR were unchanged; although continuing the high force output trials for 3 min did cause a relatively small but significant increase in LVR and a reduction in .Q(L). 5. When the breathing pattern of the fatiguing trials was mimicked with no added resistance, LVR was reduced and .Q(L) increased significantly; these changes were attributed to the negative feedback effects on MSNA from augmented tidal volume. 6. Voluntary increases in inspiratory effort, in the absence of diaphragm fatigue, had no effect on .Q(L) and LVR, whereas the two types of diaphragm-fatiguing trials elicited decreases in .Q(L) and increases in LVR. We attribute these changes to a metaboreflex originating in the diaphragm. Diaphragm and forearm muscle fatigue showed very similar time-dependent effects on LVR and .Q(L).

Fatiguing inspiratory muscle work causes reflex reduction in resting leg blood flow in humans