Most patients with severe PTS had a combination of deep and superficial reflux. Reflux in the deep proximal veins contributes significantly to the PTS.

Duplex ultrasonography is important in the diagnosis of vasculogenic erectile dysfunction. We measured the ultrasonographic parameters of cavernous blood flow in different phases of penile erection. We examined 44 volunteers with normal erectile potency. Doppler spectra of the cavernous artery were obtained in a time-dependent manner after intracavernous administration of papaverine. Following intracavernous pharmacological stimulation, the Doppler spectrum alters according to a specific pattern indicating the different hemodynamic phases of erection. Peak flow velocity and acceleration time, measured in the early post-injection phase, may be used to grade arterial inflow. The difference between resistance index in the pre-injection and late post-injection phases may be used to estimate veno-occlusive function. References values are defined.

More reflux develops in previously thrombosed vein segments. As early as after the third month, patients with superficial reflux have an increased risk of development of the first clinical signs of PTS. Within 2 years, the SVPT shows no relationship with clinical signs of PTS.

Despite the many studies on venous haemodynamics using duplex, only a few evaluated the normal values, variability and reproducibility. Therefore, the range and variability of venous diameter, compressibility, flow and reflux were measured. To obtain normal values, 42 healthy individuals (42 limbs, 714 vein segments) with no history of venous disease were scanned by duplex. To determine the reproducibility the intra-observer variability was measured in 11 healthy individuals (187 vein segments) and the inter-observer variability in 15 healthy individuals (255 vein segments) and 13 patients (169 vein segments) previously diagnosed with deep venous thrombosis. Of the 714 normal vein segments, 708 (99%) were traceable, including the crural veins. Of the traceable vein segments, 675 (95%) were compressible and in 696 (98%) flow was present. Of the 42 common femoral vein segments, in 25 (60%) the reflux duration exceeded 1.0 s, but in the other proximal vein segments the reflux duration was less than 1.0 s (95% confidence interval 3.0-10.0). With the exception of the distal long saphenous vein, in the distal vein segments the reflux duration was less than 0.5 s (95% confidence interval 3.5-8.2). The coefficient of variation of the diameter measurements ranged from 14 to 50% and that of the reflux measurements from 28 to 60%. The kappa-coefficient of the inter-observer variability in the classification of compressibility measurements in the patients was 0. 77 and that of the reflux measurements was 0.86. This study shows that almost all veins were compressible in healthy individuals, except the distal femoral veins. In healthy individuals the duration of reflux of the proximal veins was less than 1.0 s and in the distal veins it was less than 0.5 s. The inter-observer variability of the reflux and compressibility measurements in the patients was good.

Arteriovenous fistula (AVF) is a well known but rarely diagnosed complication of percutaneous biopsy of kidney allografts. In the past diagnosis was usually made when clinical signs of an AVF occurred but Doppler ultrasonography has now enabled non-invasive diagnosis. Doppler examination of kidney allografts was performed after 100 biopsies. A total of ten AVFs were diagnosed within 2 weeks of biopsy. On repeated examination 2 months later, no additional fistula was detected. All fistulas were detected by abnormal colour shading of the artery and vein of the fistula caused by high blood velocity. Quantification in the artery supplying the fistula showed a higher systolic velocity compared with that in a normal artery of comparable size and location in the graft (mean (range) 64 (25-150) versus 36 (20-65) cm/s, P < 0.05). Diastolic velocity was also higher in the artery supplying the fistula than in a normal artery (mean (range) 34 (9-72) versus 7 (0-13) cm/s, P < 0.05). In the group with an AVF the proportion with a prolonged bleeding time (> 3 min) was higher (80 versus 47 per cent, P < 0.05), as was the prevalence of a platelet count < 200 x 10(9)/l (60 versus 22 per cent, P < 0.05). After detection of the fistula, four of the grafts were lost because of rejection and two patients died from sepsis during antirejection treatment. During follow-up of the remaining four AVFs, three disappeared spontaneously and one persisted. None of the fistulas has had an impact on renal function requiring intervention. In conclusion, AVF is a complication observed frequently after kidney allograft biopsy that can be detected and monitored by Doppler ultrasonography.