After the shock of BJA Acceptance of The Revised Starling Principle hypothesis in 2012 came the post-publication anxiety that I had made a terrible mistake and would have to retract because of a glaring logical error. I had been told by someone much more famous than me that patients with sepsis would die if denied treatment with pore-blocking starch molecules, and their avoidable deaths would be my responsibility. Anxious months passed while I anticipated a ‘positive’ starch trial of which he had prior knowledge appearing in NEJM, but none came. Charles Michel gave generously of his time to talk through my ideas, and made me particularly concerned about my construction of a J curve. As you can see if you look at it, I applied the Starling equation in its simplest form at an arbitrary low capillary pressure, a normal capillary pressure, a raised capillary pressure and a very raised capillary pressure. Then I drew straight lines between them. I increased the colloid osmotic pressure difference to 2 or 3 mmHg above normal to draw the J curve for raised plasma π, then I drew the reduced plasma π J curve at 2 or 3 mmHg below normal. I chose the values as approximate to the likely effect of transfusing up to 1500 ml of a solution to an adult. In my defence, I did not take Maths beyond O level and only scored Grade C.
Charles told me he would have a grad student work on something a lot more sophisticated, and so I was pleased to see such a diagram in his latest textbook contribution. (1)
The blue line is the normal J curve (Michel calls it a hockey stick), while the red line is said to illustrate crystalloid dilution to reduce plasma COP by a massive 10mmHg. Filtration rate is increased, but so long as the return of lymph can match that rate there should be no oedema as a result. At lower pressure differences, the J curves approach one another. As I observed back in 2012, infusing fluids to achieve higher than normal blood volume and capillary pressure (and mean circulatory pressure) would be easier with a colloid solution (lower filtration rate) but I cannot think of a therapeutic application for such treatment. For resuscitation from a low blood volume the osmotic pressure of the resuscitating fluid is an irrelevance.
Figure b is another Revised Starling prediction that matches observations in clinical practice. When capillary pressure is 12 mmHg, the retention of crystalloid in an adult circulation is 80% at 40 minutes. But when capillary pressure is normal (22 mmHg) retention is only 20%.
To have my clinical hypothesis confirmed by two of the most eminent physiologists in this field has been a great relief. I wish this chapter could be required reading for clinicians who would claim to understand the science, and I wish it was Free. I’d urge you to read this truly scientific account of the clinical consequences of the Starling principle if you possibly can, currently £24.
I am afraid the full mathematics is beyond me.
(1) Michel CC, Arkill KP and Curry FE. “Chapter 2. The Revised Starling Principle and its relevance to perioperative fluid management.” pages 31-74 In: Perioperative Fluid Management. Eds Farag E and Kurz A. Springer 2016.