Sensitivity and Reproducibility of Inferior Vena Cava Diameter and Superior Vena Cava Flow Velocity Measurements to Changes in Cardiac Preload in Subjects with Hypertension.

Objectives: We investigated the sensitivity and reproducibility of inferior vena cava (IVC) diameters and superior vena cava (SVC) flow velocities in detecting changes in cardiac preload in clinically euvolemic subjects with hypertension.
Methods: Measurements were obtained during passive leg raising (PLR) and lower limb venous occlusion (LVO), interventions which respectively transiently increase and decrease cardiac preload. Measurements were made in 36 subjects and repeated on two separate occasions to examine reproducibility.
Results: During PLR, there was no significant change in IVC diameters, but peak flow velocity of the SVC S wave increased by 6.5 (95% confidence interval 1.6-11.3) cm/s (P = 0.01). During LVO, IVC diameter in expiration decreased by 3.2 (1.7-4.7) mm and the SVC S wave decreased by 9.7 (4.4-14.7) cm/s (P < 0.001). Venae cavae-derived indices can be used to assess changes in preload within the physiological range in euvolemia. Conclusions: Despite suboptimal reproducibility of baseline measurements, high agreeability between the changes in IVC diameter and SVC flow after LVO suggests that these indices can be used to monitor changes in cardiac preload. Copyright:

INTRODUCTION

In primary hypertension, an increase in intravascular volume is thought to represent one of the pathophysiological mechanisms which contribute to sustained elevation of blood pressure (BP) values. Although hypertensive subjects are usually clinically euvolemic, pharmacological treatment for hypertension often includes diuretics, which, among other actions, reduces intravascular volume. Approximately 70% of the total intravascular volume is contained within the venous compartment,[1] and active and passive changes in venous capacity can modulate heart filling pressure and central blood volume.[1] Measurements of the diameter of the inferior vena cava (IVC) and flow velocity in the superior vena cava (SVC) by cardiac ultrasound have been proposed as useful tools for assessing intravascular fluid volume and cardiac preload in acute setting and critically ill patients.[23] Outside the acute setting, venae cavae-derived indices and, in particular, IVC diameters have been also used to assess changes in cardiac preload during interventions that cause marked volume expansion[456] or contraction.[78] In controlled conditions, application of negative pressure to the lower body to redistribute fluid from the upper body to the lower extremities (lower body negative pressure [LBNP]) has been used to study cardiovascular adjustments to hypovolemia of various degrees.[9] During LBNP, SVC-derived flow velocity indices predict fluid responsiveness[10] and correlate with hemodynamic measurements derived from right heart characterization.[1011] However, a paucity of data is available regarding the interrogations of venae cavae-derived indices in hypertension and their response to interventions which modulate cardiac preload within the physiological range.

The aim of the present study was to evaluate the feasibility of using venae cavae-derived indices in detecting changes in cardiac preload within the physiological range in clinically euvolemic subjects with primary hypertension. Indices of preload were obtained during passive leg raising (PLR), an intervention which causes a transient increase in cardiac preload, and during lower limb venous occlusion (LVO). LVO is a similar but more practical procedure to LBNP, performed by inflating pressure cuffs around the thighs to decrease the venous return from the legs. Measurements were repeated after a control period of no intervention to ensure stability over time and on two separate occasions to assess reproducibility.

MATERIALS AND METHODS

Participants were 36 consecutively consenting subjects with primary hypertension attending the hypertension clinics at Guy's and St Thomas' Hospital. All participants had a diagnosis of arterial hypertension based on their medical records and/or out-of-office[12] daytime systolic ambulatory BP (or home BP averaged ≥7 days) of >135 mmHg systolic or >85 mmHg diastolic, according to current guidelines.[13] Pregnant women were excluded from the study, as were those in whom the clinical history or investigations suggested a presence of secondary hypertension. Patients with moderate or severe valvular disease and those with sustained nonsinus arrhythmias were also excluded.

The study was approved by the London Westminster Research Ethics Committee, and written informed consent was obtained from all patients. Volume status was assessed by physical examination evaluating the presence of peripheral edema, jugular venous pressure, hepatojugular reflux, the presence of extra heart sounds and lung sounds to determine whether fluid overload was present. Pulse, respiratory rate, skin turgor, and delay in capillary refill were used to assess hypovolemia. Patients with clinical signs of volume overload or hypovolemia were excluded from the study. Participants were asked to abstain from caffeine, alcohol, and strenuous exercise for at least 24 h before the measurements, which were conducted in a temperature and light-controlled laboratory following voiding. After 15 min of rest, supine BP and heart rate (HR) were recorded in the brachial artery using a validated oscillometric technique (HEM-705CP, Omron Corp, Kyoto, Japan). All BP measurements were performed in duplicate and the average of the two readings was used for the analysis.

Measurements of preload were obtained by trans-thoracic echocardiography performed using a Philips Epiq 7 ultrasound system (Koninklijke Philips N. V., Amsterdam, Netherlands Philips, Amsterdam, Netherlands) according to recommendations of the American Society of Echocardiography and the European Association of Cardiovascular Imaging.[14] All measurements were performed by a single operator and acquisitions were individually optimized for depth, gain, and frame rate to maximize image quality and minimize inconsistency in acoustic windows prior to analysis. All participants were positioned supine and IVC bidimensional diameters were recorded using the subcostal long-axis view during spontaneous breathing. A visual quality estimate was performed prior to imaging acquisition. The transducer was placed just inferior to the xiphoid process along the midline to obtain a long axis image of the IVC and images were recorded in 15-s B-mode video clips to ensure that they accounted for respiratory variation and were able to capture the points of maximal and minimal diameters. The collapsibility index of IVC was calculated as e difference between the maximum and minimum IVC diameters divided by the maximum IVC diameter, expressed as a percentage.[15] Flow velocity curves in the SVC were acquired during quiet inspiration and expiration with the transducer placed in the fossa between the sternal and clavicular heads of the sternomastoid muscle[1617] [Figure 1]. The sample volume was placed in the SVC of supine patients at a depth of approximately 5 cm to obtain an optimal Doppler flow signal. Peak values in inspiration were used to obtain peak antegrade flow during ventricular systole (S), peak antegrade flow during ventricular diastole (D), and peak retrograde flow during right atrial contraction (AR). Left ventricular end-diastolic volume (EDV) and end-systolic volume (ESV) were measured using Simpson's method and their value was used to estimate stroke volume (SV) as the difference between EDV and ESV.

Figure 1

Pulse wave Doppler of SVC flow recorded from subclavian window

Preload expansion: Passive leg raising

After baseline measurements with the subjects lying supine, the legs were passively raised to 45 degrees. After 2 min from PLR, BP measurements (in duplicate) and evaluation of venae cavae indices and cardiac volumes as detailed above were performed during quiet respiration.

Preload reduction: Lower limb venous occlusion

With subjects lying supine, pneumatic leg cuffs around the thighs were inflated to a pressure 5 mmHg greater than diastolic brachial BP to cause venous pooling in the legs and reduce the return towards the intrathoracic compartment and the general circulation. BP readings (in duplicate) and measurements of venae cavae indices and cardiac volumes as detailed above were repeated within 10 min of the start of LVO.

In a subsample of patients, measurements were repeated after a control period of no intervention to ensure stability over time (n = 10). Finally, 20 subjects were studied during LVO on two separate occasions (day 1 and day 2) at least 1 week apart to examine the reproducibility of venae cavae indices and the response to LVO.

Statistics

Statistical analysis was performed using SPSS24 (IBM Corporation, Armonk, New York, USA). Results were expressed as mean and 95% confidence intervals (95% CI). Differences in means before and after PLR and LVO were analyzed using Students t-test for normally distributed variables or Wilcoxon rank-sum test for nonnormally distributed variables. Categorical variables were analyzed by χ2 test. For all tests, a P < 0.05 was considered statistically significant. Based on previous data collected during low-level LBNP (−10/−20 mmHg), a sample size of n >30 was chosen to provide > 90% power for a type I error rate 0.05 to detect a change in SVC peak flow velocity S >10%.

RESULTS

Subject characteristics are shown in Table 1. Subjects were predominantly young to middle-aged men and women, with the proportion of black individuals in line with the demographics of referrals to the hypertension service in southeast London. Most subjects were on pharmacological treatment with one or more antihypertensive drugs at the time of the investigations. All subjects tolerated the procedures without any noticeable adverse effects.

Table 1

Characteristics of the study population

Age (years)	47.5±14.4
Gender (% male)	56.8
Ethnicity: White (%)	45.4
Black (%)	47.7
Others (%)	6.9
BMI (kg/m2)	25.1±7.8
Diabetes (%)	6.8
Dyslipidemia (%)	22.7
CV event (%)	11.3
Creatinine (µmol/L)	87.3±18.5
HbA1c (mm/M)	38.5±6.0
Total cholesterol (mm/L)	4.8±1.2

Results are reported as mean±SD or percentages. BMI=Body mass index, CV event=Previous cardiovascular event, HbA1c=Glycosylated hemoglobin, SD=Standard deviation

No significant changes in BP and HR from baseline measurements were observed after PLR but SV increased by 6.4 (95% CI: 1.2–11.6) mL (P = 0.01). IVC diameter and collapsibility index did not change significantly after PLR, but peak flow velocity of the SVC S wave increased by 6.5 (1.6–11.3) cm/s (P = 0.01), with no significant change in the D and AR waves [Table 2]. LVO did not cause a significant change in either BP or HR but reduced SV by 7.8 (3.5–12.1) mL. IVC diameter decreased in both expiration and inspiration after LVO by 3.2 (1.7–4.7) mm and 1.6 (0.8–2.4) mm, respectively (each P < 0.01) with no change in collapsibility index. Peak flow of the SVC S wave decreased by 9.7 (4.4–14.7) cm/s after LVO (P < 0.001) with no significant change in the D and AR waves [Table 2].

Table 2

Effects of passive leg raising and lower limb venous occlusion on blood pressure, stroke volume, and indices of cardiac preload

Variable	Baseline	PLR	P	LVO	P
SBP (mmHg)	141.1±17.8	139.6±19.1	0.339	138.8±16.9	0.107
DBP (mmHg)	87.1±12.8	86.3±14.2	0.464	87.4±13.2	0.790
HR (bpm)	71.8±12.9	69.4±12.0	0.053	70.2±13.2	0.205
EDV (mL)	115.4±25.1	123.74±27.3	0.008	103.4±20.2	<0.001
ESV (mL)	49.1±11.9	51.1±13.2	0.265	44.9±15.8	0.020
SV (mL)	66.3±15.4	72.7±18.2	0.017	58.5±9.9	0.001
IVCe (mm)	14.7±4.2	15.4±3.7	0.234	11.5±3.2	<0.001
IVCi (mm)	10.3±3.3	10.7±3.5	0.217	8.7±2.7	<0.001
IVC CI (%)	30.7±10.2	32.4±10.0	0.626	26.4±12.4	0.071
S (cm/s)	61.5±20.3	68.1±19.7	0.010	52.3±16.8	<0.001
D (cm/s)	42.3±17.9	43.7±21.1	0.874	40.8±16.7	0.304
AR (cm/s)	20.7±5.8	21.6±5.9	0.211	19.2±5.2	0.071

Results are reported as mean±SD. SBP=Systolic blood pressure, DBP=Diastolic blood pressure, HR=Heart rate, IVCi=Inferior vena cava diameter in inspiration, IVCe=Inferior vena cava diameter in expiration, CI=Collapsibility index, S=Superior vena cava peak flow velocity S, D=Superior vena cava peak flow velocity D, AR=Superior vena cava peak flow velocity AR, EDV=End diastolic volume, ESV=End systolic volume, SV=Stroke volume, PLR=Passive leg raising, LVO=Limb venous occlusion, SD=Standard deviation

In 10 subjects who were studied before and after a control period of 10 min during which no intervention was performed, there was no significant change in IVC or SVC-derived indices. In twenty consecutive subjects (mean age 41.6 ± 13.4 years, 75% male), BP-, HR-, and IVC- and SVC-derived indices and their change after LVO were measured on 2 days at least 1 week apart. The changes in IVC diameter and SVC S wave after LVO measured on separate occasions were similar [P > 0.1, Table 3].

Table 3

Effects of lower limb venous occlusion on blood pressure, and indices of cardiac preload performed on two separate occasion in (n=20) subjects

Variable	LVO (day 1)		LVO (day 2)
	Baseline	Change	Baseline	Change
SBP (mmHg)	141.83±11.05	−4.67±7.55	142.65±10.93	−4.82±7.88
DBP (mmHg)	91.75±10.28	1.58±4.01	92.82±9.11	1.53±4.43
HR (bpm)	68.73±12.03	−0.8±8.24	67.18±12.91	0.24±9.31
IVCe (mm)	14.56±4.76	−2.82±4.62	13.06±3.46	−2.61±2.30
IVCi (mm)	9.79±2.93	−1.71±2.05	6.7±3.13	−1.50±1.72
S (cm/s)	56.10±14.34	−6.42±7.45	61.16±15.68	−6.52±6.77
D (cm/s)	33.22±11.43	−3.87±9.40	33.89±10.53	−2.03±10.67
AR (cm/s)	21.06±6.89	−2.84±3.23	23.50±4.90	−1.87±5.10

Results are reported as mean±SD. SBP=Systolic blood pressure, DBP=Diastolic blood pressure, HR=Heart rate, IVCi=Inferior vena cava diameter in inspiration, IVCe=Inferior vena cava diameter in expiration, S=Superior vena cava peak flow velocity S, D=Superior vena cava peak flow velocity D, AR=Superior vena cava peak flow velocity, LVO=Limb venous occlusion, SD=Standard deviation

DISCUSSION

PLR is a test that can be used in critical care to predict whether cardiac output will increase with volume expansion[18] since by transferring venous blood from the lower body toward the right heart, it mimics the infusion of fluid.[19] A radionuclide method suggests that, in healthy volunteers, a volume of approximately 150 mL can be transferred from the lower body with PLR.[20] The transfer of blood from the legs operates through the splanchnic venous network to increase right ventricular (RV) preload[2122] which, depending on RV function,[23] usually increases LV preload and SV,[2425] as observed in the present study.

LVO is a novel technique to reduce cardiac preload. The basic principle of LVO is similar to LBNP and the so-called congesting cuffs or “rotating tourniquets” used to treat patients with acute pulmonary edema[26] which operate through mechanical reduction of preload.[27] LBNP has been used extensively as an experimental tool to modulate preload, but the technique has the disadvantage of being a cumbersome procedure that can only be used in patients of limited body mass (depending on the lower body enclosure) and it does not usually allow the interrogation of IVC. In our experimental protocol, LVO produced a significant decrease in SV (approximately 8 mL) with no change in BP or HR. The decrease in SV (as percentage of baseline value) is comparable to that obtained during a low level of LBNP (-15/-20 mmHg)[2829] which has been associated with a modest fluid displacement of approximately 300–500 mL.[9] PLR and LVO are therefore ideal tests to examine the sensitivity of venae cavae indices to modulation in preload, as suggested by the changes in SV.

During PLR, the only test to show a consistent increase in preload was the S wave of the SVC flow velocity waveform. IVC diameters did not increase significantly, possibly because of a limited capacity of the already filled IVC to expand further or methodological limitations as highlighted in patients undergoing fluid expansion.[30] Collapsibility index was not affected by the intervention and the mean baseline value was also similar to those reported by other authors.[3132] Conversely, after LVO, there was a significant reduction in both IVC diameters (with no change in collapsibility index) and decrease in the SVC S wave.

Reproducibility of IVC diameters and of the SVC S wave was also assessed through sham procedure and repeated measurements on separate visits. Since in hypertensive states, the main interest is in assessing response to volume depletion, we also examined the reproducibility of the response to LVO on 2 separate days and found that this to be repeatable. However, under resting conditions, reproducibility of baseline SVC and IVC measurements was suboptimal, suggesting that they might be of limited value for individual subjects, although still useful for detecting changes in cardiac preload.

The present study is subject to several important limitations. First, the conclusions relate only to hypertensive individuals with relatively modest departure from normal cardiac physiology and the effects in healthy individuals would need to be investigated. Second, the results can only be interpreted in terms of an acute change in preload and a number of compensatory homeostatic mechanisms are expected to influence hemodynamics during a long-term change in preload. The majority of subjects were on pharmacological treatment for hypertension at the time of the investigations and drug this may have affected the response to PLR and LVO, reproducibility of the measurements in untreated subjects and comparison between different operators is warranted. Suboptimal reproducibility within individuals may also be limited by methodological factors that are difficult to control despite the effort made in the present study to standardize acquisition. A comprehensive assessment of the RV function was not performed during PLR and LVO.

In conclusion, results of the present study support the concept that venae cavae derived indices can be used to assess changes in preload within physiological range in clinically euvolemic subjects with primary hypertension. LVO is also a novel simple and well-tolerated procedure to decrease preload and could be used as a test to predict hemodynamic responses to preload.

Ethical clearance

The study was approved by the London Westminster Research Ethics Committee.

Financial support and sponsorship

This work was performed as part of the AIM HY (Ancestry and biological Informative Markers in stratification of Hypertension) stratified medicines program in hypertension funded by the Medical Research Council and The British Heart Foundation (MR/M016560/1). We also acknowledge support from the Department of Health via a National Institute for Health Research (NIHR) Biomedical Research Centre and Clinical Research Facility award to Guy's & St Thomas' NHS Foundation Trust in partnership with King's College London, and the NIHR Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King's College London.

Conflicts of interest

There are no conflicts of interest.