Compensated shock

Why can a patient losing blood dangerously fast still show an entirely normal blood pressure?

Blood pressure is the number everyone reaches for first, and the intuition is simple: blood fills the system, so losing blood should lower the pressure inside it. Yet a person can lose a great deal and still show a pressure that would not raise an eyebrow — and then, without any change in the rate of loss, fall off a cliff.

That shape is the thing to explain. Not why the pressure eventually falls, which is obvious, but why it stays flat first and then collapses abruptly rather than sliding down in proportion to the loss.

The first move is to stop treating blood pressure as a measurement of blood volume. It is not a stock; it is the output of a control loop, and control loops are built to hold their output steady while everything behind them moves.

Write down what pressure is made of. Mean arterial pressure is, to a good approximation, cardiac output multiplied by systemic vascular resistance. Cardiac output is itself heart rate multiplied by stroke volume. So the pressure a cuff reads is a product of three quantities, and a product can be held constant while its factors change in opposite directions.

Now lose volume and follow it. Less blood returning to the heart means less filling, so stroke volume falls — that factor moves immediately. But the system notices. Stretch receptors in the carotid sinus and the aortic arch fire in proportion to how distended those vessels are; less distension means less firing, which releases the brainstem's restraint on sympathetic outflow. Three things follow. Heart rate rises, restoring the product of rate and stroke volume. Arterioles constrict, raising resistance. And the veins constrict too, which matters more than it sounds: the venous system holds the majority of the body's blood, so squeezing it pushes a substantial reserve back into the active circulation — the body transfusing itself from its own storage.

Two slower helpers join in: fluid moves from the interstitial spaces into the vessels as capillary pressure falls, and hormonal signals reinforce constriction while conserving salt and water.

Notice what the loop is defending. It is defending pressure, because pressure is what perfuses the brain and the heart. It will spend heart rate, spend vascular tone, spend the splanchnic reserve, and sacrifice blood flow to skin, muscle and gut in order to keep that one number where it was. So a normal blood pressure in someone who is bleeding is not evidence that little has been lost. It is evidence that the compensation is currently succeeding — which is a statement about the strain on the system, not about its safety. The naive reading of the number is close to inverted.

Then why does it end so abruptly? Because every term in the correction has a ceiling. Arterioles have a maximum constriction; beyond it, no further resistance is available. The venous reservoir, once recruited, cannot be recruited twice. And heart rate is not indefinitely useful — pushed high enough, it shortens the time available for the ventricle to fill in diastole, so stroke volume falls further and the correction begins working against itself. Up to the ceiling, each increment of loss is absorbed and the cuff reads normal. Past it, there is nothing left to absorb with, and every further millilitre lost shows up directly in the pressure. That is why this behaves as a threshold rather than a gradient, and why deterioration is so often described as sudden when the underlying loss was perfectly steady.

Which tells us where the information actually is. The variables that move early are precisely the ones the compensation is spending: heart rate, and the pulse pressure — the gap between systolic and diastolic. Rising resistance lifts the diastolic figure while falling stroke volume lowers the systolic one, so the gap narrows before the mean moves at all. Peripheral perfusion declines because it was the first thing sacrificed, and lactate rises from the beds that lost their supply. What weight to give any of those in a particular person is a clinical judgement for someone who can examine them — the point here is only mechanical: they change first because they are what is being spent.

How this could be wrong. If compensation is doing the work, then removing the compensation should abolish the plateau — and it does. A person taking a beta blocker cannot mount the tachycardia; someone with a spinal cord injury above the sympathetic outflow cannot mount the vasoconstriction. In both, pressure falls earlier and more gradually with loss. Controlled human experiments that simulate haemorrhage, such as lower-body negative pressure, reproduce the same flat-then-cliff shape. The observation that would refute this account is straightforward: if patients whose sympathetic response is pharmacologically or neurologically blocked still showed the same flat pressure across the same degree of loss, the compensation could not be what produces the flatness.

One honest caveat. The tidy staged tables of haemorrhage classes, with a percentage of blood volume attached to each set of signs, are a teaching device; studies checking real patients against them find people frequently do not progress through the categories in order. I am describing the mechanism and deliberately not quoting the bands.

The refinement that ties the reasoning together is the distinction between a controlled variable and an informative one. Blood pressure is controlled, and a well-controlled variable is by construction the last place a disturbance appears. The quantities the controller manipulates — rate, tone, regional flow — are where the disturbance is visible early, because they are the currency being spent.

A misconception worth correcting gently: compensated shock is not a mild version of shock. Shock is inadequate delivery of oxygen to tissue, and that is already happening in the compensated phase — it is happening because of the compensation, in the beds that were shut down to protect the pressure. "Compensated" describes what the blood pressure looks like, not what the tissues are receiving.

This also explains a grim inversion: young, fit people compensate hardest and hold their pressure longest, so they decompensate latest and most steeply.