The corridor to the pituitary has the optic chiasm as its roof and the carotid arteries as its walls — between 0 and 9 mm away, with little or no bone in between. That explains why immobility is not a comfort measure here. The other half of the day is that the patient's most dangerous week starts after they go home.
The distance from the pituitary gland to the internal carotid artery is between 0 and 9 mm, and once the surgeon is inside the sphenoid sinus there is very little or no bone between an instrument and that artery. Movement is therefore not a nuisance — it is a mechanism of arterial perforation, visual loss and cranial nerve injury. And the sodium problem that will actually readmit this patient does not appear on your watch: diabetes insipidus arrives in the first 24 to 48 hours, but hyponatraemia bottoms out at 7 to 9 days.
Neuro 2 day 03 was a craniotomy where the danger was pressure. This one is a much smaller operation where the danger is distance. Almost nothing about the anaesthetic is unusual. What is unusual is that a few millimetres of error in any direction hits something that cannot be repaired.
The pituitary sits in the sella turcica, a saddle in the superior aspect of the sphenoid bone. Its position on the roof of the sphenoid sinus is the whole reason a transsphenoidal approach exists — the gland can be reached without opening the skull.
Look at what borders that corridor.
Above is the diaphragma sella and, in the subarachnoid space just beyond it, the optic nerves and chiasm. Laterally are the cavernous sinuses, which contain the internal carotid arteries and cranial nerves III, IV, VI, V1 and V2. Below is the sphenoid sinus you came through. And the floor of the sella is bone, which is why a growing tumour takes the path of least resistance and goes up.
That last fact explains the presentation. As the tumour extends above the diaphragma sella it first fills the subarachnoid space, then lifts and compresses the optic structures. Bitemporal field cuts are the commonest defect, because the majority of the chiasm sits directly above the sella. Lateral extension into the cavernous sinuses, by contrast, is typically well tolerated and seldom symptomatic — the tumour may abut the carotid, but actual invasion is rare.
And the number to keep: the distance separating the carotid arteries from the pituitary gland is between 0 and 9 mm, and once inside the sphenoid sinus there is very little or no bone in between.
Pituitary tumours are 15 to 25% of intracranial tumours, and most are non-functional. When they present, they present anatomically — headache from a stretched, pain-innervated dura, visual field loss, and eventually hypopituitarism.
Hypopituitarism arrives late, and the reason is a useful piece of reserve physiology: about 10% of the gland is sufficient to maintain endocrine homeostasis, so compression is tolerated until more than 90% is non-functional. In women this shows first as menstrual abnormality, in men as loss of libido.
The functional tumours each bring their own anaesthetic problem, and it is worth knowing which problem belongs to which:
The instinct is to give hydrocortisone to everyone having pituitary surgery. The source says the opposite: steroid administration is far from routine. Most patients have enough residual normal tissue to keep ACTH and cortisol adequate and require nothing.
Cushing’s disease is the exception, and it inverts. Here the pathological ACTH has suppressed the normal corticotrophs for months or years. Remove the tumour and the suppressed gland cannot immediately take over, so these patients are at risk of severe adrenal insufficiency and cardiovascular collapse. They may be given perioperative steroid cover — which must then be tapered so glandular function can be assessed — or steroids may be deliberately withheld so that postoperative cortisol testing can confirm the pathological secretion is gone.
If steroids are withheld, the timing matters and it is not intuitive. Cortisol has a half-life of about 50 to 75 minutes, and the patient starts from a hypercortisolaemic state, so serum cortisol will not reach potentially low levels until about 20 hours after the procedure. The danger is on the ward overnight, not in recovery.
Either way this is a decision made with the surgical team.
Verify and document visual field deficits and extraocular movement abnormalities — they are the baseline against which the first postoperative examination is read. Review the imaging for size, chiasmal impingement, cavernous sinus extension and hydrocephalus. Confirm the patient is euthyroid.
Severe hyponatraemia can be present before surgery, usually with sellar arachnoid cysts, Rathke’s cleft cysts, large adenomas compressing the stalk, and pituitary apoplexy. Life-threatening hyponatraemia may represent adrenal insufficiency needing immediate steroid, or occasionally SIADH needing fluid restriction — and those want opposite management, so the cause has to be found.
One piece of consent that is easy to forget: tell the patient that occlusive packing in both nostrils will make nasal breathing difficult when they wake. It is a small thing that produces genuine panic if unexpected.
Position and tube. A RAE-type oral tube in the corner of the mouth opposite the surgeon’s dominant hand — right-handed surgeon on the patient’s right, so the tube goes to the left. Bilateral bite blocks between the molars, particularly if motor evoked potentials are used. The authors cover the tube, temperature probe, orogastric tube and bite blocks with an adhesive-edged drape so the airway can be reached without entering the sterile field.
The pharyngeal pack. Moistened gauze keeps blood out of the stomach and out of the glottis above the cuff. Leave a generous tail, inform the room, and confirm its removal before transfer. Its harms are documented — sore throat, pharyngeal plexus injury, stomatitis, massive tongue swelling, migration — and a retained pack is the reason a formal count exists.
The infiltration. Every approach uses local anaesthetic with adrenaline, and substantial rises in blood pressure can follow. Continuous communication with the surgeon, pauses if needed, and a short-acting antihypertensive ready. This is one of the reasons an arterial line is preferred in an operation that loses almost no blood.
Immobility. With no neuromonitoring, complete neuromuscular blockade is advisable — movement risks arterial perforation and haemorrhage, visual damage and cranial nerve injury. With EMG or motor evoked potentials, relaxation must be avoided and immobility comes from titrated anaesthesia and opioid, commonly TIVA with propofol and an opioid infusion.
The reflex. Asystole and bradycardia have been reported during dissection, especially during cavernous sinus exploration, attributed to the trigeminal cardiac reflex. Reported incidence spans 0.003% to 10%, which tells you the denominators are not comparable. It is usually self-limited once the surgeon stops, so saying what you are seeing is the intervention.
Coughing and emesis raise venous and arterial pressure and threaten fresh haemostasis, and matter even more if the arachnoid was violated and repaired, because of the CSF leak risk. So the emergence is planned:
Transsphenoidal closure can be quicker than a craniotomy, so infusions have to be stopped on the surgeon’s timetable, not a generic one.
And then the sleep apnoea problem. Some of these patients — acromegalic ones especially — would ordinarily go to CPAP. Pneumocephalus has been reported after non-invasive positive pressure ventilation following transsphenoidal surgery, presumably through a communication opened between the nose and the intracranial space, carrying a meningitis risk. CPAP should be avoided if possible. A nasal trumpet placed by the surgeon in place of standard packing is the alternative, and extubation should wait until you are confident the patient will ventilate without any positive pressure assistance.
Perioperative mortality is about 0.5%. Major surgical complications occur in 8 to 10%:
| Complication | Rate |
|---|---|
| CSF leak | 4.7% |
| Meningitis | 2% |
| Vision loss | 1.8% |
| Stroke | 1% |
| Vascular injury, including carotid-cavernous fistula | 0.4% |
| Carotid artery injury (adenoma resection) | 0.3% |
Carotid injury is rare and catastrophic, which is the pairing that makes immobility non-negotiable.
Vision loss or a new cranial neuropathy in the immediate postoperative period warrants immediate imaging, looking for haemorrhage or mass effect — including from a fat graft used to pack the sinus — that might justify early re-exploration. This is a complication with a surgical answer, and only if someone looks.
A CSF leak is suspected with persistent nasal discharge that tastes salty or worsens on leaning forward, and is confirmed by β2 transferrin. Treatment varies: some units re-explore early, some try lumbar drainage and bed rest first.
Three ADH-related disorders can follow, plus cerebral salt wasting. The useful organising fact is that they occupy different weeks.
Diabetes insipidus is the commonest, onset usually in the first 24 to 48 hours, occasionally intraoperatively in a long case. It is more common after surgery for macroadenomas, craniopharyngiomas and Rathke’s cleft cysts. The hallmarks are a rising serum sodium (≥145 mmol/L) and osmolarity (>300 mOsm/kg) with dilute urine (<300 mOsm/kg, or specific gravity ≤1.005). Most cases are transient and resolved by discharge. Some follow a triphasic course — early DI, antidiuresis at about a week, then permanent DI.
SIADH occurs later, with the lowest sodium at 7 to 9 days. Sodium ≤135 mmol/L, euvolaemic to slightly hypervolaemic, hypo-osmolar serum with relatively hyperosmolar urine and normal urea and creatinine. Treatment is fluid restriction to 1000 mL per day, salty food, and in severe cases intravenous sodium — at no more than 1 mmol/L/h, or 0.5 mmol/L/h if the hyponatraemia is chronic, to avoid central pontine myelinolysis. The faster of those two rates is the exception rather than the rule: Neuro 1 day 15 carries the fuller version — 0.5 mmol/L/h and 8 mmol/L in 24 hours, both ceilings applying at once — and 1 mmol/L/h belongs to the acutely symptomatic, seizing patient, not to the ward patient whose sodium has been drifting down since the operation. Read the two days as one rule. Most recover by postoperative day 28. The differential must include adrenal insufficiency, hypothyroidism, hyperglycaemia and cerebral salt wasting.
Cerebral salt wasting is worth considering when there has been significant haemorrhage or subarachnoid blood — pituitary apoplexy, for instance. It is hypovolaemic, with raised urea, creatinine and haematocrit marking a volume-contracted state. It is less common here than DI or SIADH.
The single table worth memorising is the one that separates them, and it works because the three columns disagree on volume:
| DI | SIADH | CSW | |
|---|---|---|---|
| Volume status | hypovolaemia | normo- or hypervolaemia | hypovolaemia |
| Urine output | increased | low to normal | increased |
| Serum sodium | increased | decreased | decreased |
| Urine sodium | decreased | increased | increased |
| Serum osmolality | increased | decreased | decreased |
| Urine osmolality | decreased | increased | increased |
DI and CSW share a volume state and disagree on sodium. SIADH and CSW share a sodium and disagree on volume. Nothing here is diagnosable from urine output alone — which is today’s pitfall.
Pituitary apoplexy is haemorrhage or infarction in the pituitary or a pituitary tumour: sudden severe headache, altered sensorium, visual compromise, ocular paresis, nuchal rigidity, vomiting. It looks like subarachnoid haemorrhage and must be radiologically distinguished from it.
Two things follow. These patients may have acute adrenal insufficiency from destruction of the normal gland, requiring intravenous steroid — 100 mg hydrocortisone every 8 hours — to prevent cardiovascular collapse. And most are managed conservatively; it is chiasmal compression with marked visual impairment that turns it into an emergency decompression.
Which returns the day to where it started. The corridor is defined by what sits around it, and every decision in this operation — the immobility, the smooth emergence, the urgency of a new visual deficit — is about the structures you cannot see.
No randomized trial addresses this.
The anatomy is not in dispute and it does most of the teaching. The pituitary sits in the sella turcica on the roof of the sphenoid sinus, which is what makes the transsphenoidal route possible at all. The chiasm is superior, the cavernous sinuses — carrying the internal carotid arteries and cranial nerves III, IV, VI, V1 and V2 — are lateral, and the floor of the sella is bone. Everything about the presentation and the hazards follows from that arrangement.
The endocrine physiology is equally settled. About 10% of the gland is normally sufficient to maintain endocrine homeostasis, so compression is tolerated until more than 90% is non-functional. Chronic hypercortisolism suppresses the normal corticotrophs, which is why Cushing's disease is the one situation where removing the tumour creates an endocrine emergency.
The complication rates are cohort data, not trial data. Perioperative mortality of about 0.5%, major complications in 8 to 10%, and a carotid injury risk of about 0.3% come from consecutive surgical series. They are good numbers for consenting and for calibrating vigilance; they are not comparisons between techniques.
The technique is consensus and openly variable. Sublabial versus transnasal is decided by operator experience and preference, and both have similar rates of gross total resection, adjuvant radiotherapy and mortality. Whether to give perioperative steroids is a decision the source explicitly says must be made with the surgical team rather than by protocol.
One number the source itself distrusts. Water and electrolyte disturbances after pituitary surgery are reported at up to 25% in the authors' framing, while some surveys report up to 75% with readmission rates up to 56% — and the authors say plainly that this "vastly exceeds our experience." That disagreement is left visible here rather than averaged, because a resident should know the incidence is contested and the vigilance is not.
The patient had a growth hormone-secreting adenoma resected this morning. It is now the evening, they are passing large volumes of urine, and everyone in the room knows that diabetes insipidus is the commonest problem after pituitary surgery and that it starts in the first 24 to 48 hours. The reflex is to match the output with fluid, and to reach for desmopressin.
The source is explicit that this reflex can be the injury. Physiologic diuresis — often from the fluid you gave during the case, and characteristic of acromegalic patients mobilising interstitial fluid after a successful resection — "should not be treated by matching fluid replacement as it can lead to critical volume overload." Two conditions produce the same chart entry and want opposite management.
The distinction is not clinical impression, it is numbers, and they are available:
- Diabetes insipidus is a rising serum sodium — ≥145 mmol/L — with serum osmolarity >300 mOsm/kg and dilute urine, <300 mOsm/kg or a specific gravity ≤1.005. The patient is or is becoming hypovolaemic and hypernatraemic. - Physiologic diuresis is a patient offloading fluid they did not need, with a sodium that is not climbing.
So the rule is: do not treat a urine output, treat a sodium and an osmolality. If the sodium is not rising, the polyuria is not telling you to give more. DDAVP — 1 to 2 µg subcutaneously or intravenously, or 0.1 mg orally — is considered when the patient cannot keep up and the sodium exceeds 145 mmol/L, and it carries its own risk of overcorrection into hypervolaemia and hyponatraemia.
Two further things make this trap worse than it looks.
The thirst mechanism usually solves it for you. With intact thirst and fluid available by mouth, the majority of these patients self-regulate. That is an argument for watching a conscious patient rather than intervening in one.
And the phase can reverse. Some patients follow a triphasic course: diabetes insipidus on day one, an antidiuretic phase about a week later, then permanent diabetes insipidus. A patient given DDAVP for phase one and then entering phase two on a ward that has stopped counting is the mechanism by which a sodium of 148 becomes a sodium of 122.
Which is the wider lesson of the day. Neuro 1 day 15 taught that sodium moves slowly and should be corrected slowly. This day adds that after pituitary surgery it also moves in both directions, on a schedule, and the dangerous week is the one after discharge.
Full text · Preston goes through LibKey to whatever Preston holds — the PDF where there is one — and works on a phone. The doi link always reaches the publisher's record, whether or not we hold a subscription. If you are reading on a computer, LibKey Nomad adds the same one-click PDF button to PubMed and publisher pages everywhere else; it is a desktop browser extension, so it does nothing on a phone or tablet.
During an endoscopic transnasal resection of a macroadenoma the surgeon requests transcranial motor evoked potentials because the tumour abuts the carotid artery. How should immobility be maintained?
Eight hours after transsphenoidal resection of a growth hormone-secreting adenoma, an acromegalic patient is passing 350 mL of urine per hour. Serum sodium is 139 mmol/L and serum osmolarity is 288 mOsm/kg. What should be done?
A patient undergoes transsphenoidal resection of an ACTH-secreting adenoma. The team elects to withhold perioperative steroids so that postoperative cortisol testing can be interpreted. When should hypoadrenalism be anticipated?
That is the whole orientation rule. Fat is bright on T1 and intermediate on T2, cortical bone gives no signal, and flowing blood produces a signal void. The practical summary is that anatomy is better seen on T1 and pathology on T2 — and post-contrast sequences are T1 weighted for exactly the reason given.
Ferromagnetic aneurysm clips are a contraindication; non-ferromagnetic ones are not. The risk from a ferromagnetic object is magnetically induced movement, with dislodgement and injury, and heating — image artefact is the secondary concern. The information you need is in the old operation note, not on any current imaging.