Fourteen postheatstroke (HS) subjects and nine normal control subjects (C) with similar (V̇O2max) underwent a 180-min heat-tolerance test (ambient temperature 40°C, relative humidity 40%; 12 steps.min⁻¹ on a 30-cm high bench). In group HS nine subjects (NHS) thermoregulated similar to group C, while five subjects were identified as heat intolerant (HI), characterized by significantly higher heart rate (HR), rectal temperature (Tre), and heat storage (ΔS) than groups NHS and C (at the end of the 2nd h, HR 159 ± 9, 124 ± 4, and 118 ± 3 beats.min⁻¹, Tre 38.9 ± 0.1, 37.9 ± 0.1, and 37.9 ± 0.1 °C, ΔS 24 ± 5, 10 ± 2, and 12 ± 4 W.m⁻² in groups HI, NHS, and C, respectively). The work efficiency (η) was significantly lower in groups HI and NHS when compared with C (10.2 ± 0.7, 10.4 ± 0.5, and 11.8 ± 0.6%, respectively). Skin surface area-to-body mass ratio (AD/W) in group HI was significantly lower than in group NHS or C (247 ± 7,271 ± 5, and 272 ± 4 cm².kg⁻¹, respectively). Multiple linear regression analysis between each one of the dependent variables, HR, Tre, and ΔS, mean weighted skin temperature, and sweat rate, against the two independent variables, η and AD/W, yielded significant correlations (r = 0.74, 0.73, 0.58, 0.59, and 0.44, respectively). The results suggest that a reduced η and/or a reduced AD/W play a role in heat intolerance, causing increased heat production and a concomitantly less effective heat dissipation.