The Intensive Care Society’s Webinar on Fluids yesterday was Sold Out with record number of delegates and is receiving rave reviews. Ashley Miller’s drawing of the J curve “explains it all as you’ve never seeen it” @PARADicmSHIFT. I reproduce it below.
My first attempt at the J curve looked like this in 2012;
It makes a bold assumption that the whole body filtration-pressure relationship of a patient would be a similar J curve to that observed in a single amphibian capillary. That classic experiment had been reported in 1987 by Charles Michel and Mary Philips examining “Steady-state fluid filtration at different capillary pressures in perfused frog mesenteric capillaries.” (1)
1. Michel CC, Phillips ME. J Physiol. 1987;388:421-435.)
The J curve version below has appeared in every edition of J Rodney Levick’s Introduction to Cardiovascular Physiology since 1990. The current sixth edition is in the capable hands of Neil Herring (Consultant Cardiologist) and David Paterson (Professor of Cardiovascular Physiology, University of Oxford).
Herring N, Paterson D. Circulation of fluid between plasma, interstitium and lymph. Levick’s Introduction to Cardiovascular Physiology. Boca Raton, London, New York: CRC Press; 2018. p. 191-220. Below is Figure 11.13, page 202.
Now the good news is that my friend Charles Michel accepted our invitation to contribute two chapters to Farag E, Kurz A. Perioperative Fluid Management. Springer; 2016. The book is currently in its 2nd edition, and Michel’s chapters are updated. Michel made a calculation of the whole body effect of infusing a litre of isotonic salt solution to a healthy adult. To my great relief, my hunch was affirmed, .
“Fig. 2.16 Relationships between microvascular fluid filtration (JV /A) and microvascular pressure difference (ΔP) before during and following intravenous infusion of an isotonic crystalloid solution. Point A indicates values of J V and ΔP before infusion and blue arrow, the changes in JV /A at constant ΔP to point B where infusion ends. If, as is most likely, expansion of the circulating volume leads to vasodilatation, with reduced Ra/Rv and a consequent rising PC , changes in J V are shown by the red arrow leading towards point C. Green dashed arrows indicate excretion of the fluid load. (b) Changes in Jv/A in an acutely hypovolemic patient. Initial vasoconstriction with increase in Ra/Rv, leads to a rapid fall in Pc with a consequent reversal of fluid filtration to fluid uptake from tissues to plasma (red arrow A to B). If crystalloid solution is infused at B, fluid is retained and plasma volume restored to a degree determined by dilution of plasma proteins to the stage where σΔΠ = ΔP at the value of Pc at point C. Fluid is no longer retained if infusion is continued beyond this stage.”
Figure 2.16(c) was Charles’ prediction of the intravascular retention of infused crystalloid at lower and higher capillary pressures. It fits very well with experience and published clinical observations.
“Figure 2.16(c) Predictions of the fraction of a crystalloid infusion that is retained in the circulation in a tissue such as muscle at normal and reduced PC.”
So I would like to propose that we can achieve much by using kinetic diagrams to anticipate the fluid flux from intravascular to extravascular compartments during fluid therapy. Step One is to hammer home the message that circulating plasma does not filter fluid to the tissues while capillary pressure is low. Any intravenous resuscitation fluid will ‘stay in the circulation’ until capillary pressure gets back to the J point. Colloids are unnecessary. Step Two is to appreciate that capillary pressure is greatly reduced during hypovolaemia by sympathetic arteriolar constriction that preserves mean arterial pressure. If reflex arteriolar constriction is blunted (in post-surgical hypotension or ‘sepsis’) then a trickle of infused norepinephrine is a rational choice.