Approximately 1 g of phosphorus is ingested daily in an average diet in the United States. About 300 mg is excreted in the stool, and 700 mg is absorbed (Fig. 73.1). Most of the phosphorus is absorbed in the duodenum and jejunum with minimal absorption occurring in the ileum.²⁰⁴ Phosphorus transport in proximal segments of the small intestine appears to involve both passive and active components and to be under the influence of vitamin D. The movement of phosphorus from the intestinal lumen to the blood requires (1) transport across the luminal brush-border membrane of the intestine; (2) transport through the cytoplasm; and (3) transport across the basolateral plasma membrane of the epithelium. The rate-limiting step and the main driving force of absorption is the luminal membrane step.³⁷³

The mechanism of transport across the intestinal brush border epithelial membrane involves a sodium-phosphate (NaPi) cotransport system, NaPi-IIb.²⁶⁵ The NaPi cotransporters are a secondary active form of ion transport using the energy of the Na gradient from outside to inside the cell to move phosphate ion uphill against an electrochemical gradient (Fig. 73.2).

The intestinal NaPi-IIb transporter is upregulated by a low phosphate diet and 1,25-dihydroxyvitamin D3.²⁵⁸,³¹⁴ Although low phosphate diets upregulate1,25-dihydroxyvitamin D3, studies in vitamin D-receptor (VDR) null mice indicate that the intestinal NaPi cotransport adaptation to a low phosphate diet occurs independent of vitamin D.⁵⁸⁵

Intestinal NaPi cotransport activity and NaPi-IIb protein is also regulated by several other factors—including the aging process,⁷²² glucocorticoids,²⁶ epidermal growth factor (EGF),⁷²³ and liver X receptor (LXR)¹⁰⁶—that decrease intestinal NaPi transport, and estrogen⁷²⁴ and metabolic acidosis⁶³⁸ that increase intestinal NaPi transport.

Studies of phosphorus accumulation by rat intestinal brush-border vesicles have demonstrated that it is affected by the transmembrane potential, indicating that like the renal type IIa cotransporter, NaPi-IIa, the intestinal type IIb cotransporter, NaPi-IIb, is electrogenic.²⁶⁵ The K_m(P_i) of NaPi-IIb is approximately 50 µm, similar to the renal transport protein. In contrast to the renal NaPi-IIa isoform, the intestinal NaPi-IIb cotransporter is less dependent on the pH level.

The second component of transcellular intestinal phosphorus transport involves the movement of phosphorus from the luminal to the basolateral membrane. Although little is known about the cellular events that mediate this transcellular process, evidence suggests a role for the microtubular microfilament system of intestinal cells.²⁰⁴ Microfilaments in the cell may be important in conveying phosphorus from the brush-border membrane to the basolateral membrane and may be involved in the extrusion of phosphorus at the basolateral membrane from the epithelial cell.

Little is known about the mechanisms of phosphorus extrusion at the basolateral membrane of intestinal epithelial cells.
The electrochemical gradient for phosphorus favors movement from the intracellular to the extracellular compartment because the interior of the cell is electrically negative compared with the basolateral external surface. Therefore, the presumption has been that the exit of phosphorus across the basolateral membrane represents a mode of passive transport.³²¹

Most of the inorganic phosphorus in serum (90% to 95%) is ultrafiltrable at the level of the glomerulus. At physiologic levels of serum phosphorus, approximately 7 g of phosphorus is filtered daily by the kidney, of which 80% to 90% is reabsorbed by the renal tubules and the remainder is excreted in the urine (approximately 700 mg on a 1-g phosphorus diet) equal to intestinal absorption.³³² As a result, adults are generally in balance between phosphorus intake and excretion (Fig. 73.1). Micropuncture studies have demonstrated that 60% to 70% of the filtered phosphorus is reabsorbed in the proximal tubule. However, there is also evidence that a significant amount of filtered phosphorus is reabsorbed in distal segments of the nephron.⁵⁰² When serum phosphorus levels increase and the filtered load of phosphorus increases, the capacity to reabsorb phosphorus also increases. However, a maximum rate of transport (Tm) for phosphorus reabsorption is obtained usually at serum phosphorus concentrations of 6 mg per dL. There is a direct correlation between Tm phosphorus values and glomerular filtration rate (GFR) even when the GFR is varied over a broad range. Micropuncture studies suggest two different mechanisms responsible for phosphorus reabsorption in the proximal tubule. In the first third of the proximal tubule, in which only 10% to 15% of the filtered sodium and fluid is reabsorbed, the ratio of tubular fluid (TF) phosphorus to plasma ultrafiltrable (UF) phosphorus falls to values of approximately 0.6. This indicates that the first third of the proximal tubule accounts for approximately 50% of the total amount of phosphorus reabsorbed in this segment of the nephron. In the last two thirds of the proximal tubule, the reabsorption of phosphorus parallels the movement of salt and water. In the remaining 70% of the pars convoluta, the TF:UF phosphorous ratio remains at a value of 0.6 to 0.7, whereas fluid reabsorption increases to approximately 60% to 70% of the filtered load. Thus, in the last two thirds of proximal tubule, the TF:UF phosphorus reabsorption ratio is directly proportional to sodium and fluid reabsorption. A significant amount of phosphorus, perhaps on the order of 20% to 30%, is reabsorbed beyond the portion of the proximal tubule that is accessible to micropuncture. There is little phosphorus transport within the loop of Henle, with most transport distal to micropuncture accessibility occurring in the distal convoluted tubule. In this location, Pastoriza-Munoz et al.⁵⁰² found that approximately 15% of filtered phosphorus is reabsorbed under baseline conditions in animals subjected to parathyroidectomy, but that the value falls to about 6% after administration of large doses of parathyroid hormone (PTH). The collecting duct is a potential site for distal nephron reabsorption of phosphorus.¹¹⁵,⁵⁰⁸,⁵⁹² Transport in this nephron segment may explain the discrepancy between the amount of phosphorus delivered to the late distal tubule in micropuncture studies and the considerably smaller amount of phosphorus that appears in the final urine of the same kidney. Phosphorus transport in the cortical collecting tubule is independent of regulation by PTH. This is in agreement with the absence of PTH-dependent adenylate cyclase in the cortical collecting tubule.¹¹⁵

The contribution of superficial nephrons and deep nephrons of the kidney to phosphorus homeostasis differs.
Nephron heterogeneity in phosphorus handling has been evaluated under a number of conditions by puncture of the papillary tip and the superficial early distal tubule, with the recorded fractional delivery representing deep and superficial nephron function, respectively. Microinjection of phosphorus tracer into thin ascending and descending limbs of loops of Henle reveals that only 80% of phosphorus was recovered in the urine, whereas 88% to 100% of phosphorus was recovered when the tracer was injected into the late superficial distal tubule. It was concluded that a significant amount of phosphorus must be reabsorbed by juxtamedullary distal tubules or by segments connecting the juxtamedullary distal tubules to the collecting ducts to account for the discrepancy between the results of superficial nephron injection and juxtamedullary nephron injections. These data support an increased reabsorptive capacity for phosphorus in deep as opposed to superficial nephrons and increased responsiveness to body Pi requirements.²⁵³,²⁵⁴

In summary, phosphorus transport occurs in the distal nephron, particularly in the distal convoluted tubule and cortical collecting tubular system. This transport may be considerable under certain experimental conditions, but the importance of the terminal nephron system in day-to-day phosphorus homeostasis remains to be defined. It is also evident from data obtained from various micropuncture and microinjection studies that juxtamedullary and superficial nephrons have different capacities for phosphorus transport. The increased responsiveness of the deep nephrons to phosphorus intake suggests a key regulatory role for this system in phosphorus homeostasis.

The apical membrane of renal tubular cells is the initial barrier across which phosphorus and other solutes present in the tubular fluid must pass to be transported into the peritubular capillary network. Because the electrical charge of the cell interior is negative to the exterior, and phosphorus concentrations are higher in the cytosol, phosphorus must move against an electrochemical gradient into the cell interior, whereas at the antiluminal membrane, the transport of phosphorus into the peritubular capillary is favored by the high intracellular phosphorus concentration and the electronegativity of the cell interior. Studies with apical membrane vesicles have demonstrated cotransport of Na⁺ with phosphate across the brush-border membrane, whereas the transport of phosphorus across the basolateral membrane is independent of that of Na⁺.²⁷² The apical membrane Na⁺-phosphate cotransport protein (NaPi-IIa) energizes the uphill transport of phosphate across the brush-border membrane (BBM) by the movement of Na⁺ down its electrochemical gradient. The latter gradient is established and maintained by active extrusion of Na⁺ across the basolateral cell membrane into the peritubular capillary through the action of Na⁺-K⁺-ATPase (Fig. 73.2).⁵⁷¹

Three families of NaPi cotransport proteins of the proximal tubule (types I, II, and III) have been cloned.⁴¹⁴,⁴⁷²,⁴⁷³,⁶³⁶,⁶⁶¹,⁷⁰⁹ The DNA clones encode 80- to 95-kd proteins that reconstitute Na⁺-dependent concentrative, or “uphill,” transport of phosphate.²⁰³,⁴¹⁴,⁶³⁶ The type I cotransporter, Npt1/SLC17, is expressed predominantly in the renal proximal tubule, and it accounts for about 13% of the known NaPi cotransporter mRNA in the mouse kidney.⁶⁶² Npt1 is not regulated by dietary Pi, and studies in Npt1-cRNA-injected oocytes revealed that it may function not only as a NaPi cotransporter but also as a chloride and organic anion channel.¹⁰³

The type II cotransporter NaPi-II/SLC34 proteins are similar between several species including humans.⁴⁷²,⁴⁷³,⁶⁶¹ In addition to NaPi-IIa, the predominant isoform in the renal proximal tubule, another isoform, NaPi-IIc, has been discovered.⁴⁸⁶,⁵⁸⁴,⁵⁸⁶,⁶⁶⁰ Nephron localization of NaPi-II proteins has been limited to the proximal tubule of superficial and deep nephrons (greatest in the latter, concordant with physiologic studies).⁴⁷²,⁴⁷³,⁶⁶¹ Immunolocalization studies in renal epithelial cells demonstrated apical membrane and subapical membrane vesicle staining,⁴⁷²,⁴⁷³,⁶⁶¹ suggesting that a functional pool of transporters is available for insertion into or retrieval from the BBM itself. This has been postulated to be a major mechanism of Pi transport regulation in response to acute changes in phosphorus, PTH, MEPE, and fibroblast growth factor 23 (FGF23) levels.³³,⁴⁷²,⁴⁷³,⁶⁰⁹,⁶⁶¹ The NaPi-II family is upregulated at message and protein levels by chronic feeding of low-Pi diets³⁶⁸,⁴⁰³ and downregulated at message and protein levels by PTH¹³⁶,³⁶⁸,⁴⁰³ and dietary potassium deficiency.⁹²

The type III NaPi cotransporters SLC20 were originally identified as retroviral receptors for gibbon ape leukemia virus (Glvr1) and rat amphotropic virus (Ram1).³¹⁷ They are ubiquitously expressed, and they comprise about 1% of the known NaPi cotransporter mRNAs in the mouse kidney.⁶⁶² Pit-2 protein is expressed in the apical membrane of the renal proximal tubule and the levels are regulated by dietary Pi,⁶⁷⁸ dietary potassium,⁹² and LXR.¹⁰⁶

Studies of phosphorus exit across the basolateral membrane suggest that it is accompanied by the net transfer of a negative charge and occurs down a favorable electrochemical gradient via sodium-independent mechanisms.⁵⁸²

Several proteins with PDZ domains have been identified that interact with the NaPi-IIa and NaPi-IIc protein and are localized in the BBM or the subapical compartment (Fig. 73.3). PDZ domains are modular protein interaction domains that often occur in scaffolding proteins and bind in a sequencespecific fashion to the C-terminal peptide sequence or at times the internal peptide sequences of target proteins. These domains of approximately 90 amino acids are known by an acronym of the first three PDZ-containing proteins identified including the postsynaptic protein PSD-95/SAP90, the
Drosophila septate junction protein Discs-large, and the tight junction protein ZO-1.²⁵⁵,²⁸⁸,³⁶⁷,⁵⁹⁶

PDZ domain containing proteins including NHERF-1, NHERF-2, PDZK1, CAL, and ZO-1 play an important role in: (1) the regulation of the expression and activity of renal proximal tubular BBM transport proteins including NHE-3,⁵⁹⁶,⁶⁹⁷,⁶⁹⁸,⁶⁹⁹,⁷⁰⁰ NaPiIIa,³³,³⁵⁹ and NaPi-IIc²¹⁹,⁶⁷⁷ and basolateral membrane transport proteins including Na-K-ATPase³⁵⁹ and Na-HCO₃ cotransporter (NBC)⁶⁹⁶; (2) the regulation of the expression and activity of cystic fibrosis transmembrane conductance regulator (CFTR), a cAMP-regulated chloride channel and channel regulator³⁸⁹,⁴⁷⁷,⁵⁴⁰,⁶⁵³,⁶⁸⁶; (3) parathyroid hormone 1 receptor signaling⁴¹⁷ and endocytic sorting of the β2-adrenergic receptor¹⁰⁷ and platelet-derived growth factor receptor (PDGFR)³¹³,⁴³⁶; (4) epithelial cell polarity and formation of tight junctions⁶⁴,²⁸⁹; and (5) maintaining the integrity of the glomerular barrier to proteins through interactions with podocalyxin, negatively charged sialoprotein expressed on the surface of podocytes, the glomerular visceral epithelial cells.²⁸⁴,²⁸⁵,⁴⁹⁰,⁵⁰³,⁵⁰⁴,⁶⁵⁵

In addition to their interaction with membrane proteins and receptors, the PDZ domain-containing proteins also interact with the F-actin cytoskeleton through their interactions with the ezrin-radixin-moesin (ERM) proteins (Fig. 73.3).⁹¹,⁵²⁴,⁶³³ ERM proteins are typically located peripherally in the membrane and link the cytoplasmic tails of membrane proteins and receptors to the cortical actin cytoskeleton. The ERM proteins play an important role in the formation of microvilli, cell-cell junctions, and membrane ruffles and also participate in signal transduction pathways. The ERM proteins contain an F-actin binding site within their carboxy-terminal 30 residues. In addition, the ERM proteins have FERM (four-point one, ezrin, radixin, moesin) domains, which are generally located at or near the amino terminal, and act as multifunctional protein and lipid binding sites. The FERM domain of ezrin interacts strongly with NHERF-1 and NHERF-2. NHERF-1 and NHERF-2 have 2 PDZ domains and have a carboxy-terminal sequence of 30 amino acids that bind ezrin.

Using the molecular approach (yeast two-hybrid) several proteins with PDZ domains that interact with the C terminus of NaPi-IIa have been identified including: (1) NHERF-1/EBP50, (2) NHERF-2/E3KARP, (3) PDZK1/NaPi-Cap1, (4) PDZK2/NaPi-Cap2, and (5) CAL, a CFTRassociated ligand.⁶⁰,⁶⁸,²²⁰,²²¹,²²²,²⁶³,⁴⁵⁶,⁵¹⁶,⁵⁹⁴

Different studies suggest that apical expression of NaPi-IIa depends on the presence of NHERF-1. Expression of NaPi-IIa was reduced upon introduction of dominant-negative form of NHERF-1 in OK cells.²²⁰,²²¹,²²²,²⁶³ The in vivo importance of the
NaPi-IIa interaction with NHERF-1 was also shown in a study where targeted disruption of the mouse NHERF-1 gene was associated with decreased BBM expression and increased intracellular localization of NaPi-IIa resulting in decreased renal phosphate reabsorption and renal phosphate wasting.⁵⁹⁵ On the other hand targeted disruption of NHERF-1 did not modulate the BBM expression and localization of NHE3; however, there was impaired regulation of NHE3 in response to PKA.⁵⁹⁵

In contrast to NHERF-1, targeted disruption of the PDZK1 gene failed to modulate the BBM expression of NaPi-IIa.¹⁰⁸,³³⁴ NHERF-1, therefore, plays a critical and unique role in the renal proximal tubular apical membrane targeting of NaPi-IIa protein and maintenance of phosphate homeostasis. However targeted disruption of PDZK1modulates the BBM expression of NaPi-IIc, as compared to NaPi-IIa PDZK1, which has preferential interactions with NaPi-IIc.²¹⁹,⁶⁷⁷

Recent studies have identified at least three additional proteins that may be important in the regulation of NaPi-IIa targeting and trafficking: (1) the peroxisomal protein PEX 19, a farnesylated protein that confers PTH responsiveness to NaPi-IIa²⁹⁷; (2) the calcium binding protein Vilip-3, a myristoleated protein that may be important in calcium dependent regulation of NaPi-IIa⁵¹⁷; and (3) MAP 17 which may be important for apical expression of PDZK1 (Fig. 73.3).⁵¹⁶

Of the multiplicity of factors that regulate phosphate transport in the kidney, the most important are dietary phosphate, PTH, and FG23.

Through studies of familial hypophosphatemia and tumor induced osteomalacia, a new hormone operating in a systems biology network regulating phosphate homeostasis between the skeleton and the kidney has been discovered (Fig. 73.5).³⁰⁷,⁶⁰⁰ The hormone is FGF23, secreted by skeletal osteocytes in response to changes in bone formation and serum phosphorus.⁴⁸⁷ The physiology is that the osteocyte monitors deposition of phosphorus into the skeletal reservoir and the saturation of the exchangeable phosphorus pool. When Pi levels increase within the pool due to decreased exit into the skeleton (bone formation) or due to increased plasma phosphorus, osteocytes secrete FGF23, which acts on the renal proximal tubule to decrease reabsorption and increase phosphorus excretion.²⁷⁹,⁶⁰⁴

FGF23 actions at the proximal tubule (PCT) have not been studied as extensively as the actions of PTH, which were described above. Studies indicate that FGF23 acts to decrease expression of NaPi-IIa and NaPi-IIc in the PCT.²⁰⁹,⁶⁰⁴ The actions of FGF23 on the PCT are mediated through binding to a FGF receptor—predominantly FGFR1(IIIc) and a coreceptor, Klotho.⁶⁷³ Signal transduction is stimulated through phosphorylation of extracellular signal-regulated kinase (ERK) and the immediate early response gene, early growth response-1 (Erg-1), a zinc finger transcription factor.²¹²,⁶⁷³ A matter of current uncertainty is related to KLOTHO expression which is mainly in the renal distal tubule whereas its signaling function is in the PCT. Recent studies have shown that it is expressed in the PCT which would resolve this issue (M. Kuro-o, personal communication).

Besides inhibiting PCT renal Pi transport, FGF23 signaling inhibits PCT CYP27B1, the 25-OH cholecalciferol 1α hydroxylase, and activates 24-OH hydroxylase (CYP24R1) resulting in decreased production and increased catabolism of calcitriol.⁶⁰⁰ In addition, activation of 24-OH hydroxylase results in the high prevalence of vitamin D deficiency associated with elevations of FGF23 levels, especially in chronic kidney disease (CKD). Calcitriol increases FGF23, closing the feedback loop in the system (Fig. 73.5).

Hypophosphatemia refers to serum phosphorus concentrations of less than 2.5 mg per dL. Hypophosphatemia usually results from one or a combination of the following factors (Fig. 6)³²⁸,³⁴⁰: (1) increased excretion of phosphorus in the urine; (2) decreased GI absorption of phosphorus; or (3) translocation of phosphorus from the extracellular to the intracellular space. The major causes of hypophosphatemia are shown in Table 73.1.

Several pathophysiologic conditions increase excretion of phosphorus in the urine. Some of these are characterized by elevated levels of circulating PTH or FGF23. Because PTH and FGF23 decrease phosphorus reabsorption by the kidney, elevations of the hormones increase urinary excretion (Table 73.1). Decreased tubular reabsorption of phosphorus may also occur without increased levels of PTH and may
be due to changes in the reabsorption of salt and water or to renal tubular defects specific for the reabsorption of certain solutes or phosphorus. Hypophosphatemia may also occur in the diuretic phase of acute tubular necrosis or in postobstructive diuresis, presumably due to a combination of high levels of PTH and decreased tubular reabsorption of salt and water.

Posttransplanthypophosphatemia, a commondisorder, is well described in the literature. Although described mainly in patients following renal transplantation,⁵⁶,²⁰⁸,²³⁴,²⁴⁵,⁴⁵⁸,⁴⁸⁸,⁴⁹²,⁵⁰⁰,⁵⁵⁵,⁵⁷⁴,⁶⁴³,⁶⁸⁷ posttransplant hypophosphatemia also occurs in patients undergoing bone marrow transplantation.¹⁴³,⁵³⁸ In all reports, the decrease in serum Pi concentration was associated with an increase in urinary phosphate excretion and a significant decrease in the measured or derived ratio of maximal rate of renal tubular transport of phosphate to glomerular filtration rate (TmPi/GFR).⁶⁸⁴ In addition to the impairment in renal tubular phosphate reabsorption, evidence indicates that intestinal phosphate absorption is impaired in transplant patients.¹⁸⁶,³⁶⁶,⁴²⁹,⁵⁵⁷

The mechanism for posttransplant hypophosphatemia has not been fully elucidated, but it is linked to disordered regulation of renal tubular reabsorption of Pi. As discussed earlier, PTH leads to a reduction in the expression of type II Na/Pi cotransport at the BBMs, which accounts for the phosphaturic action of PTH. Given this property of PTH, it has been postulated that increased PTH activity during chronic renal failure (CRF) may be the major mechanism responsible for maintaining Pi balance during CRF. According to this hypothesis, posttransplant hypophosphatemia has been attributed to persistent hyperparathyroidism (HPT), that is, incomplete involution of hyperplastic glands produced by renal failure prior to transplantation would cause hypophosphatemia and increased phosphaturia during the early posttransplant period.²⁶² Several studies, however, have documented that protracted HPT cannot account for the phenomenon of posttransplant hypophosphatemia since it can be seen in transplant patients with normal PTH levels. Moreover, transplant recipients failed to decrease Pi excretion in the urine even when PTH was suppressed by calcium infusion.⁵⁵⁵ In addition, the phosphaturia following kidney transplantation could not be ascribed to the effects of nephrectomy or to the influence of immunosuppressive drugs.⁵⁵⁵

A study by Green et al. determined that a non-PTH humoral mechanism accounted for the entity of posttransplant hypophosphatemia.²³⁷ The factor, however, had characteristics different from FGF-23, sFRP-4, and MEPE, phosphatonins discussed earlier.²³⁷ However, recent studies continue to focus on PTH and FGF23 as the causes of posttransplant hypophosphatemia in the early posttransplant period.

Because most of the absorption of phosphorus from the GI tract occurs in the duodenum and jejunum, gastrointestinal tract disorders such as celiac disease, tropical and nontropical sprue, and regional enteritis may decrease the absorption of phosphorus.²²⁴ Phosphorus malabsorption has also been described in patients who have undergone surgical bypass procedures for morbid obesity. The degree of hypophosphatemia varies among patients with intestinal malabsorption, being extremely mild in some and severe in others.

Most of the phosphorus ingested in the diet is present in protein, particularly meat, cheese, milk, and eggs. In many parts of the world where protein consumption is extremely low, hypophosphatemia occurs predominantly in children. Overall growth is retarded and a series of metabolic abnormalities are present.³²⁴

Certain compounds, mainly aluminum salts (aluminum hydroxide, aluminum carbonate gel) and calcium carbonate, are used in the treatment of hyperphosphatemia.⁵⁹⁸ However, when these compounds are given in excess, they may produce profound hypophosphatemia. These gels trap phosphorus in the small intestine and increase the amount of phosphorus in the stool. Patients ingesting large amounts of phosphate binders and not followed closely may develop phosphate depletion. With time, such individuals may develop severe weakness, bone pain, and osteomalacia.

Major reviews of the causes of hypophosphatemia in hospitalized patients³⁰⁹,³⁵² attributed most instances to intravenous administration of carbohydrate. However, many other causes were found, including diuretic usage, hyperalimentation, alcoholism, respiratory alkalosis, and use of phosphate binders.⁵⁵ A 31% incidence of hypophosphatemia was seen in patients admitted to a general medical ward, and a further fall in serum concentrations occurred in all patients with acute alcoholism between the second and fifth day after admission to a medical ward.⁵⁷⁰ Hypophosphatemia is also seen frequently during treatment of diabetic ketoacidosis.⁵⁸⁷ When diabetic patients develop ketoacidosis, they usually have an increase in phosphate excretion in the urine; however, the serum phosphate level may be slightly elevated due to acidosis. During the administration of insulin, there is a rapid decrease in the level of glucose with translocation of phosphate from the extracellular to the intracellular space, resulting in hypophosphatemia.

Acute respiratory alkalosis decreases urinary phosphate excretion but produces marked hypophosphatemia.⁴⁶⁵ In contrast, patients who receive sodium bicarbonate excrete large amounts of phosphate in the urine; however, the hypophosphatemia that may develop is only moderate in nature. It has been postulated that in respiratory alkalosis, there is an increase in the intracellular pH level with activation of glycolysis and increased formation of phosphate-containing sugars, leading to a precipitous fall in the concentration of serum phosphorus. The mild hypophosphatemia that may be seen during administration of sodium bicarbonate is probably secondary to increased renal phosphate excretion due to a decrease in ionized calcium and release of PTH, as well as to the consequences of ECF volume expansion.

In addition, new clinical disorders have been identified in which hypophosphatemia is an important aspect of the pathologic condition. Marked hypophosphatemia has been associated with acute leukemia or with lymphomas in the leukemic phase.⁸,⁴³⁵,⁷³⁰ These individuals typically present with hypophosphatemia, normocalcemia, and no evidence of excess PTH activity. Urinary phosphate concentration is typically extremely low. Although kinetic studies have not been performed in this setting, the facts that serum phosphate concentration correlates with a growth phase of the tumors and that hyperphosphatemia is seen when cells are destroyed by chemotherapy or radiotherapy strongly suggest that serum phosphorus was initially used in the rapid growth of new cells. Because these patients are often severely ill and under treatment with glucose infusions, as well as antacids and other drugs known to induce hypophosphatemia, they may be at great risk of developing severe acute phosphorus depletion.

Another clinical condition in which hypophosphatemia has been a prominent feature is the toxic shock syndrome. Chesney et al.¹²⁴ described 22 women with this disorder who showed hypocalcemia and hypophosphatemia as prominent manifestations. Whether respiratory alkalosis or staphylococcal sepsis-induced release of substances were responsible for acute phosphorus shifts into cells is unknown. Lindquist et al.³⁶³ studied in a prospective fashion the importance of hypophosphatemia in patients with severe burns. In 33 patients studied for 2 weeks after injury, transient hypophosphatemia was seen in the second to tenth day in all these individuals. Five of seven patients who died from complications of the terminal injury had severe hypophosphatemia. Because urinary phosphorus excretion was not increased, tissue uptake seems to be the predominant mechanism responsible for the hypophosphatemia. Levy³⁷⁸ reported the occurrence of severe hypophosphatemia during the rewarming phase in a profoundly hypothermic patient. In this individual, urinary excretion of phosphorus was minimal, suggesting that a shift of phosphate into the cells occurred as a result of rewarming. Finally, the development of hypophosphatemia resulting from refeeding clinically starved patients has been emphasized. Silvis et al.⁶¹² showed that the classic phosphorus-depletion syndrome, consisting of paresthesias, weakness, seizures, and hypophosphatemia, can occur in individuals who receive oral caloric supplements after a prolonged period of starvation. To further evaluate this issue, they performed studies in normal dogs that had been starved or had received normal diets and found that the infusion of calories through an intragastric catheter to previously starved animals resulted in a fall in serum phosphorus concentration from an average of 4.8 mg per dL to 1.6 mg per dL. Nearly 50% of starved animals developed clinical signs of phosphate depletion after oral refeeding. Weinsier and Krumdiek reported two patients who developed the phosphorus-depletion syndrome in association with cardiopulmonary decompensation following overzealous hyperalimentation after prolonged caloric deprivation.⁷⁰¹

The manifestations of hypophosphatemia are presented in Table 73.3. It has been suggested that the clinical manifestations of hypophosphatemia and severe phosphorus depletion are related to disturbances in cellular energy and metabolism. Studies have examined the effects of phosphate depletion on cellular energetics and other components of cell function. A study of glycolytic intermediates and adenine nucleotides during insulin treatment of patients with diabetic ketoacidosis emphasized the important effects of insulin-induced cellular phosphate depletion on cell metabolism.³³⁷ These results demonstrated that the reduced level of 2,3-diphosphoglycerate (2,3-DPG) seen during insulin treatment of diabetes is due to intracellular phosphorus depletion, producing a decrease in glyceraldehyde 3-phosphate dehydrogenase activity rather than inhibition of the phosphofructokinase enzyme system. Ditzel¹⁶³ has suggested that repeated transient decreases in red cell oxygen delivery due to reduced 2,3-DPG with insulin-induced hypophosphatemia could contribute over many years to the microvascular
disease seen in diabetic patients. Patients with mild degrees of hypophosphatemia are usually asymptomatic. However, if hypophosphatemia is severe—that is, if serum phosphorus levels are less than 1.5 mg per dL—a series of hematologic, neurologic, and metabolic disorders may develop. In general, the patients become anorectic and weak, and mild bone pain may be present if the hypophosphatemia persists for several months (Table 73.3).

Severe cardiomyopathy with decreased cardiac output has been described in patients and animals with severe hypophosphatemia.⁴⁸⁴,⁷³¹ Studies revealed that the resting muscle membrane potential fell, sodium chloride and water content of the tissue increased, and potassium content decreased in severe hypophosphatemia.²⁰⁵ These values returned to within the normal range after phosphate was administered. Skeletal muscle weakness and electromyographic abnormalities are associated with chronic hypophosphatemia and phosphate depletion. Dogs that were fed low-phosphate diets for several months developed changes in muscle, rhabdomyolysis, and characteristic increases in their levels of creatinine kinase and aldolase in blood.³²⁹ Rhabdomyolysis has been observed in alcoholic patients with hypophosphatemia.³³⁰ Knochel et al.³²⁹ showed that myopathy associated with phosphate depletion in dogs did lead to changes in cell water content, sodium concentration, and transmembrane potential difference. Kretz et al.³⁴¹ examined the possibility that changes in calcium transport in the sarcoplasmic reticulum of muscle were responsible for the clinical myopathy seen in acute phosphate depletion. Despite significant hypophosphatemia and a reduction in muscle phosphorus concentration, they found no significant changes in the rate of calcium uptake of calcium-concentrating ability in vesicles prepared from muscle sarcoplasmic reticulum of phosphate-depleted rats. Thus, the role of altered transcellular calcium
movements in phosphate-depleted tissues is yet to be completely resolved.

Hyperinsulinemia and abnormal glucose metabolism suggesting insulin resistance have been described in phosphate depletion. DeFronzo and Lange¹⁵⁵ have used the glucose and insulin clamp technique to study the kinetics of glucose metabolism in patients with various chronic hypophosphatemic conditions including vitamin D-resistant rickets. When glucose was infused to maintain constant glycemia at 125 mg per dL, hypophosphatemic individuals required 36% less glucose to maintain these glycemic levels than controls. Also when euglycemia was achieved by combined insulin and glucose infusion, the hypophosphatemic individuals required 40% less glucose to maintain euglycemia than controls. Insulin catabolism was apparently unaffected in these hypophosphatemic individuals. These data indicate that hypophosphatemia is associated with impaired glucose metabolism in both hyperglycemic and euglycemic patients.

Hematologic abnormalities of hypophosphatemia are a major manifestation of this syndrome.³⁸⁸,⁶⁶⁸ In addition to defects in affinity of oxyhemoglobin leading to generalized tissue hypoxia, there may be increased hemolysis.²⁹⁸,³²⁷ Quantitative and functional defects have also been described in platelets and leukocytes.¹⁴² These defects lead to diminished platelet aggregation and abnormalities in chemotaxis and phagocytosis of white blood cells. The latter may contribute to the increased risk of gram-negative sepsis reported in hypophosphatemic patients.⁵⁴⁹ This is of particular concern in immunosuppressed patients receiving phosphate-poor alimentation through a central venous line.

Manifestations at the level of the central nervous system, resulting in generalized anorexia and malaise or more severe disturbances such as ataxia, seizures, and coma, have been described in hypophosphatemia.⁴⁰⁴,⁴⁰⁵,⁵²² Neuromuscular abnormalities include paresthesias and weakness, the result of both myopathic changes and diminished nerve conduction.⁷²

The skeletal abnormalities associated with hypophosphatemia, particularly in vitamin D-resistant rickets, may be quite marked. In addition, bony abnormalities, including osteomalacia and pathologic fractures, have been described in antacid-induced phosphate depletion,³⁷,¹³⁹ as well as in hypophosphatemic patients undergoing hemodialysis who did not receive phosphate binding gels.¹¹ A rheumatic syndrome resembling ankylosing spondylitis also has been reported in hypophosphatemic patients.⁴⁶⁴

These manifestations include anorexia, nausea, and vomiting.⁴⁰⁵ It has been speculated that hypophosphatemia in the alcoholic patient may further impair hepatic function through hypoxic insult.

There is decreased phosphorus excretion and decreased tubular reabsorption of calcium, magnesium, bicarbonate, and glucose.¹³⁰,¹⁶⁵,¹⁸¹,²²⁷,²²⁸,²³⁰ The renal conservation of phosphorus occurs early in the syndrome and is the result of a primary increase in the tubular reabsorption of the anion and a decrease in the GFR and consequently in the filtered load of phosphorus.²³⁰,⁴⁶⁸ This mechanism results in complete renal conservation of phosphorus, with net losses representing only a small fraction of total body phosphorus stores.⁴⁰⁵ The increase in phosphorus reabsorption seen with phosphorus depletion is independent of several hormones known to influence phosphorus transport under other circumstances, including PTH, vitamin D, calcitonin, and thyroxine.⁶⁴⁰ The possibility that serum phosphorus concentration per se (or intracellular phosphorus) may in some manner regulate its absorption along the nephron seems plausible. Hypercalciuria of enough magnitude to produce a negative calcium balance is seen commonly in hypophosphatemic patients. Several factors contribute to this increase in calcium excretion including increased calcium mobilization from bone, enhanced GI tract calcium absorption, and inhibition of renal tubular calcium reabsorption.¹³⁰,¹⁶⁵,¹⁸¹,²²⁷,²²⁸,²³⁰ These effects appear to be independent of PTH activity and may be the result of a direct effect of phosphate on these transport processes.

Renal bicarbonate wasting, diminished titratable acid excretion, and decreased ammoniagenesis have been reported in hypophosphatemia.¹⁶⁵,⁴⁸⁵ However, these defects are counterbalanced to some extent by the mobilization of alkali from bone. Thus, steady-state pH may be near normal at the expense of skeletal buffers.¹⁸¹

In general, the cause of hypophosphatemia can be determined either from the medical history or from the clinical setting in which it occurs. When the cause is in doubt, measurement of the urinary phosphorus excretion level may be helpful. If the urinary phosphorus concentration is less than 4 mg per dL when the serum phosphorus level is less than 2 mg per dL, renal losses may be excluded.²⁵⁶ Of the three major extrarenal causes including diminished phosphorus intake, increased extrarenal losses (GI tract), and translocation into the intracellular space, the last is the most common, particularly in the hospitalized patient.³⁰⁹,³⁶⁴
When the urinary phosphorus excretion level is high, the differential diagnosis includes hyperparathyroidism, a primary renal tubular abnormality, or vitamin D-dependent or -resistant renal rickets. Measurements of serum calcium, PTH, and vitamin D and its metabolites, as well as urinary excretion of other solutes (glucose, amino acids, and bicarbonate), will usually elucidate the underlying disturbance that is responsible for the hypophosphatemia.

There are several general principles that apply to the treatment of hypophosphatemic patients. As with any predominantly intracellular ion (e.g., potassium), the state of total body phosphorus stores, as well as the magnitude of phosphorus losses, cannot be readily assessed by measurement of the concentrations in serum. In fact, under conditions in which a rapid shift of phosphorus has resulted from glucose infusion or hyperalimentation, total body stores of phosphorus may be normal, although with diminished intake and renal losses, there may be severe phosphorus depletion. Furthermore, the volume of distribution of phosphorus may vary widely, reflecting in part the intensity and duration of the underlying cause.³⁶⁴

In clinical situations in which hypophosphatemia is to be expected (e.g., glucose infusion or hyperalimentation in the alcoholic or nutritionally compromised patient during treatment of diabetic ketoacidosis), careful monitoring of the concentration of serum phosphorus is crucial. In these situations, addition of phosphorus supplementation to prevent the development of severe hypophosphatemia may prove very helpful. Certainly, other contributing causes of hypophosphatemia in this setting should be identified and treated. This is particularly true of the use of phosphate binding antacids (aluminum and magnesium hydroxide) for peptic ulcer disease, which may be replaced by aluminum phosphate antacids (Phosphagel) or cimetidine (Tagamet). It is now generally recommended that hyperalimentation solutions contain a phosphorus concentration of 12 to 15 mmol per L or 37 to 46.5 mg per dL, in order to provide an appropriate amount of phosphorus in the patient in whom renal impairment is absent.³⁶⁴ Phosphorus supplementation during glucose infusion or during the treatment of ketoacidosis is usually withheld until the serum phosphorus levels decrease to less than 1 mg per dL. Phosphorus may be given orally to these patients and others with mild asymptomatic hypophosphatemia in the form of skim milk, which contains 0.9 mg per mL, Neutra-phos (3.3 mg per mL), or phosphorus soda (129 mg per mL). However, intestinal absorption is quite variable, and diarrhea often complicates the oral administration of phosphate-containing compounds. For these reasons, parenteral administration is usually recommended in the hospitalized patient. If oral therapy is permissible, Fleet Phospho-Soda may be given at a dosage of 60 mmol daily in three doses (21 mmoL per 5 mL or 643 mg per 5 mL). A convenient method is to provide the phosphorus together with potassium replacement in these patients. Addition of 5 mL of potassium phosphate (K phosphate) into 1 L of intravenous fluid provides 22 mEq of potassium and 15 mmol (466 mg) of phosphorus.³⁶⁴ However, because potassium losses may greatly exceed the phosphorus deficit, the repletion of potassium should not be totally linked to phosphorus therapy. In patients with severe phosphate depletion, it is difficult to determine the magnitude of the total deficit of phosphorus and to calculate a precise initial dose. It is usually prudent to proceed with caution and repair the deficit slowly. The most frequently recommended regimen is 0.08 mmol per kg of body weight (2.5 mg per kg body weight) given over 6 hours for severe but uncomplicated hypophosphatemia and 0.016 mmol per kg of body weight (5 mg per kg of body weight) in symptomatic patients.³⁶⁴ Parenteral administration should be discontinued when the serum phosphorus concentration is greater than 2 mg per dL.

Calcium administration may be needed during phosphate repletion to prevent severe hypocalcemia. Calcium must not be added to bicarbonate- or phosphate-containing solutions because of the potential precipitation of calcium salts. Intravenous infusion of calcium gluconate or calcium chloride may be given until tetany abates. In addition to hypocalcemia, metastatic calcification, hypotension, hyperkalemia, and hypernatremia are potential side effects of parenteral infusion of phosphorus. These problems can be prevented by judicious use of therapy and frequent monitoring of serum electrolyte concentrations.

Hyperphosphatemia is said to occur when the serum phosphorus concentration exceeds 4.6 mg per dL in adults. In children, serum levels of phosphorus of up to 6 mg per dL may be physiologic. The most frequent cause of hyperphosphatemia is decreased excretion of phosphorus in the urine as a result of a fall in the GFR. However, increases in serum phosphorus concentration can also occur as a result of increased entry into the ECF due to excessive intake of phosphorus, increased release of phosphorus from tissue breakdown, and release of phosphorus from the skeletal reservoir through bone resorption. The major causes of hyperphosphatemia are listed in Table 73.4.