When we measure values using blood draws, we have to keep in mind that what we are measuring may have markedly different “normal” concentrations/distributions in different spaces in the body, and that even a concentration measurement is not a reflection of total amount.
There are lots and lots of ways that a typical measurement technique can give rise to a conclusion which does not accurately reflect the whole picture.
KG gives some examples around potassium.
Potassium is the major intracellular cation, and almost all of a total body store is inside the cells. We don’t measure that in a blood draw. We just measure the 2% or so that’s in the blood. (Without getting too picky, we actually measure the concentration in the serum, which is the liquid portion of the blood.) The body’s physiological mechanisms try hard to maintain the right amount of extracellular potassium, because if it gets too low or too high, the membrane potentials (the electrical charge difference between the inside and outside of a cell) get screwed up. Many processes, including the electrical system that makes nerves and muscles work depend on that membrane potential, so you’ll die from an abnormal rhythm or inability for muscles–including heart–to work if your extracellular potassium falls outside a certain range. But measuring extracellular potassium still doesn’t tell you if have a “normal” total amount because the extracellular range is maintained by taking potassium from inside the cells when necessary.
Think of this as trying to decide if a person has a lot of money in their savings account by looking at their checking account balance. A blood draw looks at the checking account, which doesn’t reflect very directly the total amount of money they have, but is nevertheless useful in deciding if they can pay for something important right that second.
Lots of other similar examples exist. Most sodium is extracellular, so we can measure that pretty easily with a blood draw. But we typically measure serum sodium (the sodium concentration in the part of blood that’s left over after we take out the cellular portion and let the proteins clot as well). Sometimes a patient will have really high levels of (lipids) “liquid” fat in their blood, and so the volume of serum (the liquid) is high. But sodium isn’t distributed into the fat portion. Therefore the measurement by concentration is way off, and gets reported as having a serum sodium that is much too low. This is called “factitious” hyponatremia. (In practice, the lab techs usually notice the fat layer and tell us the sample is unusual and suspect.)
Many other examples exist. For some analytes, like calcium, we also care whether or not it’s ionized–not just the measurement result. Total body stores of some minerals and other substances are actually quite complex to measure, although they can sometimes be inferred. We can do things like inject tracers and see how diffusely they are distributed.
But I guess the short answer to your question is that serum levels in particular can be very misleading if the wrong inference about total stores or the state of body physiology is taken from them, and that’s part of understanding what a lab value does and does not mean.