Brief training in ultrasound equips novice clinicians to accurately and reliably measure jugular venous pressure in obese patients

Abstract

Introduction/Purpose

Measurement of jugular venous pressure (JVP) by novice clinicians can be unreliable, particularly when evaluating obese patients. Measurement of JVP using ultrasound (uJVP) is simple to perform and provides accurate measurements. This study evaluated whether students and residents inexperienced with ultrasound could rapidly be taught to measure JVP using ultrasound in obese patients with the same accuracy as cardiologists measuring JVP via physical examination. Additionally, this study also evaluated the correlation between qualitative and quantitative JVP assessment.

Methods

This prospective, blinded study compared uJVP measurements performed by novice clinicians after brief training to JVP measurements performed by cardiologists (cJVP) on physical examination. Association between uJVP and cJVP was assessed using linear correlation, agreement and bias were assessed using the Bland–Altman analysis and inter‐rater reliability of uJVP was assessed using intraclass correlation coefficient (ICC). The association between qualitative and quantitative JVP assessment was assessed using linear correlation.

Results

Novice clinicians (n = 16) obtained 34 measurements from 26 patients (mean BMI 35.5) and reported moderate‐to‐high confidence in all measurements. uJVP correlated well with cJVP (r = 0.73) with an average error of 0.06 cm. The estimated uJVP ICC was 0.83 (95% CI = 0.44, 0.96). Qualitative uJVP had only a moderate correlation (r = 0.63) to quantitative uJVP.

Discussion

Novice clinicians often have difficulty assessing JVP on physical examination, particularly in obese patients. Our findings show a high degree of correlation between JVP measurements performed by novice clinicians using ultrasound compared with JVP measurements made by experienced cardiologists on physical examination. Furthermore, novice clinicians were able to be trained quickly, their measurements were determined to be accurate and precise and they expressed moderate‐to‐high confidence in their results.

Conclusions

After brief training, novice clinicians were able to accurately assess JVP in obese patients as compared to measurements made by experienced cardiologists on physical examination. Results suggest that ultrasound may greatly improve novice clinicians’ JVP assessment accuracy, particularly in obese patients.

Introduction

The assessment of a patient's central venous pressure (CVP) is vital to the diagnosis and management of heart failure. Direct measurement using right heart catheterisation (RHC) is the current gold standard for assessing CVP. ¹ Although effective, the invasive nature of RHC leaves patients at risk for complications such as infection and vascular injury. At the bedside, CVP can be estimated using jugular venous pressure (JVP) through visual inspection of the jugular vein and measuring the height of the sternal angle. Physical examination assessment of JVP has been found to correlate well with CVP measurements obtained on RHC. ² Unfortunately, assessing JVP through visual inspection can be difficult to perform.

Due to the complexity of JVP assessment, early learners are particularly prone to measurement errors and are frequently unable to assess JVP entirely. Common sources of error include mistaking the carotid artery for the jugular vein, inability to visualise JVP when markedly elevated or below the clavicle, and difficulty visualising JVP due to increased adiposity. In practice, JVP is frequently reported using relative, qualitative terms such as ‘elevated’, ‘halfway up the neck’ or ‘decreased/higher than yesterday’, which adds subjectivity that could be an additional source of error. A recent study found that medical students were only able to identify an elevated JVP 8% of the time in non‐obese patients, and obesity made JVP even more difficult to visualise. ³ Historically, studies have also shown only modest agreement between JVP measurements obtained by residents and attending physicians. ⁴ Consequently, JVP assessment is frequently unreliable when performed by novice clinicians, and additional training is warranted given the consequences of inaccurate assessment.

A growing body of literature has shown, however, that ultrasound can improve the accuracy of JVP assessment. ⁵ , ⁶ , ⁷ The correlation between the CVP measurements obtained via RHC and JVP measurements obtained with ultrasound (uJVP) has been extensively evaluated in the past. ⁶ , ⁸ , ⁹ In these studies, the correlation between CVP via RHC was found to be strong; a recent study of 114 patients found a correlation coefficient of 0.9 between uJVP and CVP via RHC and a mean difference between these values of only 0.87 cm demonstrating uJVP to be a reliable technique to accurately estimate cardiac filling pressure when performed by a trained ultrasound operator. ⁷ In addition, bedside ultrasound has been demonstrated to be simple to use and provides clear images of the jugular vein collapse point, even in obese patients. ⁵

Prior studies using uJVP have, however, exclusively used examiners who are experienced with ultrasound (e.g. cardiologists and trained sonographers). ⁵ , ⁶ , ⁷ Given that ultrasound can effectively be used to estimate the value of JVP regardless of body mass index (BMI), is accessible, and is simple to use, we reasoned that ultrasound may improve novice clinicians' ability to assess JVP, particularly in obese patients, and could be accomplished with a minimal amount of training. ² Consequently, we hypothesised that after brief training, clinicians who were inexperienced with ultrasound would be able to use ultrasound to assess JVP in obese patients as accurately and reliably as experienced cardiologists on physical examination.

Methods

This was a single‐centre study at a tertiary teaching hospital in Portland, Oregon (Providence Portland Medical Center), between August 2021 and November 2021. Patients admitted with a known diagnosis of heart failure, a BMI >30 kg/m² and a documented JVP measurement performed by the cardiology service were screened for enrolment. Of note, 16 different cardiologists participated in the study. No standard procedure for obtaining JVP was established; rather, cardiologists obtained their JVP measurements using their own personal routine practice, which typically included having the patient lay in the bed, raising the head to 40–45° and identifying a JVP collapse point on the right side of the neck. Rulers were not typically used, and measurements were usually estimated.

Fourth‐year medical students and first‐ and second‐year internal medicine residents served as participants in our study, all of whom identified themselves as novice ultrasound users. Prospective written informed consent was obtained from both patients and clinicians, and the Providence Health System Institutional Review Board reviewed and approved the study and consent form.

To systematise measurements, reduce inconsistencies and improve reliability, we used a uniform process for the ultrasound assessments replicating previous studies. ² , ¹⁰ This method of estimating JVP using ultrasound was first described in 1999 and has become the standard method. ⁵ Once informed consent was obtained, clinician participants individually watched a 4‐min instructional presentation describing how to measure JVP using ultrasound. Following the presentation, they were immediately brought to a participating patient's bedside. Clinicians were unaware of patients' clinical details. The head of the bed was raised to 40–45°, the patient's head was rotated to the left, and the right side of the neck was examined. A Butterfly iQ handheld ultrasound scanner on the vascular preset was used to obtain a transverse view of the internal jugular vein (IJV) just above the clavicle. The probe was then rotated 90°, orienting the indicator towards the angle of the mandible, providing a longitudinal view of the IJV. The probe was then slid superiorly on the neck to locate the area of vein collapse (Figure 1).

Figure 1.

Ultrasound image showing a longitudinal view of the internal jugular vein (IJV). The indicator is oriented superiorly. The bold white arrow indicates the IJV collapse point.

While viewing the IJV in the longitudinal view, the apex of the vein collapse at end expiration was noted and marked with a finger. Using a ruler and a straight edge, the vertical distance between this mark and the sternal angle was measured. Clinicians reported their measurements in 0.5 cm increments, and 5 cm was added to this measurement to obtain JVP. If the vein collapse was below the clavicle, JVP was recorded as <5 cm. If the vein collapse point was at or above the angle of the mandible, it was not included in this study given that a measurement of this point could not be obtained (typically occurring when JVP was >14 cm). The clinician was then asked to rate the confidence of their measurement as low, medium or high. To determine the inter‐rater reliability, eight patients were randomly selected to have their JVP assessed by 10 of the clinician participants. Each clinician performed their measurements alone and remained blinded to previous findings.

To determine whether the use of qualitative terms that describe uJVP leads to measurement error, this study also evaluated the ability of medical students and residents to perform qualitative assessment of JVP. While measuring uJVP, novice clinicians were asked to visually divide the area between the clavicle and the angle of the mandible into four quadrants (Figure 2) and estimate which quadrant contained the IJV collapse point. Given that novice clinicians have difficulty assessing JVP at baseline, we suspected that reporting JVP using relative terms would make it difficult for them to generate consistent results.

Figure 2.

Image illustrating the internal jugular vein (IJV) divided into quadrants.

Statistical analysis of findings was completed using JMP statistical software (JMP), Microsoft Excel and SAS Enterprise Guide 7.1. To assess the correlation between uJVP measurements performed by novice clinicians using ultrasound and those made by cardiologists on physical examination (cJVP), Pearson's correlation coefficient was computed. To evaluate the agreement and bias of the two measurements, the Bland–Altman agreement analysis was applied using cJVP as the reference, and a Bland–Altman plot was produced to show the mean difference and the 95% confidence interval of the lower and upper limits of agreement. Paired t‐tests were used to test for zero bias. The assigned JVP quadrant was plotted against uJVP measurements using a Box‐and‐Whisker plot. Eta correlation coefficients were used to determine and evaluate the strength of the association between the quadrants and the uJVP measurements. The intra‐class correlation coefficient (ICC) was calculated and used to assess inter‐rater reliability among the uJVP measurements.

Results

A total of 34 examinations were performed on 26 different patients. Patient characteristics revealed a slight male predominance (67%) with an average age of 68 and an average BMI of 35.5 (Table 1). Sixteen novice clinicians participated making one–two measurements on average with a maximum of four total measurements. They successfully measured JVP in all 34 examinations and reported moderate‐to‐high confidence levels in all measurements; moderate confidence was reported in 13 of 34 measurements, and high confidence was reported in 21 of 34 measurements. To assess the reliability of uJVP measurements, eight patients were sequentially examined by two different medical residents.

Table 1.

Baseline characteristics of the study population

Variable	All patients (n = 26)
Mean age, years (standard deviation)	68 (13)
Mean body mass index, kg/m2 (standard deviation)	35.5 (4.5)
Number of male participants (%)	17 (65)
Number of female participants (%)	9 (35)
Heart failure with preserved ejection fraction (%)	40
Heart failure with reduced ejection fraction (%)	60

The uJVP measurements correlated well with cJVP measurements (r = 0.73, 95% CI = 0.52, 0.86; Figure 3). The Bland–Altman plot showed the mean bias ± standard deviation between uJVP and cJVP measurements as 0.058 ± 1.43 cm. The limits of agreement were −2.75 and 2.86 (Figure 4). The test for zero bias using a paired t‐test showed a statistically insignificant result (P = 0.8151), indicating a lack of evidence for systematic bias. The ICC of uJVP measurements was calculated to be 0.83 with a 95% confidence interval of 0.44, 0.96. Trends of the uJVP measurements with assigned quadrants were described using a Box‐and‐Whisker plot (Figure 5). The Eta correlation coefficient was 0.63.

Figure 3.

Relationship between jugular venous pressure (JVP) measurements obtained by cardiologists (cJVP) and measurements obtained by medical students/residents (uJVP). Pearson's correlation coefficient between cJVP and uJVP was 0.73.

Figure 4.

Bland–Altman plot comparing jugular venous pressure (JVP) measurements obtained by cardiologists (cJVP) and measurements obtained by medical students/residents (uJVP). Mean bias was 0.058 cm with 95% limits of agreement between −2.75 and 2.86 cm.

Figure 5.

Neck quadrant assigned by novice clinicians vs. ultrasound jugular venous pressure (JVP) measurement in centimetre. The Eta correlation coefficient was calculated to be 0.63. n = 34.

DISCUSSION

Novice clinicians often have difficulty assessing JVP on physical examination, particularly in obese patients. Studies have shown that the use of ultrasound allows for accurate and reliable assessment of JVP even in obese patients ² , ³ , ⁴ , ¹⁰ when compared to a gold standard; however, these same studies have exclusively used trained sonographers, cardiology fellows or cardiologists as their examiners. ² , ³ , ⁴ , ¹⁰ We are not aware of any studies evaluating whether novice clinicians can accurately and reliably assess JVP in obese patients using ultrasound.

Our findings show a high degree of correlation between JVP measurements performed by novice clinicians using ultrasound compared with JVP measurements made by experienced cardiologists on physical examination. Furthermore, novice clinicians were able to be trained quickly, their measurements were determined to be accurate and precise, and they expressed moderate‐to‐high confidence in their results. These findings suggest that after brief training in ultrasound assessment of JVP, novice clinicians can accurately and reliably measure JVP using ultrasound, even in obese patients.

In this study, novice clinicians were also asked to place the JVP collapse point in 1 of 4 neck quadrants in order to evaluate their ability to make qualitative JVP assessments. Although the correlation between uJVP and assigned quadrants was moderate, there was a wide range of JVP values assigned to Quadrants 1, 2 and 3 with a significant overlap between values assigned to each quadrant (Figure 5). Only markedly elevated JVP values were assigned to Quadrant 4 (13 cm+). This suggests that qualitative assessment of JVP by novice clinicians may be inaccurate and that they may be unable to effectively estimate JVP unless it is markedly elevated. Consequently, novice clinicians should be encouraged to measure JVP to make an accurate assessment rather than providing a qualitative assessment.

Our study did have limitations. Given that this study used a convenience sample of patients admitted with heart failure and relied upon documented JVP measurements in the chart, our study only included patients with elevated JVP measurements. When JVP was 6 cm or less, cardiology providers at this institution typically documented the JVP as ‘normal’. Consequently, these patient records could not be identified for inclusion. To increase the generalisability of our findings, it would be beneficial to include patients with JVP measurements of 6 cm or less as inclusion of a wider range of measured JVP values may provide a more accurate assessment of medical student and resident abilities in a real‐world setting.

In addition, JVP measurements performed by cardiologists lack standardisation, which we used as our standard for comparison. The cardiologists in our study were not asked how they measured JVP, and their technique was not directly observed. Consequently, whether their JVP measurements were performed using a standardised technique or how specific techniques varied is unknown. However, since prior studies have shown that the visual inspection of JVP yields an accurate assessment of CVP when performed by experienced clinicians, we felt that the lack of a standardised JVP measurement technique would have a minimal impact on our study. ² Further investigation may be warranted to confirm this.

Nevertheless, the evidence obtained through this study provides support for incorporating uJVP measurements into curricula for internal medicine residents, particularly given the brief time required to familiarise clinicians with the procedure. Point‐of‐care ultrasound (POCUS) is being increasingly taught in medical schools and recognised as a valuable component of internal medicine residency programmes, yet it has been slow to take widespread hold. ¹¹ A national survey done in the United States in 2020 found that only one‐third of residency programmes had a formal curriculum to teach diagnostic POCUS to all internal medicine residents, and ‘faculty time’ was a common significant logistical barrier to expanding the curriculum. ¹² Our study shows that JVP assessment with ultrasound can be taught in less than 5 min and improves the diagnostic ability and confidence of early learners, thus bolstering their sense of self‐reliance and, subsequently, their ability to provide care.

Conclusions

This study found that brief training in ultrasound assessment of jugular venous pressure enabled medical students and residents to accurately assess jugular venous pressure in obese patients and that their measurements correlated well with jugular venous pressure measurements made by experienced cardiologists on physical examination. Results suggest that utilising ultrasound may greatly increase the ability of novice clinicians to accurately measure jugular venous pressure in obese patients. In addition, these successes helped reinforce confidence in these clinicians leaving them feeling more self‐reliant. A wider study incorporating more patients and standardised control assessments would be prudent to confirm our findings; however, our analysis does suggest that internal medicine residency programmes should consider incorporating jugular venous pressure measurement into point‐of‐care ultrasound curricula as there is significant value to novice providers and, most importantly, to their patients.

Author contributions

Nicholas Pettit was involved in conceptualization (lead), study design (equal), data collection (lead) and manuscript preparation (equal). Benjamin Pedroja was involved in study design (equal) and manuscript preparation (equal). Hsin Fang Li was involved in data analysis (lead). M. Sutcliffe was involved in manuscript preparation (equal) and draft manuscript preparation (lead).

Ethics statement

This project was approved by the Providence St. Joseph's Health – Institutional Review Board (Study ID: STUDY2021000406).

Funding

This research did not receive any grant funding or sponsorship.

Conflicts of interest

None to disclose.