Article, Ultrasound

The effect of supine versus upright patient positioning on inferior vena cava metrics

a b s t r a c t

Background: Ultrasound of the inferior vena cava is a noninvasive and rapidly obtainable method of intravascular Volume assessment; however not all patients can lie supine for this procedure. In this study we assess whether patient positioning affects IVC diameter metrics.

Methods: This was a prospective convenience sample of patients in an in-patient dialysis unit at an urban tertiary care center. IVC metrics taken in the Supine patient, and then at 45o semi-upright position, pre and post dialysis. Measurements were taken in M-mode in longitudinal plane roughly 2 cm below the level of the diaphragm. IVC-maximum and IVC-minimum diameter measurements were used to determine the IVC collapse index (IVC Max – IVC Min)/IVC Max). Statistics such as means, frequencies and percentages, intraclass correlation coefficient and Bland Altman summary statistics were calculated.

Results: Forty-five patients were enrolled. Average age was 57 years, 69% were male, 73% were African American, 82% had hypertension, 42% had diabetes. There was good to excellent agreement between supine and upright IVC measurements. Both the IVC minimum and maximum measurements had similar coefficient correlation (ri) measurements (0.917 and 0.890 respectively), whereas agreement in the collapse index was lower (ri = 0.813). Bland Altman analysis demonstrated excellent agreement and small 95% limits of agreement (+-6 mm) with minimal mean bias for both the minimum and maximum measurements.

Conclusion: IVC metrics do not change significantly based on patient position. For those patients who are unable to lay completely supine, a semi-upright measurement of the IVC for volume status may be an accurate alternative.

(C) 2014


acutely ill patients with complex medical conditions need rapid, accurate and comprehensive emergency care. Research indicates that those patients with early, aggressive, and goal-directed resuscitation do better than those who do not [1]. Invasive hemodynamic monitoring in these patients can be time consuming, labor-intensive, and is not without risk of serious complications. Furthermore, invasive hemodynamic monitoring can subject the patient to iatrogenic risk and the data from these invasive measures may itself be unreliable.

Ideally, hemodynamic monitoring tests should be non-invasive, rapidly obtainable, and accurate. Ultrasound to evaluate the inferior vena cava morphology and collapsibility index for volume status is

? Disclosures: None of the authors listed in this study have outside funding or support of any kind for the work, or any conflicts of interest to disclose.

?? Presentation of Data: Poster abstract at SAEM Annual Meeting, Boston, June 1-5, 2011.

* Corresponding author. University of Pennsylvania, Department of Emergency Medicine, 3400 Spruce Street, Philadelphia, PA, 19104.

E-mail addresses: [email protected], [email protected] (N.L. Panebianco).

non-invasive, rapidly obtainable, and repeatable. Prior research has shown that IVC diameter and IVC collapsibility index (IVC-CI) correlate with right atrial pressures [2,3].

Traditionally, sonographic IVC assessment is made in the supine patient. However, not all acutely ill patients can lie comfortably in this position. Data from a previous study of healthy volunteers showed no significant difference in IVC metrics when positioned at a 45? angle in the bed versus 5? Trendelenberg [4]. In the current study we sought to determine whether these findings could be extended to patients with coexisting medical conditions and in whom a variety of intravascular volume derangements are likely. For our study population we chose to use dialysis patients both before and after receiving dialysis. Previous studies have made extensive use of this patient population to establish the value of the inferior vena cava as an estimator of intravascular volume; in addition, they have been shown to be a good surrogate population for patients with volume overload [5]. Ultra- sound of the IVC typically is done with the patient in the supine position. Unfortunately, not all patients can tolerate supine position- ing. We investigated the effect of patient position (supine vs 45? upright) on US measurements of the IVC. Our study hypothesis was that there would be no difference in IVC metrics in the supine or upright position.

0735-6757/(C) 2014


N.L. Panebianco et al. / American Journal of Emergency Medicine 32 (2014) 13261329

Table 1

Patient demographics


This study took place at an urban hospital inpatient dialysis unit from August 2010 to August 2011. This was a prospective convenience sample of volunteer patients. One ultrasound fellowship trained attending and three ultrasound fellows collected data. All exams were performed on a Sonosite M-Turbo (Sonosite Inc., Bothell, WA) using a curvilinear transducer with an average frequency of 3.5 mHz. Patients were excluded if they could not lay supine, could not consent, or had wounds or dressings that would interfere with ultrasonography. No ultrasound data were shared with the treating team. Study sonographers consisted of one ultrasound fellowship trained attending and three ultrasound fellows, all of whom had performed over 600 ultrasound exams and over 50 exams of the IVC for volume status. The maximum number of patients enrolled by sonographer was 14 and minimum was 5.

Patients were enrolled prior to, or within 30 minutes of, the commencement of dialysis. Patients were approached for enroll- ment in no specific order. Pre-dialysis weight and vital signs as well as pertinent components of the medical history and reason for inpatient admission were obtained by patient interview and chart review and recorded. IVC metrics were obtained in the supine position, and then at 45? semi-upright position, pre- and post- dialysis. The patient’s angle in the bed was measured using an iPhone app (Plaincode GbR Clinometer on iPhone 2008-12). The patient was positioned so that their waist rested in the angle of the bed, the phone was placed on the metal side of the stretcher, and the head was raised until the angle reached 45?. For each subject, a cine clip of the IVC in transverse and longitudinal plane was recorded in the supine and upright position, pre and post dialysis (8 clips per subject). Additionally, a still image of an m-mode tracing in longitudinal plane was recorded in supine and upright position, pre and post dialysis (4 still images per subject).

The sonographer had discretion to use either the subxiphoid or perihepatic (anterior axillary line) intercostal window for IVC visualization; however, once chosen, the sonographer was instructed to use the same window for all measurements. Quantitative measurements of the IVC were taken in M-mode using the longitudinal plane approximately 2 cm below the level of the diaphragm. The vessel was scanned to identify the location of the maximal diameter. The maximum diameter at end-expiration (IVC-max) and the minimum diameter at end- inspiration (IVC-min) were obtained from the M-mode tracing, and these values were used to determine the IVC Collapsibility Index (IVC-CI = (IVC-Max – IVC-Min)/IVC-Max). During the analysis, each patient served as his or her own control both with respect to position and with respect to pre-and post dialysis measurements. Measurements recorded real- time on the ultrasound machine were rounded to the closest millimeter for purposes of reporting and statistical analysis. Immediately following dialysis the ultrasound exam was repeated using the same protocol. The volume of fluid added or removed during dialysis was recorded. Vital signs and IVC metrics as described above were again recorded.

Summary statistics such as means, frequencies and percentages were used to describe the population with regard to demographics, past medical history, and current reason for admission. To assess the agreement between IVC measurements (IVC-min, IVC-max, IVC-CI) taken in supine and upright position in both pre and post dialysis periods the intra-class correlation coefficient (ri) was calculated. Values obtained were considered to be in excellent agreement when ri N 0.9, good agreement when ri N 0.8, and fair agreement when ri N 0.7. Any ri b 0.7 was considered poor agreement. Bland Altman statistics were also generated to measure agreement and to assess for bias. All analyses were performed using SAS statistical software (version 9.3 SAS institute, Cary, NC). The Institutional Review Board approved this study and all patients provided informed consent.

Demographic parameter




Age in years Racial identity


mean (range)

African American/Black


57 (25-85)







Past medical history













Valve disease






Reason for admission










Abdominal pain




Forty-five patients were enrolled. Average age was 57 years, 69% were male, 73% were African American, 82% had hypertension, 42% had diabetes (Table 1). Pre-dialysis supine IVC measurements ranged from 1.4 mm minimum diameter at end-inspiration to 33.5 mm maximum diameter at end-expiration. The range of IVC measure- ments based on position, pre or post dialysis, is provided in Table 3. There was good to excellent intra-class correlation (ICC) agreement between supine and upright IVC measurements (Table 2, Figure A). Both the IVC minimum and maximum measurements had similar ri measurements (.917, .890 respectively) whereas agreement in the collapsibility index was lower (ri = .813). Additionally, Bland Altman statistics demonstrated excellent agreement with 95% limits of agreement of +-6 mm and a mean bias of zero for both the minimum and maximum measurements and the IVC-CI (Table 2, Figure B). Our results were not significantly different in the analysis of IVC metrics by position post-dialysis.

Similar analysis was performed in patients that had 2000 cc or more of fluid or more removed to determine if position matters in the setting of large volume shifts. Of the 45 patients, 25 were analyzed. Again, a high degree of agreement between IVC metrics in the supine and upright position, intra-class correlation = 0.927, 0.875, 0.847 for minimum, maximum, and IVC-CI, respectively.


This study was limited by the fact that it occurred in a single study center and was performed by a limited number of study sonographers, thus potentially limiting generalizability. In addition, our study population was 73% African American, which may represent many urban demographics but does not mirror the emergency department (ED) population in the majority of the country.

Table 2

Precision parameters for agreement between IVC measurements for supine and upright position

Parameter IVC-Min IVC-Max IVC-CI Intraclass correlation coefficient 0.917 0.890 0.813

Mean difference -0.13 -0.47 -0.72

Standard deviation of difference 3.50 3.54 15.41

Limits of agreement -7.00, 6.73 -7.40, 6.46 -30.93, 29.49

Standard error of the mean 0.53 0.54 2.35

95% CI for bias -1.20 0.94 -1.55 0.61 -5.42, 3.98

1328 N.L. Panebianco et al. / American Journal of Emergency Medicine 32 (2014) 13261329

Table 3 Inferior vena cava minimum and maximum measurements in the supine versus upright position, pre and post dialysis



Minimum: 1.4 mm

Maximum: 43 mm


Minimum: 1.3 mm

Maximum: 36.1 mm



Minimum: 0.4 mm

Maximum: 35.4 mm


In this study, we show that IVC metrics used to assess Intravascular volume status are not significantly affected by patient position. This data mirrors the prior work of the author on healthy volunteers. In this study we chose the inpatient dialysis unit as a surrogate patient population known to have intravascular volume derangements, while future study would include performing the study in the ED setting including patients who are clinically volume overloaded and volume deplete.

While we found no prior studies comparing the agreement

between supine and semi-recumbent IVC metrics there are a few studies describing IVC measurements in positions other than supine.


Minimum: 0.7 mm

Maximum: 38.8 mm

Fujii et al evaluated the IVC in patients undergoing a heads-up tilt table testing (HUT), first supine and then at a 70? head-up tilt. They noted that in patients with orthostatic dysfunction, the IVC-CI varied significantly with tilt table testing while healthy volunteers showed no significant variation in IVC-CI. These findings are likely to have

The study was conducted in an in-patient dialysis center providing a patient population with significant comorbid conditions with risk for intravascular volume derangements. While this patient population is not identical to the ED population, we felt that utilizing hemodialysis patients who were hospitalized created a model that closely reflects typical ED cases in which there are both acute and chronic illnesses and intravascular volume derangement.

The study was designed as a convenience sample of patients, with most studies occurring in the morning. It is possible that the afternoon dialysis patients differed from the morning patients. It is also possible that the patients approached for enrollment were different than those not approached.

Study sonographers were not blinded to the IVC metrics recorded. It is possible that this biased the measurements taken sequentially after the first measurement.



limited applicability to the patient population that we studied or indeed in the typical ED setting. With respect to the patients that we studied, it is unlikely that we would’ve seen changes in IVC-CI commensurate with those encountered in the HUT testing because we did not lower the legs of our patients. Furthermore, tilt table testing is performed in such a way as to promote venous pooling in susceptible individuals by inhibiting the use of the lower extremity muscles and thereby suppressing muscle pump which is physiologically respon- sible for the return of venous blood to the right side of the heart. The conditions created by HUT are therefore unlikely to be replicated in patients whether in the ED or the dialysis unit.

In a study by Nakao et al, the authors assessed the variation of IVC size based on the supine versus lateral position [6]. They found that in normal persons there is significant variation in IVC diameter, area and shape from the right lateral, supine and left lateral position. In our

Figure. A. Represents the Interclass correlation coefficient (ICC) of the Inferior vena cava minimum (left), maximum (center) and IVC Collapse Index (right) in the upright position. B. Is a Bland-Altman plot of the difference in millimeters between supine and upright position over the supine position of the IVC minimum (left), IVC maximum (center), and IVC Collapse Index (right).

N.L. Panebianco et al. / American Journal of Emergency Medicine 32 (2014) 13261329 1329

study, we did not test the effect of rolling a patient on IVC metrics. It is possible that the IVC is more sensitive to the compressive forces exerted by abdominal organs in right and left rotation than in supine versus upright.

In this study, the collapsibility index had a consistently worse agreement than the maximum or minimum measurements. This finding has been noted in prior published work [7]. It is most likely due to the multiplicative formula of the IVC-CI, by which data points are multiplied creating variances that are much larger than those between its constituent components. Future studies to determine which IVC metric is most accurate and useful in the acute care setting might further elucidate this issue.


In this study of 45 patients undergoing dialysis in an in-patient setting, there is good agreement between supine and upright IVC metrics. In those patients unable to lay flat for evaluation of the IVC for volume status, a semi-recumbent position may be used.


  1. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345(19):1368-77.
  2. Nagdev AD, Merchant RC, Tirado-Gonzalez A, Sisson CA, Murphy MC. Emergency department bedside ultrasonographic measurement of the Caval index for noninvasive determination of low central venous pressure. Ann Emerg Med 2010;55(3):290-5 [Epub 2009 Jun 25].
  3. Stawicki SP, Braslow BM, Panebianco NL, Kirkpatrick JN, Gracias VH, Hayden GE, Dean AJ. Intensivist use of hand-carried ultrasonography to measure IVC collapsibility in estimating intravascular volume status: correlations with CVP. J Am Coll Surg 2009;209(1):55-61 [Epub 2009 May 1].
  4. Rockwell M, Puthottile J, Russell C, Melniker L, Briggs W, Hayden G, Panebianco N. Patient positioning and inferior vena caval indices. Acad Emerg Med 2009;16(4 Suppl) s31 [Abstract only].
  5. Brennan JM, Ronan A, Goonewardena S, Blair JE, Hammes M, Shah D, Vasaiwala S, Kirkpatrick JN, Spencer KT. Handcarried Ultrasound measurement of the inferior vena cava for assessment of intravascular volume status in the outpatient hemodialysis clinic. Clin J Am Soc Nephrol 2006;1(4):749-53 [Epub 2006 May 24].
  6. Nakao S, Come PC, McKay RG, Ransil BJ. Effects of positional changes on inferior vena caval size and dynamics and correlations with right-sided cardiac pressure. Am J Cardiol 1987;59(1):125-32.
  7. Fields JM, Lee PA, Jenq KY, Mark DG, Panebianco NL, Dean AJ. The interrater reliability of inferior vena cava ultrasound by bedside clinician sonographers in emergency department patients. Acad Emerg Med 2011;18(1):98-101.

Leave a Reply

Your email address will not be published. Required fields are marked *