Use of ultrasound-measured internal jugular vein collapsibility index to determine static intracardiac pressures in patients with presumed pulmonary hypertension

We conducted a prospective study of 71 consecutive patients undergoing RHC as part of routine evaluation for pulmonary hypertension (PH) at Boston University Medical Center, Boston, MA, USA. The study was approved by the Institutional Review Board of Boston University Medical Center and the requirement for informed consent was waived.

Clinical and laboratory data were collected for all patients before and during the procedure. Collected data from the electronic medical record included past medical history and laboratory tests. Procedural data included ultrasound measurements and RHC measurements (Fig. 1).

RHC was performed in a fluoroscopy-equipped catheterization laboratory with patients positioned at the level of their respiratory comfort; 62 (87.3%) participants were supine for the entirety of the procedure. Using a bedside two-dimensional ultrasound (Sonosite, Washington, USA), the diameter of the IJ at rest, during respiratory variation (inspiration and expiration), and during manual compression (measured at 2 cm of depth from the skin) was measured. All ultrasound measurements were completed prior to the RHC. Trained Pulmonary/Critical Care physicians with certification to complete central venous and pulmonary artery catheterizations performed the ultrasound measurements and the subsequent RHC.

IJ measurements were recorded with the linear transducer probe. The IJ was visualized by placing the ultrasound transducer perpendicular to the skin in the transverse plane at a level just above the clavicle. The IJ was identified by compression as well as by color Doppler imaging and Pulse Wave Doppler. Sufficient ultrasound gel was used to prevent direct skin contact with the transducer, helping to limit the amount of pressure applied and avoid significant influence of external compression on the IJ diameter at rest [7]. Measurements were obtained by an M-mode scan on the ultrasound device. The maximum anterior–posterior diameter of the IJ was measured at rest followed by variations during the respiratory cycle, where the patient was asked to take a deep breath in, breathe out, and then hold their breath at the end of expiration for one second. Subsequently, minimal pressure (measured via ruler at 2 cm of depth from the superficial skin) via the transducer was applied on the IJ to induce extrinsic manual compression (Fig. 2). The antero-posterior diameter of the IJ was then re-measured. Accuracy of the measurements was based on agreement between the operator and the supervising physician (HWF for all patients). The collapsibility index of respiratory variation (respiratory CI) and the collapsibility index of manual extrinsic compression (compression CI) were computed using the following calculation: (maximum diameter − minimum diameter)/maximum diameter.

Following the ultrasound measurements, RHC was performed using best practice guidelines established by the European Respiratory Society (ERS) [17]. RHC measurements included right atrial systolic, diastolic, and mean pressures (mRAP); systolic, diastolic, and end diastolic right ventricular pressures (RVSP, RVDP, RVEDP, respectively); systolic, diastolic, and mean pulmonary artery pressures (PASP, PADP, mPAP, respectively); pulmonary artery occlusion pressure (PAOP). Cardiac output by both thermodilution (TD) and Fick methods. Cardiac index, pulmonary vascular resistance (PVR), and systemic vascular resistance (SVR) were calculated in standard fashion.

The primary outcome of the study was the correlation between respiratory CI, compression CI, and volume status defined by the RHC measurements. Additionally, we evaluated the correlation between respiratory CI, compression CI, and B-type natriuretic peptide (BNP) levels, measured pre-procedurally.

Assessing the ultrasound-measured compressibility of a vein and using it as a non-invasive marker for fluid status could provide yet another component to critical care ultrasound techniques. Volume status is a central component to manage both critically ill patients in an intensive care unit and chronically ill patients, such as those with underlying renal or cardiac disease [18, 19]. Moreover, the accuracy in estimating volume status is essential to improve patient care and therapeutic approaches. Based on the current observations, compression CI of the IJ, a simple, non-invasive technique that only requires basic ultrasound skills, could be used as a surrogate for CVP.

Use of the compression CI of the IJ provides a novel approach to non-invasive fluid status assessment; however, is it better or easier than use of the IVC as the target venous structure? IVC collapsibility is a well-documented measurement in critical care ultrasound and volume assessment, although evidence for its accuracy is controversial; moreover, in many situations, the IVC is poorly visualized [1, 2, 16]. Additionally, formal echocardiography uses the IVC diameter and collapsibility with sniff to generate an estimated, yet imprecise, assessment of the mRAP [3]. Nevertheless, the IVC has historically been used as the target venous structure; in particular, size of the IVC and its dynamic change during respiration is used to estimate mRAP [3‐9]. The current study suggests that use of the IJ may be an accurate and reliable estimation of mRAP during ultrasonography. Although we did not compare IJ and IVC measurements in the current study, the strong correlation observed between IJ diameter at rest and mRAP suggests the need for future studies comparing both ultrasound methods.

The current study has several limitations. 1) The operators were not blinded to the ultrasound and RHC assessments, which could have influenced the results. 2) The cohort included patients with known or suspected pulmonary vascular disease; none were critically ill patients in an intensive care unit. Unfortunately, given the paucity of pulmonary artery catheters inserted in intensive care units currently, it would be very difficult to perform a similar study in such patients. Likewise, we did not assess fluid responsiveness in these patients. 3) We only evaluated patients undergoing RHC in a catheterization laboratory, leading to issues with extrinsic validity and generalizability. 4) Ultrasound probe compression is a non-formal technique commonly used to differentiate venous and arterial structures when obtaining vascular access [20]. In the current study, we used an extrinsic compression of 2 cm at the level of the patient’s skin. Despite the outlined technique, user error in estimating 2 cm and an inability to directly measure the degree of compression could result in variability. Additionally, variability with patient’s body habitus could also affect the ability to provide adequate compression. Although a standardized process of applying compression has yet to be described in the literature, in this study, one individual supervised all the procedures and measurements to assure a standard protocol among the operators. Further studies that apply the use of a pressure manometer could mitigate this variability; such a pressure manometer has been utilized in other ultrasound-based investigations [21]. In a similar manner, standardizing the maneuver used to obtain the respiratory CI was a challenge as well [22].

Ultrasound measurements of the IJ, specifically compression CI, correlate with invasively measured biventricular hemodynamics in this non-critically ill population of patients with presumed pulmonary hypertension. Thus, compression CI may be a useful tool in the non-invasive estimation of intravascular volume status. Further studies of the compression CI of the IJ, as well as direct comparison to IVC, are warranted.