Peritoneum
A two-layered serous membrane with two completely different nerve supplies — which is why the same disease produces vague central pain one hour and a rigid, finger-pointing abdomen the next.
What it normally does
The inner layer wrapping the organs (visceral peritoneum) is supplied by autonomic afferents that travel back with the sympathetics to the spinal cord. It cannot sense cutting or burning — its adequate stimuli are stretch, distension, traction, ischaemia and chemical or inflammatory irritation — and the brain can only localise it to the midline at the level of the embryonic gut segment: foregut to the epigastrium, midgut to the umbilicus, hindgut to the suprapubic region.
This is why early appendicitis hurts around the navel and why bowel obstruction is felt as central colic, nowhere near the actual diseased organ.
The outer layer lining the abdominal wall (parietal peritoneum) is supplied by the same segmental somatic nerves as the overlying skin and muscle — T7 to L1 for the wall, and the phrenic nerve (C3, C4, C5) for the central diaphragm. Somatic means sharp, precisely localised, and wired to a reflex arc back to the muscle of that same segment.
This is why pain becomes point-tender once inflammation reaches the wall, why the muscle contracts on its own (guarding and rigidity), and why blood or pus under the diaphragm is felt at the shoulder tip.
The cavity between the two layers is only a potential space holding a small volume of lubricating fluid — usually well under 50 mL — but the membrane itself has a very large surface area, classically quoted as roughly that of the skin (about 1.7 square metres), though direct anatomical measurements give a smaller figure. It exchanges water and solutes in both directions across the capillaries of the tissues it covers, and drains via one-way lymphatic gaps (stomata) in the diaphragm.
This large exchange surface explains how litres of ascites accumulate, how peritoneal dialysis works, and how infection and fluid loss spread so fast in peritonitis.
Injured peritoneum repairs by regrowing mesothelium from islands across the whole surface, not from the edges inwards, and it normally dissolves the fibrin laid down during healing using its own plasminogen activator. Where tissue is ischaemic, dried out or handled roughly, that fibrinolysis fails and fibroblasts turn the fibrin into permanent fibrous bands.
This is why abdominal surgery leaves adhesions, and why those adhesions later strangle or kink bowel.
What goes wrong
- Appendicitis and the migrating pain← from “The inner layer wrapping the organs (visceral …”
The appendix lumen blocks (faecolith, lymphoid swelling). Mucus keeps being secreted behind the block, so the appendix distends. Stretch fires the visceral afferents, which run back with sympathetics to T10 — a midgut level — so the brain reports dull, poorly located pain around the umbilicus, often with nausea and anorexia. Over hours the wall becomes ischaemic and inflamed, and the inflammation reaches the parietal peritoneum lying against it. That layer is somatic, segmentally mapped, and now the pain is sharp, constant and precisely at McBurney point, with local guarding and rebound.
Migration of pain from periumbilical to right iliac fossa is visceral peritoneum handing over to parietal peritoneum. A retrocaecal appendix — the commonest position, not a rare variant — may never touch the anterior parietal layer, so the abdomen stays soft and the diagnosis is missed. That is the classic trap.
You would find: Pain that started around the navel and moved to the right iliac fossa. The patient reports it got worse on the drive in over speed bumps and hurts when they cough — both are the parietal layer being jarred. Percussion tenderness is a kinder test than rebound and means the same thing.
- Perforated viscus and generalised peritonitis← from “The outer layer lining the abdominal wall (par…”
A hole in the stomach, duodenum or colon spills acid, enzymes and faecal organisms across the whole peritoneal surface. Every somatic nerve from T7 to L1 is irritated at once, and each fires a reflex contraction of the muscle in its own segment, so the whole abdominal wall contracts — board-like rigidity. The inflamed membrane pours protein-rich fluid into the cavity and the gut stops moving (ileus), so litres are lost from the circulation into a space that cannot be seen. Bacteria and endotoxin cross the same huge surface into the blood.
Rigidity plus free gas is a surgical emergency, not a diagnostic puzzle. The shock is hypovolaemic before it is septic — the fluid has third-spaced into the peritoneal cavity. Resuscitate, give antibiotics, and get the source closed.
You would find: A patient lying absolutely still (movement stretches the parietal layer), rigid silent abdomen, tachycardia and hypotension. Free gas under the diaphragm on an erect chest x-ray — but this is absent in roughly a quarter to a third of perforations, so a normal film does not exclude one and CT is the next step. If the irritant reaches the underside of the central diaphragm, the pain is felt at the shoulder tip, because the phrenic nerve is C3, C4, C5 and the brain maps that to the shoulder.
- Ascites and spontaneous bacterial peritonitis← from “The cavity between the two layers is only a po…”
In cirrhosis, sinusoidal portal pressure rises. Hydrostatic pressure in the hepatic sinusoids and splanchnic capillaries now exceeds what the lymphatics can carry away, so hepatic lymph weeps from the liver capsule and splanchnic capillaries filter into the cavity faster than the diaphragmatic lymphatics can clear it; falling hepatic albumin production lowers plasma oncotic pressure and adds to this. Splanchnic vasodilation (largely nitric oxide driven) drops the effective arterial volume, so renin-angiotensin-aldosterone and ADH switch on and the kidney retains sodium and water — which feeds straight back into the ascites. That fluid is low in protein (a total protein below about 15 g/L is the risk marker) and poorly opsonised, so gut organisms that translocate across a leaky bowel wall are not cleared and multiply in it.
Every new or worsening ascites gets a diagnostic tap, and so does any decompensation in a patient who already has ascites. In Australia the common causes are alcohol-related liver disease, chronic hepatitis B and C, and increasingly metabolic dysfunction-associated fatty liver disease; chronic liver disease and its complications fall disproportionately on Aboriginal and Torres Strait Islander communities, driven by higher chronic hepatitis B prevalence and poorer access to antiviral treatment.
You would find: Distension with shifting dullness and a fluid thrill. Tap it: a serum-ascites albumin gradient of 11 g/L or more means portal hypertension. Ascitic neutrophils of 250 x 10^6/L or more is spontaneous bacterial peritonitis — and it often presents with nothing more than a low-grade fever or new confusion, with no rigidity at all, because the parietal peritoneum is barely involved.
- Adhesions and small bowel obstruction← from “Injured peritoneum repairs by regrowing mesoth…”
Surgery, blood or infection damages the mesothelial surface. Fibrin is laid down. Normally the peritoneum lyses that fibrin and regrows mesothelium in about a week, but where the tissue is ischaemic or handled roughly, fibrinolysis fails and fibroblasts move in instead — the fibrin band becomes a permanent fibrous cord between loops of bowel or bowel and wall. Later, a loop kinks over or twists around that band. Gas and fluid build up proximally, the wall distends, and stretch fires visceral afferents in waves: colic. If the loop's blood supply is caught, the wall becomes ischaemic, then inflamed through to the parietal layer, and the pain turns constant and localised.
Adhesions cause roughly two-thirds of small bowel obstruction in Australia, and previous surgery is the single most useful thing in the history. Colic that becomes constant, with fever, tachycardia and localised tenderness, means strangulation — that is the shift from visceral to parietal pain again, and it is the signal to operate.
You would find: A laparotomy scar. Colicky central pain, vomiting, distension, no flatus. Bowel sounds are high-pitched and tinkling early, then absent. Abdominal x-ray shows dilated central loops with valvulae conniventes crossing the full width.
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.
- Binds
- Penicillin-binding proteins (the transpeptidase enzymes) on the outside of the bacterial cytoplasmic membrane, where peptidoglycan is assembled. Where a beta-lactamase inhibitor is paired with the beta-lactam, the inhibitor (tazobactam) binds and suicide-inactivates bacterial beta-lactamase so the piperacillin survives long enough to reach those enzymes.
- Which does
- Cross-linking of peptidoglycan stops. The cell wall keeps being remodelled by the bacterium's own autolysins with nothing rebuilding it, so the organism swells and bursts. Killing is time-dependent, not peak-dependent — what matters is the proportion of the dosing interval spent above the MIC, which is why these drugs are given frequently or by infusion.
- So you see
- Bacterial load in the cavity and the blood falls, fever and the septic vasodilation settle over 24 to 48 hours. If the patient is not improving, the usual answer is undrained pus, not the wrong antibiotic.
- And the same mechanism causes
- The same peptidoglycan chemistry exists in the patient's own gut commensals, so the broad-spectrum drug strips out the colonic anaerobes along with the spilled organisms. That destroys colonisation resistance. Clostridioides difficile is itself a Gram-positive organism with a peptidoglycan wall, but it survives two ways the drug cannot touch: its spores are metabolically dormant, and beta-lactams kill only actively dividing cells; and its vegetative form is intrinsically resistant to many of these agents. So it germinates into an emptied colon, and its toxins A and B produce profuse watery diarrhoea days to weeks later.
- Handling
- Anaerobic cover matters because the spill is faecal — metronidazole is added when the chosen beta-lactam does not already cover anaerobes (amoxicillin plus gentamicin needs it; piperacillin-tazobactam and meropenem do not). Take blood cultures before the first dose, but never let that delay the dose in a septic patient. Choice and duration follow the current Therapeutic Guidelines and local resistance patterns, and de-escalate once cultures return.
- Binds
- The intracellular mineralocorticoid receptor in collecting duct principal cells — a nuclear hormone receptor, not a channel. Spironolactone is a competitive antagonist at it.
- Which does
- Blocks aldosterone from switching on transcription of the epithelial sodium channel (ENaC), SGK1 and the basolateral Na/K-ATPase. Fewer active sodium channels sit in the apical membrane, so less sodium is reabsorbed. Because that sodium entry is what makes the tubular lumen electrically negative, removing it also removes the electrical driving force for potassium to leave the cell through ROMK and BK channels — potassium secretion falls. It is not a one-for-one Na/K exchanger; the two are coupled through the lumen potential, which is exactly why the drug spares potassium.
- So you see
- Slow, steady natriuresis. Girth and weight fall, and the drug keeps working in a patient whose renin-angiotensin-aldosterone axis is running flat out from splanchnic vasodilation. Full effect takes several days because it depends on protein synthesis turning over, so the dose is not judged on day one.
- And the same mechanism causes
- Spironolactone is a steroid, and the mineralocorticoid receptor it blocks is a close structural relative of the androgen and progesterone receptors — so it blocks those too. Men get painful gynaecomastia and loss of libido, women get menstrual irregularity. Same molecule, wrong receptor family. Eplerenone and amiloride avoid this, since neither has the same off-target steroid receptor binding.
- Handling
- The potassium-sparing action is the other predictable consequence: serum potassium rises, which matters most when the patient also has renal impairment or is on an ACE inhibitor or ARB. Check electrolytes and creatinine before and during treatment, and pair with dietary salt restriction — diuretics cannot outrun a high-sodium diet. Diuresis is titrated against daily weight, and guidelines cap the safe rate at about 0.5 kg a day in a patient without peripheral oedema (up to about 1 kg a day if oedema is present); faster than that and you strip intravascular volume and precipitate renal impairment or encephalopathy. That is a ceiling to stay under, not a target to chase.
- Binds
- No receptor. Albumin is the main oncotic protein of plasma and also binds and buffers circulating mediators including nitric oxide, bile acids and endotoxin.
- Which does
- Adds oncotic pull to the intravascular space and mops up circulating vasodilator mediators, so the underfilled splanchnic circulation of cirrhosis is less able to drop its effective arterial volume when litres are drained off.
- So you see
- Prevents post-paracentesis circulatory dysfunction — the sudden fall in effective blood volume that triggers renin, drops renal perfusion and can tip the patient into hepatorenal syndrome. In SBP it reduces acute kidney injury and mortality; that is a trial-proven outcome benefit, not a theoretical one.
- And the same mechanism causes
- It is a volume expander, so it expands volume everywhere: in a patient with a stiff or poorly contracting left ventricle, the same oncotic pull draws fluid into the circulation faster than the heart can move it forward, and produces pulmonary oedema.
- Handling
- It is dosed by the volume of ascites removed, not matched litre for litre — the standard is roughly 6 to 8 g of albumin per litre drained, conventionally when more than about 5 litres come off. Ascitic fluid must go to the lab first: a neutrophil count of 250 x 10^6/L or more is SBP and changes the plan immediately, adding an antibiotic (a third-generation cephalosporin such as ceftriaxone is the usual empirical choice) to the albumin. Exact volumes and dosing follow local protocol and the current guidelines.
- Binds
- Mu-opioid receptors, G protein-coupled, on dorsal horn neurons, in the periaqueductal grey and rostral ventromedial medulla, and densely on neurons of the myenteric plexus.
- Which does
- Gi coupling opens potassium channels and shuts voltage-gated calcium channels, hyperpolarising the neuron. Presynaptic transmitter release falls, so the pain signal is not relayed upward, and descending inhibition is switched on.
- So you see
- The colicky visceral pain eases and the patient can tolerate the nasogastric tube, be examined properly, and lie still enough for imaging.
- And the same mechanism causes
- The identical receptor on the myenteric plexus silences the neurons that drive propulsive peristalsis. Non-propulsive segmental tone rises, coordinated propulsion stops, and water absorption goes on unopposed — so transit slows and the stool hardens. In a bowel that is already obstructed this makes distension and vomiting worse, which is exactly why an opioid never substitutes for decompression.
- Handling
- Pair it with an antiemetic, and monitor sedation score as well as respiratory rate, since sedation precedes respiratory depression and the same brainstem receptors blunt the ventilatory response to carbon dioxide. Do not give laxatives while the patient is obstructed — in mechanical obstruction they are contraindicated, and the patient is nil by mouth with a nasogastric tube on free drainage. The routine opioid-plus-aperient rule belongs to the patient with an open gut; here the bowel is reassessed once obstruction has resolved.
The diagnostic event in an acute abdomen is not how bad the pain is — it is the moment it changes character. Vague, central, crampy pain the patient cannot point to is visceral peritoneum. Sharp pain they can put one finger on, worse with coughing or a speed bump, is parietal peritoneum. When the pain migrates from the umbilicus to the right iliac fossa, the inflammation has crossed from one to the other.
Now test whether it stuck
Reading this through is not the same as being able to reconstruct it. Every question in the bank is free, with a full debrief on each option.