Potassium, Calcium, Magnesium, and Phosphate Physiology

A blood electrolyte concentration describes one compartment at one moment. Potassium, calcium, magnesium, and phosphate depend on cellular transport, binding, hormonal regulation, and renal handling. Understanding these processes explains why abnormalities can coexist, why a normal result can conceal depleted stores, and why correcting a number does not by itself establish clinical benefit.

46.1 Concentration, Body Stores, and Electrical Charge Are Different Quantities

Most body potassium is intracellular. Much calcium and phosphate is incorporated into the skeleton, while magnesium is distributed mainly in bone and cells. Blood samples therefore interrogate small, regulated pools rather than the entire inventory. Redistribution can alter a concentration rapidly without changing total body content; prolonged negative balance can reduce stores before a circulating concentration becomes conspicuously abnormal.

The measured chemical form also matters. Total calcium includes ionized, protein-bound, and complexed calcium. Total magnesium likewise includes several fractions. The routine phosphate measurement represents circulating inorganic phosphate rather than all phosphorus in adenosine triphosphate (ATP), phospholipids, and bone. A reference-range result cannot certify that these other pools are adequate.

Millimoles describe amount of substance; milliequivalents describe electrical charge:

Valence is the number of electrical charges carried by the ion. Potassium is monovalent, so 3 mmol of K⁺ represents 3 mEq. Calcium and magnesium are divalent: 3 mmol of Ca²⁺ or Mg²⁺ represents 6 mEq. These are unit examples, not administered quantities. The numerical equivalence used for potassium must not be transferred to calcium or magnesium.

Phosphate requires additional care. At physiological pH, inorganic phosphate includes H₂PO₄⁻ and HPO₄²⁻ in proportions influenced by pH. A reported mmol of phosphorus does not have one universal mEq conversion independent of chemical species. Moreover, a laboratory reporting phosphorus in mass units describes elemental phosphorus, whereas the mass of a phosphate salt includes other atoms. The analyte and units must be identified before comparing values or calculating balance.

Sample integrity precedes interpretation. Hemolysis can release intracellular potassium into the specimen. Sampling near an infusion and delayed processing can also distort results. Assess specimen conditions alongside biological plausibility, without allowing the possibility of artifact to dismiss a potentially important abnormality.

46.2 Potassium Gradients Shape Excitability and Conduction

The Na⁺/K⁺-ATPase uses ATP to maintain high intracellular potassium and low intracellular sodium. Because resting membranes are relatively permeable to potassium, the potassium gradient strongly influences resting membrane potential. The electrical gradient opposing potassium diffusion is described by its equilibrium potential.

At 37°C, a concentration-based approximation is:

E_K is the potassium equilibrium potential in millivolts, or mV. Both concentrations must use the same units. The coefficient incorporates temperature and potassium's single positive charge. The equation strictly depends on ionic activities; using concentrations is an approximation. Actual membrane potential also depends on other permeabilities and currents.

For illustration, assume intracellular potassium remains 140 mmol/L. With extracellular potassium of 4 mmol/L, 61.5 × log₁₀(4 ÷ 140) ≈ −95.0 mV. If extracellular potassium becomes 8 mmol/L, 61.5 × log₁₀(8 ÷ 140) ≈ −76.4 mV. The equilibrium potential has become approximately 18.5 mV less negative. These values demonstrate a gradient effect, not clinical safety boundaries or a prediction of the exact cardiac membrane potential.

Depolarization initially brings membrane voltage closer to the activation range of some channels. Sustained depolarization, however, reduces the availability of voltage-gated sodium channels through inactivation. Reduced inward sodium current can slow conduction and impair action-potential generation. Hodgkin and Huxley's squid axon work established the interaction of voltage-dependent conductances and sodium-channel inactivation; its quantitative model was not a human cardiac treatment model. 1

Low extracellular potassium also changes resting voltage and repolarizing potassium currents. Its effects cannot be summarized as simply making every cell less excitable. Cardiac consequences depend on tissue, rate, accompanying ions, drugs, and the evolution of the disturbance.

The surface electrocardiogram (ECG) samples the integrated electrical result, with imperfect sensitivity. Montague and colleagues' retrospective series of 90 hyperkalemic patients found that characteristic changes were inconsistently present. A reassuring tracing therefore cannot rule out potassium-related danger or replace concentration, clinical context, and serial assessment. 2

46.3 Potassium Redistribution Can Conceal the Direction of Body Balance

Insulin promotes potassium uptake into cells, including skeletal muscle, partly through stimulation of Na⁺/K⁺-ATPase activity. β₂-adrenergic signaling also favors cellular uptake. These responses buffer extracellular potassium after intake and can change plasma concentration faster than the kidney can alter total body content.

DeFronzo and colleagues studied graded insulin exposure in 29 healthy volunteers using glucose-clamp and regional metabolic measurements. Potassium uptake varied with insulin exposure, time, and tissue region. This demonstrates regulated redistribution; it does not supply a fixed potassium decrease for every insulin concentration or a predictable response in critical illness. 3

Insulin deficiency can produce the opposite combination: reduced cellular uptake and extracellular potassium elevation despite depleted total stores. Concurrent urinary and gastrointestinal losses can make the deficit substantial. Subsequent restoration of uptake can uncover low extracellular potassium without creating the preceding depletion. A change after treatment therefore needs interpretation against the earlier balance history.

Hyperosmolar extracellular conditions can favor potassium movement out of cells through changes in cellular water and ion transport. Cell injury releases intracellular potassium directly. Exercise and tissue activity can cause transient local efflux, followed by recovery through pumps and circulation. These mechanisms differ from retention caused by impaired renal excretion, although several may occur together.

Acid–base effects are similarly conditional. Mineral acid disturbances, organic acid metabolism, insulin availability, membrane transport, and renal adaptation do not produce identical potassium responses. Hydrogen and potassium are not exchanged across all cell membranes in a fixed one-for-one rule, and there is no universal potassium change per unit pH. Alkalemia can favor cellular uptake, but its accompanying cause may also independently change renal potassium losses.

The extracellular pool is too small to calculate a whole-body deficit from concentration alone. In an illustrative fixed 14-L extracellular compartment, a fall from 4 to 3 mmol/L represents 14 × (4 − 3) = 14 mmol less potassium there. The model excludes changing fluid volumes and cannot identify how much entered cells or left the body.

46.4 Renal Potassium Excretion Depends on Delivery and Distal Transport

Filtered potassium is extensively reabsorbed in the proximal tubule and thick ascending limb. Final urinary excretion is then strongly influenced by secretion and reabsorption in later segments. The distal nephron regulates the final balance rather than merely passing through a constant fraction of the filtered load.

In potassium-secreting cells, basolateral Na⁺/K⁺-ATPase brings potassium into the cell. Apical potassium channels provide an exit toward tubular fluid. Sodium entry through ENaC, the epithelial sodium channel, contributes to an electrically favorable environment for potassium secretion. Renal outer medullary potassium channels, or ROMK, support secretion. Large-conductance, calcium-activated potassium channels, or BK, become particularly relevant with increased flow and appropriate stimulation.

Aldosterone increases transport capacity through mineralocorticoid signaling. Its effect depends on available distal sodium, tubular flow, membrane voltage, and potassium supply. A high hormone concentration cannot guarantee a particular urinary potassium flux when delivery or functioning nephron mass is limited. Conversely, increased distal delivery and flow can sustain losses when extracellular potassium is already low.

Potassium itself participates in renal adaptation, influencing adrenal aldosterone release and distal sodium transport. Acid–base conditions modify multiple transport steps; potassium depletion can also promote renal hydrogen secretion and bicarbonate retention. These reciprocal effects help maintain combined disorders without requiring a fixed extracellular hydrogen–potassium exchange formula.

Measure an excretion rate by combining concentration with matched urine flow:

U_K is urine potassium concentration in mmol/L. If a representative collection contains 40 mmol/L and urine volume is 2 L/day, excretion is 40 × 2 = 80 mmol/day. With the same concentration but 0.5 L/day, it is 40 × 0.5 = 20 mmol/day. The spot concentration alone cannot distinguish these amounts.

These fluxes require interpretation against intake, gastrointestinal losses, recent therapy, and the direction of plasma change. Acute kidney injury can reduce potassium excretion, whereas a recovery phase with impaired tubular conservation and high flow can promote loss. A single urinary value does not establish steady state, quantify whole-body depletion, or identify one transport mechanism uniquely.

46.5 Ionized Calcium Must Be Interpreted at the Relevant pH

Ionized calcium participates directly in excitation–contraction coupling, secretion, coagulation, and cellular signaling. Extracellular ionized calcium also influences neuromuscular excitability. Total calcium additionally includes calcium bound to proteins, particularly albumin, and calcium complexed with smaller anions. These fractions exchange, but they are not interchangeable measurements.

Reduced albumin can lower total calcium without a corresponding fall in the ionized fraction. Alkalemia increases calcium binding to albumin and can lower ionized calcium even when total calcium changes little; acidemia tends to reduce binding. Citrate and phosphate can change complex formation. Consequently, the same total calcium can accompany different biologically active concentrations.

Albumin-adjustment equations assume a sufficiently predictable relationship between albumin and calcium binding. Critical illness changes pH, binding partners, albumin properties, and assay conditions. Slomp and colleagues prospectively compared 53 paired samples from 36 critically ill patients and found that albumin-adjusted calcium classified ionized-calcium abnormalities poorly in that population. The study supports direct measurement when the ionized fraction matters; it does not validate another universal correction coefficient. 4

Actual ionized calcium is measured at the sample's measured pH and, with appropriate collection, represents the patient's circulating condition. A pH-normalized result estimates what ionized calcium would be at a specified pH, commonly 7.40, using an assumed relationship. It is a calculated hypothetical condition rather than the free calcium actually present during patient alkalemia or acidemia.

Sampling can also change pH after blood leaves the patient. Carbon dioxide loss to air raises sample pH and may lower measured ionized calcium. Heparin binding and dilution can introduce additional error. International Federation of Clinical Chemistry (IFCC) recommendations emphasize anaerobic collection, appropriate handling, and simultaneous assessment of pH and ionized calcium. Mathematical normalization may address particular preanalytical circumstances, but it cannot reliably reconstruct every compromised specimen or erase the physiological effect of the patient's own pH. 5

Interpret the reported result only after identifying whether it is actual or normalized, how the specimen was handled, and whether major binding influences are present. Total calcium, albumin, and ionized calcium can be complementary observations without being substitutable.

46.6 Calcium Homeostasis Couples Sensing, Hormones, Kidney, Gut, and Bone

The extracellular calcium-sensing receptor, or CaSR, allows parathyroid cells to respond to extracellular calcium. Brown and colleagues cloned and functionally characterized the receptor from bovine parathyroid tissue, establishing a molecular mechanism for extracellular calcium sensing. Increased receptor activation suppresses parathyroid hormone (PTH) secretion. The cloning experiment did not establish a fixed human threshold for adequate calcium status. 6

When ionized calcium falls, the normal parathyroid response is increased PTH secretion. PTH supports renal calcium conservation, increases phosphate excretion, and stimulates renal production of calcitriol, the active vitamin D metabolite. Calcitriol promotes intestinal calcium and phosphate absorption and participates in feedback regulation. Bone supplies a large reservoir through regulated exchange and remodeling, operating on several time scales.

Only the filterable calcium fractions enter the glomerular filtrate freely; albumin-bound calcium is largely retained. The proximal tubule reabsorbs much filtered calcium with sodium and water through predominantly paracellular pathways. The thick ascending limb also supports paracellular calcium transport, influenced by its electrical environment and CaSR signaling. Distal segments provide regulated transcellular calcium reabsorption.

A change in filtration, sodium handling, tubular voltage, or hormone signaling can therefore alter urinary calcium independently of total-body calcium amount. Low urine calcium can represent active conservation, low filtered delivery, or both. Increased excretion may reflect altered transport or delivery rather than simply excessive intake.

Critical illness can disrupt several components simultaneously. Kidney dysfunction can reduce calcitriol production and phosphate excretion. Complex formation, altered hormone responsiveness, and magnesium deficiency can lower ionized calcium or impair the expected compensatory response. A PTH concentration must therefore be interpreted relative to contemporaneous ionized calcium, magnesium, phosphate, kidney function, and timing.

A reference-range PTH may be an inadequate response to low ionized calcium, while an elevated result may represent appropriate compensation. Neither observation alone determines calcium requirements or whether changing the laboratory value will improve organ function.

46.7 Magnesium Supports ATP Reactions and Has Its Own Renal Control

Magnesium is required for many enzymatic reactions, including processes that use ATP. For numerous enzymes, the functional substrate is a magnesium–ATP complex rather than uncomplexed ATP. Magnesium also interacts with ion channels, membrane stability, nucleic acids, and neuromuscular transmission. A magnesium disturbance can therefore affect energy-dependent transport and electrical behavior at the same time.

Most magnesium lies outside the circulating extracellular pool, principally in bone and cells. Total serum magnesium includes free, protein-bound, and complexed fractions. Its concentration is accessible and useful, but it is an incomplete measure of intracellular availability and total stores. A normal value after recent intake or during redistribution does not exclude antecedent depletion; a low value does not specify its magnitude.

Filterable magnesium enters the nephron, while protein-bound magnesium is largely retained in plasma. The proximal tubule reabsorbs a portion, and the thick ascending limb normally accounts for the largest reabsorptive share. Its lumen-positive voltage supports paracellular movement through selectively permeable junctions. Distal convoluted tubular transport then adjusts the remaining amount through regulated transcellular uptake.

The transient receptor potential melastatin 6 channel, or TRPM6, is important in this final control and in intestinal magnesium handling. Schlingmann and colleagues identified TRPM6 mutations in human hypomagnesemia with secondary hypocalcemia. This inherited disorder demonstrates the importance of the transport pathway; it does not imply that TRPM6 mutations explain typical magnesium disturbances in the ICU. 7

Intestinal absorption, gastrointestinal losses, nutritional supply, renal delivery, and tubular transport jointly determine balance. Drugs can alter transport, but identifying a plausible exposure does not exclude simultaneous gastrointestinal depletion or altered distribution. Renal excretion must be assessed against the filtered supply and current plasma concentration, rather than interpreted from a urinary concentration alone.

Reduced kidney function can permit magnesium accumulation when input continues. Conversely, tubular dysfunction can impair conservation despite reduced filtration. Neither creatinine nor serum magnesium alone identifies the mechanism: intake, distribution, and excretion must explain the trajectory.

46.8 Magnesium Deficiency Can Sustain Potassium and Calcium Abnormalities

Magnesium helps regulate potassium exit through ROMK channels. Intracellular magnesium can block outward channel current under relevant electrical conditions; reduced magnesium can release this restraint. Yang and colleagues demonstrated this interaction using expressed ROMK channels and rat collecting-duct preparations. The work provides a mechanism linking magnesium depletion with potassium secretion, rather than a numerical prediction from serum magnesium to urinary potassium loss. 8

Removing channel inhibition is only one component of secretion. Distal sodium delivery, ENaC activity, voltage, flow, aldosterone, and potassium availability still determine the driving forces. Magnesium deficiency therefore does not produce an identical potassium response in every patient. It can, however, help explain continued renal potassium loss when potassium intake alone seems insufficient to restore the circulating concentration.

Serum magnesium is not a direct assay of the free intracellular magnesium acting at a renal channel. A normal circulating result cannot precisely certify that this mechanism is absent, and a low result does not prove it is the only cause of potassium loss. The combined physiological explanation should remain consistent with urine output, ongoing losses, nutritional history, and renal function.

More substantial magnesium deficiency can also impair PTH secretion and responsiveness to PTH. Rude and colleagues studied 17 patients with magnesium deficiency and hypocalcemia. Their hormonal observations and selected response tests supported impaired parathyroid secretion and end-organ resistance in that setting. This was a small physiological investigation, not evidence that every low magnesium value suppresses PTH or that one replacement strategy improves ICU outcomes. 9

The interaction explains why calcium homeostasis may remain abnormal when only calcium intake is considered. It also illustrates a general principle: two abnormal concentrations need not represent two independent deficiencies. One transport or signaling disturbance can sustain both, while their external balances remain distinct.

These mechanisms justify assessing linked abnormalities and ongoing losses, without determining a universal replacement amount or route. Renal impairment, continued input, and changing transport can reverse the problem, requiring reassessment of concentration and function.

46.9 Phosphate Links Cellular Metabolism, Membranes, and Oxygen Unloading

Phosphorus is incorporated into ATP, phosphorylated metabolic intermediates, nucleic acids, phospholipid membranes, and skeletal mineral. Circulating inorganic phosphate supplies an exchangeable extracellular substrate, but its concentration is not a direct measurement of tissue ATP content. Cellular demand, uptake, release, renal handling, and bone exchange all contribute to the measured value.

Insulin and renewed anabolism promote cellular phosphate uptake and incorporation into metabolic products. Respiratory alkalosis can raise intracellular pH, stimulate glycolysis, and increase intracellular phosphate utilization. A rapid fall in circulating phosphate can therefore reflect redistribution, pre-existing depletion, or both. It need not represent an equivalent amount newly lost from the body.

Phosphate also participates in erythrocyte metabolism, including formation of 2,3-bisphosphoglycerate, or 2,3-BPG. This organic phosphate modifies hemoglobin oxygen affinity. Chanutin and Curnish demonstrated the effect of organic phosphates on oxygen equilibrium using purified adult human hemoglobin. Reduced 2,3-BPG can favor higher oxygen affinity, but the experiment does not make a serum phosphate concentration a direct measure of tissue oxygen unloading. 10

Actual oxygen transfer also depends on perfusion, hemoglobin concentration, oxygen pressures, temperature, pH, and tissue extraction. A low serum phosphate cannot alone establish that impaired oxygen unloading is limiting metabolism. Likewise, restoring its concentration does not prove that every intracellular energetic defect has resolved.

Diaphragmatic function provides a more direct physiological example. Aubier and colleagues studied eight ventilated patients with acute respiratory failure and hypophosphatemia. Transdiaphragmatic pressure elicited by phrenic stimulation increased after phosphate correction in their paired observations. This supports a potentially reversible contribution to muscle dysfunction in that population; it was not a randomized trial of ventilator liberation or survival. 11

The duration and context of deficiency matter. Acute redistribution into viable cells differs from prolonged inadequate supply, and both differ from release during cell injury. A rising phosphate concentration during tissue breakdown can accompany severe cellular dysfunction rather than improved energetic health. The laboratory direction must be connected to the process generating it.

46.10 Proximal Phosphate Transport Responds to PTH and FGF23

The kidney adjusts phosphate balance mainly by varying proximal tubular reabsorption of filtered inorganic phosphate. Apical sodium–phosphate cotransporters, including NaPi-IIa and NaPi-IIc, couple uptake to the sodium gradient maintained by basolateral Na⁺/K⁺-ATPase. Changes in transporter abundance and activity alter how much filtered phosphate reaches the urine.

PTH reduces proximal phosphate reabsorption, helping separate calcium conservation from phosphate retention. Fibroblast growth factor 23, or FGF23, produced mainly by bone cells, also promotes phosphaturia through renal receptor signaling involving α-Klotho. FGF23 reduces calcitriol availability, thereby influencing intestinal phosphate absorption as well as renal handling.

Shimada and colleagues' Fgf23-knockout mice developed abnormal phosphate and vitamin D metabolism, including increased phosphate retention and calcitriol. The experiment established the importance of this regulatory axis in an animal model. It did not identify a therapeutic FGF23 target for acute human kidney injury. 12

The filtered amount provides the starting point for interpreting excretion:

GFR is glomerular filtration rate, expressed in liters per unit time. Concentration in mmol/L gives a load in mmol per unit time. This teaching approximation neglects the small nonfilterable phosphate fraction. Consider GFR of 60 mL/min and plasma phosphate of 1 mmol/L. GFR becomes 0.060 × 1,440 = 86.4 L/day, giving 86.4 × 1 = 86.4 mmol/day filtered.

If matched urine contains 10 mmol/L at 1.5 L/day, phosphate excretion is 10 × 1.5 = 15 mmol/day. Approximate net tubular reabsorption is 86.4 − 15 = 71.4 mmol/day, assuming negligible secretion and representative measurements. This describes a hypothetical transport state, not a normal requirement or a recommended excretion target.

A falling GFR can reduce phosphate elimination even when fractional reabsorption decreases. Conversely, proximal dysfunction can produce phosphate wasting during illness or recovery. Intake, hormonal adaptation, nonsteady plasma concentrations, and collection timing can all alter interpretation. A urinary phosphate result therefore needs the concurrent filtered load and clinical trajectory; it cannot independently distinguish all redistribution from depletion.

46.11 Refeeding Exposes Interactions Between Stores and Metabolic Demand

During prolonged inadequate nutrition, potassium, magnesium, and phosphate stores can decline while circulating concentrations remain deceptively maintained. Lower intake and continued losses reduce body content; reduced anabolism and altered distribution can obscure the deficit in blood. The starting serum value therefore incompletely represents the capacity to support renewed metabolism.

Reintroduction of nutrition increases insulin signaling and anabolic demand. Potassium enters cells, phosphate is incorporated into phosphorylated compounds, and magnesium is required for many associated reactions. New demands can exceed available supply or uncover previously concealed depletion. Sodium and water retention may occur simultaneously, linking these changes to the balance mechanisms discussed in Chapter 45.

This is a coordinated physiological transition rather than an isolated low phosphate result. Organ dysfunction can reflect multiple deficiencies, changing fluid distribution, and the underlying illness. Thiamine availability also matters to renewed carbohydrate metabolism. An explanation centered only on serum phosphate misses the broader metabolic context without establishing which component is limiting in an individual patient.

Not every fall after nutrition proves that refeeding is the sole cause. Respiratory alkalosis, ongoing gastrointestinal loss, tubular wasting, insulin exposure from other indications, and changing kidney function can produce overlapping patterns. Timing increases or decreases the plausibility of a mechanism, but it does not replace assessment of competing fluxes.

Doig and colleagues randomized 339 critically ill adults who developed the trial's refeeding syndrome to continued standard nutrition or protocolized caloric restriction during management. Among assessable patients, the primary outcome of days alive after ICU discharge through 60 days did not differ statistically; some individual survival components favored restriction. The trial compared nutritional strategies within a defined population, not isolated phosphate replacement against no replacement. 13

A beneficial component outcome does not make the primary outcome positive, and a strategy comparison cannot identify one universal electrolyte target. Nutritional exposure, serial concentrations, renal function, fluid balance, and clinical function require interpretation together as the metabolic state changes.

46.12 Integrate External Balance, Redistribution, Measurement, and Function

A coherent interpretation begins with the amount entering and leaving the body over a defined interval:

Use matched amounts, such as mmol/day, rather than mixing concentrations with quantities. Oral delivery is not necessarily absorbed input, and incomplete urine or drain collections weaken the estimate. Redistribution between compartments is absent from this external balance equation because it changes location rather than total amount.

Suppose an illustrative patient absorbs or receives 80 mmol of potassium in a day, excretes 100 mmol in urine, and loses 20 mmol through the gastrointestinal tract. The external balance is 80 − 100 − 20 = −40 mmol/day. Serum potassium can nevertheless rise if concurrent cell injury or impaired cellular uptake shifts enough potassium outward. Conversely, a positive balance does not prevent a fall during strong cellular uptake. Neither concentration trajectory cancels the external accounting.

A calcium result poses a different additional question: whether total or actual ionized calcium changed. A fall in albumin may explain lower total calcium, whereas alkalemia or citrate exposure may alter the free fraction. A magnesium disturbance can simultaneously contribute to renal potassium loss and an inadequate PTH response. Falling phosphate after nutrition can reflect intracellular incorporation superimposed on prior depletion. These explanations require separate balances rather than treating all four results as one interchangeable electrolyte deficit.

Finally, biochemical plausibility and outcome evidence answer different questions. Steele and colleagues' observational study of 1,038 ICU admissions examined hypocalcemia and supplementation. Such data describe clinical trajectories and associations but cannot establish the causal benefit or harm of replacement because treatment selection and illness severity are intertwined. 14

Link serial measurements to changing input, urine output, acid–base state, renal function, cell injury, or nutritional demand. Electrical, neuromuscular, and circulatory assessment adds information beyond the number, while optimal treatment targets remain a separate question from established transport physiology.

References

1. Hodgkin AL, Huxley AF. A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol. 1952;117:500–544. doi:10.1113/jphysiol.1952.sp004764

2. Montague BT, Ouellette JR, Buller GK. Retrospective review of the frequency of ECG changes in hyperkalemia. Clin J Am Soc Nephrol. 2008;3:324–330. doi:10.2215/CJN.04611007

3. DeFronzo RA, Felig P, Ferrannini E, Wahren J. Effect of graded doses of insulin on splanchnic and peripheral potassium metabolism in man. Am J Physiol Endocrinol Metab. 1980;238:E421–E427. doi:10.1152/ajpendo.1980.238.5.E421

4. Slomp J, van der Voort PHJ, Gerritsen RT, Berk JAM, Bakker AJ. Albumin-adjusted calcium is not suitable for diagnosis of hyper- and hypocalcemia in the critically ill. Crit Care Med. 2003;31:1389–1393. doi:10.1097/01.CCM.0000063044.55669.3C

5. Boink ABTJ, Buckley BM, Christiansen TF, et al. IFCC recommendation on sampling, transport and storage for the determination of the concentration of ionized calcium in whole blood, plasma and serum. Clin Chim Acta. 1991;202:S13–S21. doi:10.1016/0009-8981(91)90267-G

6. Brown EM, Gamba G, Riccardi D, et al. Cloning and characterization of an extracellular Ca²⁺-sensing receptor from bovine parathyroid. Nature. 1993;366:575–580. doi:10.1038/366575a0

7. Schlingmann KP, Weber S, Peters M, et al. Hypomagnesemia with secondary hypocalcemia is caused by mutations in TRPM6, a new member of the TRPM gene family. Nat Genet. 2002;31:166–170. doi:10.1038/ng889

8. Yang L, Frindt G, Palmer LG. Magnesium modulates ROMK channel-mediated potassium secretion. J Am Soc Nephrol. 2010;21:2109–2116. doi:10.1681/ASN.2010060617

9. Rude RK, Oldham SB, Singer FR. Functional hypoparathyroidism and parathyroid hormone end-organ resistance in human magnesium deficiency. Clin Endocrinol (Oxf). 1976;5:209–224. doi:10.1111/j.1365-2265.1976.tb01947.x

10. Chanutin A, Curnish RR. Effect of organic and inorganic phosphates on the oxygen equilibrium of human erythrocytes. Arch Biochem Biophys. 1967;121:96–102. doi:10.1016/0003-9861(67)90013-6

11. Aubier M, Murciano D, Lecocguic Y, et al. Effect of hypophosphatemia on diaphragmatic contractility in patients with acute respiratory failure. N Engl J Med. 1985;313:420–424. doi:10.1056/NEJM198508153130705

12. Shimada T, Kakitani M, Yamazaki Y, et al. Targeted ablation of Fgf23 demonstrates an essential physiological role of FGF23 in phosphate and vitamin D metabolism. J Clin Invest. 2004;113:561–568. doi:10.1172/JCI19081

13. Doig GS, Simpson F, Heighes PT, et al. Restricted versus continued standard caloric intake during the management of refeeding syndrome in critically ill adults: a randomised, parallel-group, multicentre, single-blind controlled trial. Lancet Respir Med. 2015;3:943–952. doi:10.1016/S2213-2600(15)00418-X

14. Steele T, Kolamunnage-Dona R, Downey C, Toh CH, Welters I. Assessment and clinical course of hypocalcemia in critical illness. Crit Care. 2013;17:R106. doi:10.1186/cc12756