Scott Sparks, MD and Mark Favot, MD, FACEP
Does size matter? Wait for it… Wait for it… Yes and no. Anyone who has listened to, or read LITFL, EmCrit, Ultrasound Podcast or taken an echo exam has undoubtedly spent time memorizing the IVC table (Table 1) to estimate right atrial pressures (RAP). According to the American Society of Echocardiography, an IVC diameter >21mm in conjunction with inspiratory collapse <50% is predictive of RAP >10 mm Hg. Yet, when has anything in medicine proven to be the gospel?
As Emergency and Critical Care Physicians and POCUS users, we use the RAP to assess right-sided filling pressures in order to guide fluid management in critically ill patients. Using POCUS we can evaluate the IVC Collapsibility Indices to help us decide if our patient is in hypovolemic shock and needing IV fluid resuscitation, or if the patient has elevated right-sided filling pressures (ie, Pulmonary Embolism or Acute Respiratory Distress Syndrome - ARDS) and benefit from a fluid restrictive strategy. However, we need to pause and ask ourselves – By focusing only on the IVC, and assessing both size and respirophasic variation, are we getting it right?
Table 1: Estimation of Right Atrial Pressure on the basis of IVC Diameter and Collapse. From the American Society of Echocardiography Guidelines for Echocardiographic Assessment of the Right Heart in Adults.
Taniguchi et al recently published a study in the Journal of the American Society of Echocardiography (2015), which may answer this question. The authors performed a prospective analysis of 99 spontaneously breathing patients with heart disease comparing internal jugular vein central line or PA catheter measured end expiration CVP/RAP compared to TTE measured IVC size and collapsibility index (both passive and with a sniff). Nine patients (9.1%) were excluded due to suboptimal image quality, which is consistent with previous studies of the IVC (inadequate images in 8-12%). IVC measurements were taken just proximal to the junction of the middle hepatic vein. The primary objective of the study was to assess the impact of body surface area (BSA) on the IVC parameters of size and collapsibility index and to determine the optimal cutoff values for predicting elevated RAP (>10 mm Hg) in patients on the basis of BSA.
Receiver operating characteristic (ROC) curves were estimated to determine the optimal cutoff points of IVC variables for detecting RAP >10 mm Hg. The interaction between maximal IVC diameter and BSA was evaluated using linear regression. BSA was analyzed as both a continuous and a dichotomous (below or above the median value) variable. The difference in optimal cutoff points in higher and lower BSA subgroups was explored using bootstrapping with 100,000 resamplings. In general bootstrapping is a fairly complex statistical analysis with many different methods in which it can be carried out. In this particular instance I suspect the authors chose the case resampling method of bootstrapping because the dependent variable (RAP) could not be expected to assume a normal distribution. For more on bootstrapping, please refer to someone other than the current authors of the Cardiac Ultrasound Journal Summary.
The mean RAP in the study was 8 mm Hg (range 1-25 mm Hg). For the entire cohort the optimal cutoff points of IVC maximum diameter and IVC collapsibility index to predict RAP >10 mm Hg were 20mm & 49% respectively. When BSA was treated as a dichotomous variable there was a statistically significant correlation between IVC maximum diameter and BSA (p = .0444). BSA of 1.61 m2 was utilized as the median value distinguishing patients with a “high” BSA (n=45) and a “low” BSA (n=45). No significant difference in RAP was detected between the 2 subgroups. The mean IVC maximum diameter in the high BSA subgroup was 20mm (SD = 4mm) and in the low BSA subgroup the mean IVC maximum diameter was 16mm (SD = 5mm). The optimal cutoff point of the maximal IVC diameter in predicting elevated RAP in the higher BSA subgroup was 21mm and in the lower BSA subgroup the optimal cutoff point to predict elevated RAP was 17mm. There were no significant differences in collapsibility indices (passive respiration or sniffing) between the 2 subgroups.
So, what does this tell us?
The established IVC table is based on the ASE guidelines (2010) for IVC correlation to RAP. Goldhammer et al (1999) measured the IVC in highly trained elite athletes and found that they developed chronically dilated IVC in response to the strenuous, repetitive exercise during training. Children have smaller IVC that change with age according to Kathuria et al (2015).
But, if size really matters when measuring IVC, then are we measuring RAP incorrectly? Should we change what we are doing? The ASE guidelines have not changed to reflect the various populations. So what is the daily EM practitioner or Intensivist to do while standing at the bedside with the echo probe in hand? If, we are to follow the evidence and apply EBM to daily practice, then we must plug along with the ASE guidelines when measuring RAP for high PA pressures and begin relying on a more qualitative “eyeballing” approach to the IVC during fluid resuscitation, with this new information in mind.