We examined the influence of graded cold-water immersion (CWI) on global and regional quadriceps muscle perfusion using positron emission tomography (PET) and [15O]H2O. In thirty healthy males (33±8 yrs; 81±10 kg; 184±5 cm; percentage body fat: 13±5%; V̇O2peak: 47±8 mL·kg-1·min-1) quadriceps perfusion, thigh and calf cutaneous vascular conductance (CVC), intestinal, muscle and local skin temperatures, thermal comfort, mean arterial pressure and heart rate were assessed prior to and following 10-min of CWI at 8°C, 15°C or 22°C. Global quadriceps perfusion did not change beyond a clinically relevant threshold (0.75 mL∙100g∙min-1) in any condition, and was similar between conditions [range of the differences (95% confidence interval [CI]); 0.1 mL∙100g∙min-1 (-0.9 to 1.2 mL∙100g∙min-1) to 0.9 mL∙100g∙min-1 (-0.2 to 1.9 mL∙100g∙min-1)]. Muscle perfusion was greater in vastus intermedius (VI) compared with vastus lateralis (VL) (2.2 mL∙100g∙min-1; 95%CI 1.5 to 3.0 mL∙100g∙min-1) and rectus femoris (RF) (2.2 mL∙100g∙min-1; 1.4 to 2.9 mL∙100g∙min-1). A clinically relevant increase in VI muscle perfusion after immersion at 8°C and a decrease in RF muscle perfusion at 15°C were observed. A clinically relevant increase in perfusion was observed in the VI in 8°C compared with 22°C water (2.3 mL∙100g∙min-1; 1.1 to 3.5 mL∙100g∙min-1). There were no clinically relevant between-condition differences in thigh CVC. Our findings suggest that CWI (8-22°C) does not reduce global quadriceps muscle perfusion to a clinically relevant extent, however, colder-water (8°C) increases deep muscle perfusion and reduces (15°C) superficial muscle (RF) perfusion in the quadriceps muscle.