Simultaneous Venous-Arterial Doppler Ultrasound During Early Fluid Resuscitation to Characterize a Novel Doppler Starling Curve: A Prospective Observational Pilot Study

Abstract

Background: The likelihood of a patient being preload responsive—a state where the cardiac output or stroke volume (SV) increases significantly in response to preload—depends on both cardiac filling and function. This relationship is described by the canonical Frank-Starling curve. Research Question: We hypothesize that a novel method for phenotyping hypoperfused patients (ie, the “Doppler Starling curve”) using synchronously measured jugular venous Doppler as a marker of central venous pressure (CVP) and corrected flow time of the carotid artery (ccFT) as a surrogate for SV will refine the pretest probability of preload responsiveness/unresponsiveness. Study Design and Methods: We retrospectively analyzed a prospectively collected convenience sample of hypoperfused adult emergency department (ED) patients. Doppler measurements were obtained before and during a preload challenge using a wireless, wearable Doppler ultrasound system. Based on internal jugular and carotid artery Doppler surrogates of CVP and SV, respectively, we placed hemodynamic assessments into quadrants (Q_x) prior to preload augmentation: low CVP with normal SV (Q₁), high CVP and normal SV (Q₂), low CVP and low SV (Q₃) and high CVP and low SV (Q₄). The proportion of preload responsive and unresponsive assessments in each quadrant was calculated based on the maximal change in ccFT (ccFT_Δ) during either a passive leg raise or rapid fluid challenge. Results: We analyzed 41 patients (68 hemodynamic assessments) between February and April 2021. The prevalence of each phenotype was: 15 (22%) in Q₁, 8 (12%) in Q₂, 39 (57%) in Q₃, and 6 (9%) in Q₄. Preload unresponsiveness rates were: Q₁, 20%; Q₂, 50%; Q₃, 33%, and Q₄, 67%. Interpretation: Even fluid naïve ED patients with sonographic estimates of low CVP have high rates of fluid unresponsiveness, making dynamic testing valuable to prevent ineffective IVF administration.

Keywords

fluid responsiveness fluid tolerance carotid Doppler venous Doppler functional hemodynamic monitoring passive leg raise

Background

In the emergency department (ED), intravenous (IV) fluids are increasingly guided by venous or arterial measurements from point-of-care ultrasound.¹ Although the respiratory variation of the inferior vena cava (IVC)² is a surrogate of central venous pressure (CVP)³ and is an accepted method for estimating right ventricular systolic pressure during echocardiography,⁴ there is an inconsistent relationship between absolute CVP and an increase in cardiac output or stroke volume (SV) after IV fluid administration. For instance, 25% of patients with a CVP ≤ 5 mm Hg,⁵ 40% of patients with a CVP ≤ 8 mm Hg⁶ and 20% of patients with an IVC collapse of ≥ 25%⁷ are reported to be preload unresponsive. This means that a clinically significant fraction of patients with low venous pressure do not have the intended response to IV fluid.⁸

One arterial ultrasound measure used to guide IV fluid is the corrected flow time of the carotid artery (ccFT).^9,10 This relies on the direct relationship between the duration of systole and SV—an association made in the mid-20th century by Weissler et al.¹¹ Despite the ccFT having excellent diagnostic accuracy for predicting preload responsiveness reported in a recent meta-analysis,¹⁰ much of this data is based on an absolute ccFT threshold. In other words, a low ccFT (ie, suggesting low SV) predicts preload responsiveness and vice versa. Barjaktarevic and colleagues assessed ccFT from a dynamic approach, finding that a +7 millisecond (ms) ccFT change (ccFT_Δ) induced by passive leg raise (PLR) best-dichotomized preload responders from nonresponders.¹²

Often venous and arterial ultrasound measures described above are performed in isolation, however, it has been argued that they should be assessed in tandem by a wireless, wearable Doppler device.^13–16 Simultaneous Doppler ultrasound of the jugular vein (ie, CVP surrogate) and carotid artery (ie, SV surrogate) create a “Doppler Starling curve”^15,16 which roughly resembles the canonical “Frank-Starling” curve, but can be noninvasively derived at the bedside. By dichotomizing jugular venous Doppler morphology into “low” and “high” CVP surrogates and ccFT into “low” and “normal” SV surrogates, the Doppler Starling curve is divided into 4 quadrants that phenotype hypoperfused patients.¹⁵

Our objectives were to determine the prevalence of each hemodynamic profile and to determine the rate of preload unresponsiveness for each profile. We had four hypotheses with respect to this novel Doppler Starling curve in fluid naïve, hypoperfused ED patients. First, we predicted that most assessments would fall into the quadrants defined by a “low CVP” jugular Doppler morphology given that these were ED patients studied early in their care. Second, we anticipated that all assessments defined by low CVP would have a 20% to 40% preload unresponsiveness rate given what is known from the CVP and IVC collapse literature. Third, we expected that the assessments beginning with low CVP and normal SV Doppler surrogates would have the highest proportion of preload responsiveness. Finally, we predicted that the quadrants defined by high CVP jugular morphology would have preload unresponsive rates comparable to the intensive care unit (ICU) population (ie, at least 50% unresponsive).¹⁷

Methods

Ethics Approval and Consent

Written and informed consent was obtained from all patients or their next of kin; the study was approved by the Peoria Institutional Review Board (# 1697834-5) and performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments.

Study Design

We performed a post hoc analysis of a prospective, observational study assessing the use of a Doppler ultrasound device to determine preload responsiveness in patients in the ED. We recruited a convenience sample of patients between February and April 2021 from a community ED who were deemed to need IV fluid resuscitation by the treating clinician. The provider was blinded to the results of the wearable ultrasound. No sample size calculation was performed for this feasibility study.

Eligibility Criteria

Adult patients were enrolled if the treating clinician determined that IV fluid expansion was indicated to correct hypoperfusion. Exclusion criteria were: less than 18 years old, absent informed consent, unable to cooperate with a Doppler ultrasound assessment of the carotid artery (eg, delirium, confusion, excessive phonation, etc) or if there were anatomical contraindications precluding assessment of at least one carotid artery (eg, known bilateral carotid stenoses of at least 70%, bilateral internal jugular central lines, c-spine collar, etc). There was no exclusion for patients with atrial fibrillation, low ejection fraction, or diastolic dysfunction.

Wearable Doppler Ultrasound

The ultrasound patch (Flosonics Medical, Sudbury, ON, Canada) is a wearable, wireless, FDA-cleared, continuous-wave 4 MHz ultrasound. Adhesive straps fix the transducer angle relative to the direction of carotid blood flow.¹⁸ The wearable ultrasound displays real-time carotid corrected flow time (ccFT) (Figure 1) and, when placed over an area of anatomical overlap between the carotid and internal jugular vein, displays the internal jugular vein Doppler spectrogram.¹⁹ Placement of the device entails applying the 2 cm transducer face perpendicular to the trachea at the level of the cricoid cartilage. The device is moved laterally until the point of greatest carotid audio and visual signal is obtained via the iOS device to which the transducer is wirelessly paired. The device is then fixed in place via its incorporated adhesive.

Figure 1.

The wireless, wearable Doppler biosensor and user interface. (A) The device on a subject. (B) The user interface of the wearable Doppler showing the carotid artery and jugular Doppler simultaneously; below the spectrograms are the corrected flow time of the carotid artery (ccFT) per cardiac cycle during a preload challenge. This example shows a continuous jugular Doppler morphology with low baseline ccFT (ie, 276 milliseconds) or quadrant 3 (Q3). This example was preload responsive as the ccFT increased by 69 milliseconds. The ccFT is the duration of mechanical systole in milliseconds, corrected for heart rate by the equation of Wodey ³⁸.

Preload Augmentation

Baseline venous and arterial Doppler spectrograms were obtained in the semi-Fowler position at 45°. Prior to preload augmentation, at least 30 s of continuous Doppler spectrograms were acquired. As previously described,²⁰ preload augmentation was at the discretion of the treating clinician and could be either a 2-min PLR or rapid fluid challenge of at least 250 mL (∼4 mL/kg) administered at 100 mL/minute. Doppler spectrograms were recorded continuously throughout the entire preload challenge.

Venous Doppler Analysis

The baseline (ie, semi-Fowler) venous Doppler spectrograms were analyzed qualitatively based on the framework put forth by Kenny et al¹⁹ and Tang²¹ which is illustrated in Figure 2. Briefly, from low-to-high preload the jugular venous Doppler spectrogram progresses from a “continuous” or “pulsatile-fused” morphology toward a velocity profile that recapitulates the central venous pressure trace—with the x- and y-descents forming venous Doppler systolic “S” and diastolic “D” waves, respectively. These venous spectrogram changes have been described in the jugular vein,^22–24 superior vena cava,²⁵ IVC,²⁶ hepatic veins,²⁷ and the femoral vein.²⁸ The relationship between jugular Doppler and the central venous pressure trace has been previously reviewed.²⁴ Three experts in venous Doppler interpretation (R.P., P.R., and K.H.) identified the jugular venous morphology at baseline for each preload challenge by consensus; each expert was blind to the results of the preload challenge and each other. A Fleiss Kappa score was calculated to assess agreement between the blinded experts.

Figure 2.

Venous Doppler analysis: this figure shows the scoring system for the internal jugular vein Doppler. “Pulsatile-fused” refers to fusion of the systolic (S) and diastolic (D) waves, that is, the v-wave is not great enough to cleave the S and D waves as seen at higher CVP. This scoring system is dependent on patient position, for instance, S > D is common in healthy volunteers when supine or head down ¹⁹, however, not in the upright position.

It was previously shown¹⁹ that all healthy volunteers display a “continuous” or “pulsatile fused” jugular venous Doppler morphology with low preload, whereas systolic (S) > diastolic (D) wave jugular Doppler was only ever observed with increased preload. Given these findings, we categorized the “continuous” and “pulsatile fused” jugular venous Doppler morphology as “low” CVP; “high” CVP is considered S > D morphology or greater (ie, S > D; S = D; S < D or monophasic D) when patients are in the semirecumbent position (Figure 2). Venous Doppler spectrograms were interpreted at the point of minimum jugular velocity, which corresponds to end-expiration (ie, when the jugular vein rounds out) in spontaneously breathing patients.²⁹

Arterial Doppler Analysis

The carotid arterial Doppler spectrograms were analyzed for the ccFT at baseline (ie, to place the assessment on the Doppler Starling curve) and ccFT_Δ (ie, to determine preload responsiveness/unresponsiveness) from baseline to peak effect during preload challenge. The number of cardiac cycles averaged before and during preload augmentation was dictated by the coefficient of variation of each particular measure to ensure change could be detected with statistical confidence.³⁰ The device software is able to perform these arterial calculations in real-time, however, all data was analyzed later, offline, given that venous Doppler interpretation was required per above. The primary determinant of whether or not a patient was deemed preload responsive was the threshold reported by Barjaktarevic et al (ie, +7 ms ccFT_Δ).¹²

Doppler Starling Curve

To place patients on the Doppler Starling curve in the baseline position, each assessment was parsed into “low” or “high” CVP morphology per above; this dichotomy defined the x-axis of the Doppler Starling curve. For the y-axis, Kenny et al. have previously shown that prior to PLR in healthy volunteers, the average ccFT based upon Wodey's equation was 310 ± 10 ms in the semirecumbent position.³¹ We therefore used this value as a surrogate to dichotomize “low” from “normal” SV on the y-axis from arterial Doppler taken simultaneously with the venous Doppler. Accordingly, the “Doppler Starling curve” is composed of 4 quadrants (Figure 3). Quadrant 1 (Q₁) is defined by patients with a “low” CVP jugular morphology and a baseline ccFT ≥ 310 ms. Quadrant 2 (Q₂) comprises patients with a “high” CVP jugular morphology and baseline ccFT ≥ 310 ms. Quadrant 3 (Q₃) are patients with “low” CVP jugular Doppler and a baseline ccFT < 310 ms while quadrant 4 (Q₄) are patients with “high” CVP jugular Doppler and a baseline ccFT < 310 ms (Figure 3). Placing patients in the baseline Doppler Starling quadrant was performed offline given that human interpretation of the jugular venous morphology is required. We performed chi-squared testing to assess interactions between preload responsiveness and quadrant number and preload responsiveness versus “low” or “high” CVP jugular venous pattern.

Figure 3.

The Doppler Starling curve. See the text for how quadrants 1 to 4 are defined. The two curves depicted represent extremes, there are an infinite number of possible curves between them.

Results

Patients

In total 53 patients were enrolled. Twelve patients (23%) were entirely excluded for poor arterial or venous Doppler signal (ie, no clear dicrotic notch or absent venous Doppler signals); 41 patients and 68 assessments comprise this analysis. The disposition of the patients in this analysis from the ED included: 12% to the ICU; 61% to the general medical or surgical floor; 5% to the short-stay unit, and 22% discharged from the ED. The 28-day readmission and mortality rates for the 41 patients were 24% and 5%, respectively. The baseline clinical characteristics of the 41 included patients are listed in Table 1. The breakdown in these clinical characteristics for each of the 4 quadrants is shown in supplemental e-Table 1.

Table 1.

Baseline Patient Characteristics.

Number of patients	41
Female sex, n (%)	24 (59%)
Age, mean (SD)	62 (17.5)
Body mass index (SD)	29.3 (7.8)
IV fluids prior to enrollment (mL/kg), mean (SD)	1.6 (3)
Comorbidities, n (%)
Congestive heart failure	5 (12%)
Ejection fraction*, mean (SD)	57.2 (14.7)
Chronic kidney disease	7 (17%)
Chronic obstructive pulmonary disease	9 (22%)
Diabetes	11 (27%)
Cirrhosis	1 (2%)
Malignancy	3 (7%)
Immunosuppressed**	9 (22%)
Presentation
Infection or sepsis suspected, n (%)	33 (80%)
Heart rate (beats per minute), mean (SD)	102 (22.4)
Mean arterial pressure (mm Hg), mean (SD)	79 (18.2)
Temperature (°C), mean (SD)	37.4 (1.0)
Oxygen saturation (%), mean (SD)	96 (3)
Corrected flow time (ms), mean (SD)	296 (27)
Indication for IV fluids, n (%)
Hypotension	31 (76%)
Tachycardia	27 (66%)
Low urine output	0 (0%)
Acute kidney injury	10 (24%)
Elevated lactate	8 (20%)
Confusion	4 (10%)
Other	29 (71%)

*Ejection fraction within 12 months of ED presentation, if available.

**Defined as active malignancy, AIDS or receiving immunosuppressive medications.

Doppler Starling Curve

In the baseline, semirecumbent (ie, semi-Fowler) position, 15 (22%) assessments were in Q₁ (ie, low CVP, normal SV surrogates). Q₂ (ie, high CVP, normal SV surrogates) included 8 (12%) assessments. Q₃ (ie, low CVP, low SV surrogates) included 39 (57%) assessments. Finally, Q₄ (ie, high CVP, low SV surrogates) included 6 (9%) assessments.

Fifty-four (79%) assessments were in either Q₁ or Q₃ (ie, low CVP surrogate) while the remaining 14 (21%) assessments were in either Q₂ or Q₄ (ie, high CVP surrogate). Twenty-three (34%) assessments were initially categorized being in Q₁ or Q₂ (ie, normal ccFT, SV surrogate) while the remaining 45 (66%) were in either Q₃ or Q₄ (ie, low ccFT, SV surrogate). The Fleiss Kappa coefficient for agreement between experts was moderate at 0.44 (ie, moderate agreement).

Preload Responsiveness Versus Unresponsiveness

The fraction of assessments with preload responsiveness and unresponsiveness based on the +7 ms ccFT_Δ threshold for each quadrant is shown in Figure 4. Tables 2 and 3 break down these fractions based upon other baseline measures. By chi-squared testing, there was no statistically significant difference between preload responsiveness and quadrant number (p = .18) nor between preload responsiveness and “low” (ie, Q₁ and Q₃) versus “high” (ie, Q₂ and Q₄) CVP jugular venous patterns (p = .07).

Figure 4.

The fraction of preload responsive and unresponsive assessments per quadrant. The green (vertical) arrow represents the fraction of responsive assessments while the red (horizontal) arrow represents the fraction of unresponsive assessments in each quadrant.

Table 2.

Preload Responsiveness per Quadrant. Preload Responsiveness Was Defined by ccFT_Δ During a Preload Challenge.

	Quadrant 1 (↓CVP, $⟷$ SV)	Quadrant 2 (↑CVP, $⟷$ SV)	Quadrant 3 (↓CVP, ↓SV)	Quadrant 4 (↑CVP, ↓SV)
Responsive	80%	50%	67%	33%
Unresponsive	20%	50%	33%	67%

Abbreviations: CVP, central venous pressure; SV, stroke volume.

Table 3.

The Fraction of Responsive and Unresponsive Patients Based upon CVP or SV Surrogates Alone. Preload Responsiveness Was Defined by ccFT_Δ During a Preload Challenge.

	Q₁ and Q₃ (↓CVP)	Q₂ and Q₄ (↑CVP)	Q₁ and Q₂ ( $⟷$ SV)	Q₃ and Q₄ (↓SV)
Responsive	70%	43%	70%	62%
Unresponsive	30%	57%	30%	38%

Abbreviations: CVP, central venous pressure; SV, stroke volume.

Discussion

Precision fluid management is challenging during resuscitation; noninvasive approaches that link fundamental physiology to the bedside could enhance clinical practice. Thus, our observational, pilot study is important from three perspectives: (1) our data supports the conceptual framework of a “Doppler Starling curve,” (2) we demonstrated it is feasible to noninvasively phenotype hypoperfused ED patients based upon CVP and SV surrogates, and (3) we found that venous-arterial Doppler identified 79% of assessments in our study cohort as having “low CVP” (ie, Q₁ and Q₃) and that 30% of these assessments were preload unresponsive.

From a basic physiology perspective, our hypotheses derive from two fundamental aspects of the Frank-Starling relationship: (1) with increasing preload, there is increasing SV until a plateau is reached³² and (2) there can be a variety of different slopes defining this curve (eg, “steep,” “shallow,” “flat”) as illustrated in Figures 3 and 4. Given the general shape of the Starling relationship (ie, initial upslope followed by flattening), preload responsiveness should be more likely when CVP is low (eg, Q₁ and Q₃), however, it is not guaranteed; a plateau can be reached with low preload and this phenotype is expected to be more common in Q₃.¹⁵ For example, if we restrict our analysis to patients with low CVP jugular Doppler morphology at baseline (ie, Q₁ and Q₃), the unresponsiveness rate is 30% which is consistent with CVP and IVC collapse literature.^5,7 If we refine these patients further, unresponsiveness was more common in Q₃ than in Q₁, which is predicted by the shape of the Starling curve, though this difference was not statistically significant. On the other hand, “high CVP” jugular morphology at baseline predicted a fluid unresponsiveness rate comparable to the ICU (ie, at least 50%).⁵ While these differences were not statistically significant, our pilot investigation was underpowered and conceptualized as hypothesis-generating.

At the bedside, the 4 quadrants of the Doppler Starling curve may help the clinician narrow the pretest probability for preload responsiveness prior to performing a preload challenge (eg, PLR or rapid fluid challenge) and track the Doppler Starling curve in real-time. For example, a hypotensive patient with a baseline ccFT of 275 ms and a continuous jugular morphology is placed in Q₃. Though there is a “low CVP” jugular Doppler, this patient still has a 33% chance of being preload unresponsive, consistent with CVP data.⁵ When a PLR is performed (Supplemental Video 1), the ccFT falls consistent with preload unresponsiveness. Furthermore, during the PLR, the jugular venous morphology transitions to a “high CVP” pattern; the patient moves from Q₃ to Q₄ during the preload challenge and this simplifies the cumbersome workload typically required when performing a PLR with Doppler ultrasound.³³ Additionally, the baseline Doppler Starling quadrant could hint at the etiology of hypoperfusion. For example, a hypotensive patient in Q₁ suggests vasodilation while in Q₄, cardiac dysfunction is highly likely. Additional research should be directed at this question and especially this constellation of low ccFT with high CVP pattern.

Limitations

The key limitation of this pilot investigation is that we did not measure CVP or SV. In theory, the same framework could be used in conjunction with direct measures of CVP and SV or other ultrasonographic surrogates (eg, IVC size/variation, left ventricular outflow tract velocity time integral). Absolute ccFT as a proxy for SV is a limitation requiring specific elaborations. First, although there is a direct relation between systolic time and SV,^11,34 systolic time is also affected by afterload and contractility.^35,36,37 Accordingly, the absolute value of ccFT is analogous to pulse pressure. A low absolute ccFT (or pulse pressure) suggests diminished SV and vice versa, but these relationships are mediated by other competing variables; therefore, a dynamic measure (eg, ccFT_Δ during a preload challenge¹²) adds important hemodynamic data regarding the slope of the cardiac function curve. Additionally, when comparing absolute ccFT between investigations, the value obtained by the equations of Bazett versus Wodey is not interchangeable^38,39; the latter was used in this analysis. Only moderate agreement between the blinded observers for dichotomizing the jugular venous morphology into “low” versus “high” CVP patterns is another important limitation. Nearly all of the disagreement was distinguishing between the “pulsatile fused” and “S > D wave” jugular morphologies, that is the jugular venous signals that border the “low” versus “high” preload conditions. Future studies should include more objective criteria for distinguishing what qualifies S/D wave cleavage. Physiologically, the distinction between the “pulsatile fused” and “S > D wave” jugular morphologies is a function of the height of the right atrial pressure v-wave. Simultaneous comparison with invasive measures could codify this distinction. This would be particularly interesting in patients with elevated pulmonary pressures in acute and chronic settings (eg, chronic obstructive pulmonary disease and acute pulmonary embolism), though previous investigations have identified right heart dysfunction via Doppler of the internal jugular vein and superior vena cava.^24,40,41 There were statistical limitations to our study as well. First, repeated measures on the same patients could mean that the measures are not independent. Though physiology changes rapidly in response to various interventions in the ED and, therefore, might be considered an independent phenomenon, this is not clear cut. Future studies of this framework should include a larger sample size of patients in each of the 4 quadrants without repeated measures. Second, the sample size of our pilot investigation precludes multivariable regression to determine which demographics are associated with the presence of each hemodynamic profile. We suspect that the lack of a statistically significant difference between responsive and unresponsive subjects between the quadrants likely reflects an underpowered investigation. Another potential limitation is that B-mode imaging was not used for placement of the wearable transducer; therefore, positioning was “blind” in this sense. Nevertheless, the CW ultrasound produces a sonic curtain over 2 cm in length, making uniform insonation of all red blood cell velocities across a 6 mm to 7 mm carotid artery very likely when there is good visual and audio feedback from the device.⁴² To the extent that the internal jugular and common carotid arteries are anatomically separated, establishing simultaneous signals from both is problematic and could account for the 12 excluded subjects in whom both signals were unattainable. Because we did not track the number of patients who were excluded (ie, the total source population), bias is introduced if inherent differences between the source population and our study cohort exist. Finally, the reliance on the clinician decision to administer fluid bolus as an inclusion criterion may have introduced selection bias as we are not obtaining a comprehensive set of patients with one clinical entity, but rather, a heterogeneous group of patients.

Conclusions

This pilot data supports the construct of a Doppler Starling curve, the simultaneous assessment of venous and arterial Doppler ultrasound measures during a hemodynamic intervention, and how Doppler ultrasound can help clinicians determine the pretest probability of preload responsiveness during resuscitation. The Doppler Starling framework may also suggest hypoperfusion etiology, and further investigation and validation of this framework are ongoing.

Supplemental Material

sj-docx-1-jic-10.1177_08850666231224396 - Supplemental material for Simultaneous Venous-Arterial Doppler Ultrasound During Early Fluid Resuscitation to Characterize a Novel Doppler Starling Curve: A Prospective Observational Pilot Study

Supplemental material, sj-docx-1-jic-10.1177_08850666231224396 for Simultaneous Venous-Arterial Doppler Ultrasound During Early Fluid Resuscitation to Characterize a Novel Doppler Starling Curve: A Prospective Observational Pilot Study by Jon-Émile S. Kenny, Ross Prager, Philippe Rola, Korbin Haycock, Stanley O. Gibbs, Delaney H. Johnston, Christine Horner and Joseph K. Eibl, Vivian C. Lau, Benjamin O. Kemp in Journal of Intensive Care Medicine

Supplemental Material

Footnotes

Abbreviations

Acknowledgments

Not applicable.

Authors’ Contributions

JESK: conception, study design, analysis and interpretation, and drafting; RP: conception, analysis and interpretation, and critical revisions; PR: conception, analysis and interpretation, and critical revisions; KH: conception, analysis and interpretation, and critical revisions; SOG: data acquisition, analysis and interpretation, and critical revisions; JKE: conception, design, data acquisition, analysis, and critical revisions; CH: analysis and interpretation, and critical revisions; DHJ: data acquisition, analysis, and critical revisions; VCL: conception, data acquisition, analysis and interpretation, and critical revisions; BOK: conception, data acquisition, analysis and interpretation, and critical revisions. All authors read and approved the final manuscript.

Declaration of Conflicting Interests

JESK, SOG, JKE, CH, and DHJ work for Flosonics Medical, the start-up building the wearable Doppler ultrasound. RP, PR, KH, VCL, and BOK declare no competing interests.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Jon-Émile S. Kenny

Philippe Rola

Supplemental Material

Supplemental material for this article is available online.

References

Yoshida

Noma

Nomura

Suzuki

Mihara

. Diagnostic accuracy of point-of-care ultrasound for shock: A systematic review and meta-analysis. Crit Care. 2023;27(1):200.

Orso

Paoli

Piani

Cilenti

Cristiani

Guglielmo

. Accuracy of ultrasonographic measurements of inferior vena cava to determine fluid responsiveness: A systematic review and meta-analysis. J Intensive Care Med. 2020;35(4):354-363.

Kircher

Himelman

Schiller

. Noninvasive estimation of right atrial pressure from the inspiratory collapse of the inferior vena cava. Am J Cardiol. 1990;66(4):493-496.

Amsallem

Sternbach

Adigopula

, et al. Addressing the controversy of estimating pulmonary arterial pressure by echocardiography. J Am Soc Echocardiogr. 2016;29(2):93-102.

Magder

Bafaqeeh

. The clinical role of central venous pressure measurements. J Intensive Care Med. 2007;22(1):44-51.

Eskesen

Wetterslev

Perner

. Systematic review including re-analyses of 1148 individual data sets of central venous pressure as a predictor of fluid responsiveness. Intensive Care Med. 2016;42(3):324-332.

Corl

George

Romanoff

, et al. Inferior vena cava collapsibility detects fluid responsiveness among spontaneously breathing critically-ill patients. J Crit Care. 2017;41:130-137. doi:10.1016/j.jcrc.2017.05.008

Marik

Baram

Vahid

. Does central venous pressure predict fluid responsiveness? A systematic review of the literature and the tale of seven mares. Chest. 2008;134(1):172-178.

Beier

Davis

Esener

Grant

Fields

. Carotid ultrasound to predict fluid responsiveness: A systematic review. J Ultrasound Med. 2020;39(10):1965-1976.

10.

Singla

Gupta

Varshney

Mangla

Walikar

Jamir

. Role of carotid corrected flow time and peak velocity variation in predicting fluid responsiveness: A systematic review and meta-analysis. Korean J Anesthesiol. 2023;76(3):183-193.

11.

Weissler

Peeler

Roehll

Jr.

Relationships between left ventricular ejection time, stroke volume, and heart rate in normal individuals and patients with cardiovascular disease. Am Heart J. 1961;62:367-378. doi:10.1016/0002-8703(61)90403-3

12.

Barjaktarevic

Toppen

, et al. Ultrasound assessment of the change in carotid corrected flow time in fluid responsiveness in undifferentiated shock. Crit Care Med. 2018;46(11):e1040-e1046.

13.

Kenny

J-ÉS

Barjaktarevic

Mackenzie

, et al. Inferring the Frank–Starling curve from simultaneous venous and arterial Doppler: Measurements from a wireless, wearable ultrasound patch. Frontiers in Medical Technology. 2021;3(16):1-7.

14.

Kenny

J-ES

Eibl

Mackenzie

Barjaktarevic

. Guidance of intravenous fluid by ultrasound will improve with technology. Chest. 2021;161(2):132-133.

15.

Kenny

J-ES

. Assessing fluid intolerance with Doppler ultrasonography: A physiological framework. Med Sci (Basel). 2022;10(1):12.

16.

Kenny

Gibbs

Eibl

, et al. Simultaneous venous-arterial Doppler during preload augmentation: Illustrating the Doppler starling curve. Ultrasound J. 2023;15(1):32.

17.

Messina

Dell’Anna

Baggiani

, et al. Functional hemodynamic tests: A systematic review and a metaanalysis on the reliability of the end-expiratory occlusion test and of the mini-fluid challenge in predicting fluid responsiveness. Crit Care. 2019;23(1):264.

18.

Kenny

J-ÉS

Barjaktarevic

Eibl

, et al. A carotid Doppler patch accurately tracks stroke volume changes during a preload-modifying maneuver in healthy volunteers. Critical Care Explorations. 2020;2(1):1-6.

19.

Kenny

J-ES

Prager

Rola

McCulloch

Eibl

Haycock

. The effect of gravity-induced preload change on the venous excess ultrasound (VExUS) score and internal jugular vein Doppler in healthy volunteers. Intensive Care Med Exp. 2023;11(1):19.

20.

Kenny

Gibbs

Johnston

, et al. The time cost of physiologically ineffective intravenous fluids in the emergency department: An observational pilot study employing wearable Doppler ultrasound. J Intensive Care. 2023;11(1):7.

21.

Tang

Kitai

. Intrarenal venous flow: A window into the congestive kidney failure phenotype of heart failure? JACC: Heart Failure. 2016;4(8):683-686.

22.

Sivaciyan

Ranganathan

. Transcutaneous Doppler jugular venous flow velocity recording. Circulation. 1978;57(5):930-939.

23.

Ranganathan

Sivaciyan

Pryszlak

Freeman

. Changes in jugular venous flow velocity after coronary artery bypass grafting. Am J Cardiol. 1989;63(11):725-729.

24.

Ranganathan

Sivaciyan

. Jugular venous pulse descents patterns - recognition and clinical relevance. CJC Open. 2022;5(3):200-207. doi:10.1016/j.cjco.2022.11.016

25.

Appleton

Hatle

Popp

. Superior vena cava and hepatic vein Doppler echocardiography in healthy adults. J Am Coll Cardiol. 1987;10(5):1032-1039.

26.

Reynolds

Appleton

. Doppler flow velocity patterns of the superior vena cava, inferior vena cava, hepatic vein, coronary sinus, and atrial septal defect: A guide for the echocardiographer. J Am Soc Echocardiogr. 1991;4(5):503-512.

27.

Abu-Yousef

. Normal and respiratory variations of the hepatic and portal venous duplex Doppler waveforms with simultaneous electrocardiographic correlation. J Ultrasound Med. 1992;11(6):263-268.

28.

Abu-Yousef

Kakish

Mufid

. Pulsatile venous Doppler flow in lower limbs: Highly indicative of elevated right atrium pressure. AJR. American Journal of Roentgenology. 1996;167(4):977-980.

29.

Prager

Pratte

Kenny

Rola

. A wireless, wearable carotid Doppler ultrasound aids diagnosis and monitoring of pericardial tamponade: A case report. Crit Care Explor. 2023;5(5):e0911.

30.

Kenny

J-ÉS

Barjaktarevic

Mackenzie

, et al. Carotid Doppler measurement variability in functional hemodynamic monitoring: An analysis of 17,822 cardiac cycles. Crit. Care Explor. 2021;3(6):e0439.

31.

Kenny

J-ÉS

Horner

Elfarnawany

Eibl

. Carotid artery corrected flow time measured by wearable Doppler ultrasound accurately detects changing stroke volume during the passive leg raise in ambulatory volunteers. FBE. 2023;15(2):1-7.

32.

Berlin

Bakker

. Starling curves and central venous pressure. Crit Care. 2015;19(1):55.

33.

Kenny

Gibbs

Johnston

, et al. Continuous venous-arterial Doppler ultrasound during a preload challenge. J Vis Exp. 2023;191:1-17. doi:10.3791/64410

34.

Harley

Starmer

Greenfield

Jr.

Pressure-flow studies in man. An evaluation of the duration of the phases of systole. J Clin Invest. 1969;48(5):895-905.

35.

Chen

Liu

Fang

Xie

Zhang

Yang

. Correlation of carotid corrected flow time and respirophasic variation in blood flow peak velocity with stroke volume variation in elderly patients under general anaesthesia. BMC Anesthesiol. 2022;22(1):246.

36.

Wang

Zhao

. Carotid artery corrected flow time and respiratory variations of peak blood flow velocity for prediction of hypotension after induction of general anesthesia in elderly patients. BMC Geriatr. 2022;22(1):882.

37.

Shaver

Kroetz

Leonard

Paley

. The effect of steady-state increases in systemic arterial pressure on the duration of left ventricular ejection time. J Clin Invest. 1968;47(1):217-230.

38.

Kenny

J-ÉS

Barjaktarevic

Mackenzie

, et al. Diagnostic characteristics of 11 formulae for calculating corrected flow time as measured by a wearable Doppler patch. Intensive Care Med Exp. 2020;8(1):1-11.

39.

Mohammadinejad

Hossein-Nejad

. Calculation of corrected flow time: Wodey's formula vs. Bazett's formula. J Crit Care. 2018;44:154-155. doi:10.1016/j.jcrc.2017.10.046

40.

Ghio

Recusani

Sebastiani

, et al. Doppler velocimetry in superior vena cava provides useful information on the right circulatory function in patients with congestive heart failure. Echocardiography. 2001;18(6):469-477.

41.

Hou

Sun

D-D

Yuan

L-J

, et al. Clinical application of superior vena cava spectra in evaluation of pulmonary hypertension: A comparative echocardiography and catheterization study. Ultrasound Med Biol. 2016;42(1):110-117.

42.

Kenny

J-ÉS

Munding

Eibl

, et al. A novel, hands-free ultrasound patch for continuous monitoring of quantitative Doppler in the carotid artery. Sci Rep. 2021;11(1):1-11.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB