Safety and Feasibility Using a Fluid-Filled Wire to Avoid Hydrostatic Errors in Physiological Intracoronary Measurements

Background Using a fluid-filled wire with a pressure sensor outside the patient compared to a conventional pressure wire may avoid the systematic error introduced by the hydrostatic pressure within the coronary circulation. Aims To assess the safety and effectiveness of the novel fluid-filled wire, Wirecath (Cavis Technologies, Uppsala, Sweden), as well as its ability to avoid the hydrostatic pressure error. Methods and Results The Wirecath pressure wire was used in 45 eligible patients who underwent invasive coronary angiography and had a clinical indication for invasive coronary pressure measurement at Sahlgrenska University Hospital, Gothenburg, Sweden. In 29 patients, a simultaneous measurement was performed with a conventional coronary pressure wire (PressureWire X, Abbott Medical, Plymouth, MN, USA), and in 19 patients, the vertical height difference between the tip of the guide catheter and the wire measure point was measured in a 90-degree lateral angiographic projection. No adverse events caused by the pressure wires were reported. The mean Pd/Pa and mean FFR using the fluid-filled wire and the sensor-tipped wire differed significantly; however, after correcting for the hydrostatic effect, the sensor-tipped wire pressure correlated well with the fluid-filled wire pressure (R = 0.74 vs. R = 0.89 at rest and R = 0.89 vs. R = 0.98 at hyperemia). Conclusion Hydrostatic errors in physiologic measurements can be avoided by using the fluid-filled Wirecath wire, which was safe to use in the present study. This trial is registered with NCT04776577 and NCT04802681.


Introduction
Using intracoronary pressure measurements to evaluate the clinical signifcance of coronary stenosis is recommended in European and American guidelines [1,2].Te most common method to perform a pressure measurement is fractional fow reserve (FFR) using adenosine to accomplish hyperemia; however, resting indices, e.g., instant wave-free ratio (iFR), have been proven as efective as FFR to distinguish a signifcant from a nonsignifcant lesion [3].
Several coronary wires are available to measure pressure inside the coronary artery, all with a pressure sensor in the distal part of the wire.Te pressure sensor is calibrated to the aortic pressure by equalization of pressures when the pressure is measured at the same position, i.e., outside the guide catheter in the proximal part of the vessel or in the aorta.However, when advancing the wire into the coronary vessel, the hydrostatic pressure, due to the height diference between the coronary ostium and the distal part of the vessel, causes an unaccounted infuence on the measured pressure.In a supine position, the left anterior descending artery (LAD) and posterior descending artery from the right coronary artery (PDA) ascend on average 5 cm, and the left circumfex artery (LCX) and posterior lateral artery from the right coronary artery (PLA) descend on average 3 cm (Figure 1).Tis results in a decrease in the measured pressure in the LAD and the PDA and an increase in pressure in LCX and PLA due to the hydrostatic pressure.CT studies measuring this height diference suggest a conversion factor of 0.77 mmHg/cm in order to adjust for diferences in hydrostatic pressure [4][5][6].
Te systematic error in conventional pressure measurements introduced by the hydrostatic pressure has been described previously but has not been accounted for in studies or guidelines when suggesting clinically used pressure ratio cutofs for signifcance pre-and post-PCI regardless of type of pressure index [4].
Te Wirecath (Cavis Technologies, Uppsala, Sweden) (Figure 2) is a novel fuid-flled 0.014″ wire connected to an external pressure transducer with fxed height which makes the measurements by the Wirecath unafected by the hydrostatic pressure diference in the coronary artery [7].
Tis frst-in-human Wirecath study (ClinicalTrials.govIdentifer: NCT04776577) was designed to assess the safety and efectiveness of the novel fuid-flled wire as well as its ability to avoid the hydrostatic pressure error.Te study was approved by the Swedish Ethical Review Authority (D-nr 2020-04421).

Patient Selection.
Patients were recruited at Sahlgrenska University Hospital, Gothenburg, Sweden.Inclusion criteria were age 18 or above, signed informed consent, and clinical indication for coronary pressure assessment after coronary angiography.Exclusion criteria were known heparininduced thrombocytopenia, other allergy to heparin, contraindication for adenosine, or unstable condition by op-erator´s judgment.
An additional 16 patients were enrolled who had coronary physiology assessment only using the fuid-flled wire (Wirecath 1, Cavis Technologies, Uppsala, Sweden).After each of the 45 procedures, a usability questionnaire pertaining to the use of the fuid-flled wire was flled out by the users.
Te primary endpoint was the rate of adverse events, defned as an untoward medical occurrence, unintended disease or injury, or untoward clinical sign in any of the 45 studied subjects.Te key secondary endpoints were the agreement between Pd/Pa and FFR measured with the fuidflled (Wirecath) and sensor-tipped (Abbott PressureWire) before and after correction for the hydrostatic pressure by using the height measurements.Exploratory secondary endpoints included the correlations between fuid-flled and sensor-tipped wire-derived indices with CFR indices and the usability questionnaire score.
2.3.Procedure.Angiography was performed according to the local standard procedure.All patients received nitroglycerine prior to all physiology assessment.Hyperemia was induced by administering adenosine intravenous (140 μg/kg/min over two minutes) or intracoronary (200 μg for the left coronary vessel and 120-200 μg for the right coronary vessel).Whether the patients had simultaneous measurements with the fuid-flled and sensor-tipped wires or only had coronary physiology assessment using the fuid-flled wire was left to the discretion of the physician performing the procedure.

Simultaneous Measurements.
Te fuid-flled wire (Wirecath 1, Cavis Technologies, Uppsala, Sweden) and the sensor-tipped wire (PressureWire X, Abbott Medical, Plymouth, MN, USA) were placed simultaneously at the same location in the coronary artery.Simultaneous pressure measurements were then performed with both wires, both resting Pd/Pa and FFR.
Te vertical height diference between the tip of the guide catheter and the wire measure point was measured in a 90degree lateral angiographic projection by frst marking the position of the guide catheter tip with adhesive tape on the fuoroscopy screen, then adjusting the vertical position of the x-ray table until the measuring point of the wire on fuoroscopy reaches the tape mark on the screen, and fnally assessing the total adjusted distance in centimeter between the two table positions on a tape measure attached to the patient table (Figure 4).
When the wires were withdrawn from the artery into the tip of the guide catheter, the observed "Pd/Pa" was recorded to allow quantifcation of any drift in pressure.
By multiplying each height diference, negative or positive, with the conversion factor 0.77 mmHg/cm, a correction for each pressure measurement by the sensor-tipped wire was established.By adjusting the sensor-tipped pressure with this correction, the pressure from the sensor-tipped wire was corrected for the hydrostatic efect. 2 Cardiology Research and Practice

Coronary Flow Reserve.
A subset of patients who had simultaneous measurements by the fuid-flled and pressuretipped wires underwent bolus-thermo-derived coronary fow reserve (CFR) measurements in the LAD and was subsequently referred to transthoracic Doppler-echocardiography-derived CFR measurement (TDE-CFR), which has shown to be highly correlated with CFR derived from positron emission tomography [8] and to invasive Doppler guide wire-derived CFR [9,10].TDE-CFR was performed according to the standard procedure of the hospital [11].Te correlation between a pressure-derived CFR, bolus-thermoderived CFR, and the echocardiography-CFR was investigated.To mimic the situation during the invasive physiology measurement, all patients received the same nitroglycerin dose 5 min prior to the TDE-CFR assessment measured during adenosine hyperemia (140 μg/kg/min).Pressure-derived CFR was calculated by the formula 1-FFR/ 1-Pd/Pa [12].

Regular Use.
In the regular-use study group, the fuidflled wire was used as a regular pressure measurement tool, replacing the standard pressure wire.

Statistical Methods.
Te results are presented with descriptive statistics.Unless otherwise noted, all summed values are presented as mean ± standard deviation.In the comparison of pressure measurements, the Pearson correlation factor and the intraclass correlation coefcient were calculated and regression plots were made for paired

Results
Patients' age ranged from 54 to 82 years (mean 70 ± 6.5 years), and 31% were female.Te indications for angiography were 14 patients with stable angina, 9 with unstable angina, 7 with non-ST elevation myocardial infarction, and the remaining with other indications.
No adverse events related to the use of the pressure wires were reported.

Comparisons of Pd/Pa and FFR after Simultaneous
Measurements with Fluid-Filled and Sensor-Tipped Wires.35 coronary arteries in 29 patients were included in the simultaneous measurements study group only.
Te mean Pd/Pa at rest in the 29 simultaneous measurements, after removal of fve drift cases (Pd/Pa >0.02) and one erroneous pressure data, using the fuid-flled wire and the sensor-tipped wire were 0.96 ± 0.03 and 0.94 ± 0.04, respectively.Te mean FFR in the same 29 simultaneous measurements using the fuid-flled wire and the sensortipped wire were 0.89 ± 0.07 and 0.87 ± 0.07, respectively.
Successful height measurements were collected in 19 simultaneous measurements, after the exclusion of nine cases with missing height data and one erroneous height measurement.Data were missing due to difculties in visualizing both the guide catheter tip and the wire position in the lateral view.
Te 19 arteries consisted of 14 LAD, 2 LCX, and 3 RCA.Te mean diference in height between the coronary ostium and the pressure measure point was 2.6 cm, corresponding to a mean hydrostatic pressure diference of 2.0 mmHg, using the conversion factor 0.77 mmHg/cm [6].
Table 1 summarizes the results showing that the difference between the two wires diminished when the sensortipped wire measurements were corrected.Table 2 presents which coronary vessel was measured, measured height diference, calculated hydrostatic pressure, and aortic and coronary pressure ratios.
Figure 5 shows that after correcting for the hydrostatic efect, the sensor-tipped wire pressure correlated well with the fuid-flled wire pressure (R � 0.74 vs. R � 0.89 at rest and R � 0.89 vs. R � 0.98 at hyperemia).
In the subgroup where measurements were made in the LAD, the mean Pd/Pa at rest were 0.96 ± 0.04 and 0.92 ± 0.04 for the fuid-flled wire and the sensor-tipped wire, respectively, and the mean FFR were 0.90 ± 0.05 and 0.86 ± 0.06, respectively.
Out of the 29 valid measurements, the number of discordant classifcations between the sensor-tipped wire and the fuid-flled wire was explored for the Pd/Pa at rest and the FFR measurements.
When applying the cutof Pd/Pa ≤ 0.92 [13], two LADs were signifcant according to the sensor-tipped wire but not according to the fuid-flled wire.Two RCAs were signifcant according to the fuid-flled wire but not according to the sensor-tipped wire.
When applying the cutof FFR ≤ 0.80, three LADs were signifcant according to the sensor-tipped wire, but not according to the fuid-flled wire.
Drift was compared in 33 simultaneous measurements in 27 patients.Te fuid-flled wire demonstrated less drift than the sensor-tipped wire (standard deviation 0.11 vs. 0.18).With an increasing number of cases, less drift was observed, possibly learning curve-related.

Correlations between Pressure Indices and CFR.
Pressure-derived CFR with the fuid-flled wire correlated to TDE-CFR and thermodilution-CFR (R � 0.69 and R � 0.76, respectively (Figure 6(a))).Sensor-tipped wire pressurederived CFR did not correlate positively to the reference methods; see Figure 6(b).When using the square root of 1-

Results of the Usability
Questionnaire. 16 patients were included in the regular-use study group only.Te fuid-flled wire was used as a regular pressure measurement tool, and in two cases, the fuid-flled wire was used for balloon and stent intervention during PCI.
Usability was evaluated by using a questionnaire flled in by the interventionalists in this group.Te scoring was similar for the fuid-flled wire and the sensor-tipped wire when it came to perceived maneuverability, X-ray visibility, and signal quality.Te preparation of the fuid-flled wire had a mean score of 2.8 compared to the sensor-tipped wire 3.0, on a scale from 1 (poor) to 5 (very good).

. Discussion
Te main fndings of this study are that (i) the fuid-flled Wirecath wire (Cavis Technologies, Uppsala, Sweden) could be safely and reliably used to assess Pd/Pa and FFR for clinical decision-making in a typical patient population undergoing PCI; (ii) the diferences between FFR and resting indices obtained by the sensor-tipped versus fuid-flled wires were explained by a hydrostatic pressure component related to the height diference between the coronary ostium and the distal vessel; and (iii) correcting the sensortipped values for the height diference almost completely aligned measurements derived by the sensor-tipped versus fuid-flled wires.
Te hydrostatic pressure diference (which is the same value for resting and hyperemia measurements) has a larger impact on resting ratios since the true pressure reduction across the lesion is smaller at rest than during hyperemia (FFR).Te efect is also larger on diastolic resting ratios compared to whole-cycle ratios, since the fxed hydrostatic pressure diference is relatively larger in diastolic versus mean blood ratios [5,14].
A neglection of the hydrostatic error may have implications for clinical practice.Te hydrostatic error causes lower measured values (i.e., overestimation) in the LAD or distal PDA, since these vessels in a supine patient position are located higher than the coronary ostium, i.e., the hydrostatic pressure falls.Tis measurement error might lead to false-positive pressure ratios and thus misclassifcation as clinically signifcant.Vice versa may the rise in hydrostatic pressure cause an underestimation in non-LAD vessels.Tis could be of special importance in patients where FFR is used to evaluate the result of the PCI exposing the patient to the risk of leaving an inadequate result or the risk of complication when trying to improve an already adequate stenting [15].
Coronary CT has previously been used for the measurement of height diferences in coronary vessels [4,5].Another potential way to calculate the hydrostatic error, when a fuid-flled wire is not available, is to use quantitative coronary angiography (QCA) tools to measure the height in a lateral angiographic view and compensate for the height diference between the tip of the guide catheter and the measure point of the wire.Tis would simplify the height measurement compared to the method used in this study.Te QCA method is currently evaluated in another study (ClinicalTrials.govIdentifer: NCT04802681).
When using a fuid-flled wire, pressure-derived CFR correlates to other CFR measurement methods as opposed to when using a sensor-tipped wire.Tis is presumably due to the avoidance of the hydrostatic error, which has also been shown in other explorative studies [16].Due to the small sample size, no frm conclusion can be drawn, but the result may suggest that pressure-derived CFR can possibly be used as a screening tool for evaluating CFR and other CFR-related indices such as resistive reserve ratio (RRR) and microvascular resistance reserve (MRR).

Limitations.
A larger dataset than available in this study would be needed to statistically evaluate the amount of discordance and misclassifcation caused by hydrostatic errors.Quantifcation of the size of errors caused by hydrostatic pressure has been explored in other studies [4,5].
Te reasons for drift were not explored in this study.Drift could be caused by both technical shortcomings of the measurement devices, but also by user mistakes caused by pitfalls when doing pressure measurements [7].

Conclusions
Hydrostatic pressure introduces error in conventional intracoronary pressure measurements that has a particularly large impact on resting indices.Hydrostatic errors in physiologic measurements can be avoided by using the fuid-flled Wirecath wire, which was safe to use in the present study.

Figure 1 :
Figure 1: Hydrostatic efects in ∼99% of the population.Illustration of the infuence of the height diference between the coronary ostia and distal coronary arteries and the measured pressure.

FluidFigure 2 :Figure 3 :
Figure 2: Wirecath pressure wire with external transducer.Illustration of the design of the fuid-flled wire and transducer.

Figure 4 :
Figure 4: Height measurement between the pressure measure point and the tip of the guide catheter.Example of an observed height diference between the pressure measure point and the tip of the guide catheter.

Figure 5 :Figure 6 :
Figure5: Correlation between indices derived using the fuid-flled versus sensor-tipped wires.Te correlation for both Pd/Pa (a) and fractional fow reserve (FFR) (b) between the two wires was greater when the hydrostatic error due to height was accounted for (right versus left).Passing-Bablok regression indicates that a small proportional bias was present between FFR derived using the fuid-flled wire versus the sensor-tipped wire after correction for hydrostatic pressure (slope 1.14 (95% CI) [1.01-1.38])but not in the other comparisons.

Table 2 :
Height measurements and correction of sensor-tipped wire measurements (n �