Bone and mineral metabolism

What it normally does

• Bone is rebuilt continuously in small packets. A cell that digs (osteoclast, a fused multinucleated cell of the monocyte-macrophage line) resorbs a pit over two to three weeks; cells that fill (osteoblasts) then lay down new matrix over three to four months. About 10% of the skeleton is replaced each year, trabecular bone faster than cortical. Crucially, the digger takes its orders from the builder: osteoblasts and osteocytes display RANK ligand (RANKL), which binds RANK on osteoclast precursors and drives them to mature and survive, and they also secrete osteoprotegerin (OPG), a decoy that mops RANKL up. The RANKL:OPG ratio is the dial. Oestrogen turns it down; parathyroid hormone (when the level is continuously high), glucocorticoids and inflammatory cytokines turn it up.

  Explains why bone is lost after menopause and on steroids, why an antibody against RANKL is a bone drug at all, and why every drug that switches remodelling off buys strength today at the cost of un-repaired microdamage later.

• Bone is two things at once: a protein scaffold of type I collagen (osteoid), and mineral crystallised onto it (hydroxyapatite, essentially calcium phosphate). Osteoblasts pump out alkaline phosphatase (ALP) to strip the local inhibitor pyrophosphate so mineral can crystallise — which is why serum ALP is a marker of osteoblast activity, not of bone loss. So bone can fail two entirely different ways: too little of a normal, properly mineralised matrix, or a normal or increased amount of matrix that never got mineralised.

  Explains the whole osteoporosis-versus-osteomalacia split, and why ALP is stone normal in one and climbing in the other.

• Serum calcium is defended, and the skeleton is what pays. Of the total calcium (normal 2.10-2.60 mmol/L), roughly 40% is protein-bound — mostly to albumin — about 10% is complexed to anions, and the remaining half circulates free as ionised calcium (1.15-1.30 mmol/L), which is what the body actually senses. So a low albumin gives a falsely low total: correct it, or measure ionised calcium. Parathyroid chief cells read ionised calcium through a calcium-sensing receptor; a fall of a few hundredths releases stored parathyroid hormone (PTH) within seconds to minutes. PTH then does three things: pulls calcium out of bone (indirectly — osteoclasts carry no PTH receptor, so PTH acts by raising osteoblast and osteocyte RANKL), makes the kidney reabsorb more calcium in the distal tubule, and switches on renal 1-alpha-hydroxylase to make active vitamin D so the gut absorbs more. It also dumps phosphate in the urine.

  Explains why PTH excess gives high calcium with low phosphate, why the skeleton thins when PTH is chronically high, and why every calcium result must be read against albumin and PTH together.

• Vitamin D is a prohormone that needs three organs. UVB on skin converts 7-dehydrocholesterol to previtamin D3, which isomerises to cholecalciferol (D3); the liver adds one hydroxyl to make 25-hydroxyvitamin D — the storage form, half-life of weeks, and the one you measure; the kidney adds the second to make calcitriol (1,25-dihydroxyvitamin D), the active hormone, half-life of hours. Calcitriol acts on a nuclear receptor in the enterocyte and drives absorption of BOTH calcium and phosphate. That matters: mineralisation needs a calcium-phosphate product high enough to crystallise, and calcitriol is what supplies both halves of it.

  Explains why vitamin D deficiency starves the osteoid of mineral rather than simply lowering calcium, why 25(OH)D is the test and calcitriol is not, and why the liver and kidney sit in the middle of a bone disease.

What goes wrong

• Osteoporosis (postmenopausal and age-related)← from “Bone is rebuilt continuously in small packets.…”

  Each remodelling cycle leaves a small deficit, and after menopause the deficit widens: oestrogen loss raises RANKL relative to OPG, so more osteoclasts are recruited and each lives longer. Resorption pits are dug faster than osteoblasts can fill them. Trabecular bone, with its huge surface area, goes first — the vertebral bodies and the trabecular-rich metaphyses — and once a trabecular plate is perforated there is nothing for osteoblasts to build on, so the loss is not just quantity but architecture. The classic fracture sites are distal radius, vertebrae and proximal femur. The bone that remains is normal bone. There is simply not enough of it.

  Normal calcium, normal phosphate, normal ALP, broken bone. Normal biochemistry is the diagnosis, not a reason to stop looking.

  You would find: A fracture from a fall at standing height or less (minimal-trauma fracture) — wrist in the sixties, vertebra and hip later. Height loss and thoracic kyphosis from vertebral wedging, most of which never presented as an acute event. DXA T-score at or below -2.5. The biochemistry is normal: calcium, phosphate, ALP and PTH all sit in range, and an abnormal result means either a recent fracture (ALP can be raised for a few months as callus forms) or a secondary cause you have not found yet — so it redirects the work-up rather than confirming the diagnosis. Commonly quoted lifetime risk from age 50 is around one in three women and one in five men; about a quarter of people are dead within a year of a hip fracture and many of the survivors never walk unaided again. Aboriginal and Torres Strait Islander Australians tend to fracture at younger ages, carrying more of the risk load (smoking, diabetes, chronic kidney disease) and far less access to DXA in regional and remote areas — so it is under-diagnosed rather than uncommon.

• Osteomalacia and rickets (vitamin D deficiency)← from “Vitamin D is a prohormone that needs three org…”

  Without calcitriol the gut cannot absorb enough calcium or phosphate. Serum calcium starts to fall, PTH rises to defend it, and that secondary hyperparathyroidism succeeds — calcium is dragged out of bone and reclaimed by the kidney — but PTH also throws phosphate into the urine. The result is a normal or low-normal calcium sitting on top of a low phosphate, and a calcium-phosphate product too low to crystallise. Osteoblasts keep laying down collagen; that osteoid simply never mineralises. Soft, unmineralised bone bends and aches. In a child the same failure hits the growth plate, where cartilage cannot mineralise and the plate piles up wide and disorganised — that is rickets.

  High ALP with low phosphate and high PTH is osteomalacia until proven otherwise. Osteoporosis breaks normal bone; osteomalacia bends soft bone.

  You would find: Adults: dull generalised bone pain and tenderness, and proximal myopathy — difficulty rising from a chair, a waddling gait. Children: bowed legs, wide wrists, a rachitic rosary at the costochondral junctions, delayed walking. Biochemistry runs as a set: 25(OH)D low (Australian and New Zealand thresholds: sufficiency 50 nmol/L or above at the end of winter, mild deficiency 30-49, moderate 12.5-29, severe below 12.5), calcium low-normal, phosphate low, PTH high, and ALP HIGH — busy osteoblasts with nothing to mineralise with. Osteomalacia needs sustained, usually moderate-to-severe deficiency; a 25(OH)D of 40 nmol/L alone does not explain a bone lesion. X-ray may show Looser zones (pseudofractures) across the pubic rami, scapula or medial femoral neck. In Australia this concentrates in people with deeply pigmented skin, women who cover for religious reasons, aged-care residents and the housebound, and southern states through winter; paediatric rickets is seen most in refugee and migrant families, and in some Aboriginal and Torres Strait Islander infants who are breastfed by a deficient mother.

• Paget disease of bone← from “Bone is rebuilt continuously in small packets.…”

  In one or a few bones the osteoclasts go rogue — abnormally large, many-nucleated, wildly overactive. Osteoblasts chase them and lay down bone at frantic speed, but fast bone is woven bone: disorganised collagen, no lamellar structure, hypervascular. The affected bone ends up bigger, thicker, hotter and weaker than normal bone. Because resorption and formation stay coupled, calcium and phosphate stay normal — the whole disturbance shows up as the osteoblast marker.

  Isolated high ALP, normal calcium, normal phosphate, older patient, enlarged bone. That combination is Paget until the bone scan says otherwise.

  You would find: Most cases are silent and found as an isolated raised ALP on a routine panel with normal calcium, phosphate and liver enzymes (a normal GGT supports a bony rather than hepatic source). When symptomatic: deep bone pain that is worse at night and not relieved by rest, a bowed tibia, a skull that has quietly outgrown a hat, warmth over the bone from its own blood supply, and deafness — usually sensorineural from involvement of the otic capsule, sometimes conductive from the ossicles — when the temporal bone is involved. Pelvis, femur, lumbar spine, skull and tibia are the usual sites. Isotope bone scan lights the affected bones up; x-ray shows coarse trabeculae and expanded cortex. Rare complications are high-output cardiac failure and osteosarcoma (well under 1%, but new pain in a known Paget bone deserves imaging). Australian prevalence has more than halved over recent decades for reasons nobody has pinned down; it is a disease of people over 55, mostly of northern European ancestry.

• Primary hyperparathyroidism← from “Serum calcium is defended, and the skeleton is…”

  A parathyroid adenoma (a single gland in roughly 80-85% of cases) secretes PTH without regard for calcium. The calcium-sensing receptor loop is broken at its source. Chronically raised PTH keeps osteoblast and osteocyte RANKL high, so resorption runs continuously and cortical bone thins fastest — the opposite site pattern to postmenopausal loss, which strips trabecular bone first. The kidney reabsorbs a higher fraction of filtered calcium, but it is fed far more of it, so the filtered load wins and urine calcium climbs; phosphate is wasted in parallel.

  High calcium + low phosphate + unsuppressed PTH = primary hyperparathyroidism. High calcium with a suppressed PTH sends you looking for malignancy instead.

  You would find: Nearly always found now as an incidental high corrected calcium on a routine panel, with a PTH that is high or 'inappropriately normal' — a normal PTH with a high calcium is abnormal, because it should be suppressed. Phosphate low or low-normal, urine calcium normal or high, ALP normal or mildly raised. Symptoms when they come are the old rhyme: bones (pain, thinning, rarely the brown tumours of osteitis fibrosa cystica), stones (calcium renal calculi), abdominal moans (constipation, nausea, pancreatitis) and psychic groans (fatigue, low mood, poor concentration), plus polyuria and thirst from hypercalcaemia blunting the collecting duct's response to ADH. The must-exclude is familial hypocalciuric hypercalcaemia, a benign inactivating calcium-sensing receptor mutation that mimics it closely except that the urine calcium is low — check a urine calcium:creatinine clearance ratio (below about 0.01 suggests FHH) before anyone books an operation.

• Glucocorticoid-induced osteoporosis← from “Bone is rebuilt continuously in small packets.…”

  Glucocorticoids attack the builder and unleash the digger at the same time. They push osteoblasts and osteocytes into apoptosis and suppress new osteoblast recruitment, while raising RANKL and lowering OPG so osteoclasts live longer. They also cut gut calcium absorption, increase urinary calcium loss, and suppress the sex steroids that would otherwise restrain remodelling. Bone loss is fastest in the first three to six months, and because osteocyte death degrades bone quality independently of bone mass, fractures happen at a higher BMD than in postmenopausal osteoporosis.

  Steroids kill osteoblasts and free osteoclasts. The loss is early, the spine goes first, and the T-score under-reads the risk.

  You would find: Vertebral crush fractures — often painless, found as unexplained height loss or on a chest x-ray done for something else — in anyone on a moderate or higher dose of oral prednisolone (conventionally 7.5 mg daily or more) for three months or more: polymyalgia rheumatica, inflammatory bowel disease, rheumatoid arthritis, transplant, severe asthma. The catch is that a DXA T-score reassures falsely here, so risk is judged on the steroid exposure as much as the number. Do not forget the same patient may be developing steroid-induced proximal myopathy, which raises falls risk on top of the fragile spine.

What we give, and how it works

Open a drug to see what it binds, what that does to the cell, and what you then see in the patient.