Purpose: The prevalence of incompetent perforators increases linearly with the clinical severity of chronic venous insufficiency (CVI) and the presence of deep vein incompetence. Putative transmission of deep vein pressure to skin may cause dermal hypoxia and ulceration. Despite extensive prospective interest in the contribution of perforators toward CVI, their hemodynamic role remains controversial. The aim of this prospective study was to determine the in situ hemodynamic performance of incompetent perforating veins across the clinical spectrum of CVI, by means of duplex ultrasonography. Methods: A total of 265 perforating veins of 90 legs that had clinical signs and symptoms consistent with CVI in 67 patients referred consecutively to the blood flow laboratory were studied. The clinical distribution of the examined limbs was CEAP0, 10 limbs; CEAP1-2, 39 limbs; CEAP3-4, 21 limbs; and CEAP5-6, 20 limbs. With the use of gated-Doppler ultrasonography on real-time B-mode imaging, the flow velocity waveforms were obtained from the lumen of perforators on release of manual distal leg compression in the sitting position and analyzed for peak and mean velocities, time to peak velocity, volume flow, venous volume displaced outward, and flow pulsatility. The diameter and duration of outward flow (abnormal reflux > 0.5 seconds) were also measured. Results: Incompetent perforators had bigger diameters, higher peak and mean velocities and volume flow, longer time to peak velocity, and bigger venous volume displaced outward (VVoutward) than competent perforators (all, P <.0001). The diameter of incompetent perforators did not change significantly with CEAP class (all, P >.1). Incompetent thigh and lower-third calf perforators had a significantly bigger diameter than perforators in the upper and middle calf combined (both, P <.05), in incompetent perforators: reflux duration was unaffected by CEAP class or site (P >.3); peak velocity was higher in those in CEAP3-4 than those in CEAP1-2 (P =.024); mean velocity in those in CEAP3-6 during the first second of reflux was twice that of those in CEAP1-2 (P <.0001); both higher volume flow and VVoutward were found in the thigh perforators than those in the upper and middle calf thirds (P <.03); CEAP3-6 volume flow and VVoutward, both in the first second, were twice that in those in CEAP1-2 (P <.002); flow pulsatility in those in CEAP5-6 was lower than in those in CEAP1-2 (P =.014); in deep vein incompetence, higher peak velocity, volume flow, VVoutward, and diameter occurred than in its absence (P <.01). CEAP designation correlated significantly with mean velocity and flow pulsatility, both in the first second (r = 0.3, P <.01). The flow direction pattern in perforator incompetence was uniform across the CVI spectrum: inward on distal manual limb compression, and outward on its release; competent perforators had a smaller percentage of outward flow on limb compression (P <.01). Conclusion: In addition to an increase in diameter, perforator incompetence is characterized by significantly higher mean and peak flow velocities, volume flow, and venous volume displaced outward, and a lower flow pulsatility. Differences in early reflux enable a better hemodynamic stratification of incompetent perforators in CVI classes. In the presence of deep reflux, incompetent perforators sustain further hemodynamic impairment. In situ hemodynamics enable quantification of the function of perforators and can be used in the identification of the clinically relevant perforators and the impact of surgery. (J Vasc Surg 2001;33:773-82.)