Vascular medicine depends heavily on non-invasive measuring hemodynamic parameters, such as ankle-brachial index (ABI), peak systolic velocity (PSV), end diastolic velocity (EDV), minimum diastolic velocity (MD) and time averaged maximal velocity (TAMAX). Calculated values, such as the resistance index (RI) and the pulsatility index (PI) are also used commonly for interindividual and intraindividual comparisons. Sceptics argue that several error sources, such as media sclerosis, heart insufficiency, the Doppler probe positioning and other user-dependent factors may influence the correctness of the results [1
]. The evidence supposing a predictive role for prognosis of treatment effects is sparse at best [2
]. Even the contemporary peripheral arterial obstructive disease (PAOD) classification systems, such as Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASCII), prefer clinical over objective hemodynamic criteria (ancle brachial index [ABI]) [3
]. On the other hand, crural ulcers may have different etiologies in the presence of PAOD, and the lack of objective evidence-based criteria to differentiate PAOD from other etiologies makes it difficult to find an appropriate treatment even for these routine cases. Other non-invasive methods and parameters, such as transcutaneous partial oxygen pressure measurement, “transkutane Sauerstoffpartialdruckmessung” (tcpO2
), contrast-enhanced ultrasonography (CEUS) and toe pressure measurements try to circumvent the methodological artifacts of Doppler ultrasonography by quantifying the perfusion or oxygenation at the periphery but again, available evidence is not sufficient to recommend these methods for standard diagnostics [3
Are these hemodynamic parameters able to reflect what they are believed to, namely to quantify the real capacity of the vascular system to transport blood to the tissues and to satisfy the changing needs in oxygen? It is known that the vascular autoregulation includes constriction of the precapillary sphincters at rest and their dilatation at exercise in order to redistribute the blood between viscera and extremities. The typical triphasic pulse wave form converts to monophasic flow for some minutes after exercise in the crural arteries in healthy persons indicating a physiologic lowering of the peripheral resistance; however, the poststenotic or postocclusion flow is typically also monophasic in PAOD patients already in resting states, pointing to altered autoregulation. Logically, beside the different pulse wave morphology at Doppler ultrasonography, the blood volume per minute should be about the same at rest in healthy persons and PAOD patients (at least in Fontaine-II stage), and the functional reserve may differ between them (when measuring the blood volume per minute after exercise). Furthermore, the hemodynamic parameters and the pulse wave form may not differ at rest in patients before and after successful treatment, particularly in those patients whose crural arteries are fed via collaterals, but only after exercise. Starting from this consideration, the aim of the present investigation was to compare the Doppler ultrasonography-based hemodynamic parameters mentioned above before and after exercise in healthy persons and PAOD patients (before and after a treatment procedure). The hypothesis was that the change or course of the post-exercise hemodynamics would differentiate between healthy persons and PAOD patients and between successful and frustrating treatment while the resting state measurements would not.
Patients and methods
The investigation was performed as a prospective non-randomized, non-invasive diagnostic single-center study in a consecutive patient cohort consisting of two parts: a comparison of two independent cohorts (healthy volunteers and PAOD patients) and an intra-cohort comparison of the states before and after treatment in PAOD patients.
Patients and volunteers
In this study 19 healthy volunteers and 23 PAOD patients hospitalized in the Division of Vascular Surgery of the Magdeburg University Hospital, Germany, were included from October 2015 until June 2016 based on informed consent.
Healthy volunteers were examined by duplex ultrasonography (General Electric, Logiq E9, 9 MHz linear probe, Freiburg, Germany) and the hemodynamic parameters of the right anterior tibial artery (ATA) were measured or calculated by the integrated software after 5min of resting in supine position at the medial third of the lower leg (PSV, EDV, MD, TAMAX, RI, PI). Then, they were asked to perform dorsal flection of the feet in upright position 60 times within 1min or until claudication occurred. The ultrasound examination was repeated 1, 2, 3, 4 and 5 min after the test. Patients underwent the same test in hospital before and after the treatment; however, when they were not able to take part in the test in upright position, they were allowed to perform the test supine with slightly elevated legs. Depending on the vascular state, one of the perfused crural vessels was examined with preference for the ATA.
Clinical data and follow-up
Besides the assessed Doppler parameters, data from the medical history, such as age, weight, height, sex, predisposing factors, such as smoking, diabetes, hyperlipoproteinemia, thrombophilia, genetic predisposition and cardiovascular risk factors were asked. Furthermore, the lower extremity vessels were assessed by the clinical routine diagnostics (digital subtraction angiography [DSA], magnet resonance angiography [MRA]) and whether the examined artery was directly perfused or via collaterals was documented. The patients were followed-up after discharge from hospital until December 2016, and the censored survival, major amputation and reintervention was assessed.
According to the legal requirements of the State of Sachsen-Anhalt in Germany, for the non-invasive study using data of the routine diagnostics, approval by the local ethics committee was not necessary; however, the data safety guidelines of the Magdeburg University and the State of Sachsen-Anhalt were strictly followed.
For statistical calculations, the R software was used [4
], ANOVA was performed using the package NLME [5
], the Shapiro test with the mvnormtest package [6
], survival analysis with the survival package [7
]. Differences were considered significant when p
< 0.05 and a tendency was defined as p
: 0.05 … 0.1.
Discussion and conclusion
One of the aims of the study was to assess whether a healthy cohort can be differentiated from a PAOD patient cohort in the resting state by the mean hemodynamic parameters. Furthermore, these were compared in the PAOD patient cohort before and after treatment.
As shown in Table 1
, in particular representing “hemodynamics in resting subjects comparing healthy persons and PAOD patients”, in can be stated that based on these data it would be difficult to differentiate healthy persons from PAOD patients related solely to hemodynamic parameters of the crural arteries. Furthermore, the short-term treatment effects were not impressive with respect to the hemodynamic changes.
With respect to the “amplitudes of autoregulation in hemodynamic parameters after exercise test”, obviously healthy persons and PAOD patients can be identified much better by the autoregulation differences after exercise than by the resting values because the mean PSV, TAMAX and PI autoregulation ratios differed significantly between them. As would be expected from Fig. 1
, the hemodynamic changes after exercise are more expressed in healthy persons pointing to intact autoregulation; however, when the ratios were compared in PAOD patients before and after treatment, no significant differences could be found. This can be explained by the fact that in most patients the autoregulation remained disturbed at the microvasculature level and an improvement is visible in only three patients Fig. 1
As can be seen in the “Results” paragraph on “comparisons of the time courses after exercise by repeated-measures-ANOVA”, only the PI after exercise allows a treatment effect in the cohorts under investigation to be postulated, whereby the PI values were lower before treatment than after (negative estimate); however, as with PSV before exercise, this effect remained moderate.
Related to the “Results” paragraph entitled “influence of cofactors on PI” it can be summarized that:
It seems that the hemodynamic changes characteristic for PAOD patients (lower perfusion parameters and diminished autoregulation after exercise) are present at an early stage of PAOD when there is still only claudication.
Only sub-classifying patient into groups and direct vascular connection before treatment contributed significantly to the PI whereby the latter showed a negative influence, probably due to autoregulation after exercise.
Basically and in a general context, despite good concordance between Duplex ultrasonography and digital subtraction angiography (DSA) [9
], Doppler and Duplex ultrasonography have not yet fulfilled their promises to predict the clinical course in PAOD and to derive reliable treatment decisions [1
]. The technical evolution and the ubiquitous availability of ultrasonography equipment enables physicians nowadays to assess the supragenual and infragenual vessels with high resolution and accuracy before any invasive diagnostics while the known artifacts of non-imaging modalities remained the same (i.e., for ABI in diabetes). In this context, it is worth mentioning that 40 years ago the ABI was already measured after exercise and used for diagnosis of PAOD [10
]. Duplex ultrasonography allows diagnosis of occlusions and estimation of the percentage of stenosis in addition to the Doppler ultrasonography method [11
]. It sounds quite plausible that in PAOD the hemodynamics of the crural arteries, such as PSV, EDV, MD, TAMAX, should differ from those of healthy persons and this difference should be visible at rest; however, as our results showed (Table 1
) this is not the case. Only the RI which reflects the pulse wave form (a value of 1 corresponds to a triphasic wave) showed a tendency to lower values (monophasic pulse wave form) due to proximal occlusions or stenosis. Obviously, healthy persons differ from PAOD patients more by the functional reserve than by the resting values due to the physiological autoregulation of the arterioles. Therefore, the healthy persons and the PAOD patients underwent a standardized exercise test. The ratios of the maximum/minimum parameter values after exercise and the resting values differed significantly between healthy persons and PAOD patients (Table 2
). The amplitude of the autoregulation was much higher in healthy persons than in the latter. Therefore, we conclude that this is a much more reliable criterion to diagnose PAOD based on hemodynamics compared to resting values. When the patients would have been identified and treated according to guidelines, should the hemodynamics then return to “normal” values in order to indicate treatment success? From Table 1
, it is obvious that only an elevation of the PSV of about 5 cm/s took place after treatment, which is clearly not a reliable change. The autoregulation did also not relevantly change after treatment as seen in Table 2
(and Fig. 1
). Only the time course of the PI increased after treatment in PAOD patients as calculated by ANOVA (Table 2
) but remained significantly different to the healthy values. Most parameters did not change at all shortly after treatment. Treatment success can be differently defined and in most cases would include the follow-up with documentation of event-free survival; however, it would be of value to foresee how long this would endure for the individual patient based on changes in hemodynamics immediately after treatment. The PI is thus the parameter that might serve for further evaluations of its event-free survival prediction. A cofactor which may influence the PI independently from the treatment state was the continuous proximal inflow of the crural vessel under investigation. This should be considered for interindividual comparisons.
A possible source of bias was postoperative complications (death and major amputation in two patients) which led to case-wise exclusions of postoperative data.
We can summarize that in order to reliably assess treatment effects by Duplex and Doppler ultrasonography, the patients should undergo an exercise test, the time course after exercise should be measured and compared to the pretreatment exercise test.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.