Control of Breathing
Breathing is set by an acid sensor in the brainstem with a hypoxia alarm at the carotids — which is why respiratory depression is the opioid effect that kills, and why uncontrolled oxygen can harm a CO2 retainer.
What it normally does
Your brainstem measures acid, not air. CO2 crosses readily from blood into brain extracellular fluid and CSF (H+ and bicarbonate cross only poorly), is hydrated to carbonic acid, and chemosensitive neurons on the ventral medullary surface (retrotrapezoid nucleus) sense the resulting H+. These central chemoreceptors account for roughly 70-80% of the ventilatory response to CO2 minute to minute.
Anything that blunts these cells or their output causes underbreathing; anything that adds acid drives breathing up — this is opioid depression, obesity hypoventilation, and Kussmaul breathing.
At the bifurcation of each common carotid sits a small nest of glomus cells (peripheral chemoreceptors, carotid bodies). They discharge slowly even at normal PaO2, and their firing rises steeply once arterial PO2 falls below roughly 60 mmHg. They sense dissolved PO2, not oxygen content, and they also respond within a breath or two to arterial H+ and CO2, contributing the remaining minority of the total CO2 response.
Explains why hypoxic drive is a backup rather than the main dial, why saturation is a late warning sign, why oxygen can quieten the alarm, and why anaemia or carbon monoxide poisoning — low content but normal dissolved PO2 — do not trigger this hypoxia alarm, even though such patients may still breathe hard for another reason, such as a lactic acidosis from tissue hypoxia.
The respiratory rhythm is generated by a cluster of pacemaker-capable neurons in the ventrolateral medulla (pre-Botzinger complex), which are densely populated with mu-opioid receptors.
Explains why respiratory depression is the opioid effect that kills, why it tracks the same receptor as analgesia, and why an antagonist at that receptor reverses it.
Hold a high CO2 for days and the brain restores its own pH — bicarbonate is transported into the CSF — while the kidney retains bicarbonate in the blood. The acid signal is largely corrected even though the CO2 is still high, so the central chemoreceptor contribution is blunted.
Explains chronic CO2 retention, the compensated gas with a near-normal pH and a high bicarbonate, and why these patients depend relatively more on peripheral chemoreceptor drive — though when oxygen worsens their hypercapnia, V/Q mismatch and the Haldane effect matter more than lost hypoxic drive.
What goes wrong
- Opioid-induced respiratory depression← from “The respiratory rhythm is generated by a clust…”
An opioid binds mu receptors on pre-Botzinger complex neurons and the wider brainstem respiratory network. Gi signalling opens potassium (GIRK) channels, the cells hyperpolarise, and the rhythm slows. At the same time the ventilatory response curve to CO2 shifts right and flattens — the brainstem now tolerates a CO2 that would normally have the patient panting. Respiratory rate falls, tidal volume falls, then irregular pauses, then apnoea. Upper airway tone is lost as well, so obstruction often coexists.
Sedation precedes desaturation, which is why sedation scoring beats pulse oximetry for monitoring. Opioids are involved in close to half of all drug-induced deaths in Australia, and pharmaceutical opioids feature in more of those than heroin — the typical case is an inpatient on modified-release oxycodone plus a benzodiazepine, not someone in a laneway. Take-home naloxone is available free and without prescription through the national Take Home Naloxone Program.
You would find: A falling respiratory rate (a rate under about 12 and still falling is a warning) with a rising sedation score. Sedation comes before hypoxia — a patient you cannot rouse with a normal saturation is already in trouble. Pinpoint pupils. If they are on supplemental oxygen, the saturation probe is among the last things to change, so count the rate and score the sedation rather than watching the monitor.
- Chronic CO2 retention and oxygen-induced hypercapnia (type 2 respiratory failure in COPD)← from “Hold a high CO2 for days and the brain restore…”
Years of a high CO2 allow CSF pH to be normalised by bicarbonate transport at the choroid plexus and the kidney to retain bicarbonate. The central acid signal is blunted, and ventilation depends relatively more on the carotid bodies. Give uncontrolled high-flow oxygen and three things happen, in order of importance: hypoxic pulmonary vasoconstriction is released so perfusion returns to poorly ventilated alveoli and dead space rises; oxygenated haemoglobin offloads CO2 (Haldane effect); and, to a lesser degree, carotid body drive falls. CO2 climbs, pH falls, and the patient becomes obtunded.
Target 88-92% saturation in patients at risk of hypercapnic respiratory failure, 92-96% otherwise. Correct hypoxaemia to target — do not withhold oxygen from a hypoxaemic patient. COPD is a leading cause of potentially preventable hospitalisation in Australia, and Aboriginal and Torres Strait Islander people die from it at roughly two to three times the rate of other Australians, with hospitalisation rates higher again.
You would find: COPD patient brought in on high-flow oxygen via a reservoir mask, now drowsy with asterixis. Gas: PaCO2 high, pH low, bicarbonate already high from long-standing compensation. Saturations a comfortable 99% — on oxygen, in this patient, that is the abnormal finding.
- Obesity hypoventilation syndrome← from “Your brainstem measures acid, not air. CO2 cro…”
A heavy chest wall and abdomen make every breath mechanically more expensive and reduce functional residual capacity, and the ventilatory response to a rising CO2 is blunted (leptin resistance is the leading explanation). Patients underbreathe, worst during sleep, retain CO2, and the kidney compensates with bicarbonate. That retained bicarbonate persists into the day, so the acid signal stays blunted and the CO2 never fully corrects — awake hypercapnia. Diagnosis is BMI 30 or more with awake PaCO2 above 45 mmHg, with sleep-disordered breathing, and no other explanation for the hypoventilation.
Obstructive sleep apnoea on its own does not usually produce awake hypercapnia; obesity hypoventilation does — and around 90% of obesity hypoventilation patients also have OSA. A raised bicarbonate in a sleepy obese patient means look for it. Treatment is positive airway pressure (CPAP or NIV) plus weight management, not a drug.
You would find: BMI over 30, sleepy all day, morning headache, and a serum bicarbonate above about 27 mmol/L on a routine chemistry panel. That bicarbonate is the cheapest screening test you have — it is sitting on bloods you already ordered. Confirm with an arterial or venous blood gas, and arrange a sleep study.
- Kussmaul breathing in diabetic ketoacidosis← from “At the bifurcation of each common carotid sits…”
Ketoacids flood the blood with H+. H+ crosses the blood-brain barrier poorly, so the immediate stimulus is at the carotid bodies, which respond to arterial acid within a breath or two. The central chemoreceptors join late and reluctantly: the fall in PaCO2 that hyperventilation produces initially raises CSF pH and damps them, and only over hours, as CSF bicarbonate falls, do they add to the drive. Minute ventilation rises to blow off CO2 and defend pH. The sensor is doing exactly what it was built to do.
That breathing is the compensation, not the disease. Sedate or intubate without matching their very high minute ventilation and the CO2 rises, the pH crashes and they can arrest. Treat the ketoacidosis with fluid and insulin; leave the breathing alone.
You would find: Deep, sighing laboured breaths in a dehydrated unwell young person — the striking feature is the depth, with the rate normal or raised. Ketones on the breath, high glucose, and a gas showing a low pH with a low CO2 and a raised anion gap.
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
- Mu-opioid receptor (competitive antagonist; also antagonises kappa and delta, with highest affinity at mu)
- Which does
- Competes the agonist off the receptor. Gi signalling stops, the agonist-activated potassium (GIRK) conductance closes, the hyperpolarised respiratory neurons depolarise back toward firing threshold, and rhythm and CO2 responsiveness return.
- So you see
- Respiratory rate and depth improve within about one to two minutes after an IV dose (slower by IM or intranasal, so allow time before repeating). The patient rouses and pupils return to normal size.
- And the same mechanism causes
- You have blocked mu receptors everywhere at once, so in a tolerant patient you precipitate acute withdrawal: pain returns full force, vomiting (with aspiration risk in a drowsy patient), agitation, and a catecholamine surge that can produce tachycardia, hypertension and — rarely — flash pulmonary oedema. The size of the reaction tracks the size of the dose pushed, which is the argument for small titrated increments.
- Handling
- Naloxone's effect lasts roughly 30-90 minutes. Most opioids outlast it, and slow-release oxycodone, modified-release morphine, methadone and buprenorphine outlast it by a long way, so re-sedation after apparent reversal is expected rather than surprising. These patients need continued observation and repeat doses or an infusion — never discharge from the resus bay just because they woke once. Follow local protocols for dose and observation period.
- Binds
- Mu-opioid receptor (Gi-coupled)
- Which does
- Hyperpolarises brainstem respiratory neurons and shifts the ventilatory response curve to CO2 rightwards and flattens its slope — the brainstem tolerates a higher CO2 before it responds. The hypoxic ventilatory response is blunted too. Higher up, opioids reduce how unpleasant air hunger feels without changing the mechanics of the lung.
- So you see
- At low doses in chronic refractory breathlessness, the sensation of air hunger eases with little change in blood gases and only a modest fall in respiratory rate. At higher doses the same mechanism produces hypoventilation and apnoea.
- And the same mechanism causes
- The therapeutic effect and the lethal effect are the same effect. Blunting the CO2 response is what relieves air hunger and what kills in overdose. Add a benzodiazepine or other sedative, untreated sleep apnoea, renal impairment with metabolite accumulation, or simply sleep, and the margin between the two narrows.
- Handling
- This is for chronic refractory breathlessness that persists despite optimal treatment of the underlying disease — not for acute breathlessness, which has a cause that needs finding and treating. It is not a treatment for hypercapnic respiratory failure; in that setting it makes things worse.
- Binds
- Oxygen-sensitive background (TASK-like) potassium channels on carotid body glomus cells; the hypoxic pulmonary vasoconstriction reflex in small pulmonary arteries and arterioles; haemoglobin itself
- Which does
- A high arterial PO2 keeps those potassium channels open, so glomus cells stay hyperpolarised and their firing falls back to baseline. In the lung, a high alveolar PO2 releases hypoxic pulmonary vasoconstriction, so perfusion returns to poorly ventilated alveoli and V/Q mismatch and physiological dead space worsen. And oxygenated haemoglobin carries less CO2 than deoxygenated haemoglobin (Haldane effect), so CO2 is offloaded into the plasma.
- So you see
- Hypoxaemia corrected — and in a patient at risk of hypercapnia, a PaCO2 that climbs. Most of that rise is worsened V/Q matching plus the Haldane effect; reduced hypoxic drive is the smaller contributor, despite what the bedside teaching says.
- And the same mechanism causes
- CO2 narcosis: drowsy, confused, asterixis, falling pH — caused by the oxygen given. It is a treatment-induced respiratory acidosis, and the response is to titrate the oxygen back down to the target range while checking gases, not to keep pushing it up. Do not simply switch the oxygen off: rebound hypoxaemia is dangerous, and a patient with a persistent acidosis needs NIV, not just less oxygen.
- Handling
- Saturations drifting above target on oxygen in a COPD patient are a warning, not reassurance. Reduce to target and get an arterial blood gas. Do not withhold oxygen from a hypoxaemic patient for fear of hypercapnia — treat the hypoxaemia to target and monitor the gas.
- Binds
- Carbonic anhydrase, mainly at the proximal tubule brush border and in the proximal tubular cell
- Which does
- Bicarbonate can no longer be efficiently reclaimed, so it is lost in the urine. Plasma bicarbonate falls and a mild hyperchloraemic (normal anion gap) metabolic acidosis develops; the peripheral chemoreceptors read the rise in arterial H+ directly and the central chemoreceptors respond as the acidosis is reflected in CSF pH.
- So you see
- Minute ventilation rises, and CO2 that the alkalosis was permitting can then be blown off. The effect is self-limiting as bicarbonate stores are depleted.
- And the same mechanism causes
- The acidosis is the mechanism, so it is also the toxicity. In a patient whose respiratory pump cannot do more work, you have added a metabolic acidosis to a respiratory one and the pH falls further. Bicarbonate delivered distally acts as a non-reabsorbable anion and drives potassium loss, so hypokalaemia. Alkaline urine with reduced citrate favours calcium phosphate stones. Paraesthesia of the fingers, toes and lips is common and is usually attributed to carbonic anhydrase inhibition and the mild acidosis. The same enzyme in the ciliary body means aqueous production and intraocular pressure fall — the reason it is also used in glaucoma.
- Handling
- It is a sulfonamide derivative, so ask about severe sulfonamide reactions. Avoid or use with caution in significant renal impairment, hepatic impairment (risk of precipitating encephalopathy), hypokalaemia, hyponatraemia, and existing metabolic acidosis. A stone history matters. Check electrolytes and the gas rather than assuming the effect.
Drowsy COPD patient, arterial gas shows a high CO2. Look at the pH and the bicarbonate before you panic: a high CO2 with a near-normal pH and a high bicarbonate is chronic, and renal compensation of that size takes at least several days to build. A high CO2 with a low pH is acute or acute-on-chronic — in an exacerbation of COPD, a pH below 7.35 with hypercapnia persisting despite optimal medical therapy and controlled oxygen is the trigger for non-invasive ventilation.
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.